10 Metals v2.1

Swiss Centre
for Life Cycle

Inventories
A joint initiative
of the ETH domain and
Swiss Federal Offices
Life Cycle Inventories of Metals

Data v2.1 (2009)
Mischa Classen, Hans-Jörg Althaus, Silvio Blaser,

Wolfram Scharnhorst
EMPA, Dübendorf
Matthias Tuchschmid, Niels Jungbluth,
Mireille Faist Emmenegger
ESU-services, Uster
ecoinvent v2.1 report No. 10
Dübendorf, March 2009

Project "ecoinvent data v2.1"
Commissioners: Swiss Centre for Life Cycle Inventories, Dübendorf
Swiss Federal Office for the Environment (BAFU -
FOEN), Bern
Swiss Federal Office for Energy (BFE), Bern
Swiss Federal Office for Agriculture (BLW), Bern
ecoinvent Board: Gérard Gaillard (Chair), Agroscope Reckenholz-

Tänikon Research Station, ART, Zürich
Rainer Zah, Empa, St. Gallen
Steffanie Hellweg, ETHZ, Zürich
Stefan Hirschberg, PSI, Villigen
Institutes of the ecoinvent Centre:

Swiss Federal Institute of Technology Zürich
(ETHZ)
Paul Scherrer Institute (PSI)
Swiss Federal Laboratories for Materials Testing
and Research (Empa)
Agroscope Reckenholz-Tänikon Research Station
(ART)
Participating consultants: Doka Life Cycle Assessments, Zürich

Dr. Frank Werner Environment & Development,
Zürich
Ecointesys - Life Cycle Systems Sarl.
ESU-services Ltd., Uster
Software Support: ifu Hamburg GmbH
Project leader: Roland Hischier, ecoinvent Centre, Empa,

Dübendorf
ecoinvent v2.1 report No. 10 - ii -

Summary
Life Cycle Inventories of Metals
Editor: Hans-Jörg Althaus
Authors: Mischa Classen, Hans-Jörg Althaus, Silvio Blaser,

Wolfram Scharnhorst (Empa), Matthias
Tuchschmid, Niels Jungbluth, Mireille Faist
Emmenegger (ESU-services)
Reviewer: Roland Hischier (Empa), Roberto Dones (PSI),

Hans-Jörg Althaus, Mischa Classen (Empa), Rolf
Frischknecht(ESU-services).
Contact address: ecoinvent Centre

Empa
P.O. Box
CH-8600 Dübendorf
http://www.ecoinvent.org/
management@ecoinvent.org
Responsibility: This report has been prepared on behalf of one or

several Federal Offices listed on the opposite
page (see commissioners) and / or the ecoinvent
Centre. The final responsibility for contents and
conclusions remains with the authors of this
report.
Terms of Use: Data published in this report are subject to the

ecoinvent terms of use, in particular paragraphs 4
and 8. The ecoinvent terms of use (Version 2.0)
can be downloaded via the Internet
(www.ecoinvent.org).
Liability: Information contained herein have been compiled

or arrived from sources believed to be reliable.
Nevertheless, the authors or their organizations
do not accept liability for any loss or damage
arising from the use thereof. Using the given
information is strictly your own responsibility.
Citation:
Classen M., Althaus H.-J., Blaser S., Tuchschmid M., Jungbluth N., Doka G., Faist
Emmenegger M. and Scharnhorst W. (2009) Life Cycle Inventories of Metals. Final re-
port ecoinvent data v2.1, No 10. EMPA Dübendorf, Swiss Centre for Life Cycle Inven-
tories, Dübendorf, CH, Online-Version under: www.ecoinvent.ch
© Swiss Centre for Life Cycle Inventories / 2009

Foreword to Version 2.1
This report is a revised and extended version of the report from Althaus et al. (2004) on the datasets of
the version 1.1. It comprises
 corrections of reported errors,
 extension of inventories by new metals and the related updating of established inventories and
 some restructuring in the report structure.
The changes to version 1.2 and 1.3 as documented in Frischknecht et al. (2006) are integrated entirely
where it concerned the datasets contained in this report.
The extension of the metals database consists of primary Cadmium, primary Indium, Gold and Silver
(both primary and secondary as well as supply mix), Lead (secondary and supply mix), Tantalum (ca-
pacitor powder) and the metals Gallium, Arsenic and Tellurium as semiconductors.
Since most of these metals are by-products from the Zinc / Lead cycle, a new Part has been created
that hosts the respective content from the original Part VIII Other Metals in Althaus et al. (2004)
which in turn has been suspended. Its remaining contents have been reallocated: The copper alloys
brass, bronze and brazing solders to Part II Copper; Magnesium and Tin to new individual parts.
The original inventories as they were established in version 1.1 through 1.3 are not altered. No general
changes were made to actualise dated information such as supply and use patterns. Further the discus-
sion of selected LCI / LCIA results has not been updated with the v2.0 calculation results. The respec-
tive chapters have been deleted.
The Overview on the following page provides a schematic representation of the changes of this report
as compared to the report on data v1.1 in Althaus et al. (2004).
Compared to version 2.0, the version 2.1 contains a new part on antimony and several corrections and
updates.
References
Althaus et al. (2004) Althaus H.-J., Blaser S., Classen M. and Jungbluth N. (2004) Life Cycle Inven-
tories of Metals. Final report ecoinvent 2000 No. 10. EMPA Dübendorf, Swiss
Centre for Life Cycle Inventories, Dübendorf, CH, Online-Version under:
www.ecoinvent.ch.
Frischknecht et al. (2006) Frischknecht R., Althaus H.-J., Bauer C., Capello C., Doka G., Dones R.,
Faist Emmenegger M., Hischier R., Jungbluth N., Kellenberger D., Margni M.,
Nemecek T. and Spielmann M. (2006) Documentation of changes implemented
in ecoinvent Data v1.2 and v1.3. ecoinvent report No. 16. EMPA Dübendorf,
Swiss Centre for Life Cycle Inventories, Dübendorf, CH.
ecoinvent v2.1 report No. 10 - iv -

Overview
Following diagram indicates the relation between the contents of the preceding report on v1.1 data and
this report on v2.1 data. The report on version v2.0 was structured as v2.1 without part XVI.
Data v1.1 (2004) Data v2.1 (2009)
Part I Part I
Aluminium Aluminium
Part II Part II
Iron and Steel Iron and Steel
Part III Part III (updated)
Copper Copper
Part IV Part IV
Nickel Nickel
Part V Part V
Platinum Group Metals Platinum Group Metals
Part VI Part VI
Chromium Chromium
Part VII Part VII
Manganese Manganese
Part VIII content
Other metals transferred
Part VIII (updated)
Lead, Zinc, Cadmium and Indium
Part IX (new)
Gold and Silver
Part X (new)
Semiconductors: Tellurium, Gallium, Cadmium
Compounds
Part XI (new)
Tantalum powder
Part XII
Tin
Part XIII
Magnesium
Part IX Part XIV
Auxiliary Processes Auxiliary Processes
Part X Part XV
Metal processing Metal processing
Part XVI
Antimony
ecoinvent report v2.1 No. 10 -v-

Part I
Aluminium
Data v2.1 (2009)
Author: Hans-Jörg Althaus, EMPA, Dübendorf
Review Roland Hischier, EMPA, St. Gallen
Emmenegger M. and Scharnhorst W. (2009) Life Cycle Inventories of Metals. Final
report ecoinvent data v2.1, No 10. EMPA Dübendorf, Swiss Centre for Life Cycle In-
ventories, Dübendorf, CH, Online-Version under: www.ecoinvent.ch.
Final report of updated Swiss National Life Cycle Inventory Database "ecoinvent", version 2.1
Acknowledgement and Summary
Acknowledgement
We thank Bernard de Gélas and Christian Leroy of the European Aluminium Association (EAA) for providing
the unit process data of their study and for their help with the interpretation of the data. We also thank Eirik
Nordheim of the EAA and Gerald Reblizer of Alcan Technology & Management for their valuable comments to
the report and the data.
Comments on this version 2.1

This part corresponds to the original report in Althaus et al. (2004) that has been updated with the changes re-
ported in Frischknecht et al. (2006), where the factor for resource consumption in the bauxite mining was cor-
rected. Beside this change the CO emission in the dataset “aluminium, primary, liquid, at plant” (RER) which
was inventoried as “biogenic” in earlier versions was corrected to “fossil”.
Chapter “6 Cumulative Results and Interpretation” of the original report has not been adapted and was deleted
for this report.
Note that the flow names in the tables displaying unit processes have not been updated to reflect the name
changes with version 2.0. This namely affects the designation of transportation service flows.
Summary
The aluminium production is modelled for the European situation, primarily based on data from the European
Aluminium Association (EAA (2000) and personal communications. The mining represents a world-wide situa-
tion primarily based on Mori & Adelhardt (1998).
Some of the datasets described in this report make only sense in the context of the aluminium production and are
therefore not intended for other uses. Intermediate products like aluminium hydroxide and aluminium oxide
however can be used to represent these chemicals in other applications.
The datasets to be chosen to represent aluminium in a LCA study are either the primary aluminium (chapter 5.3),
one of the secondary aluminium datasets (chapter 5.4) or one of the production mixes (chapter 5.5). The deci-
sion whether to use primary, secondary or a mix of the two depends on the goal and scope of the LCA to be con-
ducted.
ecoinvent v2.1 report No. 10 -i-

Table of Contents
Table of Contents
ACKNOWLEDGEMENT ......................................................................................................I
COMMENTS ON THIS VERSION 2.0 .....................................................................................I
SUMMARY ......................................................................................................................I
TABLE OF CONTENTS .................................................................................................... III
1 INTRODUCTION ....................................................................................................... 1
2 RESERVES AND RESOURCES OF MATERIALS............................................................. 3
3 CHARACTERISATION OF THE MATERIALS ................................................................... 5

3.1 Bauxite .................................................................................................................................. 5
3.2 Aluminium hydroxide ........................................................................................................... 5
3.3 Aluminium Oxide.................................................................................................................. 5
3.4 Aluminium ............................................................................................................................ 6
4 USE / APPLICATION OF MATERIAL ............................................................................ 7

4.1 Bauxite, aluminium hydroxide and aluminium oxide ........................................................... 7
4.2 Aluminium ............................................................................................................................ 7
5 SYSTEM CHARACTERISATION .................................................................................. 9

5.1 Bauxite mining ...................................................................................................................... 9
5.1.1 Production process and infrastructure ................................................................................... 9
5.1.2 Emissions .............................................................................................................................. 9
5.1.3 Waste..................................................................................................................................... 9
5.1.4 Bauxite mining in ecoinvent ................................................................................................. 9
5.2 Aluminium hydroxide and aluminium oxide production .................................................... 15
5.2.1 Production process and infrastructure ................................................................................. 15
5.2.2 Emissions ............................................................................................................................ 18
5.2.3 Waste................................................................................................................................... 18
5.2.4 Aluminium hydroxide and aluminium oxide production in ecoinvent................................ 19
5.3 Primary aluminium.............................................................................................................. 30
5.3.2 Emissions ............................................................................................................................ 32
5.3.3 Waste................................................................................................................................... 32
5.3.4 Primary aluminium in ecoinvent ......................................................................................... 34
5.4 Secondary aluminium.......................................................................................................... 52
5.4.2 Emissions ............................................................................................................................ 52
5.4.3 Waste................................................................................................................................... 52
5.4.4 Secondary Aluminium in ecoinvent .................................................................................... 53
5.5 Aluminium production mix ................................................................................................. 64
5.5.1 Aluminium production mixes in ecoinvent ......................................................................... 64
6 LITERATURE ......................................................................................................... 69
ecoinvent v2.1 report No. 10 - iii -

1. Introduction
1 Introduction
This part deals with primary and secondary aluminium production. Because bauxite, aluminium hy-
drate and aluminium oxide are mainly consumed in the aluminium production, these chemicals are
also included in this report.
A considerable part of the aluminium consumed today is secondary material. Thus also the collection
of aluminium scrap and the recycling process are included.
A simplified overview of the processes is given in Fig. 1.1.
aluminium
recultivation, aluminium
mine, bauxite hydroxide,
bauxite mine oxide, plant
plant
aluminium aluminium
bauxite, at mine hydroxide, at oxide, at
plant plant
aluminium
electrolysis,
plant
aluminium,
cathode,
primary,
aluminium
liquid, at
electrolysis
plant
anode,
anode, plant aluminium
electrolysis
aluminium, aluminium,
primary, at production
plant mix, cast
alloy, at plant
aluminium,
aluminium
aluminium secondary,
casting, plant aluminium,
aluminium scrap, new, at from new
production
melting plant scrap, at plant
mix, wrought
furnace
alloy, at plant
aluminium
aluminium,
scrap, old, at
secondary,
plant aluminium,
from old
production
scrap, at plant
scrap mix, at plant
preparation
plant
Fig. 1.1 Simplified overview of the processes for the aluminium production
ecoinvent v2.1 report No. 10 -1-

2. Reserves and Resources of Materials
2 Reserves and Resources of Materials

Aluminium is the most important light metal in the outer earth crust. It never occurs in nature in its
metal form but only in compounds with other elements. Bauxite, the main resource for the aluminium
production, is used to 85% for this purpose (Mori & Adelhardt (1998)).
Bauxite, the mineral from which aluminium oxide is gained, is mined mainly in the regions north and
south of the tropical rainforests and most of the aluminium oxide is produced close to the mining sites
(Tab. 2.1). The yearly bauxite mine production was 71 Mt/a in 1983 and the estimated reserves are
22.5 Gt (Hudson et al. (1997)). The yearly bauxite consumption rose to 140 Mt/a in 2002 1 .
Tab. 2.1 World wide bauxite and aluminium oxide production (Mori & Adelhardt (1998), p. 20)
Aus- Guinea Ja- Brasil China Vene- India Russia USA Europe
trailia maica zuela
% of world bauxite
38 13 10 9 6 5 5 3
production
% of world alumin-
ium oxide produc- 30 7 5 6 30 10
tion
98% of the industrial production of aluminium is done by electrolysis of aluminium oxide (Hall-
Heroult process). This process requires a big amount of electricity. Thus most of the aluminium smelt-
ers are located in areas with cheap electricity be it from coal, nuclear or from hydro power plants. The
electricity used to produce aluminium for the European market is produced according to Tab. 2.2
(EAA (2000)).
Tab. 2.2 Electricity distributed by energy source for the production of primary aluminium for the European market
Energy re- natural nuclear hydro

brown coal hard coal crude oil
source gas electricity electricity
Share [%] 5.1 20 4.6 3.1 14.6 52.6
Because of the high energy input needed in primary aluminium production, secondary aluminium is an
important resource. The overall share of secondary aluminium on the world aluminium market is esti-
mated to be 32% 2 by industrial experts 3 .
1
Eirik Nordheim, European Aluminium Assosiation. Personal communication, 15.04.2004
2
This figure is so low because the aluminium market is growing. The share of aluminium that is recycled after use is very
vaiable depending on the application (e.g. in the building sector in Europe between 92% and 98%).
3
Dr. G. Rombach, Project manager Recycling, Hydro Aluminium Deutschland GmbH. Personal communication, 1.10.2002
http://www.aluminium.org/statistics/overview.asp, accessed 18.12.2002

3. Characterisation of the materials
3 Characterisation of the materials

3.1 Bauxite
The term bauxite [CAS Nr. 1318-16-7] is used for sedimentary rocks that contain economically recov-
erable quantities of the aluminium minerals gibbsite, boehmite, and diaspore. The average crude ore
that is mined contains 47.9% bauxite. The world average aluminium oxide content of bauxite is
44.6%, the European average content is 53%. Other important composites of bauxite are SiO2, Fe2O3
and TiO2.
3.2 Aluminium hydroxide

There are various modifications of aluminium hydroxides. The best defined crystalline forms are the
three trihydroxides, Al(OH)3 gibbsite, bayerite and nordstrandite.
Aluminium hydroxide obtained from the Bayer process is a 99.5% pure white powder. It is either cal-
cinated to aluminium oxide ore sold as wet or dry aluminium hydrate. The specifications of the prod-
ucts of the "Aluminium Oxid Stade GmbH" (AOS) are given in Tab. 3.2.
4
Tab. 3.1 Product Specification for aluminium hydroxide
Chemical and physi- Typical Value Min. Val. Max. Val. Quality
cal analysis
Moisture (20°-105° C) % 7.0 5 9 Wet
Al2O3 % 60.5 59 62 Wet
Moisture (20°-105° C) % 0.02 0.00 0.1 Dry
Al2O3 % 65 64.5 65.5 Dry
SiO2 % 0.004 0.003 0.007 Wet and dry
Fe2O3 % 0.009 0.005 0.016 Wet and dry
TiO2 % < 0.001 0.000 0.003 Wet and dry
CaO % 0.004 0.000 0.007 Wet and dry
Na2O % 0.22 0.18 0.26 Wet and dry
V2O5 % < 0.0007 0.000 0.001 Wet and dry
P2O5 % < 0.0004 0.000 0.001 Wet and dry
Ga2O3 % 0.007 0.002 0.013 Wet and dry
Ignition loss (105° - % 34.6 Wet and dry
1200° C)
Particle size < 45 µm % 4 1 12 Wet and dry
Particle size < 90 µm % 45 30 60 Wet and dry
3.3 Aluminium Oxide

Hexagonally closest packed aluminium oxide, -Al2O3, is the only stable oxide in the Al2O3-H2O-
system. The mineral is also called corundum. Corundum is common in igneous and metamorphic
rocks. Red and blue varieties of gem quality are called ruby and sapphire, respectively.
The specifications of the Al2O3 of the "Aluminium Oxid Stade GmbH" (AOS) are given in Tab. 3.2.
4
http://www.aos-stade.de/html/home_gb.html, accessed 16.12.2002

5
Tab. 3.2 Product Specification for aluminium oxide
Chemical and physical analysis Typical Value Min. Val. Max. Val.
Moisture (20°-300° C) % 0.10 0.00 0.40
Bulk density (lose) kg/m³ 970 950 1000
Al2O3 % 99.0 98.5 99.5
SiO2 % 0.006 0.004 0.010
Fe2O3 % 0.014 0.008 0.025
TiO2 % 0.001 0.000 0.004
CaO % 0.006 0.000 0.010
Na2O % 0.34 0.28 0.39
V2O5 % 0.001 0.000 0.002
P2O5 % < 0.0006 0.000 0.001
Ga2O3 % 0.011 0.003 0.020
Ignition loss (300° - 1200° C) % 0.53 0.45 0.60
Angle of repose degree ° 34 32 36
BET surface area m²/g 50 45 60
Particle size < 45 µm % 8 5 15
Particle size < 90 µm % 55 45 65
3.4 Aluminium
Aluminium is a light grey coloured light metal. It has high specific strength, low specific weight, high
electrical and thermal conductivity, good ductility or high hardness (depending on alloy) and due to a
relatively stable and dense oxide layer on its surface, it exhibits good resistance against corrosion.
Some of the most important chemical and physical properties of aluminium are given in Tab. 3.3.
Tab. 3.3 Physical properties of Aluminium
Property Value Unit

Atomic number 13 -
Atomic weight 27 g/mol
Density 2700 kg/m3
Melting point 660.4 °C
Boiling point 2467 °C
Thermal conductivity (at 20°C) 239 W/mK
Modulus of elasticity (99.99% Al) 7.1*1010 N/m2
Tensile strength (99.99% Al, soft / strain-hardened) 5*107 / 13*107 N/m2
Tensile strength (Dural) 40*107 N/m2
Specific electrical resistance 2.82*10-8 Ωm
Most technically used aluminium is alloyed. Important alloy elements for aluminium are silicon (Si),
zinc (Zn), copper (Cu) and magnesium (Mg). Basically, two types of aluminium alloys are commonly
distinguished: wrought alloys and casting alloys. As a rule of thumb, casting alloys tend to have higher
contents of non-aluminium metals, especially of iron, silicon and/or zinc, whereas wrought alloys tend
to have higher magnesium or sometimes manganese contents.
5
http://www.aos-stade.de/html/home_gb.html, accessed 16.12.2002

4. Use / Application of Material
4 Use / Application of Material

4.1 Bauxite, aluminium hydroxide and aluminium oxide
About 75% of the bauxite and 90% of the aluminium hydroxide and aluminium oxide is used for alu-
minium production. Other uses are listed in Tab. 4.1.
Tab. 4.1 Uses of aluminium precursors other than aluminium production
bauxite aluminium hydroxide aluminium oxide

Fireproof bricks Filler for paints Pigment in paints
Abrasives Filler for paper Filler
Catalysts Filler for plastics (fire retardant) Ceramics
Adsorbents Raw material for aluminium com- Cosmetics and pharmaceutics (e.g.
pounds sun protection)
Cosmetics and pharmaceutics (e.g.
toothpaste)
4.2 Aluminium
In 1995, 20.4 Million t of primary aluminium has been used world wide (Mori & Adelhardt (1998)).
The amount has increased steadily in the last decades. The biggest consumer of aluminium is the
transport industry. Also important are the building and packaging sectors. (See Fig. 4.1 and Fig. 4.2)
6
Fig. 4.1 Markets for aluminium products in Europe (Estimation EAA)
Secondary aluminium is mainly used in cast alloys because they are less sensitive to high content of
silicon and other metals in the scrap and thus can be produced more economically from scrap.
6
Fig. 4.2 Market for recycled aluminium in Europe in 2000 (estimation EAA)
6
http://www.aluminium.org/statistics/overview.asp, accessed 18.12.2002

5. System Characterisation
5 System Characterisation
5.1 Bauxite mining
This part is mainly based on Mori & Adelhardt (1998).
5.1.1 Production process and infrastructure

More than 95% of the bauxite today is gained from open mines. The thickness of the recoverable ore
deposit usually is between 3.5 and 7 meters. The cap of the deposit, which in average is less than one
meter thick, is removed as overburden and deposited nearby to be used for the recultivation when the
mine has been depleted. The ore is won in stripes using hydraulic excavators and / or belt loaders.
Drilling and blasting is only needed in some mines with exceptionally hard bauxite. The bauxite is
transported to the mobile grinder and possibly to a dryer from where it is shipped to the aluminium ox-
ide extraction plant.
According to the International Aluminium Institute, about 80% of the bauxite mines are recultivated
after their depletion and 97% of the plants have plans for rehabilitating the area currently mined.
5.1.2 Emissions
The only relevant direct emissions from bauxite mining are the dust emissions to air.
5.1.3 Waste
There is no relevant waste produced in the mining process. The overburden is used in recultivation of
the mine and not treated as waste although due to finer fractions and possibly increased weathering
higher amounts of components of the backfilled material might get into the ground water compared to
the situation before the mining activity.
5.1.4 Bauxite mining in ecoinvent

The background information to the processes are given in Tab. 5.1, flows and their sources are shown
in:
- bauxite, at mine Fig. 5.1
- mine, bauxite Fig. 5.2
- recultivation, bauxite mine Fig. 5.3
For a lack of information on this topic, the land use, the size and type of buildings and the weight of
the stationary machines used are estimated. The estimation is based on information on other mines
(gypsum, limestone).
Data quality
Data for bauxite at mine are of good overall quality. The infrastructure is roughly estimated but com-
pared to the actual mining activities the influence is negligible. The values for the mining processes
stem from a recent and comprehensive source (Mori & Adelhardt (1998)).

Tab. 5.1 Meta information for the bauxite mining processes
recultivation, bauxi-
Name
bauxite, at mine mine, bauxite te mine
Location GLO 7 GLO GLO
Infrastructure Process 0 1 1
Unit kg unit m2
Data Set Version 2.0 2.0 1.0
Includes the mining with its Includes the infrastructure
land transformation and oc- and their disposal (buildings
cupation and the first me- with their land use and ma- Includes Land transfor-
chanical grinding and sepa- chines) needed to provide the mation and the ma-
Included Processes
ration of the ore from the dried bauxite. Does not in- chines needed. No refill
gangue material. Also include the land use for the material other than the
cluded is the drying of the mine itself nor that for the overburdens from mining
bauxite to 11% moisture. roads in the mine. is included.
Amount 1 1 1
Rekultivierung, Bauxit-
Local Name
Bauxiterz, ab Abbau Mine, Bauxit mine
aluminium
Synonyms
ore//Aluminiumerz
Refers to the dry bauxite Refers to the recultiva-
General Comment to mass (11% moisture). The It is assumed that the mine is tion of a mine with baux-
reference function Al2O3 content of the bauxite in use for 20 years and has a ite content of about 50%
is 53% (w/w) yearly output of 1E6 t/a (w/w) in crude ore
Start Date 1994 1998 1998
End Date 2002 2002 2002
Data Valid For Entire
Period 1 1 1
Some European datasets
Geography text are used for global proc- European data are used for
esses. buildings and machines
Average bauxite mining Average technology for the Average bauxite mine
Technology text technology world-wide aluminium consumed in recultivation world-wide
(open mining) Europe (open mining)
Representativeness
[%] 79
111E6 t dry bauxite
Production Volume (=137E6 t crude bauxite) in
1994 unknown unknown
Sampling Procedure Literature / Survey estimation Literature / Survey
Extrapolations see Geography none none
Uncertainty is adjusted for
Uncertainty Adjust-
geographical mismatch (us-
ments
ing the pedigree matrix). none none
7
The country code should actually be RER, since the subsequent processes all are for RER. This will be changed in v2.2
ecoinvent v2.1 report No. 10 - 10 -

General Flow information Representation in ecoinvent Uncertainty

Infra
Process Sub Loca Modul name in Mean Source mean Typ StDv General
Input Output Remarks Cate gory struc Unit
Name category tion ecoinvent value value e 95% Comment
ture
Bauxite = 47.9 % in crude ore, Al2O3

Aluminium, 24% in
Bauxite = 53 % in bauxite, Al = 52.9 % in Al2O3 Mori & Adelhardt 1998
 resource in ground bauxite, 11% in crude 2.81E-01 kg 1 1.06 (1,2,2,1,1,1,12)
(resource) --> 0.281 kg Al = 1 kg Bauxite = 2.09 (p72)
ore, in ground
kg crude ore
Land use (mine) assumption that the massive size of Occupation, mineral Mori & Adelhardt 1998
 resource land 3.35E-04 m2a 1 1.50 (1,2,2,1,1,1,7)
[m2a] 3.5 to 7 m is depleated in 2 year extraction site (p. 43)
Land
Transformation, to Mori & Adelhardt 1998
transformation  resource land 1.67E-04 m2 1 2.00 (1,2,2,1,1,1,8)
mineral extraction site (p. 43)
(mine) [m2]
bauxite, at mine: input
Land
Transformation, from Mori & Adelhardt 1998
transformation  resource land 1.07E-04 m2 1 2.00 (1,2,2,1,1,1,8)
arable (p. 43)
(mine) [m2]
Land Transformation, from
Mori & Adelhardt 1998
transformation  resource land arable, non-irrigated, 3.41E-05 m2 1 2.00 (1,2,2,1,1,1,8)
(p. 43)
(mine) [m2] fallow
Land
Transformation, from Mori & Adelhardt 1998
transformation  torpical rainforest resource land 2.67E-05 m2 1 2.00 (1,2,2,1,1,1,8)
forest, extensive (p. 43)
(mine) [m2]
Recultivation recultivation, bauxite Mori & Adelhardt 1998
 80% of mines are recultivated metals extraction No GLO 1.34E-04 m2 1 2.00 (1,2,2,1,1,1,8)
mine mine (p. 43)
Other source (EAA 2001) gives 2.88E-5 Water, unspecified
Water  resource in water 5.78E-04 m3 Mori & Adelhardt 1998 1 1.06 (1,2,2,1,1,1,12)
m3 natural origin
diesel, burned in Input includes infrastructure and
construction diesel, burned in
building machine  emissions of building machines machinery No GLO 4.39E-02 MJ Mori & Adelhardt 1998 1 2.00 (1,2,2,1,1,1,5)
processes building machine
[MJ] according to their fuel consumption
electricity, medium
Electrical power  electricity production mix No UCTE voltage, production 2.80E-03 kWh Mori & Adelhardt 1998 1 1.06 (1,2,2,1,1,1,2)
UCTE, at grid
Blasting [kg Blasting includes Production of construction
 civil engineering No RER blasting 1.56E-04 kg Mori & Adelhardt 1998 1 1.06 (1,2,2,1,1,1,1)
explosive] explosive and emissions of blasting processes
Input includes infrastructure needed calculated (Mine with

Infrastructure
 at mine (buildings, machines,...) metals extraction Yes GLO mine, bauxite 8.33E-13 unit output of 1E6 t/year, 1 3.00 (1,2,2,1,1,1,9)
(mine)
(exclusive building machines) in use for 20 years)
Fig. 5.1 Flows for "bauxite, at mine" and their representation in the ecoinvent database: The percentages given in the resource name are valid for the global situation. The should,
however, be valid for the European situation and thus be 28.1% in Bauxite, 13.4% in crude ore.


Infra
Pro c e ss Cate Sub Lo c a Mo dul name in Mean So urc e mean Ty StDv Ge ne ral
Input Output Re mark s struc Unit
Name go ry c atego ry tio n e c o invent value value pe 95% Co mment
ture
low p op ulat io n
 wast e heat f rom elect ricit y use air Heat , wast e 1.0 1E-0 2 M J calculat ed 1 1.08 (1,1,2,1,1,3 ,13 )
densit y
bauxite, at mine:
low p op ulat io n
 Dust t o air [ kg ] 50% of " Dust " is > 10 um air Part iculat es, > 10 um 1.6 0E-0 3 kg 1 1.51 (1,3 ,2,3 ,1,3 ,25)
densit y t o t al amo unt : EA A
20 01 (p ers.
Output
45 % of " Dust " is b et ween 2 .5 and low p op ulat io n Part iculat es, > 2 .5 um, communicat ion),
 Dust t o air [ kg ] air 1.4 4E-0 3 kg 1 2 .01 (1,3 ,2,3 ,1,3 ,26 )
10 um densit y and < 10 um f ract io n: CEPM EIP
Dat ab ase - Emission
low p op ulat io n Fact ors
 Dust t o air [ kg ] 5 % o f " Dust " is belo w 2.5 um air Part iculat es, < 2 .5 um 1.6 0E-0 4 kg 1 3 .01 (1,3 ,2,3 ,1,3 ,27)
densit y
The wast e is used in t he

 W ast e [ kg] recult ivat io n of t he mine and t hus -4 .74E-0 1 kg mass balance
not invent o ried
B auxit e, at
 met als ext ract io n No GLO b auxit e, at mine 1.00 E+0 0 kg
mine
Fig. 5.1 Flows for "bauxite, at mine" and their representation in the ecoinvent database (cont.)


Infra
ture
Land use f or
ro ads in mine  Is invent o ried in Land use f or mine
area
est imat ion, b ased o n
Land use f or A ssumed t hat mine is in use f or 20 Occup at io n, indust rial (5,n.A .,n.A .,n.A .,n.A .,
 reso urce land 3.0 0E+05 m2a knowled ge o f o t her 1 2.2 3
build ings years area, built up n.A .,8 ); roug h guess
mines
Land use f or Transf ormat io n, t o (5,n.A .,n.A .,n.A .,n.A .,
 ro ugh assump t io n reso urce land 1.50 E+0 4 m2 knowled ge o f o t her 1 2.2 3
build ings ind ust rial area, b uilt up n.A .,8 ); roug h guess
mines
mine, bauxite
Land use f or Transf ormat io n, f rom M ori & A d elhard t (5,n.A .,n.A .,n.A .,n.A .,
 55 % f ro m agricult ure / silvicult ure reso urce land 8.2 5E+0 3 m2 1 2.2 3
build ings arable 19 98 n.A .,8 ); roug h guess
Land use f or 27% f rom ext ensive agricult ure Transf ormat io n, f rom M ori & A d elhard t (5,n.A .,n.A .,n.A .,n.A .,
 reso urce land 4.0 5E+0 3 m2 1 2.2 3
build ings (st epp e, savannah, desert ) arable 19 98 n.A .,8 ); roug h guess
Land use f or Transf ormat io n, f rom M ori & A d elhard t (5,n.A .,n.A .,n.A .,n.A .,
 18 % f ro m t rop ical rainf orest reso urce land 2.70 E+0 3 m2 1 2.2 3
build ings f orest , ext ensive 19 98 n.A .,8 ); roug h guess

mult i st orage const ruct ion (5,n.A .,n.A .,n.A .,n.A .,
 build ings Y es RER b uilding , mult i-st o rey 1.50 E+0 3 m3 knowled ge o f o t her 1 3.2 3
build ings processes n.A .,9 ); roug h guess
mines
halls f o r d ryer const ruct ion (5,n.A .,n.A .,n.A .,n.A .,
build ings Y es CH b uilding , hall 1.00 E+0 4 m2 knowled ge o f o t her 1 3.2 3
et c. processes n.A .,9 ); roug h guess
mines
machines f o r Ind ust rial machine, est imat ion, b ased o n
const ruct ion (5,n.A .,n.A .,n.A .,n.A .,
crushing ,  131.32 09 80 3 machinery Y es RER heavy, unsp ecif ied , at 1.00 E+0 4 kg knowled ge o f o t her 1 3.2 3
processes n.A .,9 ); roug h guess
washing et c. p lant mines
 M ine, bauxit e met als ext ract io n Y es GLO mine, b auxit e 1.00 E+0 0 unit
Fig. 5.2 Flows for "mine, bauxite" and their representation in the ecoinvent database


Infra
ture
diesel, b urned in half o f t he demand f o r d eplet ion as

d iesel, burned in
build ing machine  half o f t he amount of mat erial is No GLO # B EZUG! M J est imat ed 1 2 .10 (4,1,2,1,3,1,5)
b uilding machine
[ M J] moved
recultivation, bauxite mine
Land
t ransf o rmat ion Transf ormat io n, f rom
 10 0% f rom mine reso urce land 1.00 E+0 0 m2 calculat ed 1 2.0 0 (1,2 ,2,1,1,1,8)
(f ro m mine) mineral ext ract io n sit e
[ m2 ]
Land
t ransf o rmat ion Transf ormat io n, t o M ori & A d elhard t
 55% t o f arm land reso urce land 5.50E-0 1 m2 1 2.0 0 (1,2 ,2,1,1,1,8)
(f ro m mine) arable 19 98 (p .41)
[ m2 ]
Land
Transf ormat io n, t o
t ransf o rmat ion M ori & A d elhard t
 22 % t o ext ensive f arm land reso urce land arable, no n-irrig at ed , 2.2 0E-0 1 m2 1 2.0 0 (1,2 ,2,1,1,1,8)
(f ro m mine) 19 98 (p .41)
f allow
[ m2 ]
Land
 22 % t o f orest reso urce land 2.2 0E-0 1 m2 1 2.0 0 (1,2 ,2,1,1,1,8)
(f ro m mine) f orest 19 98 (p .41)
[ m2 ]
Land
 1% t o indust rial area reso urce land 1.0 0E-0 2 m2 1 2.0 0 (1,2 ,2,1,1,1,8)
(f ro m mine) ind ust rial area 19 98 (p .41)
[ m2 ]
Recult ivat ion recult ivat ion, b auxit e
 met als ext ract io n Y es GLO 1.00 E+0 0 m2
mine mine
Fig. 5.3 Flows for "recultivation bauxite mine" and their representation in the ecoinvent database

5.2 Aluminium hydroxide and aluminium oxide production

This part is mainly based on information from Aluminium Oxid Stade GmbH 8 and Queensland Alu-
mina Limited 9 and on (Mori & Adelhardt (1998)) and (Hudson et al. (1997)).

Overview
98% of the world wide production of aluminium hydroxide and aluminium oxide is based on the Bay-
er-Process. Austrian Karl-Josef-Bayer received patents in 1887 and 1892 for his process to extract a-
luminium hydroxide from bauxite using caustic soda.
The aluminium components of bauxite are dissolved by addition of caustic soda. By means of sedi-
mentation and filtration the residue of the digestion is separated. Out of the clear sodium aluminate so-
lution the aluminium hydroxide is precipitated by addition of seeding crystals. After filtration, the
aluminium hydroxide is calcined to aluminium oxide at approx.1000° C. The calcination in the circu-
lating fluidised bed calciner is to remove the water of crystallisation from the hydroxide.
The operation of the process is outlined in Fig. 5.4.
The infrastructure is estimated based on an average sized aluminium oxide refinery (the Aluminium
Oxid Stade GmbH (AOS)) in Germany. The plant is spread over an area of 55 hectares. The produc-
tion was started up at the end of 1973. The nominal production capacity is on a current level of ap-
prox. 800'000 metric tons of alumina per year. Fig. 5.5 gives an overview of the refinery.
Delivery of raw materials
Most of the refineries are located close to a bauxite mine and / or they have their own port. AOS has a
port with outside and inside berth with a usable length of the pier of about 200 m. The unloading is
performed by grab with an unloading rate of approx. 600 t/h. The bauxite is transported to the bauxite
storage area or the storage building in an enclosed conveyor system.
Bauxite storage
AOS open storage capacity is more than 1 mil metric tons. Additionally there is a storage building
with approx. 100'000 t capacity.
Grinding
The bauxite is ground to a size of maximum 20 mm under addition of recycled digestion liquor (caus-
tic soda solution). Lime can be introduced for phosphate control and mud conditioning. After grinding,
the bauxite-liquor mixture (slurry) is pumped into storage tanks.
Desilication
The silica component of the bauxite is chemically attacked by caustic soda, causing alumina and soda
losses by combining to form solid desilication products. To desilicate the slurry prior to digestion, it is
heated and held at atmospheric pressure in pre-treatment tanks, reducing the build-up of scale in tanks
and pipes. Most desilication products pass out with the mud waste as sodium aluminium silicate com-
pounds.
8
http://www.aos-stade.de, accessed 16.12.2002
9
http://www.qal.com.au, accessed 16.12.02

Fig. 5.4 Bayer process: Extraction of aluminium hydroxide from bauxite and calcination to aluminium oxide (alu-
8
mina) .
8
Fig. 5.5 Aerial view of the Aluminium Oxid Stade GmbH (AOS) alumina refinery
Digesting
The bauxite slurry is pumped through the approximately 4'000 m long heat exchanger system of four
tube digesters in parallel. For digestion, the slurry is heated to approximately 270° C at a pressure of
about 3.5 MPa. Under these conditions, the chemical reactions are rapid:
2NaOH + Al2O3*3H2O  2NaAlO2 + 4H2O (5.1)
2NaOH + Al2O3*H2O  2NaAlO2 + 2H2O (5.2)

Thus about 97% of the aluminium components are extracted from the bauxite. After digestion about
30% of the bauxite mass remains in suspension as thin red mud slurry of silicates, and oxides of iron
and titanium. The mud-laden liquor leaving the digestion vessel is flash-cooled to atmospheric boiling
point by flowing through a series of flash vessels which operate at successively lower pressures. The
energy released by flash evaporation is utilised to pre-heat the bauxite slurry. The condensate gener-
ated, which comprises approximately 75 % of the total condensate generated in the plant, is used for
washing processes.
Dilution
The bauxite slurry, concentrated by the evaporation of water during digestion, is diluted with wash
water from subsequent operations. At AOS 2 dilution tanks with each 1'000 m3 and a height of 15 m
are used. Measured addition of digestion liquor adjusts the desired molar ratio to reduce recovery los-
ses.
Red Mud Separation
The coarse fraction of the undissolved bauxite components (sand) is separated by a hydroseparator and
a rake classifier. Next, the slurry flows to the settlers where the remaining undissolved components of
the bauxite (bauxite residue (BR) or red mud) are separated by sedimentation. Flocculants might be
added to the settler feed stream to improve the rate of mud settling and achieve good clarity in the
overflow liquor. The underflow of the settlers is pumped to the first of two washers for pre-washing of
the BR. The filtrate is pumped to the washers in a countercurrent stream for dilution. Slaked lime is
added to the dilute caustic liquor in the washing process to remove carbonate (Na2CO3) which forms
by reaction with compounds in bauxite and also from the atmosphere and which reduces the effective-
ness of liquor to dissolve alumina. Lime regenerates caustic soda, allowing the insoluble calcium car-
bonate to be removed with the waste mud.
Na2CO3 + Ca(OH)2  CaCO3 + 2NaOH (5.3)

The settled residue from the second washer is pumped to BR filtration, separated by vacuum drum fil-
ters and washed with condensate. The filtrate is used in the washer while the filter cake is pumped to
the BR disposal area.
Liquor Cleaning
The rest of the undissolved contaminants, such as iron oxide, are substantially filtered out by Kelly fil-
ters (pressure filter) using polypropylene filter cloth with aluminium hydroxide as a filter aid. As the
contaminants are removed, the aluminium hydroxide is recovered as the filter cake.
Liquor Cooling
The clear supersaturated alumina solution from the top stream of the settlers is cooled down to the
temperature required by the precipitation process through multi-stage flash evaporation. Waste heat is
used to pre-heat the spent liquor from precipitation.
Precipitation
The cleaned aluminate liquor flows to the agitated precipitation tanks. In these, the super saturated liq-
uor of aluminium hydroxide is seeded (aluminium hydroxide) in two steps and aluminium hydroxide
is crystallised over 40-50 hours.
Hydroxide Filtration
The fine fraction of the hydroxide is filtered on disc filters and is returned as seed to the precipitation
process. The filtrate, after measured addition of caustic soda, flows back as digestion liquor primarily
to the tube digester and grinding facilities. After double filtration and washing, the coarse fraction of
the hydrate is calcined, dried, or transported to the hydroxide storage facility.

Calcination
The feed material is calcined to remove both free moisture and chemically-combined water. By calci-
nation, aluminium hydroxide is transformed into aluminium oxide at temperatures around 1'000° C.
Calciners can use rotary kilns or the fluid bed process. Fluid bed facilities have proven themselves ve-
ry reliable, are more energy efficient and don't contaminate the product like the rotary kilns.
Rotary or satellite coolers are used to cool the calcined alumina from the rotary kilns, and to pre-heat
secondary combustion air for the kilns. Fluidised-bed coolers further reduce alumina temperature to
less than 90°C before it is discharged on to conveyor belts which carry it to storage buildings or silos
where it is stockpiled for shipment.
5.2.2 Emissions
Air
There are no relevant direct air emissions from these processes beside dust from bauxite handling and
water from the calcination.
At some sites the area of bauxite unloading and the conveyor belt system for transportation of input
materials and products, are enclosed and feature efficient dust removing facilities.
Water
Process water is recycled. Waste water is created only by the operation of the bauxite residue disposal.
Rain water has to be drained from the disposal site. This water is treated before being emitted to na-
ture. After the treatment some NaOH, suspended particles and oxidisable compounds remain in the
water.
5.2.3 Waste
Different waste fractions are resulting from the production of alumina. The main part is the red mud or
bauxite residue, consisting to about 40 % of water and the other part of mineral matters, being undis-
solved in the caustic soda - this is principally inert iron and titanium oxides and silica compounds - de-
riving from the bauxite input. This waste is disposed in special landfills (ponds). Red mud ponds are
controlled landfills with base lining and leachate collection. The red mud disposal is therefore invento-
ried with the model for the residual landfill (Doka (2003), Part III). The red mud is not solidified with
cement (like e.g. landfilled incineration residues are). Still the calcium content of red mud is high
enough (2.2%) to estimate a carbonate phase of longer than 60'000 years (as in the residual landfill
model). No adaptation of transfer coefficients in the model is necessary.
A smaller part is sand waste. For this waste the same composition as for the red mud is assumed.
A generic red mud composition is derived from literature data for red mud from Australia, Jamaica,
India, Hungary, Yugoslavia, Turkey, and Suriname (Altundogan et al. (2002), Anonymous (2001),
McLaughlin et al. (1998), NPI (1999)). It is given in Tab. 5.2.

Tab. 5.2 Composition of red mud in this project
Element Amount Element Amount Element Amount Element Amount

[% w/w] [% w/w] [% w/w] [% w/w]
O 37.97% Cd 0.0005% Pb 0.0014% Si 10.4474%
C 0.7607% Co 0.0022% Sb 0.0000% Fe 29.6417%
S 0.0800% Cr 0.0751% Se 0.0005% Ca 2.2446%
P 0.1475% Cu 0.0037% V 0.0535% Al 8.4217%
B 0.0010% Hg 0.0000% Zn 0.0037% Na 5.9688%
F 0.1130% Mn 0.0363% Be 0.0001% Na 5.9688%
As 0.0029% Ni 0.0005% Ti 4.0203% Na 5.9688%
5.2.4 Aluminium hydroxide and aluminium oxide production in ecoinvent

The available information about the infrastructure needed for aluminium oxide production is non-
quantitative. Tab. 5.3 and Tab. 5.4 show the details of the assumptions made to estimate the infrastruc-
ture expenditures.
Tab. 5.3 Estimated infrastructure for aluminium hydroxide plant
aluminium hydroxide plant Demand Unit Life Life time

time demand
Yearly production capacity 1'220'000'000 kg 50 1'220'000'000
Area needed without calcination and aluminium oxide storage 522'500 m2 50 522'500
(95% of total) [m2]
Administration building (50x50x10 m3) 25'000 m3 50 25'000
Concrete for piers [m3] 1'200 m3 50 1'200
Grab for unloading [kg] 100'000 kg 25 200'000
conveyor to bauxite storage [m] 300 m 25 600
Storage building (hall for 100'000 t = 50'000 m3; average height 5'000 m2 50 5'000
10m) [m2]
Taking up device [kg] 10'000 kg 25 20'000
conveyor bauxite storage to grinder [m] 20 m 25 40
Rod mills for grinding [kg] 10'000 kg 25 20'000
Tube digester (4000 m length) [kg] 28'000 kg 25 56'000
High pressure diaphragm pump [kg] 10'000 kg 25 20'000
Dilution tank (2 tanks to 1000 m3 --> 4.6m radius and 15m 39'200 kg 25 78'400
height) [kg]
Red mud separation (3 settlers) 1'414 m3 50 1'414
Red mud separation (2 countercurrent washers) 705 m3 50 705
Red mud separation (12 vacuum drum filters) 63'335 kg 25 126'669
Liquor cleaning 40'000 kg 25 80'000
Precipitation tanks (30 tanks, 3000 m3 each, 15 m high --> ra- 606'956 kg 25 1'213'911
dius 8m
Hydroxide filters (11 Filters, radius 4 m, height 10 m) 135'465 kg 25 270'931
Hydroxide silo (500 t = 500 m3; height 10m radius 4m 12'315 kg 25 24'630
Hydroxide storage hall (6000 t = 6000 m3, average height 10 600 m2 50 600
m)

Tab. 5.4 Estimated infrastructure for aluminium oxide plant
aluminium oxide plant Demand Unit Life Life time

time demand
Yearly production capacity 800'000'000 kg 50 800'000'000
Area needed for calcination and aluminium oxide storage (5% 27'500 m2 50 27'500
of total) [m2]
3 fluid bed calciners 20'000 kg 25 40'000
Concrete silo (65000 m3; height 25 m, radius 30m 1'555 m3 50 1'555
Steel silo (2000 m3, height 15 m, radius 6.5m) 30'733 kg 25 61'465
conveyor belt 100 m 25 200
The other background information to the processes is given in Tab. 5.5, flows and their representation
in the database is shown in:
- aluminium hydroxide, at plant Fig. 5.6
- aluminium hydroxide, plant Fig. 5.7
- aluminium oxide, at plant Fig. 5.8
- aluminium oxide, plant Fig. 5.9
Data quality
Data for the steps in the aluminium oxide production are of good overall quality. The infrastructure is
roughly estimated but compared to the other activities the influence is negligible. The values for the
processes stem from a recent and comprehensive source (EAA (2000)).

Tab. 5.5 Meta information for the aluminium hydroxide and aluminium oxide processes
Name aluminium hydroxide, at plant aluminium hydroxide, plant

Location RER RER
Infrastructure Process 0 1
Unit kg unit
Data Set Version 2.0 2.0
Part of the Al2O3 plant needed for
Includes bauxite and ancillary materials with Al(OH)3 fabrication including port,
Included Processes their transports to the calcination plant, the storage and process infrastructure.
extraction of Al(OH)3 (grinding, digestion and Also includes Building for administra-
precipitation) and the disposal of the wastes. tion. Includes disposal.
Amount 1 1
Local Name Aluminiumhydroxid, ab Werk Aluminiumhydroxid, Fabrik
Bay-
erit//Hydragillit//Gibbsit//Tonerdehydrat//bayeri
Synonyms
te//hydragillite//gibbsite//diaspore//Al2O3*3H2
O//Aluminiumtrihydrate//aluminium hydroxide
Al(OH)3 extraction and calcination to Al2O3 is
done in the same plant in line. Only cumulated
data for the whole process are available. In-
and outputs of AL(OH)3 are calculated ac-
General Comment to cording to the stoichiometry (1.530 kg
reference function Al(OH)3 --> 1.000 kg Al2O3 + 0.530 kg H2O).
Weak data for the energy consumption frac-
tions of the two process parts were available.
It is however not known if they relate to wet or Medium sized plant. Output:
dry aluminium hydroxide. 1'220'000 t/a, life time: 50 years
Start Date 1995 2002
End Date 2002 2002
Period 1 1
Some Swiss datasets are used for European Some Swiss datasets are used for
Geography text
processes. European processes.
Average technology for the aluminium con- Technology of one medium sized
Technology text
sumed in Europe plant in Germany
Representativeness [%] 98
Production Volume unknown unknown
Sampling Procedure Literature / Survey Internet search
Extrapolations see Geography see Geography
Uncertainty is adjusted for geo-
Uncertainty Adjustments Uncertainty is adjusted for geographical mis- graphical mismatch (using the pedi-
match (using the pedigree matrix). gree matrix).

Tab. 5.5 Meta information for the aluminium hydroxide and aluminium oxide processes (cont.)
Name aluminium oxide, at plant aluminium oxide, plant

Location RER RER
Unit kg unit
Includes calcination of Al(OH)3 to Al2O3 and Part of the Al2O3 plant needed only
its infrastructure. No transports are included for Al2O3 fabrication (= calcination
Included Processes
because Al2O3 is produced in the same plant and Al2O3 storage) including dis-
as Al(OH)3. posal
Amount 1 1
Local Name Aluminiumoxid, ab Werk Aluminiumoxid, Fabrik
Synonyms Tonerde//Korund//Corundum//alumina
Al(OH)3 extraction and calcination to Al2O3 is
done in the same plant in line. Only cumulated
data for the whole process are available. In-
and outputs of AL(OH)3 are calculated ac-
General Comment to cording to the stoichiometry (1.530 kg
reference function Al(OH)3 --> 1.000 kg Al2O3 + 0.530 kg H2O).
Weak data for the energy consumption frac-
tions of the two process parts were available.
It is however not known if they relate to wet or Medium sized plant. Output: 800'000
dry aluminium hydroxide. t/a, life time: 50 years
End Date 2002 2002
Period 1 1
Some Swiss datasets are used for European Some Swiss datasets are used for
Geography text
processes. European processes.
Average technology for the aluminium con- Technology of one medium sized
Technology text
sumed in Europe plant in Germany
Production Volume unknown unknown
Sampling Procedure Literature / Survey Internet search
Uncertainty is adjusted for geo-
Uncertainty Adjustments Uncertainty is adjusted for geographical mis- graphical mismatch (using the pedi-
match (using the pedigree matrix). gree matrix).


Infra
ture
B auxit e [ kg]  met als ext ract io n No GLO b auxit e, at mine 1.40 E+0 0 kg EA A 2 00 0 (p. 4 9) 1 1.08 (1,1,2,1,1,3 ,3)
Transpo rt of
B auxit e f ro m aluminium hydroxide, at plant; Inputs part 1 t ranspo rt t ransp ort , t ranso ceanic
 Transpo rt by ship (o cean) ship No OCE 7.11E+0 0 t km EA A 2 00 0 (p. 50 ) 1 2.0 0 (1,1,2,1,1,3 ,5)
mine t o plant syst ems f reig ht ship
[ t km]
Transpo rt of
B auxit e f ro m t ranspo rt
 Transpo rt by ship (co st al) ship No RER t ransp ort , b arg e 2.0 0E-0 3 t km EA A 2 00 0 (p. 50 ) 1 2.0 0 (1,1,2,1,1,3 ,5)
mine t o plant syst ems
[ t km]
Transpo rt of
 Transpo rt by Ro ad ro ad No RER t ransp ort , lo rry 32 t 3.3 4E-0 1 t km EA A 2 00 0 (p. 50 ) 1 2.0 0 (1,1,2,1,1,3 ,5)
[ t km]
Transpo rt of
 Transpo rt by Rail t rain No RER t ransp ort , f reig ht , rail 1.10E-0 2 t km EA A 2 00 0 (p. 50 ) 1 2.0 0 (1,1,2,1,1,3 ,5)
[ t km]
Grind ing o f
Energy is includ ed in g eneral p ower
bauxit e [ kg / kg 
consump t io n
A l2O3 ]
sod ium so dium hyd ro xide, 50 %
only weig ht o f NaOH in solut ion
hydroxid e, 50%  chemicals inorganics No RER in H2O, p ro duct ion 3.9 3E-0 2 kg EA A 2 00 0 (p. 4 9) 1 1.08 (1,1,2,1,1,3 ,4)
(wit hout solvent )
in H2 O [ kg ] mix, at plant
Transpo rt of
NaOH f ro m Rail; St andard d ist ance f ro m t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
 t rain No RER t ransp ort , f reig ht , rail 4 .71E-0 2 t km st and ard dist ance 1 2.0 9
plant t o plant Eco invent 2 00 0 syst ems A .,5)
[ t km]
Transpo rt of
NaOH f ro m Road ; St andard d ist ance f ro m t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
 ro ad No RER t ransp ort , lo rry 32 t 7.85E-0 3 t km st and ard dist ance 1 2.0 9
plant t o plant Eco invent 2 00 0 syst ems A .,5)
[ t km]
Heat ing o f slurry
Heat included in general heat
and so lut io n  EA A 2 00 0
consump t io n
[ M J]
CaO f o r const ruct ion q uicklime, milled,
 add it ives No CH 3.0 1E-0 2 kg EA A 2 00 0 (p. 4 9) 1 1.08 (1,1,2,1,1,3 ,4)
desilicat ion [ kg] mat erials lo ose, at plant
Fig. 5.6 Flows for "aluminium hydroxide, at plant" and their representation in the ecoinvent database


Infra
ture
Transpo rt of
Rail; St andard d ist ance (f o r t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
CaO f rom p lant  t rain No RER t ransp ort , f reig ht , rail 3.0 1E-0 3 t km st and ard dist ance 1 2.0 9
cement ) f rom Ecoinvent 20 00 syst ems A .,5)
t o plant [ t km]
Transpo rt of
Rail; St andard d ist ance (f o r t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
CaO f rom p lant  aluminium hydroxide, at plant; Inputs part 2 ro ad No RER t ransp ort , lo rry 32 t 3.0 1E-0 3 t km st and ard dist ance 1 2.0 9
cement ) f rom Ecoinvent 20 00 syst ems A .,5)
t o plant [ t km]
Sep arat ion o f
A l(OH)4 - Not invent o ried , Energy includ ed in

So lut io n f rom general po wer / heat co nsumpt ion
resid ue solid s
W at er-co oling W at er, co oling ,

EA A 2 00 1 (pers.
f o r p ersip it at ion  Sp ec. W eight of wat er: 99 8 kg / m3 reso urce in wat er unsp ecif ied nat ural 2.4 2E-0 3 m3 1 1.08 (1,1,2,1,1,3 ,4)
communicat ion)
of A l(OH)3 o rig in
Elect rical p ower

used in p lant
(p ro cess, elect ricit y, medium
Includes all p ower consumed (9 3 % EA A 2 00 1 (pers.
light ing, o f f ice  elect ricit y prod uct io n mix No UCTE vo lt ag e, prod uct io n 1.4 0E-0 1 kW h 1 1.08 (1,1,2,1,1,3 ,2)
of t o t al alumina p ro duct ion) communicat ion)
work, ro om UCTE, at grid
coo ling...)
[ kW h]
W at er f o r
Surplus wat er is evap orat ed ,
washing o f W at er, unsp ecif ied EA A 2 00 1 (pers.
 remaining solut ion is reused in reso urce in wat er 1.0 5E-05 m3 1 1.08 (1,1,2,1,1,3 ,4)
persipit at io n & nat ural o rig in communicat ion)
process: St eam is no t invent oried.
resid ue [ kg ]
Heat energ y f o r
evapo rat ion o f Included in general heat

wash wat er consump t io n
[ M J]
Includes all heat f rom co al heat , at hard co al
General heat heat ing EA A 2 00 1 (pers.
 consumed (100 % o f t ot al alumina hard coal No RER ind ust rial f urnace 1- 1.6 3E-0 1 M J 1 1.08 (1,1,2,1,1,3 ,1)
f rom co al syst ems communicat ion)
prod uct io n) 10M W
Includes all heat f rom nat ural g as heat , nat ural gas, at
 consumed (3 6 % of t o t al alumina nat ural gas No RER ind ust rial f urnace 2 .85E-0 1 M J 1 1.08 (1,1,2,1,1,3 ,1)
f rom nat ural g as syst ems communicat ion)
prod uct io n) >100 kW
General heat Includes all heat f rom o il consumed heat ing heat , light f uel o il, at EA A 2 00 1 (pers.
 oil No CH 3.59 E+0 0 M J 1 1.08 (1,1,2,1,1,3 ,1)
f rom o il (6 5 % o f t ot al alumina prod uct io n) syst ems ind ust rial f urnace 1M W communicat ion)
Inf rast ruct ure aluminium hyd ro xide, (5,n.A .,n.A .,n.A .,n.A .,
 Includes land use met als ext ract io n Y es RER 1.6 4E-11 unit calculat ed 1 3.2 3
(p lant ) [ p iece] p lant n.A .,9 ); roug h guess
Fig. 5.6 Flows for "aluminium hydroxide, at plant" and their representation in the ecoinvent database (cont.)


Infra
ture
 wast e heat f rom elect ricit y use air unspecif ied Heat , wast e 5.0 3E-0 1 M J calculat ed 1 1.08 (1,1,2,1,1,3 ,13 )
Dust f ro m
is included in bauxit e at mine --> no t
 bauxit e
aluminium hydroxide, at plant; Outputs handling
invent o ried
Na calculat ed f rom NaOH emission EA A 2 00 1 (pers.

 NaOH in wat er wat er unspecif ied Sod ium, ion 1.2 8E-0 3 kg 1 1.51 (1,1,2,1,1,3 ,33 )
based on st oichio met ry communicat ion)
suspend ed
Suspend ed so lids, EA A 2 00 1 (pers.
 part icles in wat er unspecif ied 1.70E-0 4 kg 1 1.51 (1,1,2,1,1,3 ,33 )
unsp ecif ied communicat ion)
wat er
COD, Chemical Oxyg en EA A 2 00 1 (pers.
 COD in wat er wat er unspecif ied 4.59 E-05 kg 1 1.51 (1,1,2,1,1,3 ,32 )
Demand communicat ion)
calculat ed f rom COD, assuming B OD5, B io log ical
 B OD wat er unspecif ied 4.59 E-05 kg calculat ed 1 1.51 (1,1,2,1,1,3 ,32 )
C6H12 O6 Oxyg en Demand
TOC, To t al Organic
 TOC wat er unspecif ied 1.80 E-05 kg calculat ed 1 1.51 (1,1,2,1,1,3 ,32 )
Carbo n
DOC, Disso lved
 DOC wat er unspecif ied 1.80 E-05 kg calculat ed 1 1.51 (1,1,2,1,1,3 ,32 )
Organic Carbo n
d ispo sal, red mud f ro m
Sand t o wast e resid ual b auxit e d igest ion, 0 % EA A 2 00 1 (pers.
 comp osit ion assumed as red mud No CH 3.3 3E-0 2 kg 1 1.08 (1,1,2,1,1,3 ,6)
Dispo sal [ kg ] management mat erial landf ill wat er, t o residual communicat ion)
mat erial land f ill
solid wast e d ispo sal, inert wast e,
wast e inert mat erial EA A 2 00 1 (pers.
 unspecif ied t o No CH 5% wat er, t o inert 2.0 1E-0 2 kg 1 1.08 (1,1,2,1,1,3 ,6)
management landf ill communicat ion)
Dispo sal [ kg ] mat erial land f ill
d ispo sal, red mud f ro m
B auxit e residue
wast e resid ual b auxit e d igest ion, 0 % EA A 2 00 1 (pers.
 t o Dispo sal Red mud No CH 4 .37E-0 1 kg 1 1.08 (1,1,2,1,1,3 ,6)
management mat erial landf ill wat er, t o residual communicat ion)
[ kg ]
Heat , wast e
reco vered f o r p ro zess --> not
 (recovered)
invent o ried
[ M J]
A luminium
aluminium hyd ro xide, at
 hydroxid e, at chemicals inorganics No RER 1.00 E+0 0 kg
p lant
plant [ kg ]
Fig. 5.6 Flows for "aluminium hydroxide, at plant" and their representation in the ecoinvent database (cont.)


Infra
ture
To t al land use Transf ormat io n, f rom

 reso urce land 5.2 3E+05 m2 est imat ed 1 2.0 3 (1,4 ,1,1,1,4 ,8)
(t ransf ormat io n) unkno wn
aluminium hydroxide, plant; part 1
Land o ccupat ion

f o r b uilding s
assumpt ion: 8 0% of t o t al area is Occup at io n, indust rial
and o pen  reso urce land 2.0 9E+07 m2a est imat ed 1 1.54 (2,4 ,1,1,1,4 ,7)
built up area, built up
bauxit e st o rag e
area
Land
t ransf o rmat ion
t o build ings and  reso urce land 4.18E+05 m2 est imat ed 1 2.0 3 (2,4 ,1,1,1,4 ,8)
ind ust rial area, b uilt up
op en bauxit e
st o rag e area
Land o ccupat ion

f o r veg et at ion  reso urce land 5.23 E+0 6 m2a est imat ed 1 1.54 (2,4 ,1,1,1,4 ,7)
veget at io n area, veget at io n
area
Land
t ransf o rmat ion
 reso urce land ind ust rial area, 1.05E+05 m2 est imat ed 1 2.0 3 (2,4 ,1,1,1,4 ,8)
t o veget at io n
veg et at ion
area
B uilding , mult i const ruct ion
 build ings Y es RER b uilding , mult i-st o rey 2.50 E+0 4 m3 est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
st o ry processes
const ruct ion
B uilding , hall  build ings Y es CH b uilding , hall 5.60 E+0 3 m2 est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
processes
concret e in
const ruct ion co ncret e, normal, at
reinf orced  concret e No CH 3 .02 E+0 3 m3 est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
mat erials p lant
concret e
st eel in
reinf o rcing st eel, at
reinf orced  met als ext ract io n No RER 2 .09 E+0 6 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
p lant
concret e
Fig. 5.7 Flows for "aluminium hydroxide, plant" and their representation in the ecoinvent database


Infra
ture
Ind ust rial machine,

heavy machine, const ruct ion
 machinery Y es RER heavy, unsp ecif ied , at 2.2 0E+05 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
unspecif ic processes
p lant
Po or concret e co ncret e, sole p lat e
const ruct ion
aluminium hydroxide,
f o r ad dit ional  concret e No CH and f o undat ion, at 1.00 E+0 2 m3 est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
mat erials
f o undat ion p lant
const ruct ion
Conveyo r b elt  machinery Y es RER co nveyor belt , at p lant 6 .40 E+0 2 m est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
plant; part 2
processes
Chro mium st eel chromium st eel 18 / 8 , at
 met als ext ract io n No RER 1.8 7E+0 6 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
(st ainless) p lant
St eel low st eel, low-alloyed , at
 met als ext ract io n No RER 2 .46 E+0 4 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
alloyed p lant
reinf orced d ispo sal, build ing,
wast e build ing
 concret e t o No CH reinf o rced concret e, t o 8.73 E+0 6 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
management demo lit io n
disp osal f inal disp osal
d ispo sal, build ing,
f o undat ion
wast e build ing co ncret e, not
 concret e t o No CH 2.2 0E+05 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
management demo lit io n reinf o rced , t o f inal
disp osal
d ispo sal
Part s o f t he aluminium oxyd e plant
aluminium
needed f o r t he hydroxid e aluminium hyd ro xide,
 hydroxid e met als ext ract io n Y es RER 1.00 E+0 0 unit
prod uct io n. Out p ut 122 00 00 00 0 p lant
plant
kg/ a, lif e t ime 50 a
Fig. 5.7 Flows for "aluminium hydroxide, plant" and their representation in the ecoinvent database (cont.)


Infra
ture
A luminiumhydro aluminium hyd ro xide, at

 chemicals inorganics No RER 1.53 E+0 0 kg calculat ed 1 1.08 (1,1,2,1,1,3 ,3)
xide (A l(OH)3 ) p lant
Includes all heat f rom nat ural g as heat , nat ural gas, at
aluminium oxide, at plant
 consumed (6 4 % of t o t al alumina nat ural gas No RER ind ust rial f urnace 5.07E-0 1 M J 1 1.08 (1,1,2,1,1,3 ,1)
f rom nat ural g as syst ems communicat ion)
prod uct io n) >100 kW
General heat Includes all heat f rom o il consumed heat ing heat , light f uel o il, at EA A 2 00 1 (pers.
 oil No CH 1.94 E+0 0 M J 1 1.08 (1,1,2,1,1,3 ,1)
f rom o il (3 5 % o f t ot al alumina prod uct io n) syst ems ind ust rial f urnace 1M W communicat ion)
Elect rical p ower

used in p lant
(p ro cess, elect ricit y, medium
Includes all p ower consumed (7 % EA A 2 00 1 (pers.
light ing, o f f ice  elect ricit y prod uct io n mix No UCTE vo lt ag e, prod uct io n 1.05E-0 2 kW h 1 1.08 (1,1,2,1,1,3 ,2)
of t o t al alumina p ro duct ion) communicat ion)
work, ro om UCTE, at grid
coo ling...)
[ kW h]
Inf rast ruct ure (5,n.A .,n.A .,n.A .,n.A .,

 Includes land use met als ext ract io n Y es RER aluminium o xide, p lant 2 .50E-11 unit est imat ed 1 3.2 3
(p lant ) [ p iece] n.A .,9 ); roug h guess
 wast e heat f rom elect ricit y use air unspecif ied Heat , wast e 3 .79E-0 2 M J calculat ed 1 1.08 (1,1,2,1,1,3 ,13 )
St eam / St eam (0 .530 kg / kg A l2 O3) not

V apo ur invent o ried
A luminium
aluminium o xide, at
 oxid e, at p lant chemicals inorganics No RER 1.00 E+0 0 kg
p lant
[ kg ]
Fig. 5.8 Flows for "aluminium oxide, at plant" and their representation in the ecoinvent database


Infra
ture

 reso urce land 2 .75E+0 4 m2 est imat ed 1 2.0 3 (1,4 ,1,1,1,4 ,8)
Land o ccupat ion

f o r b uilding s
and o pen  reso urce land 1.10 E+0 6 m2a est imat ed 1 1.54 (2,4 ,1,1,1,4 ,7)
bauxit e st o rag e
area
Land
t ransf o rmat ion
t o build ings and  reso urce land 2 .20 E+0 4 m2 est imat ed 1 2.0 3 (2,4 ,1,1,1,4 ,8)
op en bauxit e
aluminium oxide, plant
st o rag e area
Land o ccupat ion

f o r veg et at ion  reso urce land 2.75E+05 m2a est imat ed 1 1.54 (2,4 ,1,1,1,4 ,7)
area
Land
t ransf o rmat ion
 reso urce land ind ust rial area, 5.50 E+0 3 m2 est imat ed 1 2.0 3 (2,4 ,1,1,1,4 ,8)
t o veget at io n
veg et at ion
area
concret e in
const ruct ion co ncret e, normal, at
reinf orced  concret e No CH 1.42 E+0 3 m3 est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
mat erials p lant
concret e
st eel in
reinf orced  met als ext ract io n No RER 9.8 0E+05 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
p lant
concret e
const ruct ion
Conveyo r b elt  machinery Y es RER co nveyor belt , at p lant 2 .00 E+0 2 m est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
processes
Chro mium st eel chromium st eel 18 / 8 , at
 met als ext ract io n No RER 4 .00 E+0 4 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
(st ainless) p lant
St eel low st eel, low-alloyed , at
 met als ext ract io n No RER 6.15E+0 4 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
alloyed p lant
reinf orced d ispo sal, build ing,
wast e build ing
 concret e t o No CH reinf o rced concret e, t o 4 .09 E+0 6 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
management demo lit io n
disp osal f inal disp osal
Only part s o f t he plant t hat are no t
aluminium needed f o r aluminium hyd ro xide.
 met als ext ract io n Y es RER aluminium o xide, p lant 1.00 E+0 0 unit
oxid e plant Out p ut 8 00 00 00 00 kg/ a, lif e t ime
50 a
Fig. 5.9 Flows for "aluminium oxide, plant" and their representation in the ecoinvent database

5.3 Primary aluminium

This part is mainly based on (Hudson et al. (1997)) and (Mori & Adelhardt (1998)).

All commercial production of aluminium today is done in Hall-Héroult cells. Two major types of cells
are currently in use: those employing prebaked carbon anodes and those employing self-baking Söder-
berg anodes. The Hall-Héroult cell with prebaked anodes is shown in Fig. 5.10. Essentially pure alu-
mina is fed into the cryolite base electrolyte. Electric current deposits aluminium into a pool of molten
aluminium held under the electrolyte in the carbon lined cavity of the cell. Oxygen from the alumina
deposits electrolytically onto the carbon anode dipping into the electrolyte and reacts with (burns) the
anode. Cells typically range from 9 to 12 m long, 3 to 4 m wide, and 1 to 1.2 m high. Thermal insula-
tion surrounds the carbon lining of the cell to control heat losses. Although carbon is the material
known to withstand best the combined corrosive action of molten fluorides and molten aluminium,
even carbon would have a very limited life in contact with the electrolyte at the sides of the cell were it
is not protected by a layer of frozen electrolyte. The thermal insulation is adjusted carefully to main-
tain a protective coating on the walls but not on the bottom, which must remain substantially bare for
electrical contact. Steel collector bars in the carbon cathode conduct electric current from the cell.
These bars are inserted into holes that have been sized carefully so that thermal expansion forms a
tight electrical contact, or cemented in place with carbonaceous cement containing metal particles, or
bonded in place with cast iron.
The electrical resistivity of prebaked anodes ranges from 0.005 to 0.006 W cm. Current density at the
anode face is 0.6 – 1.3 A/cm2.
The Söderberg anode uses a premixed "paste" of petroleum coke and coal-tar pitch. This mixture is
added at the top of a rectangular steel casing that is typically 6 – 8 m long, 2 m wide, and 1 m high.
Heat from the electrolyte and heat from the electric current passing through the anode bakes the car-
bonaceous mix as it progresses through the casing.
The baked portion extends past the casing and into the molten electrolyte. Baked mix replaces anode
being consumed at the bottom surface. Electric current enters the anode through either vertical or slop-
ing steel spikes. These spikes are pulled and reset to a higher level as they approach the lower surface.
Söderberg anodes have an electrical resistivity about 30 % higher than that of prebaked anodes. Be-
cause of the resulting lower power efficiency and the greater difficulty in collecting and disposing of
baking fumes, Söderberg anodes are being replaced by prebaked anodes, even though the former save
the capital, labour, and energy required to manufacture the latter.

Fig. 5.10 Hall-Héroult cell with prebaked anode: a) carbon anode b) electrolyte, c) insulation, d) carbon lining, e) cur-
rent collector bar, f) thermal insulation, g) steel shell, h) carbon block, i) ledge, j) crust, k) aluminium oxide
cover, l) removable covers, m) anode rods, n) fume collection, o) air cylinder, p) feeder, q) current supply, r)
crust breaker
Metal tapped from the Hall-Héroult cell may be treated in one of several ways. First, it may be trans-
ported in the molten state to a holding furnace, where, along with solid scrap, it is alloyed, fluxed, fil-
tered, and cast into a form suitable for fabrication by rolling, forging, etc., or for remelting by a foun-
dry. It may be transported further (over distances up to 200 km) on trucks, equipped with well-
insulated containers to prevent freezing, to a foundry where it is alloyed and used to make shaped cast-
ings. Both these methods of utilizing the molten metal are favoured because energy required for
remelting is conserved.
Alternatively, where ingot casting facilities or foundries are not nearby, metal is cast using cast iron
moulds into "primary" ingots for remelting. Generally, these unalloyed remelt ingots come as stack-
able 22-, 340-, or 680-kg ingots. Casting machines are often fully automated, with production rates of
up to 20000 kg/h.
Infrastructure is estimated based on information of an average sized electrolysis plant with cast house
in Norway. Production capacity is about 130'000 t/year. About 3'000 tons is secondary material (for
alloying) that can be neglected for our purpose. The aluminium is cast in 7 meter long billets with a
diameter that varies between 178 and 314 mm. Later the billets are sawed, packed and driven to the
quay for shipment. The most important data about the plant are given in Tab. 5.6.

10
Tab. 5.6 Infrastructure data for a medium sized electrolysis plant with cast house (Soral)
Production Capacity 130.000 tons

Area: 976,000 m2
Length -Pot-rooms 857 m
Breadth- Pot-rooms 23 m
Electrolysis pots 320
Cast -house 12.200 m²
Height- Silos 55 m and 62 m
30.000 tons and
Capacity-Silos
40.000 tons
Length-Conveyor Belt 900 m
Length-Quay 240 m
Capacity-Quay 70.000 tdw
Storage house 9500 m²
5.3.2 Emissions
Air
Relevant direct air emissions from the anode production (baking) and from the electrolysis itself are
CO, CO2, fluorides, dust, polycyclic aromatic hydrocarbons, Nitrogen oxides and sulphur oxides. The
casting is responsible for some direct fluoride emissions. The other processes don't emit to air directly.
Water
There are no relevant direct water emissions from these processes
5.3.3 Waste
Several different waste streams are generated. Some of them are recycled or reused and thus not inven-
toried. The others are deposited in landfills. (See Fig. 5.11, Fig. 5.12, Fig. 5.13 and Fig. 5.14)
Anode production waste
In the anode production carbon and tar wastes are produced. The compositions are unknown. 100%
carbon is assumed for tar. A certain amount of waste might be recycled. However, since no data on the
share to recycling is available, disposal in a residual material landfill is assumed.
The "carbon waste" refers to the waste of the anode butt recycling. Thus it is inventoried the same way
as the carbon fraction of spent pot lining.
Cathode production waste
Because the cathode production includes the disposal of the old cathode, the waste of this process is
basically the spent pot lining. It is separated into a carbon fraction (1st cut) and refractory (brick) frac-
tion (2nd cut). A certain amount of these wastes might be recycled. However, since no data on the share
to recycling is available, both wastes are assumed to be landfilled in a residual material landfill. The
ranges for their chemical composition that were presented by R. P. Pawlek at the 10th international Al
symposium in 1999 11 and the values used in this project are given in Tab. 5.7.
Electrolysis waste
The carbon waste of the electrolysis is basically the waste fraction from used anodes that can't be re-
cycled. It is inventoried as the carbon fraction of SPL.
10
http://www.soral.no/en/produksjon.html, accesed 16.12.2002
11
http://www.aluminium.sk/prednasky/32.htm, accessed 19.12.2002

Aluminium casting waste

The fibre and refractory wastes of the casting process are inventoried as inert material to landfill. The
dross composition given in
Tab. 5.8 stems from the environment agency of the UK 12 . Disposal in a residual material landfill is as-
sumed.
Data for the filter dust composition are given in Tab. 5.9. Disposal in a residual material landfill is as-
sumed.
Tab. 5.7 Composition of spent pot liner
Element or compound unit Carbon Refractory

(Metals mostly as oxides)
Pawlek This project Pawlek This project
11 11
SiO2 % 0–6 2.96 10 – 50 36.94

Al2O3 % 0 – 10 4.94 10 – 50 36.94
metal Al % 0–5 2.47 0 0.00
Fe2O3 % 0.00 0–3 1.80
CaO % 1–6 3.46 1–8 5.49
K2O % <0.5 0.49 <0.5 0.30
MnO % 0 – 0.02 0.01 0 – 0.02 0.01
TiO2 % 0.00 0 – 0.1 0.06
P2O5 g/t 0 – 650 0.03 0 – 300 0.02
Na % 8 – 12 9.88 6 – 10 9.79
F % 6 – 10 7.90 4 – 10 8.59
Cl % <0.06 0.06 <0.06 0.04
CN free g/t 0 – 2,000 0.10 0 – 500 0.00
CN total g/t 0 – 5,000 2.47 0 – 1,000 0.06
Carbon % - 66.46 0 0.00
12
Tab. 5.8 Composition of dross
Compound Unit Litera- This project

ture12
Aluminium % (w/w) 8-50 30
Aluminium oxide % (w/w) 30-90 61
Aluminium nitride % (w/w) 0-10 6
Aluminium carbide % (w/w) 0-5 3
12
http://www.environment-agency.gov.uk/commondata/105385/swen042.pdf?lang=_e, accessed 19.12.2002

13
Tab. 5.9 Composition of filter dust
Element Amount [% w/w] Element Amount [% w/w]

Al 33.820 Cu 0.057
Na 6.308 Ni 0.051
F 9.051 Cr 0.005
C 11.586 Pb 0.039
Fe 0.545 Zn 0.003
Si 0.046 V 0.048
Ti 0.039 P 0.058
Mg 0.090 O 38.224
Ga 0.030
5.3.4 Primary aluminium in ecoinvent

The available information about the infrastructure needed for aluminium production is non-
quantitative. Tab, Tab. 5.11 and Tab. 5.12 show the details of the assumptions made to estimate the
infrastructure expenditures.
Tab. 5.10 Estimated infrastructure for anode plant
anode, plant Demand Unit Life Life time

time demand
Area of plant (assumption based on casting plant) 300'308 m2 50 300'308
Production hall 7'508 m2 50 7'508
Production machines 615'385 kg 25 1'230'769
Storage hall 5'846 m2 50 5'846
Tab. 5.11 Estimated infrastructure for aluminium electrolysis plant
aluminium electrolysis, plant Demand Unit Life Life time

time demand
Area of plant (50% of total, rest casting plant) 488'000 m2 50 488'000
Concrete for piers [m3] 1'200 m3 50 1'200
Grab for unloading [kg] 100'000 kg 25 200'000
conveyor to storage [m] 900 m 25 1'800
Silo (30000 m3, height 55 m, radius 13m) 194'402 kg 25 388'804
Silo (40000 m3, height 62 m, radius 14m) 233'986 kg 25 467'972
Area of pot rooms (halls) 19'711 m2 50 19'711
Steel shell for 320 pots (10 by 3.5 by 1.1 m) 1'116'640 kg 25 2'233'280
13
http://sperber.lih.rwth-aachen.de/sfb/subpro/sub_pro2/entsorg/filt1_de.htm, accessed 19.12.2002

Tab. 5.12 Estimated infrastructure for aluminium casting plant
aluminium casting, plant Demand Unit Life Life time

time demand
Area of plant (50% of total, rest electrolysis plant) 488'000 m2 50 488'000
Cast house 12'200 m2 50 12'200
Casting machines 1'000'000 kg 25 2'000'000
Storage house 9'500 m2 50 9'500
The other background information to the processes is given in Tab. 5.13, flows and their representa-
tion in the database is shown in:
- anode, aluminium electrolysis Fig. 5.11
- anode, plant Fig. 5.12
- cathode, aluminium electrolysis Fig. 5.13
- aluminium, primary, liquid, at plant Fig. 5.14
- aluminium electrolysis, plant Fig. 5.15
- aluminium, primary, at plant Fig. 5.16
- aluminium casting, plant Fig. 5.17
Data quality
Data for the steps in the primary aluminium production are of good overall quality. The infrastructure
is roughly estimated but compared to the other activities the influence is negligible. The values for the
processes stem from a recent and comprehensive source (EAA (2000)).

Tab. 5.13 Meta information for the aluminium electrolysis and casting processes
anode, aluminium cathode, aluminium electro-

Name
electrolysis anode, plant lysis
Location RER RER RER
Unit kg unit kg
Includes cathode production and
removal, transports of materials to
the plant and the disposal of the
wastes. Because the cathode is
Includes anode pro- produced in the electrolysis plant,
Included Processes
duction (with its no cathode production plant is bal-
plant), transports of Production hall, machin- anced. No data have been found
materials to the plant ery and administrative on air emissions from baking. It is
and the disposal of buildings (including dis- assumed that these emissions are
the wastes. posal) included in the electrolysis.
Amount 1 1 1
Anode, Aluminium
Local Name
Elektrolyse Anode, Fabrik Kathode, Aluminium Elektrolyse
Synonyms
Medium sized plant.
Output: 80'000 t/a, life
General Comment to
time: 50 years. Assump-
reference function
tion is based on data for
casting plant.
Start Date 1995 2002 1995
End Date 2002 2002 2002
Period 1 1 1
Some Swiss data- Some Swiss datasets
Geography text sets are used for are used for European Some Swiss datasets are used for
European processes. processes. European processes.
Average technology
for the aluminium
Technology text consumed in Europe Average technology for the alumin-
(85% prebaked, 15% ium consumed in Europe (85% pre
Söderberg Assumed technology baked, 15% Söderberg
Representativeness [%] 98 98
Production Volume unknown unknown unknown
Sampling Procedure Literature / Survey Internet search Literature / Survey
Extrapolations see Geography see Geography see Geography
Uncertainty is ad- Uncertainty is adjusted
justed for geographi- for geographical mis- Uncertainty is adjusted for geo-
Uncertainty Adjustments
cal mismatch (using match (using the pedi- graphical mismatch (using the pedi-
the pedigree matrix). gree matrix). gree matrix).

Tab. 5.13 Meta information for the aluminium electrolysis and casting processes (cont.)
aluminium, pri-
Name mary, liquid, at aluminium e- aluminium, pri- aluminium
plant lectrolysis, plant mary, at plant casting, plant
Location RER RER RER RER
Infrastructure Process 0 1 0 1
Unit kg unit kg unit
Data Set Version 2.0 2.0 2.0 2.0
Includes liquid alu- Includes cast alu-
minium production minium ingot pro-
(with its plant), Port, production duction (with its
Included Processes transports of mate- hall, machinery and plant), transports of Production hall, ma-
rials to the plant and administrative build- materials to the chinery and admin-
the disposal of the ings (including dis- plant and the dis- istrative buildings
wastes. posal) posal of the wastes. (including disposal)
Amount 1 1 1 1
Aluminium, primär, Aluminiumelektro- Aluminium, primär, Aluminium giessen,
Local Name
flüssig, ab Werk lyse, Werk ab Werk Werk
Synonyms
Medium sized plant.
Medium sized plant. Output: 130'000 t/a,
Output: 130'000 t/a, life time: 50 years.
life time: 50 years. Site is shared with
General Comment to
Site is shared with electrolysis plant.
reference function
casting plant. Total Total land use is
land use is allocated allocated 50% to
50% to electrolysis, electrolysis, 50% to
50% to casting. casting.
Start Date 1995 2002 1995 2002
End Date 2002 2002 2002 2002
Period 1 1 1 1
Some Swiss data- Some Swiss data- Some Swiss data- Some Swiss data-
sets are used for sets are used for sets are used for sets are used for
Geography text
European proc- European proc- European proc- European proc-
esses. esses. esses. esses.
Average technology Average technology
for the aluminium Technology of one for the aluminium Technology of one
Technology text
consumed in medium sized plant consumed in medium sized plant
Europe in Norway Europe in Norway
Production Volume unknown unknown unknown unknown
Sampling Procedure Literature / Survey Internet search Literature / Survey Internet search
Extrapolations see Geography see Geography see Geography see Geography
Uncertainty is ad- Uncertainty is ad- Uncertainty is ad- Uncertainty is ad-
justed for geo- justed for geo- justed for geo- justed for geo-
Uncertainty Adjustments graphical mismatch graphical mismatch graphical mismatch graphical mismatch
(using the pedigree (using the pedigree (using the pedigree (using the pedigree
matrix). matrix). matrix). matrix).


Infra
ture
EA A 2 00 0 (p. 4 9) /
p et roleum co ke, at personal
Pet ro l coke [ kg]  oil f uels No RER 6.4 9E-0 1 kg 1 1.08 (1,1,2,1,1,3 ,4)
ref inery communicat ion B . d e
Gelas, EA A
Transpo rt of
pet ro l coke t o t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
 Transpo rt by road ro ad No RER t ransp ort , lo rry 32 t 6.4 9E-0 2 t km st and ard dist ance 1 2.0 9
anod e plant syst ems A .,5)
anode, aluminium electrolysis; part 1
[ t km]
Transpo rt of
pet ro l coke t o t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
 Transpo rt by rail t rain No RER t ransp ort , f reig ht , rail 1.3 0E-0 1 t km st and ard dist ance 1 2.0 9
anod e plant syst ems A .,5)
[ t km]
EA A 2 00 0 (p. 4 9) /
personal
Pit ch [ kg ]  invent o ried as bit umen oil f uels No RER b it umen, at ref inery 1.6 4E-0 1 kg 1 1.08 (1,1,2,1,1,3 ,4)
communicat ion B . d e
Gelas, EA A
Transpo rt of
t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
pit ch t o anod e  Transpo rt by road ro ad No RER t ransp ort , lo rry 32 t 1.6 4E-0 2 t km st and ard dist ance 1 2.0 9
syst ems A .,5)
plant [ t km]
Transpo rt of
pit ch t o anod e  Transpo rt by rail t rain No RER t ransp ort , f reig ht , rail 3.2 8E-0 2 t km st and ard dist ance 1 2.0 9
syst ems A .,5)
plant [ t km]
personal
recycling of used b ut t s --> not
anod e but t s  1.8 8E-0 1 communicat ion B . d e
invent o ried
Gelas, EA A
cast iron f or Repo rt et und er " Elect rolysis" b ut EA A 2 00 1 (pers.
 met als ext ract io n No RER cast iro n, at p lant 7.8 1E-0 3 kg 1 1.08 (1,1,2,1,1,3 ,4)
st ub [ kg ] used in A no des communicat ion)
Transpo rt of
iro n t o anod e  Transpo rt by road ro ad No RER t ransp ort , lo rry 32 t 7.8 1E-0 4 t km st and ard dist ance 1 2.0 9
syst ems A .,5)
plant [ t km]
Transpo rt of
iro n t o anod e  Transpo rt by rail t rain No RER t ransp ort , f reig ht , rail 1.56E-0 3 t km st and ard dist ance 1 2.0 9
syst ems A .,5)
plant [ t km]
Ref ract o ry const ruct ion ref ract o ry, f ireclay, EA A 2 00 1 (pers.
 b ricks No DE 1.0 1E-0 2 kg 1 1.08 (1,1,2,1,1,3 ,4)
mat erials mat erials p acked, at plant communicat ion)
Transpo rt of
Ref ract o ry  Transpo rt by road ro ad No RER t ransp ort , lo rry 32 t 1.0 1E-0 3 t km st and ard dist ance 1 2.0 9
syst ems A .,5)
mat erials
Transpo rt of
Ref ract o ry  Transpo rt by rail t rain No RER t ransp ort , f reig ht , rail 2.0 2E-0 3 t km st and ard dist ance 1 2.0 9
syst ems A .,5)
mat erials
Fig. 5.11 Flows for "anode, aluminium electrolysis" and their representation in the ecoinvent database


Infra
ture
heat , nat ural gas, at

General heat Calcinat io n of pet ro l coke, b aking heat ing EA A 2 00 1 (pers.
 nat ural gas No RER ind ust rial f urnace 1.90 E+0 0 M J 1 1.08 (1,1,2,1,1,3 ,1)
f rom nat ural g as of anod e syst ems communicat ion)
>100 kW
General heat Calcinat io n of pet ro l coke, b aking heat ing heat , light f uel o il, at EA A 2 00 1 (pers.
 oil No CH 9 .35E-0 1 M J 1 1.08 (1,1,2,1,1,3 ,1)
f rom o il of anod e syst ems ind ust rial f urnace 1M W communicat ion)
anode, aluminium electrolysis; part 2
W at er, co oling ,
Coo ling wat er EA A 2 00 1 (pers.
 reso urce in wat er unsp ecif ied nat ural 4.3 1E-0 3 m3 1 1.08 (1,1,2,1,1,3 ,4)
[ m3 ] communicat ion)
o rig in
Elect rical p ower elect ricit y, medium

EA A 2 00 1 (pers.
used in p lant  generic dat a f rom Ecoinvent 20 00 elect ricit y prod uct io n mix No UCTE vo lt ag e, prod uct io n 1.3 1E-0 1 kW h 1 1.08 (1,1,2,1,1,3 ,2)
communicat ion)
(p ro cess) [ kW h] UCTE, at grid
Inf rast ruct ure (5,n.A .,n.A .,n.A .,n.A .,

 met als ext ract io n Y es RER ano de, p lant 2 .50E-10 unit est imat ed 1 3.2 3
(p lant ) [ p iece] n.A .,9 ); roug h guess
 wast e heat f rom elect ricit y use air unspecif ied Heat , wast e 4 .72E-0 1 M J calculat ed 1 1.08 (1,1,2,1,1,3 ,13 )
B aP (B enzo-a- EA A 2 00 1 (pers.
 air unspecif ied B enzo (a)pyrene 3.18E-0 6 kg 1 3.0 0 (1,1,2,1,1,3 ,21)
Pyrene) t o air communicat ion)
Carbo n mono xide, EA A 2 00 1 (pers.
 CO t o air air unspecif ied 1.0 4E-0 3 kg 1 5.01 (1,1,2,1,1,3 ,17)
f ossil communicat ion)
EA A 2 00 1 (pers.
 CO2 t o air air unspecif ied Carbo n dio xide, f ossil 2 .53E-0 1 kg 1 1.08 (1,1,2,1,1,3 ,24 )
communicat ion)
t o t al EA A 20 01
(p ers.
 Dust t o air 42 % > 10 um air unspecif ied Part iculat es, > 10 um 1.3 0E-0 4 kg communicat ion), 1 1.51 (1,1,2,1,1,3 ,25)
f ract io n
www.iiasa.ac.at
t o t al EA A 20 01
(p ers.
Part iculat es, > 2 .5 um,
 Dust t o air 10 um > 30 % > 2 .5 um air unspecif ied 9 .30 E-05 kg communicat ion), 1 2.0 0 (1,1,2,1,1,3 ,26 )
and < 10 um
f ract io n
www.iiasa.ac.at
t o t al EA A 20 01
(p ers.
 Dust t o air 28 % < 2 .5 um air unspecif ied Part iculat es, < 2 .5 um 8 .68 E-05 kg communicat ion), 1 3.0 0 (1,1,2,1,1,3 ,27)
f ract io n
www.iiasa.ac.at
Fig. 5.11 Flows for "anode, aluminium electrolysis" and their representation in the ecoinvent database (cont.)


Infra
ture
f luo rid e
EA A 2 00 1 (pers.
 (g aseous) t o invent o ried as HF air unspecif ied Hydrog en f luo rid e 9 .00 E-05 kg 1 1.51 (1,1,2,1,1,3 ,31)
communicat ion)
air
anode, aluminium electrolysis; part 3 

f luo rid e (no n
gaseo us) t o air
invent o ried as part iculat e emission air unspecif ied
Part iculat es, > 2 .5 um,
and < 10 um
5.00 E-05 kg
EA A 2 00 1 (pers.
communicat ion)
1 2.0 0 (1,1,2,1,1,3 ,26 )
po lycyclic
PA H, p olycyclic EA A 2 00 1 (pers.
 aro mat ic HC t o wit ho ut B aP air unspecif ied 9 .48 E-05 kg 1 3.0 0 (1,1,2,1,1,3 ,21)
aromat ic hydrocarbo ns communicat ion)
air
EA A 2 00 1 (pers.
 NOx t o air air unspecif ied Nit ro gen o xides 1.9 0E-0 4 kg 1 1.51 (1,1,2,1,1,3 ,31)
communicat ion)
EA A 2 00 1 (pers.
 SO2 t o air air unspecif ied Sulf ur dio xide 8.4 0E-0 4 kg 1 1.08 (1,1,2,1,1,3 ,15)
communicat ion)
carb on mat erial EA A 2 00 1 (pers.
 not invent o ried 2 .65E-0 2 kg
f o r recycling communicat ion)
d ispo sal, ref ract ory

wast e resid ual SPL, A l elec.lysis, 0 % EA A 2 00 1 (pers.
 carb on wast e t o landf ill, non-hazard ous No CH 3.9 0E-0 3 kg 1 1.08 (1,1,2,1,1,3 ,6)
d ispo sal, inert wast e,
ref ract ory wast e inert mat erial EA A 2 00 1 (pers.
 t o landf ill, non-hazard ous No CH 5% wat er, t o inert 1.05E-0 2 kg 1 1.08 (1,1,2,1,1,3 ,6)
wast e management landf ill communicat ion)
d ispo sal, asphalt , 0.1%
wast e EA A 2 00 1 (pers.
 t ar wast e t o landf ill, non-hazard ous sanit ary landf ill No CH wat er, t o sanit ary 7.0 0E-0 4 kg 1 1.08 (1,1,2,1,1,3 ,6)
management communicat ion)
land f ill
 W at er / st eam not invent o ried
A nod e f o r
A luminium ano de, aluminium
 met als ext ract io n No RER 1.00 E+0 0 kg
elect rolysis elect ro lysis
[ kg ]
Fig. 5.11 Flows for "anode, aluminium electrolysis" and their representation in the ecoinvent database (cont.)


Infra
ture
To t al land use Transf ormat io n, f rom (5,n.A .,n.A .,n.A .,n.A .,

 reso urce land 3.0 0E+05 m2 est imat ed 1 2.2 3
(t ransf ormat io n) unkno wn n.A .,8 )
Land o ccupat ion

f o r b uilding s
assumpt ion: 8 0% of t o t al area is Occup at io n, indust rial (5,n.A .,n.A .,n.A .,n.A .,
and o pen  reso urce land 1.2 0E+07 m2a est imat ed 1 1.77
built up area, built up n.A .,7)
bauxit e st o rag e
area
Land
t ransf o rmat ion
Transf ormat io n, t o (5,n.A .,n.A .,n.A .,n.A .,
t o build ings and  reso urce land 2.4 0E+05 m2 est imat ed 1 2.2 3
ind ust rial area, b uilt up n.A .,8 )
op en bauxit e
st o rag e area
anode, plant
Land o ccupat ion

f o r veg et at ion  reso urce land 3 .00 E+0 6 m2a est imat ed 1 1.77
veget at io n area, veget at io n n.A .,7)
area
Land
t ransf o rmat ion (5,n.A .,n.A .,n.A .,n.A .,
 reso urce land ind ust rial area, 6 .01E+0 4 m2 est imat ed 1 2.2 3
t o veget at io n n.A .,8 )
veg et at ion
area
B uilding , mult i const ruct ion (5,n.A .,n.A .,n.A .,n.A .,
 build ings Y es RER b uilding , mult i-st o rey 1.00 E+0 4 m3 est imat ed 1 3.2 3
st o ry processes n.A .,9 )
B uilding , hall  build ings Y es CH b uilding , hall 1.34 E+0 4 m2 est imat ed 1 3.2 3
processes n.A .,9 )
heavy machine, const ruct ion (5,n.A .,n.A .,n.A .,n.A .,
 machinery Y es RER heavy, unsp ecif ied , at 1.23 E+0 6 kg est imat ed 1 3.2 3
unspecif ic processes n.A .,9 )
p lant
f o r ad dit ional  concret e No CH and f o undat ion, at 5.59 E+0 2 m3 est imat ed 1 3.2 3
mat erials n.A .,9 )
f o undat ion
wast e build ing co ncret e, not (5,n.A .,n.A .,n.A .,n.A .,
 concret e t o No CH 1.23 E+0 6 kg est imat ed 1 3.2 3
management demo lit io n reinf o rced , t o f inal n.A .,9 )
disp osal
d ispo sal
Out p ut : 8 00 00 00 0 kg / a, lif e t ime:
 anod e plant met als ext ract io n Y es RER ano de, p lant 1.00 E+0 0 unit
50 a
Fig. 5.12 Flows for "anode, plant" and their representation in the ecoinvent database


Infra
ture
Carb on b locks
 generic dat a f rom Ecoinvent 20 00 hard coal f uels No RER hard co al coke, at plant 1.29 E+0 1 M J EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,4)
[ kg ]
Transpo rt of
carb on b locks t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
 Transpo rt by road ro ad No RER t ransp ort , lo rry 32 t 1.29 E+0 0 t km st and ard dist ance 1 2.0 9
t o elect rolysis cathode, aluminium electrolysis; Inputs part 1 syst ems A .,5)
plant [ t km]
Transpo rt of
carb on b locks t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
 Transpo rt by rail t rain No RER t ransp ort , f reig ht , rail 2.58 E+0 0 t km st and ard dist ance 1 2.0 9
t o elect rolysis syst ems A .,5)
plant [ t km]
cast iron f or
EA A 2 00 1 (pers.
elect rical  met als ext ract io n No RER cast iro n, at p lant 4.4 4E-0 2 kg 1 1.08 (1,1,2,1,1,3 ,4)
communicat ion)
cont act s [ kg]
Transpo rt of
iro n t o t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
elect rolysis syst ems A .,5)
plant [ t km]
Transpo rt of
iro n t o t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
plant [ t km]
Ref ract o ry const ruct ion ref ract o ry, f ireclay, EA A 2 00 1 (pers.
 bricks No DE 4 .78E-0 1 kg 1 1.08 (1,1,2,1,1,3 ,4)
mat erials [ kg ] mat erials p acked, at plant communicat ion)
Transpo rt of
ref ract ory
mat erials t o  Transpo rt by road ro ad No RER t ransp ort , lo rry 32 t 4 .78E-0 2 t km st and ard dist ance 1 2.0 9
syst ems A .,5)
elect rolysis
plant [ t km]
Transpo rt of
ref ract ory
mat erials t o  Transpo rt by rail t rain No RER t ransp ort , f reig ht , rail 9.57E-0 2 t km st and ard dist ance 1 2.0 9
syst ems A .,5)
elect rolysis
plant [ t km]
Collar/ ramming EA A 2 00 1 (pers.
 invent o ried as bit umen oil f uels No RER b it umen, at ref inery 1.9 4E-0 1 kg 1 1.08 (1,1,2,1,1,3 ,4)
past e [ kg ] communicat ion)
Transpo rt of
past e t o t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
p lant [ t km]
Transpo rt of
past e t o t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
plant [ t km]
Fig. 5.13 Flows for "cathode, aluminium electrolysis" and their representation in the ecoinvent database


Infra
Pro c e ss Cate Sub Lo c a Mo dul name in Me an So urc e mean Ty StDv Gene ral
Input Output Remark s struc Unit
Name go ry c ate go ry tio n ec o invent value value pe 95% Co mment
ture
A luminium o xid e aluminium o xid e, at EA A 2 0 0 1 (p ers.

 chemicals ino rg anics No RER 2 .2 2 E-0 2 kg 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
[ kg ] p lant co mmunicat io n)
Transp o rt o f EA A 2 0 0 0 (p . 50 ) /
A l2 O3 t o t ransp o rt p erso nal
 Transp o rt b y ro ad ro ad No RER t ransp o rt , lo rry 3 2 t 1.50 E-0 2 t km 1 2 .0 0 (1,1,2 ,1,1,3 ,5)
elect ro lysis syst ems co mmunicat io n B . d e
p lant [ t km] cathode, aluminium electrolysis; Inputs part 2 Gelas, EA A
 Transp o rt b y rail t rain No RER t ransp o rt , f reig ht , rail 4 .2 0 E-0 2 t km 1 2 .0 0 (1,1,2 ,1,1,3 ,5)
p lant [ t km] Gelas, EA A
A l2 O3 t o t ransp o rt t ransp o rt , t ranso ceanic p erso nal
 Transp o rt b y ship (o cean) ship No OCE 3 .74 E+0 0 t km 1 2 .0 0 (1,1,2 ,1,1,3 ,5)
elect ro lysis syst ems f reig ht ship co mmunicat io n B . d e
 Transp o rt b y ship (co st al) ship No RER t ransp o rt , b arg e 2 .0 4 E-0 1 t km 1 2 .0 0 (1,1,2 ,1,1,3 ,5)
co nst ruct io n civil
B last ing [ kg ]  No RER b last ing 2 .2 2 E-0 2 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,1)
p ro cesses eng ineering
Transp o rt o f
t ransp o rt (4 ,5,n.A .,n.A .,n.A .,n.
b last ing  Transp o rt b y ro ad ro ad No RER t ransp o rt , lo rry 3 2 t 2 .2 2 E-0 3 t km st and ard d ist ance 1 2 .0 9
syst ems A .,5)
mat erials [ t km]
Transp o rt o f
t ransp o rt (4 ,5,n.A .,n.A .,n.A .,n.
b last ing  Transp o rt b y rail t rain No RER t ransp o rt , f reig ht , rail 4 .4 4 E-0 3 t km st and ard d ist ance 1 2 .0 9
syst ems A .,5)
mat erials [ t km]
St eel b ars
(elect rical  met als ext ract io n No RER 2 .8 3 E-0 1 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
p lant
co nt act s) [ kg ]
Transp o rt o f t ransp o rt (4 ,5,n.A .,n.A .,n.A .,n.
 Transp o rt b y ro ad ro ad No RER t ransp o rt , lo rry 3 2 t 2 .8 3 E-0 2 t km st and ard d ist ance 1 2 .0 9
st eel b ars [ t km] syst ems A .,5)
Transp o rt o f t ransp o rt (4 ,5,n.A .,n.A .,n.A .,n.
 Transp o rt b y rail t rain No RER t ransp o rt , f reig ht , rail 5.6 7E-0 2 t km st and ard d ist ance 1 2 .0 9
st eel b ars [ t km] syst ems A .,5)
heat , nat ural g as, at
General heat heat ing
 nat ural g as No RER ind ust rial f urnace 1.4 5E+0 0 M J EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,1)
f ro m nat ural g as syst ems
>10 0 kW
General heat heat ing heat , lig ht f uel o il, at

 o il No CH 2 .78 E-0 1 M J EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,1)
f ro m o il syst ems ind ust rial f urnace 1M W
Elect rical p o wer elect ricit y, med ium

used in p lant  g eneric d at a f ro m Eco invent 2 0 0 0 elect ricit y p ro d uct io n mix No UCTE vo lt ag e, p ro d uct io n 2 .2 2 E-0 1 kW h EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,2 )
(p ro cess) [ kW h] UCTE, at g rid
Inf rast ruct ure

 Includ ed in Elect ro lysis p lant
(p lant ) [ p iece]
Fig. 5.13 Flows for "cathode, aluminium electrolysis" and their representation in the ecoinvent database (cont.)


Infra
ture
 wast e heat f rom elect ricit y use air unspecif ied Heat , wast e 8.0 1E-0 1 M J calculat ed 1 1.08 (1,1,2,1,1,3 ,13 )
Direct
emissions o f
cat ho de EA A 2 00 1 (pers.
 invent o ried wit h elect ro lysis
prod uct io n communicat ion)
cathode, aluminium electrolysis;
(CO2 , CO,
SOx, ...) ??
A l2O3 resid ue EA A 2 00 1 (pers.
 reuse --> no t invent oried 2.2 1E-0 2 kg
[ kg ] communicat ion)
So lid wast e
 unspecif ied t o landf ill (co nt rolled ) No CH 5.52E-0 3 kg 1 1.08 (1,1,2,1,1,3 ,6)
[ kg ]
Outputs
Carb on wast e wast e resid ual SPL, A l elec.lysis, 0 % EA A 2 00 1 (pers.

 t o landf ill, non-hazard ous No CH 4 .25E-0 1 kg 1 1.08 (1,1,2,1,1,3 ,6)
[ kg ] management mat erial landf ill wat er, t o residual communicat ion)
Carb on wast e EA A 2 00 1 (pers.
 reuse as f uel --> not invent o ried 3 .65E-0 1 kg
St eel scrap EA A 2 00 1 (pers.
 recycling --> not blanced 2 .87E-0 1 kg
ref ract ory wast e resid ual SPL, A l elec.lysis, 0 % EA A 2 00 1 (pers.
 t o landf ill, non-hazard ous No CH 8.4 0E-0 1 kg 1 1.08 (1,1,2,1,1,3 ,6)
mat erials [ kg ] management mat erial landf ill wat er, t o residual communicat ion)
ref ract ory EA A 2 00 1 (pers.
 reuse --> no t invent oried 9.3 9E-0 2 kg
mat erials [ kg ] communicat ion)
Heat , wast e
 reco vered --> not invent o ried
[ M J]
Cat ho de f or
A luminium cat hod e, aluminium
elect rolysis elect ro lysis
[ kg ]
Fig. 5.13 Flows for "cathode, aluminium electrolysis" and their representation in the ecoinvent database (cont.)


Infra
Pro c e ss Cate Sub Lo c a Mo dul name in Me an So urc e mean Ty StDv Gene ral
Name go ry c ate go ry tio n ec o invent value value pe 95% Co mment
ture
A luminium
aluminium o xid e, at
o xid e, at p lant  chemicals ino rg anics No RER 1.9 2 E+0 0 kg EA A 2 0 0 0 (p . 4 9 ) 1 1.0 8 (1,1,2 ,1,1,3 ,3 )
p lant
[ kg ]
A l2 O3 f ro m t ransp o rt t ransp o rt , t ranso ceanic p erso nal
 Transp o rt b y ship (o cean) ship No OCE 3 .74 E+0 0 t km 1 2 .0 0 (1,1,2 ,1,1,3 ,5)
p lant t o p lant syst ems f reig ht ship co mmunicat io n B . d e
[ t km] Gelas, EA A
aluminium, primary, liquid, at plant; part 1
A l2 O3 f ro m t ransp o rt p erso nal
p lant t o p lant syst ems co mmunicat io n B . d e
[ t km] Gelas, EA A
 Transp o rt b y Ro ad ro ad No RER t ransp o rt , lo rry 3 2 t 1.50 E-0 2 t km 1 2 .0 0 (1,1,2 ,1,1,3 ,5)
[ t km] Gelas, EA A
 Transp o rt b y Rail t rain No RER t ransp o rt , f reig ht , rail 4 .2 0 E-0 2 t km 1 2 .0 0 (1,1,2 ,1,1,3 ,5)
[ t km] Gelas, EA A
A no d e f o r
ano d e, aluminium
A luminium  met als ext ract io n No RER 4 .4 8 E-0 1 kg EA A 2 0 0 0 (p . 4 9 ) 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
elect ro lysis
elect ro lysis [ kg ]
Transp o rt o f
ano d e t o t ransp o rt t ransp o rt , t ranso ceanic EA A 2 0 0 1 (p ers.
 Transp o rt b y ship (o cean) ship No OCE 6 .4 0 E-0 2 t km 1 2 .0 0 (1,1,2 ,1,1,3 ,5)
elect ro lysis syst ems f reig ht ship co mmunicat io n)
p lant [ t km]
Transp o rt o f
ano d e t o t ransp o rt EA A 2 0 0 1 (p ers.
elect ro lysis syst ems co mmunicat io n)
p lant [ t km]
Transp o rt o f
 Transp o rt b y Ro ad ro ad No RER t ransp o rt , lo rry 3 2 t 1.6 1E-0 2 t km 1 2 .0 0 (1,1,2 ,1,1,3 ,5)
p lant [ t km]
Transp o rt o f
 Transp o rt b y Rail t rain No RER t ransp o rt , f reig ht , rail 1.6 1E-0 2 t km 1 2 .0 0 (1,1,2 ,1,1,3 ,5)
p lant [ t km]
Cat ho d e f o r
cat ho d e, aluminium
A luminium  met als ext ract io n No RER 1.8 1E-0 2 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
elect ro lysis
elect ro lysis [ kg ]
Cryo lit e [ kg ]  g eneric d at a f ro m Eco invent 2 0 0 0 chemicals ino rg anics No RER cryo lit e, at p lant 1.6 0 E-0 3 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
Fig. 5.14 Flows for " aluminium, primary, liquid, at plant" and their representation in the ecoinvent database


Infra
Process Sub Loca Modul name in Mean Source mean Typ StDv General
Input Output Remarks Cate gory struc Unit
Name category tion ecoinvent value value e 95% Comment
ture
Transport of
cryolite to transport transport, lorry >16t, (4,5,n.A.,n.A.,n.A.,n.
 Transport by Road road No RER 1.60E-04 tkm standard distance 1 2.09
electrolysis plant systems fleet average A.,5)
[tkm] aluminium, primary, liquid, at plant; part 2
Transport of
cryolite to transport (4,5,n.A.,n.A.,n.A.,n.
 Transport by Rail train No RER transport, freight, rail 3.20E-04 tkm standard distance 1 2.09
electrolysis plant systems A.,5)
[tkm]
Aluminium aluminium fluoride, at
 metals extraction No RER 1.87E-02 kg EAA 2000 1 1.08 (1,1,2,1,1,3,4)
fluoride [kg] plant
Transport of AlF3
transport transport, lorry >16t, (4,5,n.A.,n.A.,n.A.,n.
to electrolysis  Transport by Road road No RER 1.87E-03 tkm standard distance 1 2.09
systems fleet average A.,5)
plant [tkm]
Transport of AlF3
transport (4,5,n.A.,n.A.,n.A.,n.
to electrolysis  Transport by Rail train No RER transport, freight, rail 3.74E-03 tkm standard distance 1 2.09
systems A.,5)
plant [tkm]
heat, natural gas, at
General heat
 natural gas heating systems No RER industrial furnace 8.40E-02 MJ EAA 2000 1 1.08 (1,1,2,1,1,3,1)
from natural gas
>100kW
General heat heat, light fuel oil, at
 oil heating systems No CH 8.90E-02 MJ EAA 2000 1 1.08 (1,1,2,1,1,3,1)
from oil industrial furnace 1MW
Electrical power electricity, medium

EAA 2001 (pers.
used in plant  Aluminium industry mix electricity supply mix No GLO voltage, aluminium 1.56E+01 kWh 1 1.08 (1,1,2,1,1,3,2)
communication)
(process) [kWh] industry, at grid
Infrastructure aluminium electrolysis, (5,n.A.,n.A.,n.A.,n.A.

 metals extraction Yes RER 1.54E-10 unit estimated 1 3.23
(plant) [piece] plant ,n.A.,9); rough guess
 waste heat from electricity use air unspecified Heat, waste 5.60E+01 MJ calculated 1 1.08 (1,1,2,1,1,3,13)
BaP (Benzo-a- EAA 2001 (pers.
 air unspecified Benzo(a)pyrene 1.30E-06 kg 1 3.00 (1,1,2,1,1,3,21)
Pyrene) to air communication)
Methane, tetrafluoro-, R- EAA 2001 (pers.
 CF4 90% of polyfluorinated carbon air unspecified 2.52E-04 kg 1 1.51 (1,1,2,1,1,3,31)
14 communication)
Ethane, hexafluoro-, EAA 2001 (pers.
 C2F6 10% of polyfluorinated carbon air unspecified 2.80E-05 kg 1 1.51 (1,1,2,1,1,3,31)
HFC-116 communication)
EAA 2001 (pers.
 CO to air air unspecified Carbon monoxide, fossil 9.17E-02 kg 1 5.01 (1,1,2,1,1,3,17)
communication)
Fig. 5.14 Flows for " aluminium, primary, liquid, at plant" and their representation in the ecoinvent database (cont.)


Infra
ture
EA A 2 00 1 (pers.
 CO2 t o air air unspecif ied Carbo n dio xide, f ossil 1.50 E+0 0 kg 1 1.08 (1,1,2,1,1,3 ,24 )
communicat ion)
part icules t hat are not ret ained b y
EA A 2 00 1 (pers.
 Dust t o air t he cyclo ne --> Part icule size < 2 .5 air unspecif ied Part iculat es, < 2 .5 um 2.6 1E-0 3 kg 1 3.0 0 (1,1,2,1,1,3 ,20 )
communicat ion)
aluminium, primary, liquid, at plant; part 3 um
f luo rid e
EA A 2 00 1 (pers.
 (g aseous) t o invent o ried as HF air unspecif ied Hydrog en f luo rid e 5.3 9E-0 4 kg 1 1.51 (1,1,2,1,1,3 ,31)
communicat ion)
air
f luo rid e (no n Part iculat es, > 2 .5 um, EA A 2 00 1 (pers.
 invent o ried as part iculat e emission air unspecif ied 6.0 9E-0 4 kg 1 2.0 0 (1,1,2,1,1,3 ,26 )
gaseo us) t o air and < 10 um communicat ion)
po lycyclic
PA H, p olycyclic EA A 2 00 1 (pers.
 aro mat ic HC t o wit ho ut B aP air unspecif ied 4 .57E-05 kg 1 3.0 0 (1,1,2,1,1,3 ,21)
aromat ic hydrocarbo ns communicat ion)
air
EA A 2 00 1 (pers.
 NOx t o air air unspecif ied Nit ro gen o xides 6 .39 E-05 kg 1 1.4 1 (1,1,2,1,1,3 ,29 )
communicat ion)
EA A 2 00 1 (pers.
 SO2 t o air air unspecif ied Sulf ur dio xide 8.8 3E-0 3 kg 1 1.08 (1,1,2,1,1,3 ,15)
communicat ion)
carb on mat erial EA A 2 00 1 (pers.

 not invent o ried 3.0 0E-0 3 kg
f o r recycling communicat ion)

 carb on wast e t o landf ill (co nt rolled ) No CH 1.9 0E-0 3 kg 1 1.08 (1,1,2,1,1,3 ,6)
EA A 2 00 1 (pers.
 cast iron recycled --> not invent o ried 1.8 0E-0 3 kg
communicat ion)
crushed bat h EA A 2 00 1 (pers.
 recycled --> not invent o ried 4.10E-0 3 kg
sold communicat ion)
d ispo sal, f ilt er d ust A l
dust f rom wast e resid ual elect ro lysis, 0% wat er, EA A 2 00 1 (pers.
 t o landf ill (co nt rolled ) No CH 2.0 0E-0 3 kg 1 1.08 (1,1,2,1,1,3 ,6)
Cyclones/ ESP management mat erial landf ill t o residual mat erial communicat ion)
land f ill
So lid wast e wast e inert mat erial EA A 2 00 1 (pers.
 t o landf ill, non-hazard ous No CH 5% wat er, t o inert 5.0 0E-0 3 kg 1 1.08 (1,1,2,1,1,3 ,6)
unspecif ied management landf ill communicat ion)
d ispo sal, bit umen, 1.4 %
wast e EA A 2 00 1 (pers.
 soo t t o landf ill (co nt rolled ) sanit ary landf ill No CH wat er, t o sanit ary 1.2 0E-0 3 kg 1 1.08 (1,1,2,1,1,3 ,6)
management communicat ion)
land f ill
A luminium, aluminium, p rimary,
liquid [ kg ] liq uid, at plant
Fig. 5.14 Flows for " aluminium, primary, liquid, at plant" and their representation in the ecoinvent database (cont.)


Infra
Pro c ess Cate Sub Lo c a Mo dul name in Mean So urc e mean Ty StDv General
Name go ry c atego ry tio n ec o inve nt value value pe 95% Co mment
ture
To t al land use Transf o rmat io n, f ro m

 reso urce land 4 .8 8 E+0 5 m2 est imat ed 1 2 .0 3 (2 ,4 ,1,1,1,4 ,8 )
(t ransf o rmat io n) unkno wn
Land o ccup at io n
f o r b uild ing s
assump t io n: 8 0 % o f t o t al area is Occup at io n, ind ust rial
and o p en  reso urce land 1.9 5E+0 7 m2 a est imat ed 1 1.54 (2 ,4 ,1,1,1,4 ,7)
b uilt up area, b uilt up
b auxit e st o rag e
area
Land
t ransf o rmat io n
Transf o rmat io n, t o
t o b uild ing s and  reso urce land 3 .9 0 E+0 5 m2 est imat ed 1 2 .0 3 (2 ,4 ,1,1,1,4 ,8 )
o p en b auxit e
st o rag e area
Land o ccup at io n
assump t io n: 2 0 % o f t o t al area is Occup at io n, ind ust rial
aluminium electrolysis, plant
f o r veg et at io n  reso urce land 4 .8 8 E+0 6 m2 a est imat ed 1 1.54 (2 ,4 ,1,1,1,4 ,7)
veg et at io n area, veg et at io n
area
Land
Transf o rmat io n, t o
t ransf o rmat io n
 reso urce land ind ust rial area, 9 .76 E+0 4 m2 est imat ed 1 2 .0 3 (2 ,4 ,1,1,1,4 ,8 )
t o veg et at io n
veg et at io n
area
B uild ing , mult i co nst ruct io n
 b uild ing s Y es RER b uild ing , mult i-st o rey 2 .50 E+0 4 m3 est imat ed 1 3 .0 4 (3 ,4 ,1,1,1,4 ,9 )
st o ry p ro cesses
co nst ruct io n
B uild ing , hall  b uild ing s Y es CH b uild ing , hall 1.9 7E+0 4 m2 est imat ed 1 3 .0 4 (3 ,4 ,1,1,1,4 ,9 )
p ro cesses
co ncret e in
co nst ruct io n co ncret e, no rmal, at
reinf o rced  co ncret e No CH 1.0 9 E+0 3 m3 est imat ed 1 3 .0 4 (3 ,4 ,1,1,1,4 ,9 )
mat erials p lant
co ncret e
st eel in
reinf o rced  met als ext ract io n No RER 7.56 E+0 5 kg est imat ed 1 3 .0 4 (3 ,4 ,1,1,1,4 ,9 )
p lant
co ncret e
heavy machine, co nst ruct io n
 machinery Y es RER heavy, unsp ecif ied , at 2 .0 0 E+0 5 kg est imat ed 1 3 .0 4 (3 ,4 ,1,1,1,4 ,9 )
unsp ecif ic p ro cesses
p lant
Po o r co ncret e co ncret e, so le p lat e
co nst ruct io n
f o r ad d it io nal  co ncret e No CH and f o und at io n, at 9 .0 9 E+0 1 m3 est imat ed 1 3 .0 4 (3 ,4 ,1,1,1,4 ,9 )
mat erials
f o und at io n p lant
co nst ruct io n
Co nveyo r b elt  machinery Y es RER co nveyo r b elt , at p lant 1.8 0 E+0 3 m est imat ed 1 3 .0 4 (3 ,4 ,1,1,1,4 ,9 )
p ro cesses
St eel lo w st eel, lo w-allo yed , at
 met als ext ract io n No RER 8 .57E+0 5 kg est imat ed 1 3 .0 4 (3 ,4 ,1,1,1,4 ,9 )
allo yed p lant
reinf o rced d isp o sal, b uild ing ,
wast e b uild ing
 co ncret e t o No CH reinf o rced co ncret e, t o 3 .16 E+0 6 kg est imat ed 1 3 .0 4 (3 ,4 ,1,1,1,4 ,9 )
manag ement d emo lit io n
d isp o sal f inal d isp o sal
d isp o sal, b uild ing ,
f o und at io n
wast e b uild ing co ncret e, no t
 co ncret e t o No CH 2 .0 0 E+0 5 kg est imat ed 1 3 .0 4 (3 ,4 ,1,1,1,4 ,9 )
manag ement d emo lit io n reinf o rced , t o f inal
d isp o sal
d isp o sal
aluminium
Out p ut : 13 0 0 0 0 0 0 0 kg / a, lif e t ime: aluminium elect ro lysis,
 elect ro lysis met als ext ract io n Y es RER 1.0 0 E+0 0 unit
50 a p lant
p lant
Fig. 5.15 Flows for "aluminium electrolysis, plant" and their representation in the ecoinvent database


Infra
Pro c e ss Cate Sub Lo c a Mo dul name in Mean So urc e me an Ty StDv Ge neral
Name go ry c atego ry tio n e c o inve nt value value pe 95% Co mme nt
ture
A luminium, aluminium, p rimary, EA A 2 00 1 (pers.

 met als ext ract io n No RER 1.0 0E+00 kg 1 1.08 (1,1,2,1,1,3,3)
liquid [ kg] liq uid , at plant co mmunicat ion)
W at er, cooling ,
Coo ling wat er EA A 2 00 1 (pers.
 Sp ec. W eig ht o f wat er: 9 98 kg / m3 reso urce in wat er unspecif ied nat ural 4 .71E-0 3 m3 1 1.08 (1,1,2,1,1,3,4)
[ m3 ] co mmunicat ion)
origin
Ref ract ory const ruct ion ref ract ory, f ireclay, EA A 2 00 1 (pers.
 bricks No DE 7.0 0E-04 kg 1 1.08 (1,1,2,1,1,3,4)
mat erials [ kg] mat erials packed, at p lant co mmunicat ion)
A llo ying EA A 2 00 1 (pers.
 met als ext ract io n No NO M G-silicon, at plant 1.08E-0 2 kg 1 1.08 (1,1,2,1,1,3,4)
element s [ kg] co mmunicat ion)
Nit rogen g as nit rog en, liq uid , at EA A 2 00 1 (pers.
 chemicals ino rg anics No RER 6.00E-0 4 kg 1 1.08 (1,1,2,1,1,3,4)
aluminium, primary, at plant; Inputs
[ kg] plant co mmunicat ion)

EA A 2 00 1 (pers.
A rg on g as [ kg ]  chemicals ino rg anics No RER argon, liq uid , at plant 1.50E-0 3 kg 1 1.08 (1,1,2,1,1,3,4)
co mmunicat ion)
chlorine, liq uid ,
EA A 2 00 1 (pers.
Clorine gas [ kg]  chemicals ino rg anics No RER prod uct ion mix, at 1.00E-0 4 kg 1 1.08 (1,1,2,1,1,3,4)
co mmunicat ion)
plant
Fibre mat erial insulat ion EA A 2 00 1 (pers.
 product ion No CH ro ck woo l, at p lant 1.10E-04 kg 1 1.08 (1,1,2,1,1,3,4)
[ kg] mat erials co mmunicat ion)
Fluxing salt s EA A 2 00 1 (pers.
 chemicals ino rg anics No RER cryolit e, at p lant 4.00E-0 4 kg 1 1.08 (1,1,2,1,1,3,4)
[ kg] co mmunicat ion)
M ineral / EA A 2 00 1 (pers.
 chemicals organics No MY crude palm oil, at p lant 8.00E-0 5 kg 1 1.08 (1,1,2,1,1,3,4)
veget al oils [ kg] co mmunicat ion)
cardb oard & co rrugat ed bo ard ,

Packag eing pap er & EA A 2 00 1 (pers.
 cardb oard assumed co rrugat ed No RER mixed f ib re, single wall, 1.80E-0 3 kg 1 1.08 (1,1,2,1,1,3,4)
mat erials [ kg] card bo ard co mmunicat ion)
bo ard at plant
Transpo rt of t ransp ort (4,5,n.A .,n.A .,n.A .,n.
 Transpo rt b y Road ro ad No RER t ransport , lo rry 32 t 1.61E-03 t km st and ard dist ance 1 2.09
mat erials [ t km] syst ems A .,5)
Transpo rt of t ransp ort (4,5,n.A .,n.A .,n.A .,n.
 Transpo rt b y Rail t rain No RER t ransport , f reig ht , rail 3.22E-0 3 t km st and ard dist ance 1 2.09
mat erials [ t km] syst ems A .,5)
Elect rical p ower
used in plant
(pro cess, elect ricit y, med ium
EA A 2 00 1 (pers.
light ing , of f ice  A luminium indust ry mix elect ricit y supply mix No GLO vo lt age, aluminium 1.60E-0 2 kW h 1 1.08 (1,1,2,1,1,3,2)
co mmunicat ion)
work, ro om ind ust ry, at grid
coo ling ...)
[ kW h]
General heat heat ing heat , light f uel oil, at EA A 2 00 1 (pers.

 oil No CH 4 .41E-0 1 M J 1 1.08 (1,1,2,1,1,3,1)
f rom oil syst ems ind ust rial f urnace 1M W co mmunicat ion)

 nat ural gas No RER ind ust rial f urnace 6 .41E-0 1 M J 1 1.08 (1,1,2,1,1,3,1)
f rom nat ural g as syst ems co mmunicat ion)
>100kW
Inf rast ruct ure aluminium cast ing, (5,n.A .,n.A .,n.A .,n.A .,
 Incl. Land use met als ext ract io n Y es RER 1.54 E-10 unit est imat ed 1 3.23
(plant ) [ unit ] plant n.A .,9); ro ugh guess
Fig. 5.16 Flows for "aluminium, primary, at plant" and their representation in the ecoinvent database


Infra
ture
 wast e heat f rom elect ricit y use air unspecif ied Heat , wast e 5.76E-0 2 M J calculat ed 1 1.08 (1,1,2,1,1,3 ,13 )
f luo rid e
aluminium, primary, at plant;
EA A 2 00 1 (pers.
 (g aseous) t o invent o ried as HF air unspecif ied Hydrog en f luo rid e 3.0 0E-0 6 kg 1 1.51 (1,1,2,1,1,3 ,31)
communicat ion)
air
f luo rid e (no n Part iculat es, > 2 .5 um, EA A 2 00 1 (pers.

 invent o ried as part iculat es air unspecif ied 7.0 0E-0 6 kg 1 2.0 0 (1,1,2,1,1,3 ,26 )
gaseo us) t o air and < 10 um communicat ion)
d ispo sal, dross f ro m

wast e resid ual A l elect ro lysis, 0% EA A 2 00 1 (pers.
 Dro ss (f ines) t o landf ill (co nt rolled ) No CH 1.10E-0 4 kg 1 1.08 (1,1,2,1,1,3 ,6)
Outputs

wast e (f ib er wast e inert mat erial EA A 2 00 1 (pers.
 t o landf ill, non-hazard ous No CH 5% wat er, t o inert 9 .00 E-05 kg 1 1.08 (1,1,2,1,1,3 ,6)
mat erial) management landf ill communicat ion)
ref ract ory wast e inert mat erial EA A 2 00 1 (pers.
 t o landf ill, non-hazard ous No CH 5% wat er, t o inert 9.0 0E-0 4 kg 1 1.08 (1,1,2,1,1,3 ,6)
wast e management landf ill communicat ion)
skimmings and
EA A 2 00 1 (pers.
 dross f or recycled --> not invent o ried 1.8 6E-0 2 kg
communicat ion)
recycling
swarf / EA A 2 00 1 (pers.
 recycled --> not invent o ried 8.4 0E-0 4 kg
t urnings communicat ion)
aluminium, 0%
aluminium, p rimary, at
 recycling , at met als ext ract io n No RER 1.00 E+0 0 kg
p lant
plant
Fig. 5.16 Flows for "aluminium, primary, at plant" and their representation in the ecoinvent database (cont.)


Infra
ture

 reso urce land 4.8 8E+05 m2 est imat ed 1 2.0 3 (2,4 ,1,1,1,4 ,8)
Land o ccupat ion

f o r b uilding s
and o pen  reso urce land 1.95E+07 m2a est imat ed 1 1.54 (2,4 ,1,1,1,4 ,7)
bauxit e st o rag e
area
Land
aluminium casting, plant
t ransf o rmat ion

t o build ings and  reso urce land 3.9 0E+05 m2 est imat ed 1 2.0 3 (2,4 ,1,1,1,4 ,8)
op en bauxit e
st o rag e area
Land o ccupat ion

f o r veg et at ion  reso urce land 4 .88 E+0 6 m2a est imat ed 1 1.54 (2,4 ,1,1,1,4 ,7)
area
Land
t ransf o rmat ion
 reso urce land ind ust rial area, 9.76 E+0 4 m2 est imat ed 1 2.0 3 (2,4 ,1,1,1,4 ,8)
t o veget at io n
veg et at ion
area
B uilding , mult i const ruct ion
 build ings Y es RER b uilding , mult i-st o rey 2 .00 E+0 3 m3 est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
st o ry processes
const ruct ion
B uilding , hall  build ings Y es CH b uilding , hall 2.17E+0 4 m2 est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
processes
heavy machine, const ruct ion
 machinery Y es RER heavy, unsp ecif ied , at 2 .00 E+0 6 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
unspecif ic processes
p lant
const ruct ion
f o r ad dit ional  concret e No CH and f o undat ion, at 9 .09 E+0 2 m3 est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
mat erials
f o undat ion
wast e build ing co ncret e, not
 concret e t o No CH 2 .00 E+0 6 kg est imat ed 1 3.0 4 (3,4 ,1,1,1,4 ,9)
management demo lit io n reinf o rced , t o f inal
disp osal
d ispo sal
aluminium Out p ut : 130 00 00 00 kg/ a, lif e t ime: aluminium cast ing,
 met als ext ract io n Y es RER 1.00 E+0 0 unit
cast ing plant 50 a p lant
Fig. 5.17 Flows for "aluminium casting, plant" and their representation in the ecoinvent database

5.4 Secondary aluminium

This part is mainly based on EAA (2000).

A wide variety of aluminium scrap is processed in the secondary aluminium industry. A basic differ-
entiation is made between "new" and "old" scrap.
New scrap is surplus material that arises during production and fabrication of aluminium products up
to the point where they are sold to the final consumer. It is usually of known quality and alloy and of-
ten uncoated. It can then be remelted with little preparation. Some new scrap is coated with paints or
plastics. This scrap has to be de-coated in an oven or a mesh conveyer where the coatings are volatil-
ised or burned.
Old scrap or post-consumer scrap is the aluminium recovered after the product has been used by the
final consumer. It comes into the secondary aluminium industry via metal merchants and waste man-
agement companies which recover aluminium from e.g. vehicles, household goods etc. This is usually
done by shredding, magnetic separation and sink-and -float installations or eddy current installations
successively. Old scrap is usually recycled together with two-thirds of new scrap from aluminium
processing.
Different melting processes are used depending on composition of the scrap, process availability and
economics amongst other variables. Molten metal fluxing and filtration technologies are used for qual-
ity control of the resulting alloys.
5.4.2 Emissions
Air
If new scrap is clean and uncoated no emissions arise from cleaning or de-coating. Melting and alloy-
ing of new scrap emits HCl and chlorides. Old scrap preparation leads to dust and also HCl and chlo-
ride emission. Melting and alloying of old scrap leads to dust and different inorganic emissions.
The size fractions of the dust emissions are derived by the medium level emissions in cepmeip emis-
sion factor database 14 .
Water
There are no reports of direct water emissions from these processes. The preparation of old scrap how-
ever is expected to emit certain substances to water.
5.4.3 Waste
New scrap production waste
In the preparation of uncoated new scrap no waste is generated.
Aluminium from new scrap production waste
The secondary aluminium production from new scrap generates only some refractory material waste
which is inert and deposited in landfills.
Old scrap production waste
The preparation of old scrap generates different waste fractions:
14
http://www.air.sk/tno/cepmeip/em_factors_results.php?, accessed 08.01.03

- Waste oil is assumed to be deposited as hazardous waste in an underground deposit.
- No composition data could be found for ball mill dust, waste filter materials and waste sediments.
All the fractions are assumed to consist mainly of organic coatings of the scrap and to a lower ex-
tent of aluminium oxide. Deposition as hazardous waste in an underground deposit is assumed.
- Dirt is assumed to be an inert waste deposited in inert material landfills.
- Filter dust composition is assumed to be the same as in the electrolysis process (see Tab. 5.9).
Disposal in a residual material landfill is assumed.
Aluminium from old scrap processing waste
The secondary aluminium production from old scrap also generates "ball mill dust" and "filter dust".
These wastes are treated as the respective wastes from the old scrap production.
5.4.4 Secondary Aluminium in ecoinvent

The available information about the infrastructure needed for aluminium production is non-
quantitative. Tab. 5.14 and Tab. 5.15show the details of the assumptions made to estimate the infra-
structure expenditures.
Tab. 5.14 Estimated infrastructure for aluminium melting furnace
aluminium melting furnace Unit Life Life time

time demand
[a]
Production hall 938 m2 50 938
Furnace 76'923 kg 25 153'846
Tab. 5.15 Estimated infrastructure for scrap preparation plant
scrap preparation plant Demand Unit Life Life time

time demand
[a]
Production hall 938 m2 50 938
conveyor belt [m] 100 m 25 200
Production machines 76'923 kg 25 153'846
Floating tanks 20 x 5 x 1 m 21 m3 50 21
Storage hall 731 m2 50 731
The other background information to the processes are given in, flows are shown in Tab. 5.16:
- aluminium scrap, new, at plant Fig. 5.18
- aluminium, secondary, from new scrap, at plant Fig. 5.19
- aluminium melting furnace Fig. 5.20

- aluminium scrap, old, at plant Fig. 5.21
- aluminium, secondary, from old scrap, at plant Fig. 5.22
- scrap preparation plant Fig. 5.23
Tab. 5.16 Meta information for the secondary aluminium production processes
aluminium, secondary,
Name aluminium scrap, new, from new scrap, at aluminium melting
at plant plant furnace
Unit kg kg unit
Production hall and ma-
chinery (including dis-
posal). Does not include
Included Processes
Includes only transports of Melting, alloying and cast- administrative buildings
new scrap to secondary ing of new scrap to secon- because they are included
aluminium production. dary aluminium billets. in casting plant.
Amount 1 1 1
Aluminiumschrott, neu, ab Aluminium, sekundär, aus
Local Name
Werk neuem Schrott, ab Werk Aluminium Schmelzofen
Synonyms
Medium sized plant. Out-
It is assumed that the new put: 10'000 t/a, life time: 50
General Comment to
scrap is not coated and years. Assumption is
reference function
can be used without treat- based on data for casting
ment. plant.
Start Date 1995 1995 2002
End Date 2002 2002 2002
Period 1 1 1
Some Swiss datasets are Some Swiss datasets are Some Swiss datasets are
Geography text used for European proc- used for European proc- used for European proc-
esses. esses. esses.
Average technology for the Average technology for the
Technology text aluminium recycled in aluminium recycled in
Europe Europe Assumed technology
Representativeness [%]
Sampling Procedure Literature / Survey Literature / Survey Internet search
Uncertainty is adjusted for Uncertainty is adjusted for Uncertainty is adjusted for
geographical mismatch geographical mismatch geographical mismatch
(using the pedigree ma- (using the pedigree ma- (using the pedigree ma-
trix). trix). trix).

Tab. 5.16 Meta information for the secondary aluminium production processes (cont.)
aluminium, secondary,
Name aluminium scrap, old, scrap preparation from old scrap, at
at plant plant plant
Unit kg unit kg
Collecting, sorting and
preparing (cleaning, press- Production hall, machinery Melting, alloying and cast-
Included Processes
ing...) of post consumer and administrative building of old scrap to secon-
aluminium scrap ings (including disposal) dary aluminium billets.
Amount 1 1 1
Aluminiumschrott, alt, ab Aluminium, sekundär, aus
Local Name
Werk Sortieranlage Schrott altem Schrott, ab Werk
Synonyms post consumer aluminium
Medium sized plant. Out-
put: 10'000 t/a, life time: 50
General Comment to
years. Assumption is
reference function
based on data for casting
plant.
Start Date 2002 2002 1995
End Date 2002 2002 2002
Period 1 1 1
Some Swiss datasets are Some Swiss datasets are Some Swiss datasets are
Geography text used for European proc- used for European proc- used for European proc-
esses. esses. esses.
Average technology for the Average technology for the
Technology text aluminium recycled in aluminium recycled in
Europe Assumed technology Europe
Sampling Procedure Literature / Survey Internet search Literature / Survey
Uncertainty is adjusted for Uncertainty is adjusted for Uncertainty is adjusted for
geographical mismatch geographical mismatch geographical mismatch
(using the pedigree ma- (using the pedigree ma- (using the pedigree ma-
trix). trix). trix).
Data quality
Data for the steps in the secondary aluminium production are of satisfactory overall quality. The val-
ues for the smelting processes stem from a recent and comprehensive source. The proxies chosen to
represent the alloying elements in ecoinvent however are decreasing the overall quality. The collection
of the scrap and the infrastructure are roughly estimated. The infrastructure is of minor importance
compared to the other activities but the transports (scrap collection) could have a noticeable impact.


Infra
ture
New scrap f rom

 f rom recycling --> no t invent oried 1.00 E+0 0 kg calculat ed
prod uct io n
scrap, new,
aluminium
at plant
Transpo rt of t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
ro ad: 100 km ro ad No RER t ransp ort , lo rry 32 t 1.0 0E-0 1 t km st and ard dist ance 1 2.0 9
input s t o plant syst ems A .,5)
rail: 2 00 km t rain No RER t ransp ort , f reig ht , rail 2.0 0E-0 1 t km st and ard dist ance 1 2.0 9
aluminium aluminium scrap, new,
scrap , new at plant
Fig. 5.18 Flows for "aluminium scrap, new, at plant" and their representation in the ecoinvent database

Infra
ture
New aluminium aluminium scrap, new,

 met als ext ract io n No RER 1.01E+0 0 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,3)
aluminium, secondary, from
scrap at plant
new scrap, at plant; part 1
A lloying
element s (pure  Silico n as proxy met als ext ract io n No NO M G-silicon, at plant 1.14E-0 2 kg EA A 2 00 0 1 1.22 (1,1,2,1,3,3 ,4)
f o rm)
A lloying
co pp er, at reg ional
element s  Cop per as proxy met als ext ract io n No RER 1.0 4E-0 2 kg EA A 2 00 0 1 1.22 (1,1,2,1,3,3 ,4)
st orage
(mast er allo ys)
elect ricit y, medium
elect ricit y  elect ricit y prod uct io n mix No UCTE vo lt ag e, prod uct io n 1.74E-0 1 kW h EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,2)
UCTE, at grid
Thermal energ y heat ing
 nat ural gas No RER b oiler mod ulat ing 3 .04 E+0 0 M J EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,1)
f rom g as syst ems
>100 kW
Thermal energ y
heat ing heat , heavy f uel o il, at
f rom heavy f uel  oil No RER 7.0 0E-0 2 M J EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,1)
syst ems ind ust rial f urnace 1M W
oil
Thermal energ y
heat ing heat , light f uel o il, at
f rom lig ht f uel  oil No CH 3.3 2E-0 3 M J EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,1)
syst ems ind ust rial f urnace 1M W
oil
Fig. 5.19 Flows for "aluminium, secondary, from new scrap, at plant" and their representation in the ecoinvent database


Infra
ture
W at er, co oling ,
Coo ling wat er  reso urce in wat er unsp ecif ied nat ural 7.8 0E-0 2 m3 EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,4)
o rig in
arg on  1.6 64 kg/ m3 chemicals inorganics No RER argo n, liquid , at p lant 2 .50E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,4)
aluminium, secondary, from new
nit ro gen, liq uid, at

nit rog en  1.2 51 kg / m3 chemicals inorganics No RER 1.25E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,4)
p lant
chlo rine, liq uid,
scrap, at plant; part 2
chlorine  3.2 14 kg/ m3 chemicals inorganics No RER p ro duct ion mix, at 3 .86 E-05 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,4)
p lant
so dium chlo rid e,
salt  chemicals inorganics No RER 8.9 0E-0 4 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,4)
p owd er, at plant
const ruct ion ref ract o ry, f ireclay,
ref ract ories  bricks No DE 1.4 0E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,4)
mat erials p acked, at plant
Inf rast ruct ure aluminium melt ing (5,n.A .,n.A .,n.A .,n.A .,
 met als ext ract io n Y es RER 2.0 0E-0 9 unit est imat ed 1 3.2 3
melt ing f urnace n.A .,9 ); roug h guess
Inf rast ruct ure aluminium cast ing, (5,n.A .,n.A .,n.A .,n.A .,
 met als ext ract io n Y es RER 1.54E-10 unit est imat ed 1 3.2 3
cast ing p lant n.A .,9 ); roug h guess
 HCl air unspecif ied Hydrog en chloride 3 .30 E-05 kg EA A 2 00 0 1 1.51 (1,1,2,1,1,3 ,31)
Halog enat ed
 chlorides air unspecif ied hyd ro carb ons, 3.9 0E-0 6 kg EA A 2 00 0 1 1.51 (1,1,2,1,1,3 ,31)
chlo rinat ed
non hazardo us wast e inert mat erial
 No CH 5% wat er, t o inert 1.50E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
wast e management landf ill
 skimmings t o recycling --> not invent o ried 3 .27E-0 2 kg EA A 2 00 0
second ary aluminium, seco ndary,
 aluminium f rom met als ext ract io n No RER f ro m new scrap , at 1.00 E+0 0 kg
new scrap p lant
Fig. 5.19 Flows for "aluminium, secondary, from new scrap, at plant" and their representation in the ecoinvent database (cont.)


Infra
ture
To t al land use Transf ormat io n, f rom (5,n.A .,n.A .,n.A .,n.A .,

 reso urce land 3 .75E+0 4 m2 est imat ed 1 2.2 3
(t ransf ormat io n) unkno wn n.A .,8 )
Land o ccupat ion

f o r b uilding s
and o pen  reso urce land 1.50 E+0 6 m2a est imat ed 1 1.77
built up area, built up n.A .,7)
bauxit e st o rag e
area
aluminium melting furnace
Land
t ransf o rmat ion
Transf ormat io n, t o (5,n.A .,n.A .,n.A .,n.A .,
t o build ings and  reso urce land 3 .00 E+0 4 m2 est imat ed 1 2.2 3
op en bauxit e
st o rag e area
Land o ccupat ion

f o r veg et at ion  reso urce land 3.75E+05 m2a est imat ed 1 1.77
veget at io n area, veget at io n n.A .,7)
area
Land
t ransf o rmat ion (5,n.A .,n.A .,n.A .,n.A .,
 reso urce land ind ust rial area, 7.51E+0 3 m2 est imat ed 1 2.2 3
t o veget at io n n.A .,8 )
veg et at ion
area
B uilding , hall  build ings Y es CH b uilding , hall 9 .38 E+0 2 m2 est imat ed 1 3.2 3
processes n.A .,9 )
heavy machine, const ruct ion (5,n.A .,n.A .,n.A .,n.A .,
 machinery Y es RER heavy, unsp ecif ied , at 1.54E+05 kg est imat ed 1 3.2 3
unspecif ic processes n.A .,9 )
p lant
f o r ad dit ional  concret e No CH and f o undat ion, at 6 .99 E+0 1 m3 est imat ed 1 3.2 3
f o undat ion
wast e build ing co ncret e, not (5,n.A .,n.A .,n.A .,n.A .,
 concret e t o No CH 1.54E+05 kg est imat ed 1 3.2 3
management demo lit io n reinf o rced , t o f inal n.A .,9 )
disp osal
d ispo sal
aluminium Out p ut : 100 00 00 0 kg / a, lif e t ime: aluminium melt ing
 met als ext ract io n Y es RER 1.00 E+0 0 unit
melt ing f urnace 50 a f urnace
Fig. 5.20 Flows for "aluminium melting furnace" and their representation in the ecoinvent database


Infra
ture
old scrap f rom

consumers
 f rom recycling --> no t invent oried 1.23 E+0 0 kg EA A 2 00 0
(includ ing
wat er)
ro ad: everyt hing 10 0 km ro ad No RER t ransp ort , lo rry 32 t 1.2 3E-0 1 t km st and ard dist ance 1 2.0 9
Transpo rt of rail: met als / lime: 2 00 km, rest 60 0 t ranspo rt (4,5,n.A .,n.A .,n.A .,n.
aluminium scrap, old, at plant; part 1
t rain No RER t ransp ort , f reig ht , rail 2.4 6E-0 1 t km st and ard dist ance 1 2.0 9
input s t o plant km syst ems A .,5)
elect ricit y, medium
elect rical  elect ricit y prod uct io n mix No UCTE vo lt ag e, prod uct io n 6.4 0E-0 2 kW h EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,2)
UCTE, at grid
t hermal f rom heat ing heat , heavy f uel o il, at
 oil No RER 1.6 6E-0 1 M J EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,1)
heavy f uel o il syst ems ind ust rial f urnace 1M W

t hermal f rom heat ing
 nat ural gas No RER b oiler mod ulat ing 1.02 E+0 0 M J EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,1)
gas syst ems
>100 kW
chemicals o rg anic, at
det erg ent  chemicals organics No GLO 2.6 0E-0 4 kg EA A 2 00 0 1 1.22 (1,1,2,1,3,3 ,4)
p lant
hydraulic oil  chemicals organics No RER lub ricat ing o il, at p lant 7.3 0E-0 6 kg EA A 2 00 0 1 1.22 (1,1,2,1,3,3 ,4)
lig ht f uel oil, at
light oil  oil f uels No RER 8 .30 E-05 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,4)
regio nal st o rag e
alloying co pp er, at reg ional
 Cop per as proxy met als ext ract io n No RER 9.6 0E-0 4 kg EA A 2 00 0 1 1.22 (1,1,2,1,3,3 ,4)
add it ives st orage
f erro silicon  Silico n as proxy met als ext ract io n No NO M G-silicon, at plant 5.8 0E-0 4 kg EA A 2 00 0 1 1.22 (1,1,2,1,3,3 ,4)
const ruct ion lime, hyd rat ed, p acked,
lime  bind er No CH 2.8 0E-0 4 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,4)
mat erials at plant
Inf rast ruct ure Out p ut : 100 00 00 0 kg / a, lif e t ime: scrap p rep arat ion (5,n.A .,n.A .,n.A .,n.A .,
 met als ext ract io n Y es RER 2.0 0E-0 9 unit est imat ed 1 3.2 3
(p lant ) [ p iece] 50 a p lant n.A .,9 ); roug h guess
Halog enat ed
chloride-gas in
 air unspecif ied hyd ro carb ons, 4 .60 E-07 kg EA A 2 00 0 1 1.51 (1,1,2,1,1,3 ,31)
air
chlo rinat ed
t o t al d ust : 0 .00 00 29 , 32 % <2 .5 um, EA A 2 00 0 /
 dust air unspecif ied Part iculat es, > 10 um 9.2 8E-0 6 kg 1 1.51 (1,1,2,1,1,3 ,25)
48 % >2 .5<10 um, 20 % > 10 um cepmeip
Part iculat es, > 2 .5 um, EA A 2 00 0 /
 dust air unspecif ied 1.39 E-05 kg 1 2.0 0 (1,1,2,1,1,3 ,26 )
and < 10 um cepmeip
EA A 2 00 0 /
 dust air unspecif ied Part iculat es, < 2 .5 um 5.8 0E-0 6 kg 1 3.0 0 (1,1,2,1,1,3 ,27)
cepmeip
Fig. 5.21 Flows for "aluminium scrap, old, at plant" and their representation in the ecoinvent database


Infra
ture
HCl(hydrog en
 air unspecif ied Hydrog en chloride 1.10 E-05 kg EA A 2 00 0 1 1.51 (1,1,2,1,1,3 ,31)
chloride)
d ispo sal, hazard ous
wast e underground
 oil (wast e) No DE wast e, 0% wat er, t o 2.6 0E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
management dep osit
und erg ro und d epo sit
ball mill d ust comp osit ion unkno wn, hazard ous wast e underground
 No DE wast e, 0% wat er, t o 4.8 1E-0 2 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
(wast e) wast e d epo sit io n assumed management dep osit
aluminium scrap, old, at plant; part 2
by-prod uct : A l-
 t o ot her use --> no t invent oried 8.3 0E-0 4 kg EA A 2 00 0
Mg
by-prod uct :
 t o ot her use --> no t invent oried 9 .52E-0 3 kg EA A 2 00 0
iro n scrap
wast e inert mat erial
 dirt No CH 5% wat er, t o inert 1.87E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
management landf ill
comp osit ion and t reat ment as f ilt er wast e resid ual elect ro lysis, 0% wat er,
 f ilt er d ust No CH 3.2 8E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
dust of elect rolysis management mat erial landf ill t o residual mat erial
land f ill
ref ract ory wast e inert mat erial
 No CH 5% wat er, t o inert 8.70 E-05 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
d ispo sal, rub ber,
wast e municipal unsp ecif ied , 0% wat er,
 rub ber No CH 2.3 6E-0 2 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
management incinerat ion t o municip al
incinerat io n
d ispo sal, municipal
solid wast e - wast e municipal so lid wast e, 22 .9%
 No CH 3.2 3E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
unspecif ied management incinerat ion wat er, t o municip al
incinerat io n
wast e inert mat erial
 st o nes No CH 5% wat er, t o inert 4.16E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
wast e f ilt er comp osit ion unkno wn, hazard ous wast e underground
 No DE wast e, 0% wat er, t o 4.4 0E-0 6 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
mat erial wast e d epo sit io n assumed management dep osit
comp osit ion unkno wn, hazard ous wast e underground
 wast e sed iment No DE wast e, 0% wat er, t o 3.3 3E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
wast e d epo sit io n assumed management dep osit
Non f ero us
 t o ot her use --> no t invent oried 4.0 4E-0 2 kg EA A 2 00 0
met al scrap
aluminium aluminium scrap, o ld, at
scrap , old p lant
Fig. 5.21 Flows for "aluminium scrap, old, at plant" and their representation in the ecoinvent database (cont.)


Infra
Pro c ess Cate Sub Lo c a Mo dul name in Mean So urc e me an Ty StDv Ge neral
Name go ry c atego ry tio n ec o inve nt value value pe 95% Co mme nt
ture
Prep ared o ld aluminium p rep ared scrap (net f ro m aluminium scrap , o ld , at

 met als ext ract io n No RER 1.0 3 E+0 0 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,3 )
scrap 1110 kg - 76 .6 5 kg al met allics) p lant
elect ricit y, med ium

elect rical  elect ricit y p ro d uct io n mix No UCTE vo lt ag e, p ro d uct io n 2 .8 8 E-0 1 kW h EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,2 )
UCTE, at g rid
aluminium, secondary, from old scrap, at plant; Inputs

t hermal f ro m heat ing heat , heavy f uel o il, at
 o il No RER 5.13 E-0 1 M J EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,1)
heavy o il syst ems ind ust rial f urnace 1M W
heat , nat ural g as, at

t hermal f ro m heat ing
 nat ural g as No RER b o iler mo d ulat ing 8 .2 7E+0 0 M J EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,1)
g as syst ems
>10 0 kW
chlo rine, liq uid ,
chlo rine  chemicals ino rg anics No RER p ro d uct io n mix, at 1.6 0 E-0 3 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
p lant
nit ro g en, liq uid , at
nit ro g en  chemicals ino rg anics No RER 1.8 0 E-0 3 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
p lant
hyd ro chlo ric hyd ro chlo ric acid , 3 0 %
 chemicals ino rg anics No RER 2 .0 0 E-0 4 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
acid in H2 O, at p lant
lig ht f uel o il, at
lig ht o il  o il f uels No RER 2 .4 0 E-0 6 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
reg io nal st o rag e
so d ium hyd ro xid e, 50 %
so d ium
 chemicals ino rg anics No RER in H2 O, p ro d uct io n 1.59 E-0 3 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
hyd ro xid e
mix, at p lant
sulp huric acid , liq uid ,

sulp huric acid  chemicals ino rg anics No RER 7.9 6 E-0 3 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
at p lant
wat er W at er, unsp ecif ied
 reso urce in wat er 7.9 6 E-0 3 m3 EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
(co nsumed ) nat ural o rig in
Silico n as
 met als ext ract io n No NO M G-silico n, at p lant 1.2 3 E-0 2 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
allo ying ad d it ive
Co p p er as co p p er, at reg io nal

 met als ext ract io n No RER 7.50 E-0 4 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
allo ying ad d it ive st o rag e
o t her allo ying zinc f o r co at ing , at

 Zinc as p ro xy met als ext ract io n No RER 6 .50 E-0 2 kg EA A 2 0 0 0 1 1.2 2 (1,1,2 ,1,3 ,3 ,4 )
ad d it ives reg io nal st o rag e
co nst ruct io n lime, hyd rat ed , p acked ,
lime  b ind er No CH 7.4 1E-0 3 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
mat erials at p lant
so d ium chlo rid e,
salt  chemicals ino rg anics No RER 1.3 7E-0 2 kg EA A 2 0 0 0 1 1.0 8 (1,1,2 ,1,1,3 ,4 )
p o wd er, at p lant
Transp o rt o f ro ad : met als, lime & chemicals 10 0 t ransp o rt (4 ,5,n.A .,n.A .,n.A .,n.
 ro ad No RER t ransp o rt , lo rry 3 2 t 1.15E-0 1 t km st and ard d ist ance 1 2 .0 9
inp ut s t o p lant km, g ases 50 km syst ems A .,5)
Transp o rt o f rail: met als & lime 2 0 0 km, t ransp o rt (4 ,5,n.A .,n.A .,n.A .,n.
 t rain No RER t ransp o rt , f reig ht , rail 2 .3 9 E-0 1 t km st and ard d ist ance 1 2 .0 9
inp ut s t o p lant chemicals 6 0 0 km, g ases 10 0 km syst ems A .,5)
Inf rast ruct ure aluminium melt ing (5,n.A .,n.A .,n.A .,n.A .,
 met als ext ract io n Y es RER 2 .0 0 E-0 9 unit st and ard d ist ance 1 3 .2 3
melt ing f urnace n.A .,9 ); ro ug h g uess
Inf rast ruct ure aluminium cast ing , (5,n.A .,n.A .,n.A .,n.A .,
 met als ext ract io n Y es RER 1.54 E-10 unit est imat ed 1 3 .2 3
cast ing p lant n.A .,9 ); ro ug h g uess
Fig. 5.22 Flows for "aluminium, secondary, from old scrap, at plant" and their representation in the ecoinvent database


Infra
ture
ammonia (
 emissions t o air air unspecif ied A mmo nia 2 .00 E-05 kg EA A 2 00 0 1 1.51 (1,1,2,1,1,3 ,31)
NH3 )
Halog enat ed
chloride-gas in
 air unspecif ied hyd ro carb ons, 5.00 E-05 kg EA A 2 00 0 1 1.51 (1,1,2,1,1,3 ,31)
air
chlo rinat ed
aluminium, secondary, from old scrap, at plant;
chlorine-g as in
 air unspecif ied Chlo rine 4 .90 E-07 kg EA A 2 00 0 1 1.51 (1,1,2,1,1,3 ,31)
air
t o t al d ust : 0 .00 00 24 , 32 % <2 .5 um, EA A 2 00 0 /
 dust air unspecif ied Part iculat es, > 10 um 7.68 E-07 kg 1 1.51 (1,1,2,1,1,3 ,25)
48 % >2 .5<10 um, 20 % > 10 um cepmeip
Part iculat es, > 2 .5 um, EA A 2 00 0 /
 dust air unspecif ied 1.15E-0 6 kg 1 2.0 0 (1,1,2,1,1,3 ,26 )
and < 10 um cepmeip
EA A 2 00 0 /
 dust air unspecif ied Part iculat es, < 2 .5 um 4 .80 E-07 kg 1 3.0 0 (1,1,2,1,1,3 ,27)
cepmeip
H2 S (hydrog en
 air unspecif ied Hydrog en sulf id e 2.8 0E-0 6 kg EA A 2 00 0 1 1.51 (1,1,2,1,1,3 ,31)
sulphid e)
HCl(hydrog en
 air unspecif ied Hydrog en chloride 4.10E-0 6 kg EA A 2 00 0 1 1.51 (1,1,2,1,1,3 ,31)
chloride)
HF (hyd ro gen
 air unspecif ied Hydrog en f luo rid e 3 .50E-0 6 kg EA A 2 00 0 1 1.51 (1,1,2,1,1,3 ,31)
f luo rid e)
Outputs
 nit rog en not invent o ried 2 .45E-0 3 kg EA A 2 00 0
 pho sphine Pho spho rus as p ro xy air unspecif ied Phosp horus 5.20 E-07 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
comp osit ion unkno wn, hazard ous wast e underground
 ball mill d ust No DE wast e, 0% wat er, t o 1.45E-0 2 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
wast e d epo sit io n assumed management dep osit
by-prod uct :
 not invent o ried 2.18E-0 3 kg EA A 2 00 0
iro n scrap
wast e resid ual elect ro lysis, 0% wat er,
 f ilt er d ust No CH 9.6 3E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
management mat erial landf ill t o residual mat erial
land f ill
ref ract ory wast e inert mat erial
 No CH 5% wat er, t o inert 2.0 2E-0 3 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
d ispo sal, municipal
solid wast e - wast e municipal so lid wast e, 22 .9%
 No CH 8 .00 E-05 kg EA A 2 00 0 1 1.08 (1,1,2,1,1,3 ,6)
unspecif ied management incinerat ion wat er, t o municip al
incinerat io n
A luminium
 not invent o ried 1.19E-0 1 kg EA A 2 00 0
oxid e
 aluminium f rom met als ext ract io n No RER f ro m old scrap , at 1.00 E+0 0 kg
old scrap p lant
Fig. 5.22 Flows for "aluminium, secondary, from old scrap, at plant" and their representation in the ecoinvent database (cont.)


Infra
Pro c ess Cate Sub Lo c a Mo dul name in Me an So urc e me an Ty StDv Ge neral
Name go ry c ate go ry tio n e c o inve nt value value pe 95% Comment
ture
To t al land use Transf o rmat ion, f ro m (5,n.A .,n.A .,n.A .,n.A .,

 reso urce land 3 .75E+0 4 m2 est imat ed 1 2 .2 3
(t ransf o rmat io n) unkno wn n.A .,8 )
Land occup at ion

f o r b uild ings
assump t io n: 8 0 % o f t o t al area is Occup at io n, ind ust rial (5,n.A .,n.A .,n.A .,n.A .,
and o p en  reso urce land 1.50 E+0 6 m2 a est imat ed 1 1.77
b uilt up area, b uilt up n.A .,7)
b auxit e st o rag e
area
Land
t ransf o rmat io n
Transf o rmat ion, t o (5,n.A .,n.A .,n.A .,n.A .,
t o b uild ing s and  reso urce land 3.0 0 E+0 4 m2 est imat ed 1 2 .2 3
o p en b auxit e
st o rage area
Land occup at ion

assump t io n: 2 0 % o f t o t al area is Occup at io n, ind ust rial (5,n.A .,n.A .,n.A .,n.A .,
f o r veg et at io n  reso urce land 3 .75E+0 5 m2 a est imat ed 1 1.77
veg et at io n area, veg et at io n n.A .,7)
scrap preparation plant
area
Land
Transf o rmat ion, t o
t ransf o rmat io n (5,n.A .,n.A .,n.A .,n.A .,
 reso urce land ind ust rial area, 7.51E+0 3 m2 est imat ed 1 2 .2 3
t o veg et at io n n.A .,8 )
veg et at io n
area
B uild ing , mult i co nst ruct io n (5,n.A .,n.A .,n.A .,n.A .,
 b uild ing s Y es RER b uild ing , mult i-st o rey 2 .50 E+0 4 m3 est imat ed 1 3 .2 3
st o ry p ro cesses n.A .,9 )
co nst ruct io n (5,n.A .,n.A .,n.A .,n.A .,
B uild ing , hall  b uild ing s Y es CH b uild ing , hall 1.6 7E+0 3 m2 est imat ed 1 3 .2 3
p ro cesses n.A .,9 )
co ncret e in
co nst ruct io n co ncret e, no rmal, at (5,n.A .,n.A .,n.A .,n.A .,
reinf o rced  co ncret e No CH 1.9 1E+0 1 m3 est imat ed 1 3 .2 3
mat erials p lant n.A .,9 )
co ncret e
st eel in
reinf o rcing st eel, at (5,n.A .,n.A .,n.A .,n.A .,
reinf o rced  met als ext ract io n No RER 1.3 2 E+0 4 kg est imat ed 1 3 .2 3
p lant n.A .,9 )
co ncret e
heavy machine, co nst ruct io n (5,n.A .,n.A .,n.A .,n.A .,
 machinery Y es RER heavy, unsp ecif ied , at 1.54 E+0 5 kg est imat ed 1 3 .2 3
unsp ecif ic p ro cesses n.A .,9 )
p lant
Po o r co ncret e co ncret e, so le p lat e
f o r ad d it io nal  co ncret e No CH and f o und at io n, at 6 .9 9 E+0 1 m3 est imat ed 1 3 .2 3
f o und at io n p lant
Conveyo r belt  machinery Y es RER co nveyo r b elt , at p lant 2.0 0 E+0 2 m est imat ed 1 3 .2 3
p ro cesses n.A .,9 )
reinf o rced d isp o sal, b uild ing ,
wast e b uild ing (5,n.A .,n.A .,n.A .,n.A .,
 co ncret e t o No CH reinf o rced co ncret e, t o 5.53 E+0 4 kg est imat ed 1 3 .2 3
manag ement d emo lit io n n.A .,9 )
d isp o sal f inal d isp osal
d isp o sal, b uild ing ,
f o und at io n
wast e b uild ing co ncret e, no t (5,n.A .,n.A .,n.A .,n.A .,
 co ncret e t o No CH 1.54 E+0 5 kg est imat ed 1 3 .2 3
manag ement d emo lit io n reinf o rced , t o f inal n.A .,9 )
d isp o sal
d isp o sal
scrap
Out p ut : 10 0 00 0 0 0 kg/ a, lif e t ime: scrap p rep arat io n
 p rep arat io n met als ext ract io n Y es RER 1.0 0 E+0 0 unit
50 a p lant
p lant
Fig. 5.23 Flows for "scrap preparation plant" and their representation in the ecoinvent database

5.5 Aluminium production mix

The share of primary and secondary metal on the world wide production of different aluminium alloys
is given in Tab. 5.17. These mixes are proposed to be used in attributional LCA, if the scope of the
study does not imply a special share of secondary material for a certain product. They are used as de-
fault within the ecoinvent database. In consequential LCA the value corrected substitution as proposed
in Werner (2003); Werner & Richter (2000) might be better suited for modelling the life cycles of
aluminium products.
15
Tab. 5.17 Production mix of "average" aluminium alloys
secondary aluminium secondary aluminium Share on produc-

Alloy primary aluminium
from new scrap from old scrap tion
Cast alloy 20 % 47 % 33 % 31 %
Wrought alloy 90 % 10 % 0% 69 %
Unspecific 68 % 22 % 10 % 100 %
5.5.1 Aluminium production mixes in ecoinvent

The background information to the processes are given in, flows are shown in Tab. 5.18:
- aluminium, production mix, cast alloy, at plant Fig. 5.24
- aluminium, production mix, wrought alloy, at plant Fig. 5.25
- aluminium, production mix, at plant Fig. 5.26
Data quality
The shares are estimates. Nevertheless they are believed to give a good representation of the overall
situation.
15
Dr. G. Rombach, Project manager Recycling, Hydro Aluminium Deutschland GmbH. Personal communication, 1.10.2002

Tab. 5.18 Meta information for the aluminium production mixes
aluminium, production aluminium, production

Name mix, cast alloy, at mix, wrought alloy, at aluminium, production
plant plant mix, at plant
Unit kg kg kg
Mix of primary and secon- Mix of primary and secon- Mix of primary and secon-
dary aluminium according dary aluminium according dary aluminium according
Included Processes
to their share on world- to their share on world- to their share on world-
wide production. wide production. wide production.
Amount 1 1 1
Aluminium, Produktions- Aluminium, Produktions-
Local Name mix, Gusslegierung, ab mix, Knetlegierung, ab Aluminium, Produk-
Werk Werk tionsmix, ab Werk
Synonyms
General Comment to
reference function
Start Date 2002 2002 2002
End Date 2002 2002 2002
Period 1 1 1
World-wide production mix World-wide production mix World-wide production mix
Geography text as proxy for European as proxy for European as proxy for European
consumption consumption consumption
Technology text Mix Mix Mix
Sampling Procedure Interview with expert Interview with expert Interview with expert
Extrapolations None None None
Uncertainty Adjustments None None None


Infra
Proc ess Cate Sub Lo c a Mo dul name in Mean So urc e mean Ty StDv General
Name go ry c atego ry tio n ec o invent value value pe 95% Co mment
ture
Dr. G. Ro mb ach,
Project manager
mix, cast alloy, at plant

Recycling, Hyd ro
p rimary
aluminium, production aluminium, p rimary, at A luminium
 met als ext ract io n No RER 2.0 0 E-0 1 kg 1 1.2 4 (3 ,1,1,1,1,5,4)
aluminium p lant Deut schland Gmb H.
Personal
co mmunicat ion,
1.10.20 0 2
seco nd ary aluminium, second ary,
aluminium f ro m  met als ext ract io n No RER f rom o ld scrap , at 3.3 0 E-0 1 kg 1 1.2 4 (3 ,1,1,1,1,5,4)
o ld scrap p lant
aluminium f ro m  met als ext ract io n No RER f rom new scrap, at 4.70 E-0 1 kg 1 1.2 4 (3 ,1,1,1,1,5,4)
new scrap p lant
aluminium mix aluminium, p ro duct io n
 met als ext ract io n No RER 1.0 0E+00 kg
f o r cast alloys mix, cast allo y, at p lant
Fig. 5.24 Flows for "aluminium, production mix, cast alloy, at plant" and their representation in the ecoinvent database

Infra
Proc ess Cate Sub Lo c a Mo dul name in Mean So urc e mean Ty StDv General
Name go ry c atego ry tio n ec o invent value value pe 95% Co mment
ture
Dr. G. Ro mb ach,
Project manager
mix, wrought alloy, at
aluminium, production
Recycling, Hyd ro
p rimary aluminium, p rimary, at A luminium
 met als ext ract io n No RER 9.0 0 E-0 1 kg 1 1.2 4 (3 ,1,1,1,1,5,4)
aluminium p lant Deut schland Gmb H.
Personal
plant
co mmunicat ion,
1.10.20 0 2
aluminium f ro m  met als ext ract io n No RER f rom new scrap, at 1.00 E-0 1 kg 1 1.2 4 (3 ,1,1,1,1,5,4)
new scrap p lant
aluminium mix aluminium, p ro duct io n
 f o r wroug ht met als ext ract io n No RER mix, wro ug ht allo y, at 1.0 0E+00 kg
alloy p lant
Fig. 5.25 Flows for "aluminium, production mix, wrought alloy, at plant" and their representation in the ecoinvent database


Infra
ture
Dr. G. Ro mbach,
Project manager
Recycling, Hydro
primary aluminium, p rimary, at A luminium

 met als ext ract io n No RER 6.8 0E-0 1 kg 1 1.24 (3,1,1,1,1,5,4 )
aluminium p lant Deut schland GmbH.
mix, at plant
Personal
communicat ion,
1.10.2 00 2
aluminium f rom  met als ext ract io n No RER f ro m old scrap , at 1.0 4E-0 1 kg 1 1.24 (3,1,1,1,1,5,4 )
old scrap p lant
aluminium f rom  met als ext ract io n No RER f ro m new scrap , at 2.16E-0 1 kg 1 1.24 (3,1,1,1,1,5,4 )
new scrap p lant
aluminium, p ro duct ion
 aluminium mix met als ext ract io n No RER 1.00 E+0 0 kg
mix, at plant
Fig. 5.26 Flows for "aluminium, production mix, at plant" and their representation in the ecoinvent database

6. Literature
6 Literature
www.ecoinvent.ch.
Altundogan et al. (2002) Altundogan H. S., Altundogan S., Tümen F. and Bildik M. (2002) Arsenic ad-
sorption from aqueous solutions by activated red mud. In: Waste Management,
22, pp. 357-363, Online-Version under: http://www.environmental-center.com/
magazine/elsevier/waste/art11.pdf.
Anonymous (2001) Anonymous (2001) Local Vegetable Fibres + Industrial and Mineral Wastes for
Composite Materials. Building Materials & Technology Promotion Council
BMTPC of the Ministry of Urban Development and Poverty Alleviation, New
Delhi. Online-Version under: http://www.bmtpc.org/fibre.pdf.
Doka (2003) Doka G. (2003) Life Cycle Inventories of Waste Treatment Services. Final re-
port ecoinvent 2000 No. 13. EMPA St. Gallen, Swiss Centre for Life Cycle In-
ventories, Dübendorf, CH, Online-Version under: http://www.ecoinvent.ch.
EAA (2000) EAA (2000) Environmental Profile Report for the European Aluminium Indus-
try. European Aluminium Association (EAA), Brussels.
Hudson et al. (1997) Hudson L. K., Misra C. and Wefers K. (1997) Aluminium Oxide. In: Ullmann's
encyclopedia of industrial chemistry (ed. Anonymous). 5th edition on CD-ROM
Edition. Wiley & Sons, London.
McLaughlin et al. (1998) McLaughlin M. J., Maier N. A., Correll R. L., Smart M. K. and Grant C. D.
(1998) In-situ immobilisation techniques to remediate cadmium-contaminated
agricultural soils. pp. 453-460 In: Contaminated Soil '98. Proceedings of the
Sixth International FZK/TNO, London. Online-Version under: http://www.sardi.
sa.gov.au/pages/horticulture/potato/insitu.htm:sectID=319&tempID=152.
Mori & Adelhardt (1998) Mori G. and Adelhardt W. (1998) Stoffmengenflüsse und Energiebedarf bei der
Gewinnung ausgewählter mineralischer Rohstoffe; Teilstudie Aluminium. In:
Geologisches Jahrbuch, Vol. Sonderhefte SH 2. Bundesanstalt für Geowissen-
schaften und Rohstoffe, Hannover. ISBN 3-510-95824-1.
NPI (1999) NPI (1999) Emission Estimation Technique Manual for Alumina Refining. In:
National Pollutant Inventory. Environment Australia, Online-Version under:
http://www.npi.gov.au/handbooks/approved_handbooks/pubs/falref.pdf.
Werner (2003) Werner F. (2003) Interdependencies Between LC-modelling And The Use Of
LCA In Product Design-related Decisions; With Special Emphasis On The In-
fluence Of Cognitive Models And Values On The Modelling Of Reuse & Recy-
cling And Other End-of-life Options. In: Diss. 14750. Eidg. Technische
Hochschule (ETH), Zürich.
Werner & Richter (2000) Werner F. and Richter K. (2000) Economic Allocation in LCA: A Case Study
About Aluminium Window Frames. In: International Journal for Life Cycle As-
sessment, 5(2), pp. 79-83.

6. Literature

Part II
Iron and Steel

Data v2.1 (2009)
Author: Hans-Jörg Althaus, EMPA, Dübendorf
Update Hans-Jörg Althaus, Mischa Classen, EMPA, Dübendorf
tories, Dübendorf, CH, Online-Version under: www.ecoinvent.ch.
Acknowledgement and Summary
Comments on version 2.1

This part corresponds to the original report in Althaus et al. (2004) that has been updated with the
changes reported in Frischknecht et al. (2006). The changes concerned emission factors for dioxine
(factor 1000 reduced in sinter, pellet and pig iron), PAH (factor 1000 lower in pellet) and CO2 (factor
1000 in pig iron), as electricity demand (factor 3.6 lower in cast iron, electric steel, both chromium
and un- / low alloyed) and coke consumption for pig iron (factor 28.6 higher).
In version 2.0, changes have been made to the following datasets:
 iron ore, 65% Fe, at beneficiation / GLO [kg]:
all water emissions decreased approx. a factor of 100
 sinter, iron, at plant / GLO [kg]:
PAH and PCB emissions decreased by a factor of 1000. VOC emissions increased slightly due
to decreased PAH and PCB.
 pig iron, at plant / GLO [kg]:
additional demand of 0.12 MJ natural gas,
higher waste heat generation (accordingly),
slightly higher transports due to increased coke consumption,
higher amount of waste water to treatment (factor 1000)
and higher CO2 emission (approx. factor 2; newly calculated from C-balance)
 steel, electric, un- and low-alloyed, at plant / RER [kg]
transports decreased by a factor of approx. 8
 steel, electric, chromium steel 18/8, at plant/ RER [kg]
 cast iron, at plant/ RER [kg]
Chapter “6 Cumulative Results and Interpretation” of the original report has not been adapted and was
deleted for this report. The flow names in the tables displaying unit process data have been updated
with the actual names of version 2.0. This affects the designation of namely the transportation service
flows that have been changed respective to the preceding version.
Comments on this version 2.1

This version corresponds to the version 2.0 with a changed composition of low alloyed steel.
Summary
The steel production is modelled for the European situation, primarily based on data from the description of the
current situation in the “Best Available Techniques Reference Document on the Production of Iron and Steel”
IPPC (2000) and on process descriptions.
The mining, beneficiation and partly the sinter and pellet production represent a world-wide situation primarily
based on Roth et al. (1999).
Some of the datasets described in this report make only sense in the context of the steel production and are there-
fore not intended for other uses.
The datasets to be chosen to represent iron or steel in a LCA study are either one of the primary (converter) steel
datasets (chapter 5.6), one of the secondary (electric) steel datasets (chapter 5.7), the cast iron (chapter 5.8) or
one of the production mixes (chapter 5.9). The decision whether to use primary, secondary or a mix of the two
depends on the goal and scope of the LCA to be conducted.

Table of Contents
Table of Contents
COMMENTS ON THIS VERSION 2.0 .....................................................................................I
SUMMARY ......................................................................................................................I
1 INTRODUCTION ....................................................................................................... 1

3.1 Iron ore .................................................................................................................................. 3
3.2 Sinter and Pellets ................................................................................................................... 3
3.3 Hot metal / Pig iron ............................................................................................................... 3
3.4 Sponge iron ........................................................................................................................... 3
3.5 Cast iron ................................................................................................................................ 3
3.6 Steel....................................................................................................................................... 4
3.6.1 Unalloyed steel...................................................................................................................... 4
3.6.2 Low-alloyed steel .................................................................................................................. 4
3.6.3 Stainless steel ........................................................................................................................ 5

4.1 Iron ore .................................................................................................................................. 6
4.2 Pig iron / hot metal ................................................................................................................ 6
4.3 Cast iron ................................................................................................................................ 6
4.4 Steel....................................................................................................................................... 6
4.4.1 Unalloyed and low-alloyed steel ........................................................................................... 6
4.4.2 Stainless steel ........................................................................................................................ 6

5.1 General overview .................................................................................................................. 7
5.2 Iron ore mining and beneficiation ......................................................................................... 7
5.2.1 Production process and infrastructure ................................................................................... 7
5.2.2 Emissions .............................................................................................................................. 8
5.2.3 Waste..................................................................................................................................... 8
5.2.4 Iron ore mining and beneficiation in ecoinvent..................................................................... 8
5.3 Agglomeration: sinter and pellet production....................................................................... 15
5.3.2 Emissions ............................................................................................................................ 15
5.3.3 Waste................................................................................................................................... 16
5.3.4 Transports............................................................................................................................ 16
5.3.5 Sinter and pellet production in ecoinvent............................................................................ 17
5.4 Blast furnace process: pig iron / hot metal production........................................................ 23
5.4.2 Emissions ............................................................................................................................ 24
5.4.3 Waste................................................................................................................................... 25
5.4.4 Pig iron / hot metal in ecoinvent ......................................................................................... 26
5.5 Making available of iron scrap............................................................................................ 32

Table of Contents
5.5.1 Iron scrap in ecoinvent ........................................................................................................ 32

5.6 Basic oxygen furnace (BOF) steel making.......................................................................... 35
5.6.2 Emissions ............................................................................................................................ 37
5.6.3 Waste................................................................................................................................... 37
5.6.4 Converter steel in ecoinvent................................................................................................ 39
5.7 Electric arc furnace (EAF) steel making ............................................................................. 52
5.7.2 Emissions ............................................................................................................................ 53
5.7.3 Waste................................................................................................................................... 54
5.7.4 Electric steel in ecoinvent ................................................................................................... 55
5.8 Cast iron .............................................................................................................................. 64
5.8.2 Emissions ............................................................................................................................ 64
5.8.3 Waste................................................................................................................................... 64
5.8.4 Cast iron in ecoinvent.......................................................................................................... 64
5.9 Consumer mixes for carbon steel, low-alloy steel and stainless steel ................................. 69
5.9.1 Steel in ecoinvent ................................................................................................................ 69
6 LITERATURE ......................................................................................................... 73

1. Introduction
1 Introduction
This part deals with primary and secondary iron and steel production. A rough overview of the cradle
to gate processes of the cast iron and steel production are given in Fig. 1.1.
Mining of
iron ore
Sorting
Benefication
Pellet feed Sinter feed
Transport to
Lump ore Pellets iron work
Sinter
Transport to iron
work
Direct Blast furnace Other melt-

reduction reduction reduction
Hot metal/
Sponge iron Pig iron
Casting
Pig iron
Transport to
steel work
Iron scrap
Transport to steel work

Electric steel Basic
furnace oxygenfurnace
Cast iron Steel
Fig. 1.1 Overview of cast iron and steel production (The grey materials and processes are of minor relevance and
not considered in the ecoinvent data)


Iron is the 4th most common element in the outer earth crust. It occurs in nature in compounds with
other elements. The economically most important iron minerals are hematite and magnetite.
Iron ore is mined mainly in China, Brazil and in Australia. (Tab. 2.1). The yearly iron ore production
was about 1*109 t/a in 1996 and the estimated reserves are about 150 *109 t. (Roth et al. (1999))
Tab. 2.1 World wide iron ore production in 1996 (Roth et al. (1999), appendix 2 & 4)
Western Eastern Asia Africa Northern South Australia World

Europe Europe America America
Iron ore [% of world 3% 12% 34% 5% 9% 22% 15% 100%
production]
Iron in ore [% of 4% 12% 25% 5% 11% 26% 17% 100%
world production]
Iron concentration in 57% 54% 39% 61% 64% 65% 63% 54%
ore [%]
Most of the iron ore is reduced to iron in blast furnaces. About 0.5 kg of coke and smaller amounts of
other fossil energy carriers and electricity are needed for the reduction.
Iron and steel scrap are also important resources because of the lower primary energy consumption
and lower CO2 emissions, compared to the blast furnace – converter production route. The share of
secondary iron on the European market is 34% according to Frischknecht et al. (1996) and the world
average of steel scrap used in steel production is about 40% Steffen (1997).


3.1 Iron ore
Iron ores are mixtures of up to 400 different minerals. The most important ones are the iron oxides and
sulphides. Economically interesting are those ores with high iron content. Two third of the ore mined
today come from banded iron formations (BIF). The average iron content of BIF's is between 15% and
45% but the more interesting ores contain up to 68% iron. Although the world average iron content in
the ores mined in the years 1975 to 1996 is 56% (calculated from Roth et al. (1999), appendices 1 - 4),
the iron content of raw ores used for the input- output tables in Roth et al. (1999) is only 46%. The au-
thors don't mention this discrepancy. It is likely that different methods of calculation are responsible
for it. In the ecoinvent inventories we use an iron content of 46% to be compatible with the flow data
given in Roth et al. (1999).
An enrichment of this ore to an iron content of 65% in world average is done at the mine. 13% of the
enriched ore are big lumps that can be fed to the blast furnace directly. The rest is more or less fine
grade ore that has to be agglomerated to sinter or pellets for use in the blast furnace.
3.2 Sinter and Pellets

As a rule, blast-furnace burden contains lump ore, sinter, and pellets. Pellets are made from both con-
centrate and iron-rich fine ore. If the beneficiation process yields very fine-grained concentrates (e.g.,
<0.1 mm), these are usually converted to pellets at the mine and transported in this form. Sinter is gen-
erally produced at the ironworks. The varying loads imposed on material during transport to the blast
furnace mean that pellets must satisfy stringent mechanical strength requirements. Pellets are also
nearly always made from one well-defined ore or concentrate, whereas sinter is produced from pre-
designed mixtures of ores and additives.
Sinter and pellets differ in the way the particles are bound into the agglomerate:
In pellets, the ore particles are connected chiefly by bridges of ore. The formation of iron oxide
bridges is supported by additional slag bridges; it must be remembered that the quantity of slag in ma-
ny pellet qualities is only 3 – 4 % (w/w) (corresponding to 6 – 7 % (v/v)).
Sinter consists of a calcium ferrite matrix, with siliceous and aluminous constituents dissolved and
primary hematite embedded in it. Whereas pellets are thoroughly oxidised (a maximum of 0.5 %
(w/w) FeO), sinter contains 4 – 6 % (w/w) FeO, which is linked to silica and aluminium oxide in the
matrix. (Meiler (1997))
3.3 Hot metal / Pig iron

Pig iron is the iron produced in the blast furnace. It is an alloy of 94% Fe and more than 2% C (world
average 4.5%). Liquid pig iron is also referred to as "hot metal".
3.4 Sponge iron

Sponge iron is produced by direct reduction of iron ore in the solid state. This process is only possible
with special ores with high iron content and low content on acid slag components. About 6% of the
total world pig iron and sponge iron production is sponge iron.
3.5 Cast iron

The term cast iron is used for unalloyed, low-alloy, and highly alloyed iron – carbon cast material with
carbon content ca. 2.8 – 4 %. The carbon is for the most part not chemically bonded to the iron, but is

present in elemental form. The properties of cast iron can be modified within wide limits by adding
elements such as nickel, chromium, manganese, copper, and silicon, which change the metallic struc-
ture. Corrosion-resistant and high-temperature grades are used, e.g., in the chemical industry, in fur-
naces, and in the automobile industry. The cryogenics industry uses low-temperature grades that do
not become brittle at temperatures far below 0 °C. Other important characteristics include its property
of damping vibration, its good resistance to corrosion by weathering, and its good machining proper-
ties. (Kucharcik (1997))
3.6 Steel
Steels are forgeable iron alloys with a carbon content of less than 2.1% (w/w). More than 90% of the
world iron production is used for steel making and more than 3500 different grades of steel are avail-
able today 1 .
The following iron-bearing materials are used for crude steel production:
- Iron melt from blast furnaces (hot metal), mainly used for steel production by the basic oxygen
furnace (BOF) process.
- Scrap, sponge iron, and solidified blast-furnace iron (pig iron), mainly used for melting steel in
electric furnaces.
According to the production process, steels can be classified as converter (BOF) or electric steel.
3.6.1 Unalloyed steel

Unalloyed steels or carbon steels are those in which the level of an individual element in no case rea-
ches or exceeds the limiting concentration given in Tab. 3.1 (Kalla & Kalwa (1997)). Typical unal-
loyed steels are the reinforcement steels used in the building industry.
Tab. 3.1 Limiting concentration of elements in unalloyed steel according to EN 10 020 Kalla & Kalwa (1997)
Element Al B Bi Co Cr 1 Cu 1 La Mn Mo 1 Nb 2
max. con- 3
0.10 0.0008 0.10 0.10 0.30 0.40 0.05 1.65 0.08 0.06
centration
Element Ni 1 Pb Se SI Te Ti 2 V2 W Zr 2 Others 4
max. con-
0.30 0.40 0.10 0.50 0.10 0.05 0.10 0.10 0.05 0.05
centration
1
If two, three or four of these elements are present in concentrations less than the maximum permit-
ted, their total concentration must not exceed 70% of the sum of the maximum
2
The same rule applies to these elements
3
If the manganese content is quoted as minimum, this value applies
4
Except C, N, O and S
3.6.2 Low-alloyed steel

Low-alloy steel contains more than the minimum given in Tab. 3.1 of at least one of the elements and
less than 5% alloying elements in total.
1
http://www.worldsteel.org/understanding/18; accessed 15.01.2003

For this project a composition of the average low-alloyed steel is derived from literature and from the
production volumes of the different alloying elements. Tab. 3.2 shows the composition and the sources
of the individual numbers.
Tab. 3.2 Composition of low-alloyed steel
Element concentration Source

Cr 2.15% Papp 1994 2 , Tab 19
Ni 1% Estimation
Mn 1.1% Jones 1994 3
Calculated from Mo content in recycled steel (26700 t Mo in 78 Mio t steel according
Mo 0.034%
to Sibley 2004 4
Estimated for this study. Mean value between
0.004%, calculated from Nb content in recycled steel (2900 t Nb in 78 Mio t steel ac-
Nb 0.066% cording to Sibley 2004) and
0.129%, calculated from Nb use in steel making (80% of 10000 t used in 6.19 mio t
steel according to USGS Mineral commodity summaries 2008 5 )
Calculated from W use in steel making (5200 t used in 6.19 mio t steel according to
W 0.084%
USGS Mineral commodity summaries 2008)
Calculated from V use in steel making (91% of 4030 t used in 6.19 mio t steel accord-
V 0.059%
ing to USGS Mineral commodity summaries 2008)
Cu 0% Assumption. Cu is not wanted in steel
3.6.3 Stainless steel

Stainless steel is the generic name for a number of different steels used primarily for their resistance to
corrosion. The one key element they all share is a minimum of 10.5% (w/w) of chromium. Although
other elements, particularly nickel and molybdenum, are added to improve corrosion resistance, chro-
mium is always the deciding factor. The vast majority of steel produced in the world is carbon and al-
loy steel, with the more expensive stainless steels representing a small, but valuable niche market 6 . Of
all the stainless steels the grade 304 steel (18/8 stainless, 18% chromium 8% nickel) is most com-
monly used. Typical applications for this grade are cutlery, food and chemical production infrastruc-
ture and containers and more and more architectural applications.
In this project we balance only 18/8 stainless steel.
2
Chromium Life Cycle Study. http://pubs.usgs.gov/usbmic/ic-9411/ic-9411.pdf; accessed 17.12.2008
3
Manganese Materials Flow Patterns. http://pubs.usgs.gov/usbmic/ic-9399/9399.pdf; accessed 17.12.2008
4
Flow Studies for Recycling Metal Commodities in the United States. http://pubs.usgs.gov/circ/2004/1196am/c1196a-
m_v2.pdf; Accessed 17.12.2008
5
http://minerals.usgs.gov/minerals/pubs/mcs/2008/mcs2008.pdf ; accessed 17.12.2008
6
http://www.worldstainless.org/what_basic.php; accessed 15.01.2003


4.1 Iron ore
Most of the iron ore is used for iron and steel production.
4.2 Pig iron / hot metal

Most of the pig iron is used for steel production before it is solidified. A smaller part is cast to ingots
that are used for cast iron production later.
4.3 Cast iron

About 40 % of cast iron is used in the automobile industry. Other applications include mechanical en-
gineering, the building industry (e.g. radiators, boilers, sanitary ware and pipes), chemical plant, ship-
building, and mining. The former demand by steelworks for cast iron ingot moulds has been greatly
diminished by the introduction of the continuous casting process. (Kucharcik (1997))
4.4 Steel
Steel is one of the most important materials in the industrialised world. It is widely used in all indus-
tries and in many products. The yearly world steel production has risen exponentially between 1940
and 1970 to about 700 million tonnes. Since then it has shown a slight increase to about 850 million
tonnes in the year 2000.
4.4.1 Unalloyed and low-alloyed steel

Unalloyed steel is mainly used as reinforcing steel in the building industry. There are many other ap-
plications in industrial machines, vehicles etc. for unalloyed and low-alloyed steel.
4.4.2 Stainless steel

Stainless steel is often used in the chemical and food industry. Again there are many other applications
for stainless steel.

5.1 General overview
A rough overview of the cradle to gate processes of the cast iron and steel production are given in Fig.
5.1.
Mining of
iron ore
Sorting
Benefication
Pellet feed Sinter feed
Transport to
Lump ore Pellets iron work
Sinter
Transport to iron
work
Direct Blast furnace Other melt-

reduction reduction reduction
Hot metal/
Sponge iron Pig iron
Casting
Pig iron
Transport to
steel work
Iron scrap
Transport to steel work

Electric steel Basic
furnace oxygenfurnace
Cast iron Steel
Fig. 5.1 Overview of cast iron and steel production (The grey materials and processes are of minor relevance and
not considered in the ecoinvent data)
5.2 Iron ore mining and beneficiation

This part is mainly based on (Roth et al. (1999)) and (Meiler (1997)).

Most of the iron ore today is gained from open mines. The thickness of the recoverable ore deposit
usually is in the range of 100 meters. The cap of the deposit is removed as overburden and deposited
nearby to be used for the recultivation when the mine has been depleted. The ore is won in stripes us-

ing hydraulic excavators and / or belt loaders. Drilling and blasting is needed in some mines with hard
ore.
The only processing steps used for ores mined from rich deposits (rich ores) are crushing and screen-
ing. Such ores come mainly from the weathering zones of itabirite deposits. At some mines, washing is
combined with the screening of fines (< 6 – 8 mm) for sintering; very fine material containing kaolin
is thus separated, and the aluminium oxide content is reduced. Low-grade ore is concentrated to lower
transport costs and bring about a chemical composition suitable for the subsequent reduction step
(lower silica content). The concentration of iron ore is generally based on physical differences be-
tween the iron-bearing and gangue minerals (spec. weight, magnetic properties).
Harmful components present in the ore must be removed. Such components include phosphorus, arse-
nic, chromium, copper, vanadium, alkali, sulphur and titanium; these either make metallurgical proc-
essing more difficult or degrade the iron and steel quality. A small quantity of titanium (<0.5 %) is,
however, added to the burden at some ironworks, because it has been found to increase the life of the
blast-furnace brickwork.
5.2.2 Emissions
Air
The only relevant direct emissions to air from iron ore mining and beneficiation are the dust emissions.
Water
Heavy metal and hydrocarbon emissions to water occur in the beneficiation process. They are esti-
mated according to the maximum values given in Anonymous (1998)
5.2.3 Waste
There is no relevant waste produced in the mining and beneficiation process. The overburden is used
in recultivation of the mine and not treated as waste.
5.2.4 Iron ore mining and beneficiation in ecoinvent

The background information to the processes are given in Tab. 5.1, flows are shown in:
- iron ore, 46% Fe, at mine Fig. 5.2
- mine, iron Fig. 5.3
- recultivation, iron mine Fig. 5.4
- iron ore, 65% Fe, at beneficiation Fig. 5.5
For a lack of information on this topic, the land use, the size and type of buildings and the weight of
the stationary machines used in the iron mines are estimated. The estimation is based on information
on other mines (gypsum, limestone).

Tab. 5.1 Meta information for the iron ore mining and beneficiation processes
iron ore, 46% Fe, recultivation, iron iron ore, 65% Fe, at
Name
at mine mine beneficiation mine, iron
Location GLO GLO GLO GLO
Unit kg m2 kg unit
Mining with its Infrastructure of
direct land use the mine (without
and transport of land use of the
Included Processes
the crude ore to Recultivation and open mine itself)
the enrichment the land trans- Milling and sorting of and of the en-
plant. formation. crude ore. richment plant.
Amount 1 1 1 1
Eisenerz, 46% Rekultivierung, Eisenerz, 65% Fe, ab
Local Name
Fe, ab Abbau Eisenmine Anreicherung Mine, Eisen
Synonyms
This ore has different
It is assumed, grain sizes. It can be
that the crude ore lump ore that can be
residues are used directly in the blast
used for reculti- furnace or it can be ore
vation. The same of smaller grain size that
General Comment to ref-
amount of energy is used for sinter and
erence function
used in building pellet production. Water These data are a
The dust emis- machines that is emissions are estimated rough assump-
sions are esti- used per kg according to the maxi- tion. Yearly out-
mated to be the crude ore is used mal levels for financial put is
same as in baux- per kg residue for support of mining activi- 40000000000 kg,
ite mining. recultivation. ties by the world bank. life Time 30 years
Start Date 1999 2000 1998 2000
End Date 2000 2000 2000 2000
Data Valid For Entire Pe-
riod 1 1 1 1
Other Period Text
Data relate to the Data relate to the global
Geography text
global average average
Data relate to the Milling and mechanical Buildings and
Technology text Data relate to recultivation of sorting. Average iron stationary ma-
open mining open mines yield is 84%. chines.
Production Volume 1.02E12 kg/a unknown 6.14E11 kg/a unknown
based on litera-
Sampling Procedure
ture estimation based on literature estimation
Extrapolations none none none none
Uncertainty Adjustments none none none none


Infra
Process Sub Loca Modul name in Mean Typ StDv General
Input Output Remarks Cate gory struc Unit Source mean value
Name category tion ecoinvent value e 95% Comment
ture
crude iron ore = 35.1 % in mined material,

iron in ore Fe = 46.19 % in iron ore --> 0.462 kg Fe in Iron, 46% in ore, 25% in
 resource in ground 4.62E-01 kg Roth et al. 1999 1 1.05 (1,2,1,1,1,1,12)
(resource) ore corresponds to 1 kg ore and to 2.85 kg crude ore, in ground
mined material
land use (mine) assumption that mine is depleated in 30 Occupation, mineral calculated from data for 1
 resource land 1.20E-04 m2a 1 1.54 (2,4,1,1,1,4,7)
[m2a] years. extraction site mine in Roth et al. 1999
land
mine of 4.8 km2 with an anual production Transformation, from calculated from data for 1
transformation  resource land 2.13E-06 m2 1 2.03 (2,4,1,1,1,4,8)
of 40 Mio. t/a. unknown mine in Roth et al. 1999
from unknown
land
mine of 4.8 km2 with an anual production Transformation, to calculated from data for 1
transformation  resource land 2.13E-06 m2 1 2.03 (2,4,1,1,1,4,8)
of 40 Mio. t/a. mineral extraction site mine in Roth et al. 1999
to mine
recultivation Assumption that 80% of mine area is
metals extraction No GLO recultivation, iron mine 1.70E-06 m2 estimated / calculated 1 3.05 (4,nA,1,1,1,nA,9)
iron ore, 46% Fe, at mine

mine recultivated
recycled water from mine --> not
water  1.15E-05 m3 Roth et al. 1999
inventoried
diesel, burned in
building  machinery No GLO 2.55E-02 MJ Roth et al. 1999 1 2.00 (1,2,1,1,1,1,5)
machine [MJ]
electricity, medium
electricity  electricity production mix No UCTE voltage, production 1.42E-03 kWh Roth et al. 1999 1 1.06 (1,2,1,3,1,1,2)
UCTE, at grid
blasting [kg construction
 civil engineering No RER blasting 2.71E-04 kg Roth et al. 1999 1 1.06 (1,2,1,3,1,1,1)
explosive] processes
includes infrastructure needed at mine
infrastructure estimation, based on
 (buildings, access roads,...) (exclusive metals extraction Yes GLO mine, iron 8.33E-13 unit 1 3.05 (4,nA,1,1,1,nA,9)
(mine) [unite] knowledge of other mines
building machines)
low population
 waste heat from electricity use air Heat, waste 5.13E-03 MJ calculated 1 1.06 (1,2,1,3,1,1,13)
density
low population estimated according to

 dust, coarse 50% of "dust" is > 10 um air Particulates, > 10 um 1.44E-03 kg 1 1.63 (2,5,1,3,3,3,25)
density bauxite mining
low population Particulates, > 2.5 um,

 dust, medium 45 % of "dust" is between 2.5 and 10 um air 1.30E-03 kg 1 2.10 (2,5,1,3,3,3,26)
density and < 10um
low population
 dust, fine 5 % of "dust" is below 2.5 um air Particulates, < 2.5 um 1.44E-04 kg 1 3.10 (2,5,1,3,3,3,27)
density
from mine, partly recycled --> not
 water 1.07E-04 m3 Roth et al. 1999
inventoried
the waste is used in the recultivation of
 waste 8.82E-04 kg Roth et al. 1999
the mine and thus not inventoried
unrefined iron
 46.19% Fe metals extraction No GLO iron ore, 46% Fe, at mine 1.00E+00 kg
ore
Fig. 5.2 Flows for "iron ore, 46% Fe, at mine" and its representation in the ecoinvent database


Infra
ture
diesel, burned in
recultivation, iron
the same amount per kg refilled material diesel, burned in (2,4,1,1,1,4,5); rough
building  No GLO 5.97E+03 MJ estimated 1 5.00
as per kg crude ore in mining building machine estimation
machine [MJ]
waste form mining and enrichment --> not
refill material  2.34E+05 kg
inventoried
land
mine
transformation Transformation, from (2,4,1,1,1,4,8); rough

 resource land 1.00E+00 m2 calculated 1 5.00
(from mine) mineral extraction site estimation
[m2]
land
transformation land occupation is included in the mining Transformation, to (2,4,1,1,1,4,8); rough
(to unknown) activity unknown estimation
[m2]
recultivation
 metals extraction No GLO recultivation, iron mine 1.00E+00 m2
mine
Fig. 5.3 Flows for "mine, iron" and its representation in the ecoinvent database


Infra
ture
diesel, burned in
building  recultivation, iron the same amount per kg refilled material
as per kg crude ore in mining
No GLO
diesel, burned in
building machine
5.97E+03 MJ estimated 1 5.00
(2,4,1,1,1,4,5); rough
estimation
machine [MJ]
waste form mining and enrichment --> not
refill material  2.34E+05 kg
inventoried
land
mine
transformation Transformation, from (2,4,1,1,1,4,8); rough
(from mine) mineral extraction site estimation
[m2]
land
transformation land occupation is included in the mining Transformation, to (2,4,1,1,1,4,8); rough
(to unknown) activity unknown estimation
[m2]
recultivation
 metals extraction No GLO recultivation, iron mine 1.00E+00 m2
mine
Fig. 5.4 Flows for "recultivation iron mine" and its representation in the ecoinvent database


Infra
ture
unrefined iron
 raw ore, 46.19% Fe metals extraction No GLO iron ore, 46% Fe, at mine 1.66E+00 kg Roth et al. 1999 1 1.05 (1,2,1,1,1,1,3)
ore
only fresh water. 1.329E-4 m3 recycled Water, unspecified
water  resource in water 1.52E-03 m3 Roth et al. 1999 1 1.05 (1,2,1,1,1,1,12)
beneficiation; part 1 water from mine is not inventoried. natural origin
electricity, medium
iron ore, 65% Fe, at
UCTE, at grid
infrastructure included in infrastructure of mine --> not

(refinery) [unite] inventoried here
low population
 waste heat from electricity use air Heat, waste 6.74E-02 MJ calculated 1 1.06 (1,2,1,3,1,1,13)
density
included in mining --> not inventoried
 dust
here
 water water itself not inventoried 1.16E-03 m3 Roth et al. 1999
water emissions based on maximum levels
estimated according to
to be achieved for world bank support of a
 oil and grease water unspecified Oils, unspecified 5.79E-06 kg world bank (Anonymous 1 3.18 (4,5,1,3,3,5,34)
project. Half of maximum assumed as
1998)
mean value
 cyanide water unspecified Cyanide 5.79E-07 kg 1 1.73 (4,5,1,3,3,5,33)
COD, Chemical Oxygen
 COD water unspecified 8.69E-05 kg 1 1.73 (4,5,1,3,3,5,32)
Demand
Fig. 5.5 Flows for "iron ore, 65% Fe, at beneficiation" and its representation in the ecoinvent database


Infra
ture
 As water unspecified Arsenic, ion 5.79E-08 kg 1 1.73 (4,5,1,3,3,5,33)
 Cd water unspecified Cadmium, ion 5.79E-08 kg 1 5.21 (4,5,1,3,3,5,35)
 Cr (VI) water unspecified Chromium VI 5.79E-08 kg 1 5.21 (4,5,1,3,3,5,35)

beneficiation; part 2
 Cu water unspecified Copper, ion 2.90E-07 kg 1 5.21 (4,5,1,3,3,5,35)
 Fe water unspecified Iron, ion 2.03E-06 kg 1 5.21 (4,5,1,3,3,5,35)
 Pb water unspecified Lead 1.16E-07 kg 1 5.21 (4,5,1,3,3,5,35)
 Hg water unspecified Mercury 5.79E-09 kg 1 5.21 (4,5,1,3,3,5,35)
 Ni water unspecified Nickel, ion 2.90E-07 kg 1 5.21 (4,5,1,3,3,5,35)
 Zn water unspecified Zinc, ion 1.16E-06 kg 1 5.21 (4,5,1,3,3,5,35)

BOD5, Biological Oxygen
 BOD calculated from COD, assuming C6H12O6 water unspecified 8.69E-05 kg calculated 1 1.73 (4,5,1,3,3,5,32)
Demand
TOC, Total Organic
 TOC water unspecified 3.14E-06 kg calculated 1 1.73 (4,5,1,3,3,5,32)
Carbon
DOC, Dissolved Organic
 DOC water unspecified 3.14E-06 kg calculated 1 1.73 (4,5,1,3,3,5,32)
Carbon
the waste is used in the recultivation of
 waste [kg] 6.58E-01 kg Roth et al. 1999
the mine and thus not inventoried
64.19 % Fe content. 13% as lump ore, 70% iron ore, 65% Fe, at
 refined iron ore metals extraction No GLO 1.00E+00 kg
to sinter and 27% to pelet fabrication beneficiation
Fig. 5.5 Flows for "iron ore, 65% Fe, at beneficiation" and its representation in the ecoinvent database (cont.)

Data quality
Data uncertainty is derived using the Pedigree matrix as described in (Frischknecht et al. (2003)) ex-
cept for the recultivation of the mine and for the mine infrastructure where this method would result in
values far too low.
Data for iron ore mining are of good overall quality although the infrastructure is roughly estimated.
Water emission data for beneficiation are based on threshold values and thus exhibit a greater uncer-
tainty.
5.3 Agglomeration: sinter and pellet production

This part is mainly based on information from (Meiler (1997)).

Very fine material is undesirable in the reduction equipment. Modern high-performance blast furnaces
require physical and metallurgical preparation of the burden. Agglomeration on one hand is used as a
process of size enlargement aimed at improving the permeability of the blast-furnace burden. On the
other hand slagging materials are introduced via the agglomerates.
As a rule, blast-furnace burden contains lump ore, sinter, and pellets. Pellets are made from both con-
centrate and iron-rich fine ore. If the beneficiation process yields very fine-grained concentrates (e.g.,
<0.1 mm), these are usually converted to pellets at the mine and transported in this form. Sinter is gen-
erally produced at the ironworks. The varying loads imposed on material during transport to the blast
furnace mean that pellets must satisfy stringent mechanical strength requirements. Pellets are also
nearly always made from one well-defined ore or concentrate, whereas sinter is produced from pre-
designed mixtures of ores and additives.
Sinter and pellets differ in the way the particles are bound into the agglomerate. Pellets are burned at a
temperature below the melting points of the constituents. Their strength gain is adequate only when
the reaction temperature is controlled to ±10 °C over long time periods. For this reason, hot gas is used
to supply heat; it can be generated by the combustion of fuel gas, oil, or even coal. In sintering, on the
other hand, the constituents of the mixture are melted. This happens at temperatures so high that the
reaction itself goes to completion within minutes, but in a thin layer that continuously advances
through the material bed concurrently to the gas.
In pelletising spherical green pellets (green balls) of appropriate size (ca. 9 – 16 mm in diameter) are
first formed by a pelletising disk or drum. Then the pellets are thermally hardened of at 1200 – 1320
°C.
Iron-ore sinters are virtually all produced in travelling-grate installations. Along with ore and concen-
trate, the mixture for sintering also contains recycle material from the ironworks, provided no build-up
of undesirable or detrimental substances results.
Coke is the dominant sinter plant energy input (about 85%), with electricity and gas (COG and/or BF
gas and/or natural gas) supplying the remainder in equal shares (IPPC (2000)).
5.3.2 Emissions
Air
Sintering and pelletising leads to the emission of dust, various heavy metals and hydrocarbons, hydro-
gen halogenides, NOx, SOx and CO2 from the burning of the fuel gas and some of the coke. These pro-

cesses are also a major source of dioxide emissions. The dioxin emission value for the sinter produc-
tion stems from (Anonymous (2002)).
Dust grain size analysis before treatment shows two distinctive peaks. The first and higher peak is at
about 1 m while the second one is at 100 m (IPPC (2000)). The abatement used in sinter plants has
a high efficiency in removing the larger particles while the smaller alkali and lead chlorides are hard to
remove in the electrostatic precipitators due to their high resistivity. Thus a particle size of < 2.5 m is
assumed for the dust emission after abatement.
Water
The water used in the agglomeration processes evaporates. Thus no relevant water emissions occur.
5.3.3 Waste
Normally all solid wastes originating from the agglomeration plant (dusts from de-dusting devices and
sieving fractions) are recycled to the strand. However, in sinter plants there can be two exceptions:
The first concerns sludge from wet waste gas treatment systems, which is put to landfill. At present,
there are only two plants in Europe operating such a system. The second exception is filter dust from
the last field of electrostatic precipitators (ESP). Most European sinter plants are operated with fully
closed dust cycles. However, some plants exclude fine dust from the last field of the ESP. This dust
mainly consists of alkali and metal chlorides. This so-called partly open filter dust cycle is carried out
in order to improve the operation of ESP or (in one case) of the bag filter, or to reduce alkali and metal
chloride emissions. (IPPC (2000))
In this project we neglect the wastes to landfill because:
- their amount is not known (but small)
- their composition is not known
- their impact is well below the uncertainty of the impact of the air emissions
5.3.4 Transports
Sinter usually is produced near the blast furnace while pellets are produced near the beneficiation
plant. This is because pellets are made of fraction of the ore so fine that it would ask for elaborate
technologies for transportation (especially loading and unloading).
Masses for transportation of the ore to the sinter plant and of the lump ore and pellets to the blast fur-
nace are estimated according to the supply and demand situation in the different regions of the world.
It is assumed, that as much ore as possible is used within these regions. Thus Western Europe and Asia
are net importers while Eastern Europe / GUS, South America, Australia and Africa are net exporters.
North America produces almost exactly as much iron ore as it consumes. An export mix is generated
according to the export capacity of the net exporters and it is assumed that the imports of the net im-
porters correspond to this mix. Distances within the regions are neglected and those between the re-
gions and the means of transpiration are assumed. Data are given in Tab. 5.2.

Tab. 5.2 Transport masses and distances for iron ore and pellets
Region Iron ore pro- Pig iron pro- Ore needed for Import Export
duction 1995 duction 1995 pig iron produc- demand compo-
tion 1995 total sition
Mio t % of world Mio t % of world Mio t % of world Mio t %
West Europe 34 3.4 125 23.8 211 23.8 177
East Europe /GUS 131 13.1 60 11.3 100 11.3 -31 9.1
Asia 352 35.1 235 44.5 394 44.5 43
Africa 50 4.9 9 1.6 14 1.6 -35 10.4
North America 98 9.7 59 11.3 100 11.3 2
South America 193 19.3 32 6.1 54 6.1 -139 41.2
Australia 145 14.5 7 1.4 13 1.4 -133 39.3
Import Import [Mio t] Distance Train Distance ship Distance ship

[km] (ocean) [km] (coastal) [km]
to from East Europe /GUS
West Europe 16 2500 0 0
Asia 4 1000 14366.761 0
to from Africa
West Europe 18 1000 6009.615 173.772
Asia 4 1000 17444.778 0
to from South America
West Europe 73 1000 8495.52 173.772
Asia 18 1000 18344.209 0
to from Australia
West Europe 70 1000 15526.85 173.772
Asia 17 1000 7108.562 0
5.3.5 Sinter and pellet production in ecoinvent

- sinter, iron, at plant Fig. 5.6
- pellets, iron, at plant Fig. 5.7

Where (IPPC (2000)) is given as the source, arithmetic means of the low and high values are used in
this project.
No data on the infrastructure are available. Thus the aluminium oxide plant (calcination plant) was
chosen to represent the sinter and the pelletising plant.

Tab. 5.3 Meta information for the iron ore agglomeration processes
Name sinter, iron, at plant pellets, iron, at plant

Location GLO GLO
Unit kg kg
Blending, mixing and sintering. Blending, mixing and sintering.
Included Processes
Emissions are abated Emissions are abated
Amount 1 1
Local Name Sinter, Eisen, ab Werk Pellets, Eisen, ab Werk
Synonyms
Inputs and air emissions from dif- Inputs and air emissions from dif-
General Comment to reference ferent sources. Transports of iron ferent sources. No transport of iron
function ore calculated according to informa- ore because pellets are fabricated
tion about exports and imports. at mine
End Date 2002 2002
Data Valid For Entire Period 1 1
Other Period Text
Inputs relate to the global average. Inputs relate to the global average.
Geography text
Emissions relate to European plants Emissions relate to European plants
Down draft sintering on continuous
Technology text
travelling grates Strate grate and grate kiln process
Production Volume 3.83E11 kg/a 1.54E11 kg/a
Main in- and outputs based on Main in- and outputs based on
world average, emissions based on world average, emissions based on
Sampling Procedure
some plants in the EU. Everything some plants in the EU. Everything
based on literature. based on literature.
Uncertainty Adjustments none none
Data quality
cept for the infrastructure where this method would result in values far too low.
Data of input and main output materials for sintering and pelletising are of good overall quality.
Emission data for pelletising stem from all 5 pelletising plants in Europe. These 5 plants produced a-
bout 15 Mt of the 35 Mt pellets used in Europe in 1995. Thus data for European consumption are of
rather high uncertainty.
Emission data for sintering stem from 5 of the 47 sinter plants in Europe. Thus data for European con-
sumption are of rather high uncertainty.
A rough proxy is used for the infrastructure.


Infra
ture
refined iron ore iron ore, 65% Fe, at

 metals extraction No GLO 1.05E+00 kg Roth et al. 1999 1 1.05 (1,2,1,1,1,1,3)
[kg] beneficiation
transport ship refined ore from mine to plant by ship transport transport, transoceanic
 ship No OCE 2.84E+00 tkm estimated 1 2.01 (2,nA,1,1,1,3,5)
(ocean) (ocean) systems freight ship
transport ship refined ore from mine to plant by ship transport
 ship No RER transport, barge 3.15E-02 tkm estimated 1 2.01 (2,nA,1,3,1,3,5)
(costal) (costal) systems
transport
transport rail  refined ore from mine to plant by rail train No RER transport, freight, rail 2.89E-01 tkm estimated 1 2.01 (2,nA,1,3,1,3,5)
systems
construction quicklime, in pieces,
lime  additives No CH 5.00E-02 kg Roth et al. 1999 1 1.06 (1,2,1,3,1,1,3)
materials loose, at plant
transport
sinter, iron, at plant; part 1
transport rail  limestone & coke: 200 km train No RER transport, freight, rail 2.00E-02 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
systems
transport transport, lorry >16t,
transport road  20 km road No RER 2.00E-03 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
systems fleet average
natural gas, COG and BFgas are used, only
fuel gas for consumption of natural gas is inventoried natural gas, high
 natural gas fuels No RER 3.63E-02 MJ Roth et al. 1999 1 1.06 (1,2,1,3,1,1,1)
heating with the assumption that half of the pressure, at consumer
energy is supplied by natural gas
compressed air  not inventoried 3.00E-03 Nm3 Roth et al. 1999
suction air  not inventoried 2.00E+00 Nm3 Roth et al. 1999
coke chippings  28.6 MJ/kg hard coal fuels No RER hard coal coke, at plant 1.43E+00 MJ Roth et al. 1999 1 1.06 (1,2,1,3,1,1,3)
Water, unspecified
natural origin
includes power consumed for air electricity, medium
electricity  compression and suction and for exhaust electricity production mix No UCTE voltage, production 1.00E-02 kWh Roth et al. 1999 1 1.06 (1,2,1,3,1,1,2)
gaz cleaning UCTE, at grid
infrastructure (5,5,nA,nA,nA,nA,9);
 aluminium oxide plant used as proxy metals extraction Yes RER aluminium oxide, plant 2.50E-11 unit estimated 1 5.00
(plant) [piece] rough estimation
from electricity fuel gas and coke use
 waste heat air unspecified Heat, waste 1.54E+00 MJ calculated 1 1.06 (1,2,1,3,1,1,13)
(assumption all coke is burned)
 dust, fine all "dust" is below 2.5 um air unspecified Particulates, < 2.5 um 2.06E-04 kg calculated after IPPC 2000 1 3.01 (2,3,2,3,1,3,27)
 Cd air unspecified Cadmium 1.93E-08 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,22)
 Cr air unspecified Chromium 2.52E-08 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,22)
 Cu air unspecified Copper 7.66E-08 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,22)
 Hg air unspecified Mercury 7.57E-08 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,22)
Fig. 5.6 Flows for "sinter, iron, at plant" and its representation in the ecoinvent database


Infra
ture
 Mn air unspecified Manganese 1.93E-07 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,22)
 Ni air unspecified Nickel 1.93E-08 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,22)
 Pb air unspecified Lead 3.23E-06 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,22)
 Ti air unspecified Titanium 1.60E-08 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,22)
 V air unspecified Vanadium 1.15E-08 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,22)
sinter, iron, at plant; part 2
 Zn air unspecified Zinc 8.26E-07 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,22)
 HCl air unspecified Hydrogen chloride 3.76E-05 kg calculated after IPPC 2000 1 1.52 (2,3,2,3,1,3,31)
 HF air unspecified Hydrogen fluoride 2.25E-06 kg calculated after IPPC 2000 1 1.52 (2,3,2,3,1,3,31)
 NOx air unspecified Nitrogen oxides 5.27E-04 kg calculated after IPPC 2000 1 1.52 (2,3,2,3,1,3,16)
 SO2 air unspecified Sulfur dioxide 1.26E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,15)
 CO air unspecified Carbon monoxide, fossil 2.57E-02 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,17)
 CO2 air unspecified Carbon dioxide, fossil 2.04E-01 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,14)
Hydrocarbons, aliphatic,
 VOC VOC without PAH and PCB air unspecified 1.38E-04 kg calculated after IPPC 2000 1 1.52 (2,3,2,3,1,3,16)
alkanes, unspecified
PAH, polycyclic aromatic

 PAH air unspecified 4.72E-10 kg calculated after IPPC 2000 1 3.01 (2,3,2,3,1,3,21)
hydrocarbons
Polychlorinated
 PCB air unspecified 6.42E-12 kg calculated after IPPC 2000 1 3.01 (2,3,2,3,1,3,21)
biphenyls
Quass gives a range of 1 to 10 ng / m3 flue Dioxins, measured as
gas (-->2.1 to 21 ng/kg sinter). The 2,3,7,8-
 dioxin (I-TEQ) air unspecified 7.00E-12 kg Anonymus 2002 1 3.01 (1,3,2,3,1,2,21)
European Dioxine Inventory uses 7 ng/kg tetrachlorodibenzo-p-
sinter as average for Europe) dioxin
totally recycled in the wast majority of
 dust to landfill plants. The small part that is not recycled
in few plants is neglected
applies only for two plants that have a
 sludge
wet scrubber --> neglected.
 sinter metals extraction No GLO sinter, iron, at plant 1.00E+00 kg
Fig. 5.6 Flows for "sinter, iron, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture

refined iron ore  metals extraction No GLO 1.05E+00 kg Roth et al. 1999 1 1.05 (1,2,1,1,1,1,3)
beneficiation
construction
bentonite  additives No DE bentonite, at processing 4.00E-02 kg Roth et al. 1999 1 1.06 (1,2,1,3,1,1,4)
materials
transport
transport rail  bentonite: 600 km; coal: 200km train No RER transport, freight, rail 2.41E-02 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
systems
pellets, iron, at plant; part 1

transport road  bentonite: 100 km; coal: 20km road No CH 4.01E-03 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
hard coal mix, at
coal dust  28.6 MJ/kg hard coal fuels No UCTE 1.30E-02 kg Roth et al. 1999 1 1.06 (1,2,1,3,1,1,11)
regional storage
Water, unspecified
natural origin
generic heat module is inventoried natural gas, burned in
fuel gas for
 (because no detailed information about natural gas heating systems No RER industrial furnace 6.90E-02 MJ Roth et al. 1999 1 1.06 (1,2,1,3,1,1,11)
heating
emissions is available) >100kW
electricity, medium
UCTE, at grid
infrastructure (5,5,nA,nA,nA,nA,9);
 aluminium oxide plant used as proxy metals extraction Yes RER aluminium oxide, plant 2.50E-11 unit estimated 1 5.00
(plant) [piece] rough estimation
from electricity fuel gas and coal use
 waste heat air unspecified Heat, waste 5.31E-01 MJ calculated 1 1.06 (1,2,1,3,1,1,13)
(assumption all coal is burned)
 dust, fine all "dust" is below 2.5 um air unspecified Particulates, < 2.5 um 7.50E-05 kg calculated after IPPC 2000 1 3.01 (2,3,2,3,1,3,27)
Fig. 5.7 Flows for "pellets, iron, at plant" and its representation in the ecoinvent database


Infra
ture
pellets, iron, at plant; part 2
 V air unspecified Vanadium 8.55E-08 kg calculated after IPPC 2000 1 5.01 (2,3,2,3,1,3,22)
Hydrocarbons, aliphatic,
 VOC VOC without PAH air unspecified 2.25E-05 kg calculated after IPPC 2000 1 1.52 (2,3,2,3,1,3,16)
alkanes, unspecified

hydrocarbons
Dioxins, measured as
2,3,7,8-
 dioxin (I-TEQ) air unspecified 5.70E-15 kg calculated after IPPC 2000 1 3.01 (2,3,2,3,1,3,21)
tetrachlorodibenzo-p-
dioxin
pellets, iron, at
 metals extraction No GLO pellets, iron, at plant 1.00E+00 kg
plant [kg]
Fig. 5.7 Flows for "pellets, iron, at plant" and its representation in the ecoinvent database (cont.)

5.4 Blast furnace process: pig iron / hot metal production

This part is mainly based on Roth et al. (1999) and IPPC (2000).

The blast furnace remains by far the most important process for the production of pig iron. The tech-
nique is likely to continue to dominate pig iron production for at least the next 20 years.
A blast furnace is a closed system into which iron bearing materials (iron ore lump, sinter and/or pel-
lets), additives (slag formers such as limestone) and reducing agents (coke) are continuously fed from
the top of the furnace shaft through a charging system that prevents escape of blast furnace gas (BF
gas). Fig. 5.8 shows a simplified layout of a blast furnace consisting of the furnace itself, the cast
house, the hot stoves and two-stage treatment of BF gas.
Fig. 5.8 Simplified scheme of a blast furnace (IPPC (2000))
No data have been found on the infrastructure needed for the process. Assumptions made in this pro-
ject are based on different pictures and information on the aluminium production infrastructure (see
Tab. 5.4).
The infrastructure is modelled to be used for 50 years. The replacement of the machines with a life-
time of only 25 years is taken into account by doubling the amount. Thus the values given in Tab. 5.5
are inventoried. For the waste disposal it is assumed that the conveyer belts and the machines are
completely recycled. The disposal of the buildings is already inventoried in the building modules.
Thus only the concrete and the refractory materials are disposed.

Tab. 5.4 Assumptions for infrastructure of blast furnaces
blast furnace amount unit lifetime [a] Material / Module

yearly production capacity 1'500'000'000 kg 50 Yearly production capacity
area of plant 976'000 m2 50 Transformed area
administration building (50x50x10 m3) 25'000 m3 50 Building, multi story
concrete for piers [m3] 1'200 m3 50 reinforced concrete
grab for unloading [kg] 100'000 kg 25 heavy machine, unspecific
conveyor to storage [m] 900 m 25 conveyor belt
silo (40000 m3, height 62 m, radius 14m) 233'986 kg 25 Steel low alloyed
furnace & cowper, concrete 30 m3 25 reinforced concrete
furnace & cowper, refractory materials 100'000 kg 25 firebrick
cast house 10'000 m2 50 Building, hall
casting machines 10'000'000 kg 25 heavy machine, unspecific
storage house 9'500 m2 50 Building, hall
Tab. 5.5 Infrastructure of the blast furnace
blast furnace Unit Remarks

Yearly production capacity 1'500'000'000 kg
Transformed area 976'000 m2
Building, multi story 25'000 m3
Building, hall 19'500 m2
reinforced concrete 1'260 m3
concrete in reinforced concrete (91 % (v/v)) 1'147 m3
steel in reinforced concrete (9 % (v/v)) 793'800 kg
heavy machine, unspecific 20'200'000 kg
Poor concrete for additional foundation 9'182 m3 same weight as machines
conveyor belt 1'800 m
Steel low alloyed 467'972 kg
firebrick 200'000 kg
By-products of the pig iron production in blast furnaces are:
- Slag that is used in road building, concrete aggregate, thermal insulation (mineral wool) and as a
cement replacement. Total utilisation of blast furnace slag is a target which has already been met
in many cases.
- BF gas that contains about 20-28% CO, 1–5% H2, inert compounds (50-55% N2, 17-25% CO2),
some sulphur and cyanide compounds and large amounts of dust from the burden. After treatment
and enrichment, BF gas can be used in the hot blast stove or e.g. in the sinter production.
- Electricity recovered of the top gas pressure energy. Many plants with adequate top gas pressure
(> 1.5 bar gauge) and sufficient pressure drop are suitable for installation of expansion turbines.
5.4.2 Emissions
Air
Relevant direct air emissions from the pig iron production are CO, CO2, nitrogen oxides, sulphur ox-
ides and heavy metals. The casting is responsible for most of the direct dust emissions.

CO2 emission given in IPPC (2000) are considerably lower than the calculated C-balance suggests
(Tab. 5.6). The calculated value from the C-balance (0.849 kg/kg) is used in ecoinvent v2.0 Data.
Tab. 5.6 C-balance of blast furnace process
Flow comment Amount Unit kg C/ unit flow kg C/ kg pig iron

coke 0.34 kg/kg @ 28.6 MJ/kg 9.72E+00 MJ 0.031 0.306
coal 1.50E-01 kg 0.900 0.135
energy content = 3 MJ/Nm3;
BF gas 9.00E-01 Nm3 0.241 0.217
Input
C content = 0.241kg/Nm3
coke oven energy content = 3 MJ/Nm3;
1.20E-02 Nm3 0.241 0.003
gas C content = 0.241kg/Nm3
natural
0.003 Nm3 @ 40 MJ/Nm3 1.20E-01 MJ 0.020 0.002
gas
CO 1.34E-03 kg 0.429 0.001
CO2 8.49E-01 kg 0.273 0.232
Output
pig iron 1.00E+00 kg 0.045 0.045

energy content = 3 MJ/Nm3;
BF gas 1.60E+00 Nm3 0.241 0.386
C content = 0.241kg/Nm3
Water
There are no relevant direct water emissions from these processes
5.4.3 Waste
Solid waste
Several different waste streams are generated. Some of them are recycled or used for different pur-
poses and thus not inventoried. Others can not be used due to a high contamination and are deposited
in landfills.
The sludge from BF gas treatment can't be reused because the zinc and lead compounds are mainly
separated in the scrubber thus leading to a high contamination of the sludge. Sludge composition data
are average data from IPPC (2000). The difference to 100 % is assumed to be water. (see Tab. 5.7).
There is no representative information available on the fate of the rubble waste/by-product. In some
steelworks it is partly recycled to the BOF or reused for the production of new refractory material. In
other cases it is put to landfill.

Tab. 5.7 Typical composition in [% (w/w)] of sludge from BF gas treatment (IPPC (2000), Tab. 7.6)
(IPPC (2000), Tab. 7.6) This project

Min Max
C 15 47 31
Fe 7 35 21
Pb 0.8 2 1.4
Zn 0.1 10 5.05
Mn 0.12 0.14 0.13
Al2O3 0.8 4.6 2.7
S 2.4 2.5 2.45
SiO2 3 9 6
P2O5 0.1 0.44 0.27
CaO 3.5 18 10.75
MgO 3.5 17 10.25
Na2O 0.15 0.24 0.195
K2O 0.08 0.36 0.22
water 8.585
Waste water
Overflow of water from slag granulation primarily depends on water availability and is in the range
0.125 – 10 m3/t pig iron produced. Information on chemical composition from one German plant is
given in Tab. 5.8 (IPPC (2000), Tab 7.8).
Tab. 5.8 Waste water composition from pig iron production (IPPC (2000), Tab 7.8)
Parameter Unit Number of Average con- Median con- Emission Unit

measure- centration centration factor
ments
Flow [m3/d] 16000 1.96 [m3/t]
Pb [µg/l] 19 3.45 2.0 3.92 [mg/t pig iron]
Cr [µg/l] 19 3.32 2.0 3.92 [mg/t pig iron]
Cu [µg/l] 19 5.99 4.20 8.232 [mg/t pig iron]
Zn [µg/l] 19 37.63 20.00 39.2 [mg/t pig iron]
Ni [µg/l] 19 4.91 3.00 5.88 [mg/t pig iron]
TOC [mg/l] 19 4.40 4.10 8.036 [g/t pig iron]
COD [mg/l] 19 20.62 20.30 39.788 [g/t pig iron]
Emission factor is based on the median
5.4.4 Pig iron / hot metal in ecoinvent

The background information to the processes is given in Tab. 5.9, flows are shown in:
- pig iron, at plant Fig. 5.9
- blast furnace Fig. 5.10

Where (IPPC (2000)) is given as the source, arithmetic means of the low and high values are used in
this project.
Transport distances for iron ore and pellets are shown in Tab. 5.2.

Data quality
Data uncertainty is derived using the Pedigree matrix as described in Frischknecht et al. (2003) except
for the infrastructure where this method would result in values far too low.
Data of input and main output materials for pig iron production are of good overall quality.
Emission data for pig iron production stem from 4 of 81 blast furnaces in Europe. Thus data for Euro-
pean consumption are of rather high uncertainty.
Data for the infrastructure is roughly estimated.
Tab. 5.9 Meta information for the blast furnace processes
Name pig iron, at plant blast furnace

Location GLO RER
Unit kg unit
Infrastructure for input material
unloading and storage, blast fur-
Included Processes
Blast furnace process. Emissions nace processes and casting and for
are abated administrative buildings
Amount 1 1
Local Name Roheisen, ab Werk Hochofen
Synonyms hot metal
Inputs and air emissions from dif- Assumptions for medium sized
General Comment to reference ferent sources. Transports of iron plant. Output: 1'500'000 t/a, life
function ore calculated according to informa- time: 50 years. Site is shared with
tion about exports and imports. BOF converter.
End Date 2002 2002
Other Period Text
Inputs relate to the global average.
Geography text Emissions relate to European Some Swiss datasets are used for
plants European processes.
Assumed technology of medium
Technology text
Blast furnace process sized plant.
Production Volume 5.30E11 kg/a unknown
Main in- and outputs based on
world average, emissions based on
Sampling Procedure
some plants in the EU. Everything
based on literature. Internet search


Infra
ture
sinter  metals extraction No GLO sinter, iron, at plant 1.05E+00 kg Roth et al. 1999, Tab 24 1 1.05 (1,2,1,1,1,1,3)
pellets  metals extraction No GLO pellets, iron, at plant 4.00E-01 kg Roth et al. 1999, Tab 24 1 1.05 (1,2,1,1,1,1,3)
lump ore  metals extraction No GLO 1.50E-01 kg Roth et al. 1999, Tab 24 1 1.05 (1,2,1,1,1,1,3)
beneficiation
transport ship pellets and lump ore from mine to plant transport transport, transoceanic
 ship No OCE 1.49E+00 tkm estimated 1 2.01 (2,nA,1,1,1,3,5)
(ocean) by ship (ocean) systems freight ship
transport ship pellets and lump ore from mine to plant transport
 ship No RER transport, barge 1.65E-02 tkm estimated 1 2.01 (2,nA,1,3,1,3,5)
(costal) by ship (costal) systems
pellets and lump ore from mine to plant transport
transport rail  train No RER transport, freight, rail 1.51E-01 tkm estimated 1 2.01 (2,nA,1,3,1,3,5)
by rail systems
construction
additives  limestone, according to IPPC 2000 additives No CH limestone, at mine 1.00E-02 kg Roth et al. 1999, Tab 24 1 1.06 (1,2,1,3,1,1,4)
materials
refractory construction refractory, fireclay,
 bricks No DE 2.00E-03 kg Roth et al. 1999, Tab 24 1 1.06 (1,2,1,3,1,1,4)
materials materials packed, at plant
pig iron, at plant; part 1
slag  almost total utilisation --> not inventoried 2.70E-01 kg Roth et al. 1999, Tab 24
Water, unspecified
water  resource in water 6.00E-03 m3 Roth et al. 1999, Tab 24 1 1.05 (1,2,1,1,1,1,12)
natural origin
coke  0.34 kg/kg @ 28.6 MJ/kg hard coal fuels No RER hard coal coke, at plant 9.72E+00 MJ Roth et al. 1999, Tab 24 1 1.06 (1,2,1,3,1,1,11)
hard coal mix, at
coal  hard coal fuels No UCTE 1.50E-01 kg Roth et al. 1999, Tab 24 1 1.06 (1,2,1,3,1,1,11)
regional storage
not inventoried because the same amount
electricity total  1.00E-01 kWh Roth et al. 1999, Tab 24 1 1.06 (1,2,1,3,1,1,11)
is generated (cf. outputs)
steam  not inventoried as internally generated 3.00E-02 kg Roth et al. 1999, Tab 24
BF gas Output higher than input 9.00E-01 Nm3 Roth et al. 1999, Tab 24
coke oven gas cut off from coke oven 1.20E-02 Nm3 Roth et al. 1999, Tab 24
natural gas, high
natural gas 0.003 Nm3 @ 40 MJ/Nm3 natural gas fuels No RER 1.20E-01 MJ Roth et al. 1999, Tab 24 1 1.06 (1,2,1,3,1,1,11)
pressure, at consumer
limestone, refractories, coal & coke: 20 transport transport, lorry >16t,
transport road  road No RER 1.00E-02 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
km; systems fleet average
limestone, refractories, coal & coke: 200 transport
transport rail  train No RER transport, freight, rail 1.00E-01 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
km; systems
infrastructure
 metals extraction Yes RER blast furnace 1.33E-11 unit estimated 1 3.23 (5,nA,nA,nA,nA,nA,9)
(blast furnace)
from burning of coke, coal and natural
gas. Heat from burning of BFgas and from
electricity use is not included because net
energy consumption is zero or negative
5% > 10 um, 5% between 2.5 and 10 um
 dust, coarse air unspecified Particulates, > 10 um 1.60E-06 kg calculated after IPPC 2000 1 1.52 (2,3,2,3,1,3,25)
and 90% < 2.5 um (CEPMEIP database)
Particulates, > 2.5 um,
 dust, medium air unspecified 1.60E-06 kg calculated after IPPC 2000 1 2.01 (2,3,2,3,1,3,26)
and < 10um
 dust, fine air unspecified Particulates, < 2.5 um 2.87E-05 kg calculated after IPPC 2000 1 3.01 (2,3,2,3,1,3,27)
Fig. 5.9 Flows for "pig iron, at plant" and its representation in the ecoinvent database


Infra
ture
 SOx air unspecified Sulfur dioxide 1.33E-04 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,15)
 H2S air unspecified Hydrogen sulfide 1.07E-05 kg calculated after IPPC 2000 1 1.52 (2,3,2,3,1,3,31)
Average given in IPPC 2000 = 0.415 kg/kg.

pig iron, at plant; part 2
 CO2 air unspecified Carbon dioxide, fossil 8.49E-01 kg calculated from C-balance 1 1.11 (2,3,2,3,1,3,14)
Value used calculated from C-balance
2,3,7,8-
dioxin
used in road construction --> not
 slags 2.61E-01 kg calculated after IPPC 2000
inventoried
 top gas dust used in sinter plant --> not inventoried 1.17E-02 kg calculated after IPPC 2000
disposal, sludge, pig iron
waste residual production, 8.6% water,
 BF gas sludge No CH 4.26E-03 kg calculated after IPPC 2000 1 3.01 (2,3,2,3,1,3,21)
management material landfill to residual material
landfill
dust from cast
 used in sinter plant --> not inventoried 1.06E-03 kg calculated after IPPC 2000
house de-dusting
rubble allthough partly recycled total amount is disposal, inert waste, 5%
waste inert material
 (refractory assumed to be disposed in inert material No CH water, to inert material 2.07E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,6)
management landfill
material) landfill. landfill
treatment, pig iron
waste wastewater production effluent, to
 waste water No CH 1.81E-03 m3 calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,6)
management treatment wastewater treatment,
class 3
 pig iron 94 % Fe metals extraction No GLO pig iron, at plant 1.00E+00 kg
net output 0.7 Nm3 (input=900,
 BF gas output=1600 Nm3); energy content = 3 1.60E+00 Nm3 Roth et al. 1999
MJ/Nm3; C content = 0.241kg/Nm3
only if of top gas turbine is present. It is
assumed that this is the case for 50% of
 electricity the blast furnaces. Thus the electricity 2.08E-01 kWh IPPC 2000
consumption of the average process is
covered by this production.

Fig. 5.9 Flows for "pig iron, at plant" and its representation in the ecoinvent database (cont.)

Infra
ture
Total land use Transformation, from (5,nA,nA,nA,nA,nA,8);

 resource land 9.76E+05 m2 estimated 1 5.00
(transformation) unknown rough estimation
Land occupation
for buildings and Occupation, industrial (5,nA,nA,nA,nA,nA,7);
 resource land 3.90E+07 m2a estimated 1 5.00
open storage area, built up rough estimation
area
blast furnace; part 1
Land
transformation
Transformation, to (5,nA,nA,nA,nA,nA,8);
to buildings and  assumption: 80% of total area is built up resource land 7.81E+05 m2 estimated 1 5.00
industrial area, built up rough estimation
open bauxite
storage area
Land occupation
Occupation, industrial (5,nA,nA,nA,nA,nA,7);
for vegetation  resource land 9.76E+06 m2a estimated 1 5.00
area, vegetation rough estimation
area
Land
Transformation, to
transformation assumption: 20% of total area is (5,nA,nA,nA,nA,nA,8);
 resource land industrial area, 1.95E+05 m2 estimated 1 5.00
to vegetation vegetation rough estimation
vegetation
area
Building, multi construction (5,nA,nA,nA,nA,nA,9);
 buildings Yes RER building, multi-storey 2.50E+04 m3 estimated 1 5.00
story processes rough estimation
construction (5,nA,nA,nA,nA,nA,9);
Building, hall  buildings Yes CH building, hall 1.95E+04 m2 estimated 1 5.00
processes rough estimation
concrete in
construction concrete, normal, at (5,nA,nA,nA,nA,nA,9);
reinforced  concrete No CH 1.15E+03 m3 estimated 1 5.00
materials plant rough estimation
concrete
steel in
reinforcing steel, at (5,nA,nA,nA,nA,nA,9);
reinforced  metals extraction No RER 7.94E+05 kg estimated 1 5.00
plant rough estimation
concrete
Fig. 5.10 Flows for "blast furnace" and its representation in the ecoinvent database


Infra
ture
industrial machine,
heavy machine, construction (5,nA,nA,nA,nA,nA,9);
 machinery Yes RER heavy, unspecified, at 2.02E+07 kg estimated 1 5.00
unspecific processes rough estimation
plant
Poor concrete
construction concrete, sole plate and (5,nA,nA,nA,nA,nA,9);
blast furnace; part 2
for additional  concrete No CH 9.18E+03 m3 estimated 1 5.00

materials foundation, at plant rough estimation
foundation
Conveyor belt  machinery Yes RER conveyor belt, at plant 1.80E+03 m estimated 1 5.00
Steel low steel, low-alloyed, at (5,nA,nA,nA,nA,nA,9);
 metals extraction No RER 4.68E+05 kg estimated 1 5.00
alloyed plant rough estimation
construction refractory, basic, (5,nA,nA,nA,nA,nA,9);
firebrick  bricks No DE 2.00E+05 kg estimated 1 5.00
materials packed, at plant rough estimation
reinforced disposal, building,
waste building (5,nA,nA,nA,nA,nA,9);
 concrete to No CH reinforced concrete, to 3.32E+06 kg estimated 1 5.00
management demolition rough estimation
disposal final disposal
disposal, building,
foundation
waste building concrete, not (5,nA,nA,nA,nA,nA,9);
 concrete to No CH 2.02E+07 kg estimated 1 5.00
management demolition reinforced, to final rough estimation
disposal
disposal
firebricks to waste building disposal, building, brick, (5,nA,nA,nA,nA,nA,9);
 No CH 2.00E+05 kg estimated 1 5.00
disposal management demolition to final disposal rough estimation
 blast furnace Output: 1500000000 kg/a, life time: 50 a metals extraction Yes RER blast furnace 1.00E+00 unit
Fig. 5.10 Flows for "blast furnace" and its representation in the ecoinvent database (cont.)

5.5 Making available of iron scrap

About 35% of the cast iron and steel production is based on scrap as iron source. While in the BOF
converter scrap is only used to control the temperature of the melt, it is the only (or at least main) iron
bearing input in the electric arc furnace.
A wide variety of iron scrap is processed in the secondary iron and steel industry. A basic differentia-
tion is made between "new" and "old" scrap.
New scrap is surplus material that arises during production and fabrication of iron and steel products
up to the point where they are sold to the final consumer. It is usually of known quality and alloy and
often uncoated. Some new scrap is coated with paints or plastics.
Old scrap or post-consumer scrap is the iron and steel recovered after the product has been used by the
final consumer. It comes into the secondary iron and steel industry via metal merchants and waste
management companies which recover aluminium from e.g. vehicles, household goods etc. This is u-
sually done by shredding, magnetic separation and "sink-and-float" installations or eddy current instal-
lations successively.
Because of the high process temperature and the addition of slag builder old and new scrap can be
remelted with little preparation (no de-coating). No information on the share of the two types on the
overall scrap market and on specific treatment or emissions form remelting of one or the other was
found.
5.5.1 Iron scrap in ecoinvent

The background information to the processes are given in Tab. 5.10, flows are shown in Fig. 5.11.
Data quality
Data are based on rough assumptions and of poor quality.
Data uncertainty is not derived using the Pedigree matrix as described in (Frischknecht et al. (2003))
because the resulting values would seem to low.

Tab. 5.10 Meta information for the iron scrap collection and sorting processes
Name iron scrap, at plant

Location RER
Infrastructure Process 0
Unit kg
Data Set Version 2.0
Collecting of new and old iron scrap, transport to scrap-yard, sorting
Included Processes
and pressing to blocks.
Amount 1
Local Name Eisenschrott, ab Werk
Synonyms
General Comment to reference function Data based on assumptions.
Start Date 2002
End Date 2002
Data Valid For Entire Period 1
Other Period Text
Geography text Data relate to plants in the EU
Technology text Assumed technology of medium sized plant.
Production Volume unknown
Sampling Procedure Internet search
Extrapolations none
Uncertainty Adjustments none


Infra
ture
scrap from
production and  from recycling --> not inventoried 1.00E+00 kg calculated
end-user
iron scrap, at plant

transport road  road: 100 km road No RER 1.00E-01 tkm estimated 1 5.00 rough estimation
transport
transport rail  rail: 200 km train No RER transport, freight, rail 2.00E-01 tkm estimated 1 5.00 rough estimation
systems
diesel powered
equipent for construction diesel, burned in
 machinery No GLO 1.00E-01 MJ estimated 1 5.00 rough estimation
sorting & processes building machine
pressing
electricity use electricity, medium
for sorting and  electricity production mix No UCTE voltage, production 1.00E-02 kWh estimated 1 5.00 rough estimation
pressing UCTE, at grid
aluminium scrap plant as proxy (output:
Infrastructure  metals extraction Yes RER scrap preparation plant 1.00E-09 unit estimated 1 5.00 rough estimation
20000000 kg/a, life time: 50 a)
 waste heat from electricity air unspecified Heat, waste 3.60E-02 MJ calculated 1 5.00 rough estimation
 iron scrap metals extraction No RER iron scrap, at plant 1.00E+00 kg
Fig. 5.11 Flows for "iron scrap" and its representation in the ecoinvent database

5.6 Basic oxygen furnace (BOF) steel making

This part is mainly based on (IPPC (2000)).
The objective in oxygen steel making is to burn (i.e., oxidise) the undesirable impurities contained in
the metallic feedstock. The main elements thus converted into oxides are carbon, silicon, manganese,
phosphorus, and sulphur. The purpose of this oxidation process, therefore, is:
- to reduce the carbon content to a specified level (from approximately 4% to less than 1 %, but of-
ten lower)
- to adjust the contents of desirable foreign elements
- to remove undesirable impurities to the greatest possible extent

The production of steel by the BOF process is a discontinuous process which involves the following
steps:
- transfer and storage of hot metal
- pre-treatment of hot metal (desulphurisation)
- oxidation in the BOF (decarburisation and oxidation of impurities)
- secondary metallurgical treatment
- casting (continuous or/and ingot)

The individual steps and their associated emissions are summarised in Fig. 5.12. More information on
the individual steps is given in (IPPC (2000)).
Fig. 5.12 Individual steps in BOF steel making (IPPC (2000))
Desulphurisation can be done in the torpedo ladles used for transporting the hot metal from the blast
furnace to the BOF. The most widespread hot metal desulphurisation method used in Europe today is

that based on calcium carbide, which has superseded the previous soda process for waste disposal and
air quality management reasons.
There are several types of reactors used for the basic oxygen steel making process. The most com-
monly used type is the LD converter (Linz-Donawitz) applied for pig iron with low phosphorous con-
tent; in the case of high phosphorous content a modified process is used (LD/AC process = Linz-
Donawitz/Arbed-CRM-process). The converter is a pear-shaped, refractory-lined reactor into which a
water-cooled oxygen lance is lowered. Through this lance pure oxygen (>99%) from an air separation
plant is blown onto the liquid pig iron. The amount of oxygen consumed depends on the composition
of the hot metal (i.e. C, Si, P content). Thus undesired impurities are oxidised with subsequent re-
moval with the off-gas or slag in the BOF furnace. The energy required to raise the temperature and
melt the input materials is supplied by the exothermic oxidation reactions, so that no additional heat
input is required, on the one hand, and scrap or ore have to be added to balance heat on the other hand.
Usually, approximately 10-20% of the converter charge is scrap, but values up to 40% are sometimes
used. The amount of scrap charged depends on the pre-treatment given to the pig iron and the required
liquid steel tapping temperature. Variations in the market value of scrap and the required steel specifi-
cations also have an influence.
The oxidising process in the converter is usually followed by post-treatment comprising a number of
metallurgical operations referred to as "secondary metallurgy". The main objectives of secondary met-
allurgy are:
- mixing and homogenising
- adjustment of chemical compositions to close analysis tolerances
- temperature adjustment in time for the downstream casting process
- deoxidisation
- removal of undesirable gases such as hydrogen and nitrogen
- improvement of the oxidic purity by separating non-metallic inclusions.

An overview of secondary metallurgy operations is given in (IPPC (2000)).
Once the final steel quality has been achieved the steel is conveyed in a casting ladle to the casting
machines. Today, the method of choice (>95% in Europe) is continuous casting, whereby the steel is
cast in a continuous strand.
Tab. 5.11).
Tab. 5.11 Assumptions for infrastructure of BOF converters
blast oxygen furnace converter amount unit lifetime [a] Material / Module
yearly production capacity (5000h; 300 t/h) 1'500'000'000 kg 50 Yearly production capacity
area of plant (500 x 100 m) 500'000 m2 50 Transformed area
production hall (70% of total area) 400'000 m2 50 Building, hall
torpedo ladle (without lining) 30'000 kg 10 heavy machine, unspecific
furnace & ladle 1'000'000 kg 25 heavy machine, unspecific
lining for furnace & ladle 100'000 kg 25 firebrick

time of only 25 years is taken into account by doubling the amount, the amount of torpedo ladles is in-
ventoried five times. The infrastructure is disposed after its useful life in inert material landfills. Thus
the values given in Tab. 5.12 are inventoried. For the waste disposal it is assumed that the conveyer
belts and the machines are completely recycled. The disposal of the buildings is already inventoried in
the building modules. Thus only the concrete and the refractory materials are disposed.
Tab. 5.12 Infrastructure of BOF converters
blast oxygen furnace Unit Remarks

converter
Yearly production capacity 1'500'000'000 kg
5.6.2 Emissions
Air
The gases produced during oxygen blowing (converter gas) contain large amounts of carbon monox-
ide. In many steel making plants, measures have been taken to recover the converter gas and use it as
an energy source. Both "open combustion" and "suppressed combustion" systems are in use. Open
combustion systems introduce air into the converter flue gas duct, thus combusting the carbon monox-
ide. The heat generated is later recovered in a waste heat boiler. In suppressed combustion, a skirt is
lowered over the converter mouth during oxygen blowing. Thus, ambient oxygen can not enter the
flue gas duct and the combustion of carbon monoxide is prevented. The CO-rich flue gas can be col-
lected, cleaned and stored for subsequent use as fuel. A main advantage of suppressed combustion is
the smaller flue gas flow since no combustion occurs and no additional air-nitrogen is introduced. This
results in higher productivity since oxygen blowing speed can be increased.
The oxygen steel making process also generates considerable quantities of particulate matter, during
charging of scrap and hot metal, blowing and during tapping of slag and liquid steel. All steel making
shops in the EU have taken measures to reduce particulate matter emissions. The overall emissions
given are in the lowest range in (Berdowski et al. (1995)) thus all particles are assumed to be below
2.5 m.
Water
There are no reports of direct water emissions from these processes.
5.6.3 Waste
During the steel making process, slag is formed. Slag control is intended to effectively reduce the a-
mount of undesirable substances contained in the hot metal and to generate slag of high quality that
will be suitable for subsequent processing and usage. Usually, the slag is cooled and crushed, after
which metallic iron is recovered by magnetic separation. The technical properties of the slag make it
suitable for many kinds of application in civil and hydraulic engineering. Because of its structure, LD

slag has high abrasion resistance and is therefore often used for road construction. It is also put to o-
ther uses or disposed of in landfills.
Desulphurisation slag is a heterogeneous slag which is only partially melted. The composition of de-
sulphurisation slag strongly depends on the used desulphurisation agents. The typical composition of
such a slag are given in Tab. 5.13.
Tab. 5.13 Chemical composition of slag from pig iron desulphurisation [% (w/w)] (IPPC (2000), Tab.8.3)
CaO SiO2 Al2O3 MgO Total Metal- MnO P2O5 Cr2O3 Free S CaO/SiO2
Fe lic Fe CaO
27.0 18 8 10 20 15 0.5 0.2 0.1 5 4 1.5
In this project x is interpreted as x/2
The relatively high sulphur content and unsatisfactory mechanical properties do not make desulphuri-
sation slag ideal for reuse. Partly they are used for landfill construction or for noise protection barriers.
In Europe 41% is landfilled.
The slag from basic oxygen steel making makes up the largest share of residues from BOF steel mak-
ing. The chemical composition of this slag depends on the processes employed (see Tab. 5.14).
BOF slag can be reused by returning it to the iron-making process. It can also be used for fertiliser
manufacture or for road construction. Lime phosphate slag (from LD/AC or OBM processes) is used
entirely for fertiliser production. BOF slag is normally used in civil and hydraulic engineering, road
construction and cement industry. Thereby the relatively high free CaO content has to be considered.
Nevertheless, in the EU a considerable percentage (26%) of BOF slag is still put to landfill. For the
composition of BOF slag used in this project see Tab. 5.14. Katz & Salem (1994) give a chromium
content for steel mill slag of 100-200 ppm and 9’000-48’000 ppm. This indicates a rather wide range
in which the value used in this project is included.
Tab. 5.14 Chemical composition of BOF slag [% (w/w)] (IPPC (2000), Tab.8.8)
Process LD/AC LD AOD This project

CaO 50.0 50.0 53.0 51
SiO2 9.0 15.0 28.0 16
Al2O3  2  2 3.0 1.6
MgO 3 3 5.0 2.6
Total Fe 12.0 16.0 2 8.9
Metallic Fe 1 1 1 0.5
MnO 2.0 4 1 1.5
P2O5 15.0 2  0.5 8.7
Cr2O3 1 1 2 0.66
Free CaO 7  10 5 3.3
S - - 1 0.16
CaO/SiO2 4 2.5 1.8 3.1
Legend: LD/AC = Linz-Donawitz/Arbed-CRM process; LD = Linz-Donawitz process;
AOD = Argon-Oxygen-Decarburisation
This project = 0.57 * LD/AC + 0.11 * LD + 0.32 * AOD based on capacity in Europe (IPPC (2000),
Tab.8.1) with  x interpreted as x/2

Coarse dust waste is usually reused while fine dust, due to high Pb and Zn contamination is partly
landfilled. About 60% of the total dust waste is fine dust of which 42% is landfilled (IPPC (2000),
Tab.8.6, Fig. 8.15). Thus it is assumed that 25% of the total dust is landfilled. The composition of this
dust is assumed according to the information for fine dust of one plant given in Tab. 5.15. The compo-
sition of this dust seems rather incomplete. Especially the lack of Cr and Ni in the dust also for
stainless steel probably is due to these elements not being measured. Considering the high environ-
mental relevance of these elements in landfills, this data gap would lead to considerable underestima-
tion of environmental effects. Data for the electric arc furnace (EAF) dusts are much more comprehen-
sible (cf. Tab. 5.21). For this reason, the disposal modules of EAF dusts are inventoried as proxy for
the disposal of the BOF dust.
Tab. 5.15 Chemical composition of the BOF dust fraction that is landfilled [% (w/w)]. Data not used in this project (cf
text) (IPPC (2000), Tab.8.9)
Parameter Coarse dust Fine dust

Total Fe 30 – 85 54 – 70
Metallic Fe 72 20
CaO 8 – 21 3 – 11
Zn 0.01 – 0.4 1.4 – 3.2
Pb 0.01 – 0.04 0.2 – 1.0
S 0.02 – 0.06 0.07 – 0.12
C 1.4 0.7
5.6.4 Converter steel in ecoinvent

The background information to the processes is given in Tab. 5.17, flows are shown in:
- steel, converter, unalloyed, at plant Fig. 5.13
- steel, converter, low-alloyed, at plant Fig. 5.14
- steel, converter, chromium steel 18/8, at plant Fig. 5.15
- blast oxygen furnace converter Fig. 5.16

If not stated otherwise, arithmetic means of the low and high values given in (IPPC (2000)) are used in
this project.
Because the different slag fractions of the BOF process that are not recycled or reused can be assumed
to be landfilled together, these wastes are treated as one having an average composition according to
Tab. 5.16.

Tab. 5.16 Chemical composition of the average BOF slag to landfill [% (w/w)]
desulphurisation slag BOF slag BOF slag, average

CaO 27 51 48.3
SiO2 18 16 16.5
Al2O3 8 1.6 2.51
MgO 10 2.6 3.66
Total Fe 20 8.9 10.5
Metallic Fe 15 0.5 2.51
MnO 0.25 1.5 1.35
P2O5 0.1 8.7 7.62
Cr2O3 0.05 0.66 0.584
Free CaO 0.25 3.3 2.92
S 2 0.16 0.417
CaO/SiO2 1.5 3.1 2.92
BOF waste, average = 0.14 * desulphurisation slag + 0.86 * BOF slag (according to amount to landfill),
linear extrapolation to 100% (from 102%)
For the elemental composition, CaOSiO2 is interpreted as 50% CaO and 50% SiO2
Data quality
Data of input and main output materials as well as emission data for BOF steel production stem from 4
of 95 basic oxygen furnaces in Europe. Thus data for European consumption are of rather high uncer-
tainty.

Tab. 5.17 Meta information for the BOF steel production processes
steel, converter, steel, converter,

Name steel, converter, low-alloyed, at chromium steel blast oxygen fur-
unalloyed, at plant plant 18/8, at plant nace converter
Location RER RER RER RER
Unit kg kg kg unit
Infrastructure for
input material
Transports of hot Transports of hot Transports of hot unloading, for BOF
metal and other metal and other metal and other and secondary
Included Processes
input materials to input materials to input materials to metallurgical proc-
converter, steel converter, steel converter, steel esses and casting
making process making process making process and for administra-
and casting. and casting. and casting. tive buildings
Amount 1 1 1 1
Blasstahl, nie- Blasstahl, Chrom- Sauerstoff-
Local Name Blasstahl, unle- driglegiert, ab stahl 18/8, ab blaskonverter,
giert, ab Werk Werk Werk Stahl
Synonyms
Assumptions for
medium sized
This process pro- This process pro- This process pro- plant. Output:
General Comment to refer-
duces primary duces primary duces primary 1'500'000 t/a, life
ence function
steel. Scrap is only steel. Scrap is only steel. Scrap is only time: 50 years.
used for cooling used for cooling used for cooling Site is shared with
the liquid steel. the liquid steel. the liquid steel. blast furnace.
Start Date 2001 2001 2001 2002
End Date 2001 2001 2001 2002
Data Valid For Entire Period 1 1 1 1
Other Period Text
Some Swiss data-
sets are used for
Geography text
Data relate to Data relate to Data relate to European proc-
plants in the EU plants in the EU plants in the EU esses.
EU technology mix EU technology mix EU technology mix Assumed technol-
Technology text (mainly LD con- (mainly LD con- (mainly LD con- ogy of medium
verter) verter) verter) sized plant.
Production Volume unknown unknown unknown unknown
Sampling Procedure based on literature based on literature based on literature Internet search
Extrapolations none none none see Geography
Uncertainty Adjustments none none none none


Infra
ture
pig iron / hot

 metals extraction No GLO pig iron, at plant 9.00E-01 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
steel, converter, unalloyed, at plant; part
metal
iron scrap  metals extraction No RER iron scrap, at plant 2.13E-01 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
iron ore  metals extraction No GLO 1.35E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
beneficiation
other Fe iron ore, 65% Fe, at
 as iron ore metals extraction No GLO 8.50E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
material beneficiation
coke  28.6 MJ/kg hard coal fuels No RER hard coal coke, at plant 2.50E-04 MJ calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,11)
lime  additives No CH 4.25E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,4)
construction
dolomite  others No RER dolomite, at plant 2.75E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,4)
materials
ferronickel, 25% Ni, at
alloys  as feronickel metals extraction No GLO 6.00E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
plant
oxygen (99%)  1.429 kg/m3 chemicals inorganics No RER oxygen, liquid, at plant 7.15E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,4)
1
Energy from natural gas, high

 natural gas fuels No RER 3.75E-02 MJ calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,11)
natural gas pressure, at consumer
electricity, medium
electricity  electricity production mix No UCTE voltage, production 2.19E-02 kWh calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,2)
UCTE, at grid
net steam production --> consumptionnot
steam  8.50E-02 MJ calculated after IPPC 2000
inventoried
not inventoried. Energy for compressing
compressed air  1.10E-02 m3 calculated after IPPC 2000
included in electricity consumption
Water, unspecified
water  resource in water 2.70E-03 m3 calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,12)
natural origin
transport ship iron ore from mine to plant by ship transport transport, transoceanic
 ship No OCE 5.94E-02 tkm calculated (Tab. 4.2) 1 2.01 (2,nA,1,1,1,3,5)
(ocean) (costal) systems freight ship
transport ship iron ore from mine to plant by ship transport
 ship No RER transport, barge 6.60E-04 tkm calculated (Tab. 4.2) 1 2.01 (2,nA,1,3,1,3,5)
(costal) (ocean) systems
Fig. 5.13 Flows for "steel, converter, unalloyed, at plant" and its representation in the ecoinvent database


Infra
ture
iron scrap: 100 km; coke, lime, dolomite,

steel, converter, unalloyed, at plant; part 2
transport road  alloys: 20 km; iron ore from mine to plant road No RER 2.23E-02 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
by train
transport
transport rail  alloys: 200 km; iron ore from mine to train No RER transport, freight, rail 1.44E-01 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
systems
plant by train
for transport of hot metal, BOF process, blast oxygen furnace
infrastructure  metals extraction Yes RER 1.33E-11 unit estimated 1 3.23 (5,nA,nA,nA,nA,nA,9)
secondary metallurgy and casting converter
from coke, natural gas and electricity.
Waste heat from oxidation of C in pig iron
to CO and CO2 already included in waste
heat of pig iron.
all dust is fine according to CEPMEIP
database

hydrocarbons
2,3,7,8-
dioxin
Fig. 5.13 Flows for "steel, converter, unalloyed, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture
disposal, basic oxygen

amount inventoried corresponds to the
steel, converter, unalloyed, at plant; part 3 
desulphurisation
slag
41% of the slag that is landfilled. The rest
waste residual
management material landfill
No CH
furnace wastes, 0%
water, to residual
4.39E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,6)
is recycled or reused
material landfill

waste residual furnace wastes, 0%
 BOF slag 26% of the slag that is landfilled. The rest No CH 2.54E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,6)
management material landfill water, to residual
material landfill
assumed to be composed and used / disposal, basic oxygen

slag from
disposed as BOF slag --> amount waste residual furnace wastes, 0%
 secondary No CH 2.34E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,6)
inventoried represents only 26% of total management material landfill water, to residual
metallurgy
slag material landfill
 spittings recycled --> not inventoried 4.50E-03 kg calculated after IPPC 2000
disposal, dust, unalloyed

waste residual EAF steel, 15.4% water,
 dust 25% of the total dust that is landfilled. No CH 1.06E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,6)
The rest is recycled or reused
landfill
 slag from casting recycled --> not inventoried 4.50E-03 kg calculated after IPPC 2000
 mill scale recycled --> not inventoried 3.60E-03 kg calculated after IPPC 2000
management landfill
 waste water amount and composition not known calculated after IPPC 2000
 BOFgas internally used --> not inventoried 4.20E-01 MJ calculated after IPPC 2000
 steam internally used --> not inventoried 1.35E-01 MJ calculated after IPPC 2000
unalloyed BOF steel, converter,
 metals extraction No RER 1.00E+00 kg
steel slab unalloyed, at plant
Fig. 5.13 Flows for "steel, converter, unalloyed, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture
pig iron / hot

 metals extraction No GLO pig iron, at plant 9.00E-01 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
metal
iron scrap  metals extraction No RER iron scrap, at plant 1.25E-01 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
steel, converter, low-alloyed, at plant; part 1
ferronickel  1% Ni in alloy metals extraction No GLO 4.00E-02 kg calculated 1 1.07 (2,nA,nA,nA,1,nA,3)
plant
ferrochromium, high-
ferrochromium  2.15% Cr in alloy metals extraction No GLO 3.16E-02 kg calculated 1 1.07 (2,nA,nA,nA,1,nA,3)
carbon, 68% Cr, at plant
ferromanganese, high-
ferromanganese  1.1% Mn in alloy metals extraction No RER coal, 74.5% Mn, at 1.53E-02 kg calculated 1 1.07 (2,nA,nA,nA,1,nA,3)
regional storage
0.034% Mo in alloy (molybdenite contains
molybdenite  metals extraction No GLO molybdenite, at plant 5.96E-04 kg calculated 1 1.07 (2,nA,nA,nA,1,nA,3)
57% Mo)
Niobium  0.066% Nb in alloy (ferronickel as proxy) metals extraction No GLO 2.64E-03 kg calculated 1 1.07 (2,nA,nA,nA,1,nA,3)
plant
0.084% W in alloy (ferrochromium used as ferrochromium, high-
Tungsten  metals extraction No GLO 1.24E-03 kg calculated 1 1.07 (2,nA,nA,nA,1,nA,3)
proxy) carbon, 68% Cr, at plant
Vanadium  0.059% V in alloy (ferronickel as proxy) metals extraction No GLO 2.36E-03 kg calculated 1 1.07 (2,nA,nA,nA,1,nA,3)
plant
iron ore  metals extraction No GLO 1.35E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
beneficiation
other Fe material  as iron ore metals extraction No GLO 8.50E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
beneficiation
coke  28.6 MJ/kg hard coal fuels No RER hard coal coke, at plant 2.50E-04 MJ calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,11)
lime  additives No CH 4.25E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,4)
construction
dolomite  others No RER dolomite, at plant 2.75E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,4)
materials
oxygen (99%)  1.429 kg/m3 chemicals inorganics No RER oxygen, liquid, at plant 7.15E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,4)
Energy from natural gas, high
 natural gas fuels No RER 3.75E-02 MJ calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,11)
electricity, medium
electricity  electricity production mix No UCTE voltage, production 2.19E-02 kWh calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,2)
UCTE, at grid
steam  8.50E-02 MJ calculated after IPPC 2000
inventoried
compressed air  1.10E-02 m3 calculated after IPPC 2000
Water, unspecified
water  resource in water 2.70E-03 m3 calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,12)
natural origin
Fig. 5.14 Flows for "steel, converter, low-alloyed, at plant" and its representation in the ecoinvent database


Infra
ture
transport ship iron ore from mine to plant by ship transport transport, transoceanic
 ship No OCE 5.94E-02 tkm calculated (Tab. 4.2) 1 2.01 (2,nA,1,1,1,3,5)
(ocean) steel, converter, low-alloyed, at plant; part 2 (costal) systems freight ship
transport road  alloys: 20 km; iron ore from mine to plant road No RER 1.53E-02 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
by train
transport
transport rail  alloys: 200 km; iron ore from mine to train No RER transport, freight, rail 1.08E-01 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
systems
plant by train
heat of pig iron.
database

hydrocarbons
2,3,7,8-
dioxin
Fig. 5.14 Flows for "steel, converter, low-alloyed, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture
steel, converter, low-alloyed, at plant; part 3

desulphurisation waste residual furnace wastes, 0%
 41% of the slag that is landfilled. The rest No CH 4.39E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,6)
slag management material landfill water, to residual
material landfill

material landfill

slag from
metallurgy

landfill
management landfill
low-alloyed BOF steel, converter, low-
steel slab alloyed, at plant
Fig. 5.14 Flows for "steel, converter, low-alloyed, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture
pig iron / hot

 metals extraction No GLO pig iron, at plant 5.28E-01 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
metal
steel, converter, chromium steel
ferronickel  8% Ni in alloy metals extraction No GLO 3.20E-01 kg calculated 1 1.07 (2,nA,nA,nA,1,nA,3)

plant
ferrochromium  18% Cr in alloy metals extraction No GLO 2.65E-01 kg calculated 1 1.07 (2,nA,nA,nA,1,nA,3)
18/8, at plant; part 1

iron ore  metals extraction No GLO 1.35E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
beneficiation
other Fe iron ore, 65% Fe, at
 as iron ore metals extraction No GLO 8.50E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
material beneficiation
coke  28.6 MJ/kg hard coal fuels No RER hard coal coke, at plant 2.50E-04 MJ calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,11)
construction
dolomite  others No RER dolomite, at plant 2.75E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,4)
materials
oxygen (99%)  1.429 kg/m3 chemicals inorganics No RER oxygen, liquid, at plant 7.15E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,4)
energy from natural gas, high
 natural gas fuels No RER 3.75E-02 MJ calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,11)
electricity, medium
UCTE, at grid
steam  8.50E-02 MJ calculated after IPPC 2000
inventoried
compressed air  1.10E-02 m3 calculated after IPPC 2000
Water, unspecified
water  resource in water 2.70E-03 m3 calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,12)
natural origin
Fig. 5.15 Flows for "steel, converter, chromium steel 18/8, at plant" and its representation in the ecoinvent database


Infra
ture
transport ship
(ocean)
 steel, converter, chromium steel 18/8, at plant; iron ore from mine to plant by ship
(costal)
transport
systems
ship No OCE
transport, transoceanic
freight ship
5.94E-02 tkm calculated (Tab. 4.2) 1 2.01 (2,nA,1,1,1,3,5)

coke, lime, dolomite, alloys: 20 km; iron transport transport, lorry >16t,
transport road  road No RER 1.26E-02 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
ore from mine to plant by train systems fleet average
coke, lime, dolomite, alloys: 200 km; iron transport
transport rail  train No RER transport, freight, rail 1.32E-01 tkm estimated 1 2.09 (4,5,nA,nA,nA,nA,5)
ore from mine to plant by train systems
heat of pig iron.
part 2
database

hydrocarbons
2,3,7,8-
dioxin
Fig. 5.15 Flows for "steel, converter, chromium steel 18/8, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture

desulphurisation waste residual furnace wastes, 0%
steel, converter, chromium steel 18/8, at
slag management material landfill water, to residual
material landfill

material landfill

slag from
metallurgy
plant; part 3
disposal, dust, alloyed
landfill
management landfill
steel, converter,
BOF chromium
 metals extraction No RER chromium steel 18/8, at 1.00E+00 kg
steel slab
plant
Fig. 5.15 Flows for "steel, converter, chromium steel 18/8, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture
total land use Transformation, from (5,nA,nA,nA,nA,nA,8);

land occupation
area
land
blast oxygen furnace converter
transformation
open bauxite
storage area
land occupation
area
land
Transformation, to
vegetation
area
building, multi construction (5,nA,nA,nA,nA,nA,9);
building, hall  buildings Yes CH building, hall 4.00E+05 m2 estimated 1 5.00
industrial machine,
plant
poor concrete
foundation
disposal, building,
foundation
disposal
disposal
blast oxygen furnace
 BOF converter Output: 1500000000 kg/a, life time: 50 a metals extraction Yes RER 1.00E+00 unit
converter
Fig. 5.16 Flows for "blast oxygen furnace converter" and its representation in the ecoinvent database

5.7 Electric arc furnace (EAF) steel making

This part is mainly based on (IPPC (2000)).
The direct smelting of iron-containing materials, such as scrap is usually performed in electric arc fur-
naces (EAF) which play an important and increasing role in modern steel works concepts. Today the
percentage of electric arc furnace steel of the overall steel production in the EU is 35.3 %.
The major feed stock for the EAF is ferrous scrap, which may comprise of scrap from inside the steel-
works, cut-offs from steel product manufacturers (e.g. vehicle builders) and capital or post-consumer
scrap (e.g. end of life products). Direct-reduced iron (DRI) is also increasingly being used as a feed-
stock due both to its low gangue content and variable scrap prices. As in the BOF, a slag is formed
from lime to collect undesirable components in the steel.

The production of steel by the EAF process is a discontinuous process which involves the following
steps:
- raw material handling and storage
- furnace charging with/without scrap preheating
- EAF scrap melting
- steel and slag tapping
- ladle furnace treatments for quality adjustment
- slag handling
- continuous casting
For high alloyed and special steels (15% of European steel production), the operation sequence is mo-
re complex and tailor-made for the end-products. In addition to the mentioned operations for carbon
steels various ladle treatments (secondary metallurgy) are carried out like
- desulphurisation
- degassing for the elimination of dissolved gases like nitrogen and hydrogen
- decarburisation (AOD=Argon-Oxygen-Decarburisation or VOD=Vacuum-Oxygen-

Decarburisation)
The actual melting is done by lowering graphite electrodes to the scrap until they strike an arc that
melts the scrap. More information on the individual steps is given in (IPPC (2000)).
Tab. 5.18).

Tab. 5.18 Assumptions for infrastructure of EAF converters
electric arc furnace converter amount unit lifetime [a] Material / Module
yearly production capacity (5000h; 100 t/h) 500'000'000 kg 50 Yearly production capacity
area of plant (500 x 100 m) 500'000 m2 50 Transformed area
production hall (50% of total area) 250'000 m2 50 Building, hall
furnace & ladle 500'000 kg 25 heavy machine, unspecific
lining for furnace & ladle 50'000 kg 25 firebrick
time of only 25 years is taken into account by doubling the amount. The infrastructure is disposed af-
ter its useful life in inert material landfills. Thus the values given in Tab. 5.21 are inventoried. For the
waste disposal it is assumed that the conveyer belts and the machines are completely recycled. The
disposal of the buildings is already inventoried in the building modules. Thus only the concrete and
the refractory materials are disposed.
Tab. 5.19 Infrastructure of EAF converters
electric arc fur- Unit Remarks

nace converter
Yearly production capacity 500'000'000 kg
5.7.2 Emissions
Air
Relevant emissions of heavy metals varying largely with the scrap quality occur in electric steel mak-
ing. Also varying with the scrap quality are chlorinated hydrocarbon emissions such as PCB and diox-
ins. Other VOC emissions can be remarkably high and correspond with the use off coal.
The overall particle emissions given are in the medium range in (Berdowski et al. (1995)) thus 42.5%
of particles are assumed to be smaller than 2.5 m, 42.5% of particles are assumed to be between 2.5
m and 10 m and 15% of particles are assumed to be larger than 10 m.
Water
Drainage water form unpaved scrap-yards can be contaminated, especially in case of oil/emulsion con-
taining scrap like turnings. There is no information available on quantities and pollution of drainage
water. Usually it is at least treated in an oil separator prior to be discharged.
Soil
Soil contamination may arise from contaminated scrap in scrap-yards. No information on quantities
and pollutants is available.

5.7.3 Waste
The main waste generated in EAF steel making is slag. Their composition slightly depends on the al-
loy and on the sub-process they are generated in. The differences are not significant and therefore ne-
glected in this study. One average composition is established (see Tab. 5.20). The recycling fraction
also depends on the alloy. While slag from carbon steel and low alloyed steel production are landfilled
to 69% and 59% respectively, only 53% of the slag from high alloyed steel production is landfilled.
Tab. 5.20 Chemical composition of slag from EAF steel making [% (w/w)]
CaO SiO2 Al2O3 MgO Total TiO2 MnO P2O5 Cr2O3 Free S C
Fe CaO
36 14 5.6 8.7 22 0.31 8.2 0.31 1.5 2.1 0.021 0.34
Na2O K2O V2O5 ZnO CuO NiO
0.47 0.11 0.18 0.021 0.031 0.21
Arithmetic mean of values in (IPPC (2000), Tab. 9.4), linear adjusted to 100% (from 97.5%). In this pro-
ject x is interpreted as x/2
As already mentioned the treatment of off gases (mostly primary off gases together with secondary off
gases) is very often performed in bag filters. The composition of dusts from production of carbon, low
alloyed and high alloyed steel can be seen from Tab. 5.21.
Wastewater also results from off gas treatment in a wet scrubber and from direct cooling from con-
tinuous casting. For the former wastewater no information on applied treatment techniques and dis-
charged quantities and its pollution is available. Latter wastewater contains mill scale (1–3 g/l) and
oil/grease. Usually this wastewater is treated together with other streams from the rolling mills.

Tab. 5.21 Chemical composition of dust from EAF steel making [% (w/w)]
Component Dust from car- Dust from high- Dust from car- Dust from high-
bon/low-alloyed alloyed / stain- bon/low-alloyed alloyed EAF
steel produc- less steel pro- EAF steel in steel in this
tion duction this project project
(IPPC (2000), Tab. 9.6) [% (w/w)] [% (w/w)]
Fe tot 25 – 50 30 – 40 3.59E+01 3.25E+01
SiO2 1.5 – 5 7 – 10 3.11E+00 7.88E+00
CaO 4 – 15 5 – 17 9.10E+00 1.02E+01
Al2O3 0.3 – 0.7 1–4 4.79E-01 2.32E+00
MgO 1–5 2–5 2.87E+00 3.25E+00
P2O5 0.2 – 0.6 0.01 – 0.1 3.83E-01 5.10E-02
MnO 2.5 – 5.5 3–6 3.83E+00 4.17E+00
Cr2O3 0.2 – 1 10 – 20 5.75E-01 1.39E+01
Na2O 1.5 – 1.9 n/a 1.63E+00 1.63E+00
K2O 1.2 – 1.5 n/a 1.29E+00 1.29E+00
Zn 10 – 35 2 – 10 2.16E+01 5.56E+00
Pb 0.8 – 6 0.5 – 2 3.26E+00 1.16E+00
Cd 0.02 – 0.1 0.01 – 0.08 5.75E-02 4.17E-02
Cu 0.15 – 0.4 0.01 – 0.3 2.64E-01 1.44E-01
Ni 0.02 – 0.04 2–4 2.87E-02 2.78E+00
V 0.02 – 0.05 0.1 – 0.3 3.35E-02 1.85E-01
Co 0.001 – 0.002 n/a 1.44E-03 1.44E-03
As 0.003 – 0.08 n/a 3.98E-02 3.98E-02
Hg 0.0001 – 0.001 n/a 5.27E-04 5.27E-04
Cl 1.5 – 4 n/a 2.64E+00 2.64E+00
F 0.02 – 0.9 0.01 – 0.05 4.41E-01 2.78E-02
S 0.5 – 1 0.1 – 0.3 7.19E-01 1.85E-01
C 0.5 – 2 0.5 – 1 1.20E+00 6.96E-01
Basicity 2.0 – 6.5 n/a 4.25E+00 4.25E+00
Moisture 6 – 16 n/a 1.05E+01 1.05E+01
(Energetics (2000), Tab. 5-5)
Fe 28.5 31.7
Zn 19 1
Cd < 0.1 0.16
Cr 0.39 10.2
CaO / MgO 10.7 3.1
This project = arithmetic mean values, linear adjusted to 100% (from 104% and 108% for carbon/low-
alloyed steel and high-alloyed steel respectively)
5.7.4 Electric steel in ecoinvent

- steel, electric, un- and low-alloyed, at plant Fig. 5.17
- steel, electric, chromium steel 18/8, at plant Fig. 5.18
- electric arc furnace converter Fig. 5.19

If not stated otherwise, arithmetic means of the low and high values given in (IPPC (2000)) are used in
this project.

Tab. 5.22 Meta information for the EAF steel production processes
steel, electric, un- and steel, electric, chromium electric arc furnace con-
Name
low-alloyed, at plant steel 18/8, at plant verter
Unit kg kg unit
Transports of scrap metal Transports of scrap metal Infrastructure for input
and other input materials and other input materials material unloading and
Included Processes to electric arc furnace, to electric arc furnace, storage, for EAF proc-
steel making process and steel making process and esses and casting and for
casting. casting. administrative buildings
Amount 1 1 1
Elektrostahl, un- und Elektrostahl, Chromstahl Elektrolichtbogenofen,
Local Name
niedriglegiert, ab Werk 18/8, ab Werk Stahl
Synonyms
This process produces This process produces Assumptions for medium
General Comment to refer- secondary steel. Only secondary steel. Only sized plant. Output:
ence function scrap is used as iron scrap is used as iron 500'000 t/a, life time: 50
bearing input. bearing input. years.
Start Date 2001 2001 2002
End Date 2001 2001 2002
Data Valid For Entire Period 1 1 1
Other Period Text
Some Swiss datasets are
Geography text Data relate to plants in Data relate to plants in used for European proc-
the EU the EU esses.
EU technology mix EU technology mix
(mainly furnace with 4th (mainly furnace with 4th
hole, partly with additional hole, partly with additional
Technology text evacuation of building evacuation of building
atmosphere). An average atmosphere). An average
for different alloys is rep- for different alloys is rep- Assumed technology of
resented. resented. medium sized plant.
Sampling Procedure based on literature based on literature Internet search
Extrapolations none none see Geography
Uncertainty Adjustments none none none
Data quality
Data of input and main output materials as well as emission data for EAF steel production stem from
an unknown number of 203 electric arc furnaces in Europe. Thus data for European consumption are
of rather high uncertainty.


Infra
ture
iron scrap  metals extraction No RER iron scrap, at plant 1.11E+00 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
steel, electric, un- and low-

alloyed, at plant; part 1
hard coal mix, at

coal  hard coal fuels No UCTE 1.40E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,11)
regional storage
graphite anode, aluminium
 metals extraction No RER 3.00E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,4)
electrode electrolysis
refractory
construction refractory, basic,
materials  bricks No DE 1.35E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,4)
materials packed, at plant
(lining)
oxygen (99%)  density 1.429 kg/m3 chemicals inorganics No RER oxygen, liquid, at plant 5.07E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,4)
natural gas, high
other energy  all other energy assumed as natural gas natural gas fuels No RER 9.75E-01 MJ calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,11)
electricity, medium
UCTE, at grid
scrap, lime, refractories, coal & transport transport, lorry >16t, estimated (standard
transport road  road No RER 1.19E-01 tkm 1 2.09 (4,5,nA,nA,nA,nA,5)
electrode: 100 km; systems fleet average distance)
scrap, lime, refractories & coal: 200 km; transport estimated (standard
transport rail  train No RER transport, freight, rail 1.21E-01 tkm 1 2.09 (4,5,nA,nA,nA,nA,5)
electrode: 600km systems distance)
water  closed cycle --> not inventoried 0.00E+00 IPPC 2000
for EAF process, secondary metallurgy and electric arc furnace
casting converter
Fig. 5.17 Flows for "steel, electric, and- and low-alloyed, at plant" and its representation in the ecoinvent database


Infra
ture
from coal, electrode (29 MJ/kg), other

steel, electric, un- and low-alloyed, at
energy and electricity

 dust, coarse 15% > 10 um air unspecified Particulates, > 10 um 5.86E-05 kg calculated after IPPC 2000 1 1.52 (2,3,2,3,1,3,25)
 dust, medium 42,5% between 2.5 um and 10 um air unspecified 1.66E-04 kg calculated after IPPC 2000 1 2.01 (2,3,2,3,1,3,26)
and < 10um
 dust, fine 42.5% < 2.5 um air unspecified Particulates, < 2.5 um 1.66E-04 kg calculated after IPPC 2000 1 3.01 (2,3,2,3,1,3,27)
plant; part 2
Fig. 5.17 Flows for "steel, electric, and- and low-alloyed, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture
as aromatic hydrocarbons (92% C);

TOC (total
 individualy mentioned sources of organic air unspecified Hydrocarbons, aromatic 7.70E-05 kg calculated after IPPC 2000 1 1.52 (2,3,2,3,1,3,18)
organic carbon)
steel, electric, un- and low-alloyed, at
carbon are subtracted

 benzene (92% C) air unspecified Benzene 2.29E-06 kg calculated after IPPC 2000 1 2.01 (2,3,2,3,1,3,23)
hexachlorobenze
 (25 %C) air unspecified Benzene, hexachloro- 2.00E-08 kg calculated after IPPC 2000 1 2.01 (2,3,2,3,1,3,23)
ne
(92%C); benzene and hexachlorobenzene PAH, polycyclic aromatic
are not included hydrocarbons
Polychlorinated
 PCB (50%C) air unspecified 2.33E-08 kg calculated after IPPC 2000 1 3.01 (2,3,2,3,1,3,21)
biphenyls
plant; part 3
2,3,7,8-
 dioxin (I-TEQ) (45% C) air unspecified 4.54E-12 kg calculated after IPPC 2000 1 3.01 (2,3,2,3,1,3,21)
dioxin
disposal, slag, unalloyed
slag from waste residual electr. steel, 0% water,
furnace management material landfill to residual material
landfill
waste residual electr. steel, 0% water,
 slag from laddle 64% of the slag that is landfilled. The rest No CH 1.28E-02 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,6)
landfill
landfill
allthough partly recycled total amount is disposal, inert waste, 5%
refractory waste inert material
 assumed to be disposed in inert material No CH water, to inert material 5.00E-03 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,6)
material management landfill
landfill. landfill
un-/low-alloyed steel, electric, un- and
EAF steel slab low-alloyed, at plant
Fig. 5.17 Flows for "steel, electric, and- and low-alloyed, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture
iron scrap  metals extraction No RER iron scrap, at plant 5.20E-01 kg calculated after IPPC 2000 1 1.11 (2,3,2,3,1,3,3)
steel, electric, chromium steel 18/8,

nickel  8% in alloy metals extraction No GLO 3.20E-01 kg calculated 1 1.07 (2,nA,nA,nA,1,nA,3)
plant
chromium  18% in alloy metals extraction No GLO 2.65E-01 kg calculated 1 1.07 (2,nA,nA,nA,1,nA,3)
hard coal mix, at
at plant; part 1
regional storage
refractory
(lining)
natural gas, high
electricity, medium
UCTE, at grid
scrap, alloys, lime, refractories, coal & transport transport, lorry >16t, estimated (standard
scrap, alloys, lime, refractories & coal: transport estimated (standard
200 km; electrode: 600km systems distance)
casting converter
Fig. 5.18 Flows for "steel, electric, chromium steel 18/8, at plant" and its representation in the ecoinvent database


Infra
ture

steel, electric, chromium steel 18/8,
and < 10um
at plant; part 2
Fig. 5.18 Flows for "steel, electric, chromium steel 18/8, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture

TOC (total
steel, electric, chromium steel 18/8, at
organic carbon)
hexachlorobenze
ne
Polychlorinated
biphenyls
plant; part 3
2,3,7,8-
dioxin
landfill
landfill
disposal, dust, alloyed
landfill
landfill. landfill
steel, electric,
EAF chromium
 metals extraction No RER chromium steel 18/8, at 1.00E+00 kg
steel slab
plant
Fig. 5.18 Flows for "steel, electric, chromium steel 18/8, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture
Total land use Transformation, from (5,nA,nA,nA,nA,nA,8);

Land occupation
area
Land
transformation
electric arc furnace converter
open bauxite
storage area
Land occupation
area
Land
Transformation, to
vegetation
area
Building, multi construction (5,nA,nA,nA,nA,nA,9);
Building, hall  buildings Yes CH building, hall 2.50E+05 m2 estimated 1 5.00
industrial machine,
plant
Poor concrete
foundation
disposal, building,
foundation
disposal
disposal
electric arc furnace
 EAF converter Output: 500000000 kg/a, life time: 50 a metals extraction Yes RER 1.00E+00 unit
converter
Fig. 5.19 Flows for "electric arc furnace converter" and its representation in the ecoinvent database

5.8 Cast iron

Cast iron is made by remelting pig iron in small furnaces (e.g. cupola furnaces or electric arc furnaces)
and poured into moulds to make castings. It usually contains 2% to 6% carbon. Scrap iron or steel is
often added to vary the composition. Cast iron is used extensively to make machine parts, engine cyl-
inder blocks, stoves, pipes, steam radiators, and many other products. Grey cast iron, or grey iron, is
produced when the iron in the mould is cooled slowly. Part of the carbon separates out in plates in the
form of graphite but remains physically mixed in the iron. Grey iron is brittle but soft and easily ma-
chined. White cast iron, or white iron, which is harder and more brittle, is made by cooling the molten
iron rapidly. The carbon remains distributed throughout the iron as cementite (iron carbide, Fe3C). A
malleable cast iron can be made by annealing white iron castings in a special furnace. Some of the
carbon separates from the cementite; it is much more finely divided than in grey iron. A ductile iron
may be prepared by adding magnesium to the molten pig iron; when the iron is cast the carbon forms
tiny spherical nodules around the magnesium. Ductile iron is strong, shock resistant, and easily ma-
chined. 7
5.8.2 Emissions
Air
Air emissions depend mainly on the type of furnaces and the fuel used in them. Emission composition
of the production in a cupola furnace is similar to that of the blast furnace while emissions of the cast
iron production in electric arc furnaces is similar to those of the steel making in those furnaces.
Water
There are no reports of direct water emissions from these processes.
5.8.3 Waste
As for the steel making the main wastes from cast iron production is slag and filter dust. No specific
information was found.
5.8.4 Cast iron in ecoinvent

The background information to the process "cast iron, at plant" are given in Tab. 5.23, and flows are
shown in Fig. 5.20
For a lack of specific data the cast iron production is inventoried using the data of the EAF steel mak-
ing but assuming a mix of iron input of 65% pig iron and 35% scrap. Since cast iron usually is un- or
low-alloyed and melted in electric furnaces, this assumptions seem justified.
7
http://www.infoplease.com/ce6/sci/A0858908.html, accessed 12.02.03

Tab. 5.23 Meta information for the cast iron production processes
Name cast iron, at plant

Location RER
Unit kg
Transports of metal and other input materials to electric arc furnace,
Included Processes
melting and refining process and casting.
Amount 1
Local Name Gusseisen, ab Werk
Synonyms grey iron//white iron//Grauguss
General Comment to reference function 35% scrap and 65% pig iron assumed as iron input
Start Date 2001
End Date 2002
Other Period Text
Geography text Data relate to plants in the EU
Electric arc furnace for melting. Energy consumption and emissions
Technology text
from EAF steel making
Production Volume unknown
Sampling Procedure Internet search
Extrapolations see technology
Data quality
Data are estimates based on a technology that is similar to one of different technologies used in the
cast iron production. Thus data uncertainty is rather high.


Infra
ture
iron scrap  assumption: 35% scrap, 65% primary iron metals extraction No RER iron scrap, at plant 3.87E-01 kg calculated after IPPC 2000 1 1.23 (2,3,2,3,3,3,3)
pig iron  assumption: 35% scrap, 65% primary iron metals extraction No GLO pig iron, at plant 7.18E-01 kg calculated after IPPC 2000 1 1.23 (2,3,2,3,3,3,3)
cast iron, at plant; part 1

hard coal mix, at
regional storage
refractory
(lining)
natural gas, high
electricity, medium
UCTE, at grid
scrap, pig iron, lime, refractories, coal & transport transport, lorry >16t, estimated (standard
scrap, pig iron, lime, refractories & coal: transport estimated (standard
200 km; electrode: 600km systems distance)
casting converter
Fig. 5.20 Flows for "cast iron, at plant" and its representation in the ecoinvent database


Infra
ture

and < 10um
Fig. 5.20 Flows for "cast iron, at plant" and its representation in the ecoinvent database (cont.)


Infra
ture

TOC (total
organic carbon)
hexachlorobenze
ne

Polychlorinated
biphenyls
2,3,7,8-
dioxin
landfill
landfill
landfill
landfill. landfill
 cast iron metals extraction No RER cast iron, at plant 1.00E+00 kg
Fig. 5.20 Flows for "cast iron, at plant" and its representation in the ecoinvent database (cont.)

5.9 Consumer mixes for carbon steel, low-alloy steel and

stainless steel
5.9.1 Steel in ecoinvent
The different steel alloys inventoried in ecoinvent are modelled by mixtures of differently produced
steel and alloy materials. In Europe in 2000 39.8% of the steel was produced in electric furnaces and
the remaining 60.2% in basic oxygen furnaces. The world production mix was 58.5% basic oxygen
furnace, 34% electric furnace and 4.4% open hearth furnace (IISI (2002)). For the consumption mix in
Europe we assume 63% BOF and 37% EAF production.
Because steel for every application is hot rolled, this process is also included in the modules. For a de-
scription of the process see Part IXX Metalprocessing.
The background information to the processes is given in Tab. 5.24, and flows are shown in:
- reinforcing steel, at plant Fig. 5.21
- steel, low-alloyed, at plant Fig. 5.22
- chromium steel 18/8, at plant Fig. 5.23
Data quality
Data uncertainty is derived using the Pedigree matrix as described in (Frischknecht et al. (2003)).
The shares of the different technologies stem from a statistical yearbook. Thus the data are reliable.

Tab. 5.24 Meta information for the consumer steels
chromium steel 18/8, at

Name
reinforcing steel, at plant
steel, low-alloyed, at plant plant
Unit kg kg kg
Mix of differently pro-
Mix of differently pro- Mix of differently pro-
Included Processes duced steels and hot roll-
duced steels and hot roll- duced steels and hot roll-
ing ing ing
Amount 1 1 1
Stahl, niedriglegiert, ab Chromstahl 18/8, ab
Local Name
Armierungsstahl, ab Werk Werk Werk
carbon steel//unalloyed high alloyed
Synonyms
steel steel//stainless steel
Represents Average of Represents Average of Represents Average of
World and European pro- World and European pro- World and European pro-
General Comment to refer- duction mix. This is as- duction mix. This is as- duction mix. This is as-
ence function sumed to correspond to sumed to correspond to sumed to correspond to
the consumption mix in the consumption mix in the consumption mix in
Europe Europe Europe
Start Date 2000 2000 2000
End Date 2002 2002 2002
Other Period Text
Data relate to plants in Data relate to plants in Data relate to plants in
Geography text
the EU the EU the EU
Technology text technology mix technology mix technology mix
Sampling Procedure literature, statistics literature, statistics literature, statistics
Extrapolations


Infra
ture
unalloyed basic
steel, converter, estimation, based on IISI
reinforcing steel,
oxygen furnace  metals extraction No RER 6.30E-01 kg 1 1.11 (2,3,2,3,1,3,3)
unalloyed, at plant 2002
steel
unalloyed
at plant
steel, electric, un- and estimation, based on IISI

electric arc  metals extraction No RER 3.70E-01 kg 1 1.11 (2,3,2,3,1,3,3)
low-alloyed, at plant 2002
furnace steel
relates directly to
hot rolling  metals processing No RER hot rolling, steel 1.00E+00 kg calculated 1 1.00
output
reinforcing steel reinforcing steel, at
(carbon steel) plant
Fig. 5.21 Flows for "reinforcing steel, at plant" and its representation in the ecoinvent database

Infra
ture
low-alloyed
alloyed, at plant
steel, converter, low- estimation, based on IISI

basic oxygen  metals extraction No RER 6.30E-01 kg 1 1.11 (2,3,2,3,1,3,3)
alloyed, at plant 2002
steel, low-
furnace steel
low-alloyed
steel, electric, un- and estimation, based on IISI
electric arc  metals extraction No RER 3.70E-01 kg 1 1.11 (2,3,2,3,1,3,3)
low-alloyed, at plant 2002
furnace steel
relates directly to
output
low alloyed steel, low-alloyed, at
steel plant
Fig. 5.22 Flows for "steel, low-alloyed, at plant" and its representation in the ecoinvent database


Infra
ture
basic oxygen steel, converter,

estimation, based on IISI
chromium steel
18/8, at plant
furnace  metals extraction No RER chromium steel 18/8, at 6.30E-01 kg 1 1.11 (2,3,2,3,1,3,3)
2002
chromium steel plant
electric arc steel, electric,
estimation, based on IISI
furnace  metals extraction No RER chromium steel 18/8, at 3.70E-01 kg 1 1.11 (2,3,2,3,1,3,3)
2002
chromium steel plant
relates directly to
output
chromium steel chromium steel 18/8, at
18/8 plant
Fig. 5.23 Flows for "chromium steel 18/8, at plant" and its representation in the ecoinvent database

6. Literature
6 Literature
www.ecoinvent.ch.
Anonymous (1998) Anonymous (1998) Base Metal and Iron Ore Mining. World Bank. Retrieved
from
http://lnweb18.worldbank.org/essd/essd.nsf/GlobalView/PPAH/$File/48_b-
asem.pdf
Anonymous (2002) Anonymous (2002) European Dioxin Inventory. European Commission. Re-
trieved from http://europa.eu.int/comm/environment/dioxin/stage1/sinter_plants
.pdf. Date of last revision: 22.08.2002.
Berdowski et al. (1995) Berdowski J., Visschedijk A., Creemers E., Pulles T., Pacyna J., Fudala J. and
Querreveld D. (1995) Co-ordinated European Programme on Particulate Matter
Emission Inventories, Projections and Guidance (CEPMEIP). CEPMEIP. Re-
trieved from http://www.air.sk/tno/cepmeip/
Energetics (2000) Energetics I. (2000) Energy and Environmental Profile of the U.S. Iron and Steel
Industry. U.S. Department of Energy, Office of Industrial Technologies, Online-
Version under: http://www.oit.doe.gov/steel/pdfs/steel_profile.pdf.
Frischknecht et al. (2003) Frischknecht R., Jungbluth N., Althaus H.-J., Doka G., Dones R., Hellweg S.,
Hischier R., Nemecek T., Rebitzer G. and Spielmann M. (2003) Overview and
Methodology. Final report ecoinvent 2000 No. 1. Swiss Centre for Life Cycle
Inventories, Dübendorf, CH, Online-Version under: www.ecoinvent.ch.
Frischknecht et al. (1996) Frischknecht R., Suter P., Bollens U., Bosshart S., Ciot M., Ciseri L., Doka G.,
Hischier R., Martin A., Dones R. and Gantner U. (1996) Ökoinventare von Ener-
giesystemen, Grundlagen für den ökologischen Vergleich von Energiesystemen
und den Einbezug von Energiesystemen in Ökobilanzen für die Schweiz. 3.
Aufl. Edition. Bundesamt für Energiewirtschaft (BEW/PSEL), Bern.
IISI (2002) IISI (2002) Steel Statistical Yearbook 2002. I nternational Iron and Steel Insti-
tute (IISI). Retrieved from
http://www.worldsteel.org/media/ssy/iisi_ssy_2001.pdf
IPPC (2000) IPPC (2000) Integrated Pollution Prevention and Control (IPPC); Best Available
Techniques Reference Document on the Production of Iron and Steel. European
Commission. Retrieved from http://www.jrc.es/pub/english.cgi/0/733169
Kalla & Kalwa (1997) Kalla U. and Kalwa G. (1997) Steel; Uses. In: Ullmann's encyclopedia of indus-
trial chemistry (ed. Anonymous). 5th edition on CD-ROM Edition. Wiley &
Sons, London.
Katz & Salem (1994) Katz S. A. and Salem H. (1994) The Biological and Environmental Chemistry of
Chromium. New York. ISBN 1-56081-629-5.
Kucharcik (1997) Kucharcik L. (1997) Cast Iron. In: Ullmann's encyclopedia of industrial chemis-
try (ed. Anonymous). 5th edition on CD-ROM Edition. Wiley & Sons, London.

6. Literature
Meiler (1997) Meiler H. (1997) Iron Production Process; Ores. In: Ullmann's encyclopedia of
industrial chemistry (ed. Anonymous). 5th edition on CD-ROM Edition. Wiley
& Sons, London.
Roth et al. (1999) Roth W., Atmaca T. and Mori G. (1999) Stoffmengenflüsse und Energiebedarf
bei der Gewinnung ausgewählter mineralischer Rohstoffe; Teilstudie Eisen. In:
Steffen (1997) Steffen R. (1997) Steel; Raw materials. In: Ullmann's encyclopedia of industrial
chemistry (ed. Anonymous). 5th edition on CD-ROM Edition. Wiley & Sons,
London.

6. Literature

Part III
Copper, Molybdenum & Telluride

and Silver containing byproducts
Data v2.1 (2009)
Authors: Mischa Classen, Empa Dübendorf

Hans-Jörg Althaus, Empa Dübendorf
Review Roberto Dones, PSI

Roland Hischier, Empa, St. Gallen
Update v2.0 Mischa Classen, Empa Dübendorf

Update v2.1 Hans-Jörg Althaus, Empa Dübendorf
Summary

This part corresponds to the original report in Althaus et al. (2004) that has been updated with the
changes reported in Frischknecht et al. (2006). The changes were implemented with v1.2 and affect
mainly the datasets for primary and secondary copper by increasing the emission factors to air for di-
oxin, carbon monoxide, particulate matter and volatile organic compounds.
The inventories covering copper alloys (brass, bronze and brazing solder), which were treated in the
now removed “Part VIII Other Metals” in Althaus et al. (2004), are now included in this Part III.
Additionally the generation of telluride and silver containing anode slime has been introduced in this
version to answer the related demand from the inventories of gold and silver (Part IX) and tellurium
(Part X). This extension was part of the mandate commissioned by the Swiss Federal Office for the
Environment (FOEN) to improve the overall quality and applicability of the metals inventories.
In version 2.1 we corrected an error in the dataset “molybdenum, at regional storage (RER)”. See Fig.
5.21.
Chapter “6 Cumulative Results and Interpretation” of the original report has been deleted for the cur-
rent update.
Summary
The primary copper and molybdenum production is modelled for the global situation, based on data
from various sources. Additionally, the situations for different concentrate-producing regions are
modelled separately. Since most of tellurium and considerable parts of silver originate from copper
production, a specific chapter treats the generation of silver and telluride containing anode slime.
The mining and beneficiation are modelled as coupled processes with copper and molybdenit concen-
trate as outputs.
For copper also the secondary production in Europe is inventoried. Finally a mix is generated to be
used for copper consumed in Europe.
Some of the datasets described in this report make only sense in the context of the copper and molyb-
denum production and are therefore not intended for other uses.
The datasets to be chosen to represent copper in a LCA study are either one of the primary copper
datasets, the secondary copper dataset or the European mix (at regional storage). The decision whether
to use primary, secondary or a mix of the two depends on the goal and scope of the LCA to be con-
ducted.
If molybdenum is needed as alloying element in an LCA study, molybdenite should be chosen as in-
put. The metallic molybdenum is only used in very specific applications.

Table of Contents
Table of Contents
COMMENTS ON VERSION 2.0 ............................................................................................I
SUMMARY ......................................................................................................................I
1 INTRODUCTION ....................................................................................................... 1
2 RESERVES AND RESOURCES OF COPPER AND MOLYBDENUM .................................... 2

2.1 Primary copper ...................................................................................................................... 2
2.2 Primary Molybdenum ........................................................................................................... 7
2.3 Secondary Copper ................................................................................................................. 8

3.1 Copper ................................................................................................................................... 9
3.2 Copper alloys: brass bronze and brazing solders .................................................................. 9
3.2.1 Brass.................................................................................................................................... 10
3.2.2 Bronze ................................................................................................................................. 10
3.2.3 Solders................................................................................................................................. 10
3.3 Molybdenite concentrate and Molybdenum trioxide .......................................................... 10
4 USE / APPLICATION OF MATERIAL .......................................................................... 12

4.1 Copper ................................................................................................................................. 12
4.2 Molybdenite concentrate ..................................................................................................... 13
4.3 Molybdenum trioxide and Ferromolybdenum .................................................................... 14
5 SYSTEM CHARACTERISATION ................................................................................ 15

5.1 Exploitation of copper ore................................................................................................... 22
5.1.1 Production process .............................................................................................................. 22
5.1.2 Emissions & Waste ............................................................................................................. 22
5.1.3 Exploitation of copper-ore in ecoinvent.............................................................................. 23
5.2 Winning primary copper ..................................................................................................... 31
5.2.2 Emission.............................................................................................................................. 32
5.2.3 Waste................................................................................................................................... 33
5.2.4 Copper, blister copper, at primary smelter in ecoinvent ..................................................... 33
5.2.5 Winning of primary copper in ecoinvent ............................................................................ 33
5.3 Hydrolytic winning of primary copper................................................................................ 41
5.3.2 Emissions and wastes .......................................................................................................... 42
5.3.3 Hydrolytic winning of primary copper in ecoinvent. .......................................................... 42
5.4 Winning of secondary copper ............................................................................................. 46
5.4.2 Emissions and waste ........................................................................................................... 46
5.4.3 Winning of secondary copper in ecoinvent......................................................................... 47
5.5 Copper, at regional storage.................................................................................................. 52
5.5.1 Copper, from imported concentrates, at refinery in ecoinvent............................................ 52
5.5.2 Copper, at regional storage in ecoinvent............................................................................. 52

Table of Contents
5.6 Copper alloys: brass bronze and brazing solders ................................................................ 55

5.6.1 Copper alloy production in ecoinvent ................................................................................. 55
5.6.2 Data quality ......................................................................................................................... 56
5.7 Telluride and Silver containing slime as by-product........................................................... 64
5.7.1 Anode slime generation in ecoinvent .................................................................................. 64
5.7.2 Allocation and resource correction ..................................................................................... 66
5.7.3 Data quality ......................................................................................................................... 67
5.8 Molybdenum ....................................................................................................................... 69
5.8.1 Exploitation of molybdenite in ecoinvent ........................................................................... 69
5.8.2 Molybdenite, at plant in ecoinvent...................................................................................... 70
5.8.3 Molybdenum, at regional storage in ecoinvent ................................................................... 70
6 LITERATURE ......................................................................................................... 72

1. Introduction
1 Introduction
This chapter describes the production of copper from mining of the ore to consumer in Europe and the
coupled production of molybdenum. Inventory data for the copper production is taken mainly from
Krauss et al. (1999) together with background data from IPPC (2002), IPPC (2001), EEA (1999) and
Ayres et al. (2002) (primary copper) and Rentz et al. (1999) and Bruch et al. (1995) (secondary cop-
per).
Since copper is always associated with other metals, the production yields a fraction of other metals
than copper, mainly nickel (Ni), molybdenum (Mo) and platinum group metals. The process of pri-
mary production described in this chapter is formally a multioutput-process with the coupled products
Cu and Mo. No other potential co-products have been considered in this study. This makes it neces-
sary to allocate the inventory to these products. The demand for copper-containing ore is allocated by
mass. The rest of the inventory is allocated by revenue.
Copper is used as pure metal and as alloying element with other metals and in steels. On the other
hand, molybdenum is almost exclusively used as alloying element in steels. Its marketed product form
is molybdenum trioxide and ferromolybdenum.
According to EEA (1999), emissions of sulphur dioxide from non-ferrous metal production, particu-
larly copper production, contribute less than 10% to the total global emissions of this pollutant. How-
ever, copper smelting can be the most important source of sulphur dioxide emissions in certain re-
gions, such as the Kola Peninsula, the Urals, the Norilsk area, and the Fergana region in Russia, Lower
Silesia in Poland, and the Gijon region in Spain. During the copper production process various trace
elements originating from impurities in the ore are emitted to the atmosphere. This process is the ma-
jor source of atmospheric arsenic and copper (about 50 % of the global emissions of the element), and
indium (almost 90 %), and a significant source of atmospheric antimony, cadmium and selenium (ca.
30 %), and nickel and tin (ca. 10 %). The production of copper together with nickel is the major source
of pollution from these and other elements in regions with non-ferrous metal production.
According to mythology, the goddess Venus (or Aphrodite) was born on the Mediterranean island of
Cyprus, formerly Kypros (Greek), where copper was exploited millennia before Christ. Therefore, in
early times the Romans named it cyprium, later called cuprum. This name is the origin of copper and
of the corresponding words in most Romance and Germanic languages, e.g., cobre (Spanish and Por-
tuguese), cuivre (French), Kupfer (German), koper (Dutch), and koppar (Swedish).

2. Reserves and Resources of Copper and Molybdenum
2 Reserves and Resources of Copper and

Molybdenum
Metals are a special class of materials. They are not consumed but rather used during a product life cy-
cle. When recovered the properties are theoretically identical to those in the raw material. The metal
can be re-used in another product life cycle, provided that its concentration makes a recovery eco-
nomic. The recovered metal is referred to as secondary metal, whereas the metal originating directly
from the ore is referred to as primary metal. In LCIs these two types – primary and secondary metal –
are often differentiated because they are assigned with different ecological burden.
Molybdenum and Copper are coexisting in porphyry deposits of the copper-molybdenum type, as mo-
lybdenite (MoS2) and chalcopyrite (CuFeS2). About half of the world-wide produced molybdenum is a
co-product of the primary copper industry while for another substantial part copper is the co-product
(Krauss et al. (1999) and Sebenik et al. (1997)). Hence almost all of the molybdenum is produced in a
process similar to the copper primary production. Molybdenum secondary production – mainly from
spent petroleum catalysts – is not remarkable, and no secondary production is considered in this study.
USGS (2003) estimates that 30% of the molybdenum is re-used in the form of the molybdenum con-
tent steel alloys which are recycled to the foundries. Secondary production of copper from scrap plays
an important role. The resources of primary copper are limited, a continuous depletion within 60 years
is estimated (Ayres et al. (2002), p. 10).
2.1 Primary copper

Recoverable copper deposits are now mostly (90-95%) in the form of disseminated sulphide minerals,
such as chalcopyrite (CuFeS2), chalcocite (Cu2S), bornite (Cu5FeS4) and enargite (Cu3AsS4). On aver-
age, a tonne of sulphur is associated with each tonne of copper. By far the most important deposits are
of the disseminated porphyry copper type, in which the dominant mineral is chalcopyrite. Globally,
90%-95% of copper is derived from sulphide ores. Arsenic is associated with some porphyry copper
deposits, such as Butte, Montana, and El Indio (Chile).
Another type of porphyry deposit is the copper-molybdenum type, in which the mineral molybdenite
(MoS2) is interspersed with chalcopyrite, as in Henderson Colorado, Quartz Hill Canada and a few
other locations in Chile and the Philippines. These ore bodies are mainly mined for molybdenum, with
copper as a by-product.
So called oxide ores, which account for 5%-10% of global output, are weathered products of sulphide
minerals, notably malachite (CuCO3·Cu(OH)2), azurite (2CuCO3·Cu (OH)2), cuprite (Cu2O) and chry-
sacolla (CuSiO3·2H2O) (Ayres et al. (2002)).
Overall more than 200 minerals contain copper in definable amounts, but only ca. 20 are of impor-
tance as copper ores (Tab. 2.1) or as semiprecious stones (turquoise and malachite). Secondary miner-
als are formed in sulphide ore bodies near the earth's surface in two stages. In the oxidation zone, oxy-
gen-containing water forms copper oxides, subsalts (subcarbonates and subsulphates), and silicates. In
the deeper cementation zone, copper-bearing solutions from these salts are transformed into secondary
copper sulphides (chalcocite and covellite) and even native copper of often high purity, e.g., in the
Michigan copper district (Keweenaw Peninsula). The grade of copper ore at a mine or in a mining re-
gion decreases as ore is removed, sometimes dropping below that of gangue left from earlier mining.
The grade of the copper ore mined and milled in the U.S. dropped from 3.5% in 1900 to nowadays
under 0.7% which was the content of the 1900’s tailings (Fig. 2.1, Graedel et al. (2002)).

Fig. 2.1 Development of copper ore grade in U. S. since 1880 (Ayres et al. (2002)).
Other metallic elements frequently found in copper ores are iron, lead, zinc, antimony, and arsenic;
less common are selenium, tellurium, bismuth, silver, and gold. Substantial enrichments sometimes
occur in complex ores. For example, ores from Sudbury, Ontario, in Canada contain nickel and copper
in nearly the same concentrations, as well as considerable amounts of platinum metals. The copper
ores from Zaire and Zambia are useful sources of cobalt. Many porphyry copper ores in America con-
tain significant amounts of molybdenum and are the most important single source of rhenium. The ex-
traction of precious metals and other rare elements can be decisive for the profitability of copper mi-
nes, smelters, and refineries (Fabian (1997)).
Land-based resources are estimated to be 1.6 billion tons of copper, and resources in deep-sea nodules
are estimated to be 700 million tons (USGS (2003)). A detailed overview over the global refinery pro-
duction and a statistic on US use and production of copper is available in the online version of the
USGS “Mineral Commodity Summary” (USGS (2003)). The world-wide mine production by country
in 2002 and the estimated reserves are listed in Tab. 2.3, copper production over the last seven thou-
sand years is shown in Fig. 2.2.
Exploitable reserves of recoverable copper were estimated at about 100 million tons in 1935; new dis-
coveries raised this to 212 million tons in 1960. Reserves grew again sharply to 340 million metric
tons (MMT) in 1984, but since then they have declined slowly to 321 MMT in 1990 and 310 MMT in
1994. Estimates vary according to prices and assumptions. Total potential resources have increased
somewhat over the same period, from 500 MMT to around 590 MMT. As a matter of interest, cumula-
tive global production of copper between 1970 and 1996 was 216 MMT (Ayres et al. (2002), p. 10).
According to USGS in 2002 (Tab. 2.3) estimated resources amount to 480 MMT (“reserves”) and po-
tential resources to 950 MMT (“reserve base”). With the actual mine production of around 13.5
MMT/a, the reserves would last 36 years and the reserve base 70 years.
Production of tellurium and silver are tightly related to the production of copper. Tellurium occurs in
tellurium-bearing minerals such as sylvanite ((Au, Ag),Te2) in copper and gold deposits. Silver and
tellurium are contained in small traces in the copper ore and arise in the anode slime of the electrolytic

refining of copper. Whereas tellurium is won almost exclusively as by-product from the copper pro-
duction 1 , copper ores contribute to 25-30% to the world silver supply (Klapwijk et al. (2006)).
Tab. 2.1 Resources. The most important copper minerals (Fabian (1997)).
Mineral Formula Copper wt-% Crystal system Density, g/cm3

Native copper Cu ¢ 99.92 Cubic 8.9
Chalcosite Cu2S 79.9 Orthorhombic 5.5-5.8
Digenite Cu9S5 78.0 cubic 5.6
Covellite CuS 66.5 hexagonal 4.7
Chalcopyrite CuFeS2 34.6 tetragonal 4.1-4.3
Bornite Cu5FeS4/Cu3FeS3 55.5-69.7 tetragonal 4.9-5.3
Tennantite Cu12As4S13 42-52 cubic 4.4-4.8
Tetraedrite Cu12Sb4S13 30-45 cubic 4.6-5.1
Enargite Cu3AsS4 48.4 orthorhombic 4.4-4.5
Bournonite CuPbSbS3 13.0 orthorhombic 5.7-5.9
Cuprite Cu2O 88.8 cubic 6.15
Tenorite CuO 79.9 monoclinic 6.4
Malachite CuCO3 · Cu(OH)2 57.5 monoclinic 4.0
Azurite 2 CuCO3 · Cu(OH)2 55.3 monoclinic 3.8
Chrysocolla CuSiO3 · n H2O 30-36 (amorphous) 1.9-2.3
Dioptase Cu6[Si6O18] · 6 H2O 40.3 rhombohedral 3.3
Brochantite CuSO4 · 3 Cu(OH)2 56.2 monoclinic 4.0
Antlerite CuSO4 · 2 Cu(OH)2 53.8 orthorhombic 3.9
Chalcanthite CuSO4 · 5 H2O 25.5 triclinic 2.2-2.3
Atacamite CuCl2 · 3 Cu(OH)2 59.5 orthorhombic 3.75
1
http://www.mmta.co.uk/economicsFacts/miningJournalReview.aspx, “Tellurium” in the Mining Journal Review, The Minor
Metals Trade Association, accessed in July 2007

Tab. 2.2 Occurrence. Typical copper contents of natural materials (Fabian (1997)).
Mineral Content (ppm)

Basalt 85
Diorite 30
Granite 10
Sandstone 1
Copper ores (poor) 5‘000
Copper ores (rich) 50‘000
Native copper 950‘000
Seawater 0.003
Deep-sea clays 200
Manganese nodules 10‘000
Marine ore sludges 10‘000
Earth's crust (average) 50
Meteorites (average) 180
Tab. 2.3 Mine production and reserves of copper in 2001. “Reserves” is referred to as “the part of the reserve base
which could be economically extracted or produced at the time of determination”, while “Reserve base” de-
notes “that part of an identified resource that meets specified minimum (...) criteria related to current mining
and production practices” (USGS (2003)).
Mine production Reserves Reserve base

2001 [kt Cu / a] 2002 [kt Cu / a] [kt Cu] [kt Cu]
(estimated)
United States 1340 1130 35000 70000
Australia 869 850 24000 43000
Canada 633 625 10000 23000
Chile 4740 4450 160000 370000
China 590 580 26000 63000
Indonesia 1050 1100 28000 38000
Kazakhstan 470 450 14000 20000
Mexico 367 330 27000 40000
Peru 722 850 35000 60000
Poland 474 500 31000 50000
Russia 620 680 20000 30000
Zambia 300 320 19000 35000
Other countries 1510 1500 60000 110000
World total (may be 13700 13400 480000 950000
rounded)

Tab. 2.4 World production of copper, silver and tellurium in tonnes per year, tpa. Data compiled after USGS (2003)
through 2004 and USGS (2007) for 2005/6.
Copper Silver Tellurium

year tpa world mine tpa world refin- tpa world refin-
production ery production ery production
1986 7'940'000 13'000 85.4
1987 8'240'000 14'000 73.8
1988 8'720'000 15'500 69.3
1989 9'040'000 16'400 65.0
1990 9'200'000 16'600 67.4
1991 9'330'000 15'600 82.9
1992 9'470'000 14'900 97.6
1993 9'490'000 14'100 88.2
1994 9'500'000 14'000 117.0
1995 10'000'000 14'900 175.0
1996 11'000'000 15'100 122.0
1997 11'500'000 16'500 109.0
1998 12'100'000 17'200 123.0
1999 12'800'000 17'600 116.0
2000 13'200'000 18'100 110.0
2001 13'700'000 18'900 109.0
2002 13'600'000 18'500 89.0
2003 13'700'000 18'400 95.2
2004 14'600'000 19'700 92.7
2005 15'000'000 19'300 119.0
2006 15'300'000 19'500 128.0
Fig. 2.2 History of copper production in logarithmic scale (Ayres et al. (2002)).

2.2 Primary Molybdenum

About the half of the world-wide produced molybdenum originates as co-product from the copper in-
dustry. The chief mineral values for the coupled production of molybdenum and copper are chalcopy-
rite (CuFeS2) and chalcocite (Cu2S), associated with pyrite (FeS2), and small amounts of other sul-
phides, including molybdenite. Generally there are five genetic types of molybdenum deposits
(Sebenik et al. (1997)):
1. porphyry deposits in which metallic sulphides are disseminated throughout large volumes of
altered and fractured rock,
2. contact-metamorphic zones and bodies in which silicated limestone is adjacent to intrusive

granites,
3. quartz veins,
4. pegmatites, and
5. deposits bedded in sedimentary rocks.

Types (1) – (3) are of hydrothermal origin and provide almost all of the molybdenum ore currently
mined. The last two are of no economic importance at present. Average molybdenite concentrations in
primary porphyry deposits range from 0.05 to 0.25 %; in secondary copper-molybdenum porphyry
deposits molybdenite concentrations are much lower (0.01 – 0.05 %) so that the mineral can only be
recovered as a by-product.
Molybdenum is obtained commercially almost exclusively from molybdenite, which is either mined
and concentrated as the primary product of the mine or open pit, or recovered as a concentrate during
the processing of ore from a copper mine. In the latter case the molybdenite can be either a by- or a co-
product, depending on its economic importance to the output of the mine. Molybdenum is also recov-
ered from solutions obtained by leaching scheelite ores in the production of tungsten. It is extracted as
a by- or co-product of copper production in Canada, Chile, Iran, Mexico, Papua New Guinea, Peru,
Philippines, and in the United States (Arizona, Nevada, New Mexico, and Utah).
Molybdenum supply in the Western world can be broken down into four production segments
(Sebenik et al. (1997)):
1) primary mine production (ca. 40 %),
2) by- or co-production output from copper and scheelite mines (ca. 55 %),
3) imports from the People's Republic of China (3 %), and
4) molybdenum recovered from the processing of spent petroleum catalyst (2 %).
The world mine production of molybdenum amounted to 130 kt in 2001 (USGS (2003)).
Reserves of molybdenite in the market economy countries have been estimated in a survey evaluating
identified ore bodies, i.e., those which have been explored as well as those which have been exploited.
The results indicate the following amounts of recoverable molybdenum: United States, 4100 kt; Chile,
1770 kt; Canada, 928 kt; Mexico, 306 kt; Peru, 288 kt; other countries, 356 kt. Ore bodies producing
primarily molybdenum contain 55 % of the reserves identified; only 29 % of these ore bodies were be-
ing exploited at the time of the survey (January 1985). Operations producing molybdenum as a by-
product contained the remaining 45 % of reserves and 67 % of them were producing molybdenite
(Sebenik et al. (1997)). The estimated total recoverable molybdenum from primary and byproduct re-
serves is listed in Tab. 2.5.

Tab. 2.5 Global reserves of molybdenum. Total recoverable reserves from primary production and as by-product
(Sebenik et al. (1997)).
Source of reserve Amount in [kt]

Primary producing properties 1'265
Primary nonproducing properties 3'036
Byproduct producing properties 2'316
Byproduct nonproducing properties 1'142
2.3 Secondary Copper

The recycling of copper is as old as the usage of the metal itself. World-wide refinery production in
year 2000 was 14.9 106 tonnes copper, of which 1.92 106 tonnes copper (=13.2%) originated from sec-
ondary copper sources (USGS (2003)). In Germany in 1994, 1.48 106 tonnes copper was consumed, of
which 0.65 106 tonnes were from secondary materials, of which again roughly 50% was re-cycled in
the metallurgical step (Krauss et al. (1999)).
Scrap is classified as prompt scrap (new scrap, manufacturing scrap), which is generated during the
fabrication of copper products, and old scrap, which is generated from products exiting the use phase.
These differ greatly in their quality. Because of its high purity, most of the prompt scrap can be di-
rectly re-melted, while a good deal of old scrap has to return to the refinery and the smelter stage.
About 36% of the copper in old scrap comes from pure copper products, and about 64% from alloy
products (Ayres et al. (2002)). Bertram et al. (2002) suggest to assume that the old scrap consists of
the waste categories ELV, C&D, and WEEE (Tab. 2.6).
Depending on the quality of the recycled material, the blending of primary and secondary substance
flows takes place at different production stages. In general, secondary copper used in smelters and re-
fineries is composed of 70% old and 30% prompt scrap. For the year 1994 Bertram et al. (2002) calcu-
lated an input of 540 kt of old scrap and 230 kt (=29.9%) of prompt scrap into the smelters / refineries
and an input of 200 kt of old scrap and 750 kt (=78.9%) of prompt scrap into the production of semis
(semi-fabricated products, such as shapes and ingots).
Tab. 2.6 Sources of copper waste for Europe in 1994 (Bertram et al. (2002)).
Source Total waste Copper content

Mt / a Uncer- % Mt Cu / a % Cu in Range % % of copper
tainty waste frac- waste
% tion
IW Industrial waste 350 ± 30 45.7 0.07 0.02 0.01-0.05 7.6
C&D Construction & demolition waste 184 ± 100 24.0 0.129 0.07 0.05-1 14
MW Municipal solid waste 165 ± 20 21.5 0.083 0.05 0.04-0.06 9.0
HW Hazardous waste 30 ± 10 3.9 0.015 0.05 0.02-0.08 1.6
ELV End of life vehicles 27 ± 20 3.5 0.243 0.9 0.8-1.0 26.4
WEEE Electrical & electronic equip- 2.9 ± 40 0.4 0.377 13 6-14 41
ment
SS Sewage sludge 7.7 ± 20 1.0 0.0028 0.037 0.02-0.04 0.3
Total 767 100 0.92 0.12 100

3.1 Copper
Copper [7440-50-8], the red metal, apart from gold the only metallic element with a colour different
from a grey tone, has been known since the early days of the human kind. It has always been one of
the significant materials, and today it is the most frequently used heavy non-ferrous metal. The utility
of pure copper is based on its physical and chemical properties, above all its electrical and thermal
conductivity (exceeded only by silver), its outstanding ductility and thus excellent workability, and its
corrosion resistance (a chemical behaviour making it a half-noble metal). Copper compounds and ores
are distinguished by bright coloration, especially reds, greens, and blues. Copper in soil is an essential
trace element for most creatures, including humans. Its common alloys, particularly brass and bronze,
are of great practical importance. Commercial grades of cathode copper today range from 99.95% to
99.97% pure.
Most properties of metallic copper depend on the degree of purity and on the source of the metal. Va-
riations in properties are caused by
- Grade of copper, i.e., the oxygen content: tough-pitch copper, deoxidised copper, oxygen-free
copper;
- Content of native impurities (e.g., arsenic) or remnants of additives (e.g., phosphorus), which form
solid solutions or separate phases at the grain boundaries;
- Thermal and mechanical pre-treatment of the metal, which lead to states such as cast copper, hot-
rolled copper, cold-worked (hard) copper, annealed (soft) copper, and sintered copper.
The atomic number of copper is 29, and the atomic mass Ar is 63.546±0.003. Copper consists of the
two natural, stable isotopes 63Cu (68.94 %) and 65Cu (31.06 %). There are also nine synthetic radioac-
tive isotopes with atomic masses between 58 and 68, of which 67Cu has the longest half-life, ca.
58.5 h.
The theoretical density at 20 °C, computed from lattice constant and atomic mass is 8.93 g/cm3. The
international standard was fixed at 8.89 g/cm3. The density of commercial copper depends on its com-
position, especially the oxygen content, its mechanical and thermal pre-treatment, and the temperature.
At 20 °C, a range of values from 8.30 g/cm3 to 8.93 g/cm3 are found.
The corresponding electrical resistivity (ρ) is 1.7241×10–8 W·cm. The theoretical conductivity at 20 °C
is nearly 60.0 MS/m or 103.4 % IACS, and today commercial oxygen-free copper (e.g., C10200 or Cu
– OF) has a conductivity of 101 % IACS.
3.2 Copper alloys: brass bronze and brazing solders

This chapter is mainly based on Isbell (1997) and on a internet source 2
There are scores of different copper alloys ranging from high copper alloys with copper content of
>99% (w/w) to brasses with only about 50% (w/w) copper. A differentiation of wrought and cast cop-
per alloys is common.
2
http://www.stahl-info.de/deutsch/schriften/MB235-WeichundHartloeten.pdf; accessed on 16.04.03

3.2.1 Brass
The series of alloys from which a majority of the industry's products are made are the copper-zinc al-
loys  the brasses. The most popular alloy of the group is cartridge brass (70/30 brass). Zinc is by far
the most important alloying element in the copper alloys. Lead can be alloyed to brasses to enhance
the machinability. If secondary material is used to produce copper alloys, contamination with un-
wanted heavy metals (e.g. lead, cadmium, beryllium...) can be a problem.
3.2.2 Bronze
Also of technical relevance are the copper-tin alloys  the tin bronzes. Beside about 5% (w/w) (maxi-
mum 10%) Sn, tin bronzes often contain some phosphorus and sometimes lead or silver. These alloys
provide the best combination of strength, electrical conductivity, and resistance to corrosion, as well as
great versatility.
The probably most important family of the speciality alloys is that of the copper-aluminium alloys -
the aluminium bronzes. Aluminium bronzes contain between 2 and 15% (w/w) Al, possibly with other
alloying elements. When cold finished products are to be manufactured, the aluminium content is lim-
ited to 9% (w/w). Aluminium bronze alloys may also contain up to 4.5% Fe, 2 % Sn, 6%, Ni, 3.5%
Mn and 3% Si to achieve special properties. These alloys combine good formability, good conductiv-
ity and high strength.
3.2.3 Solders
Alloys or pure metals are used as solders or brazing alloys. When their liquidus temperature is
<450 °C the process is called soft soldering, and when it is >450 °C it is known as brazing. In high-
temperature brazing, the operation is carried out in the absence of a fluxing agent and air, and the
bonding metal has a liquidus temperature >900 °C.
The main elements used in solders are copper, tin, zinc, silver, lead, cadmium and sometimes nickel,
manganese or cobalt. An important class of brazing solders are the "General-purpose, cadmium-free
brazing alloys". Their melting point is between 700 °C und 1100 °C.
Fluxes are organic or inorganic substances that remove oxides and other harmful layers from the sur-
face to be joined by soldering or brazing. They thereby provide the conditions that allow wetting by
the molten filler metal. Copper, copper - tin, or silver can be brazed in air without fluxes if phospho-
rus-containing copper-based brazing alloys are used. The phosphorus has a similar effect to that of a
flux.
3.3 Molybdenite concentrate and Molybdenum trioxide

Molybdenite concentrate is the purified product of the flotation of molybdenite ore and consists of 90
– 95 % Molybdenum disulphide. Molybdenum disulphide is the raw material for molybdenum triox-
ide, which is the starting point of most applications of molybdenum. Some properties of Mo-
compounds are listed in Tab. 3.1.
Molybdenum disulphide (CAS 1317-33-5; MoS2) is a black crystalline solid that occurs in two forms:
hexagonal, which is found in the mineral molybdenite (CAS 1309-56-4) and rhombohedral. The solid
is diamagnetic and a semiconductor. Lubricant-grade molybdenum disulphide consists of 99 % MoS2.
Molybdenum trioxide (CAS 1313-27-5; MoO3) is the ultimate product of the oxidation of molybde-
num compounds. It forms white orthorhombic crystals that are photosensitive, i.e., turn blue in light.
The compound is weakly paramagnetic and an n-type semiconductor. It is produced by roasting mo-

lybdenum disulphide in air at 600 °C and may be purified by sublimation. It can also be prepared by
reacting ammonium molybdate at 550 °C with oxygen or by precipitation from an aqueous ammonium
molybdate solution treated with concentrated nitric acid (Sebenik et al. (1997)).
Tab. 3.1 Properties of molybdenum trioxide and dioxide and molybdenum disulphide (Sebenik et al. (1997))
Property MoO3 MoO2 MoS2

Crystal structure orthorhombic monoclinic
Mo coordination octahedral, distorted octahedral, distorted
rutile
Hf°, kJ/mol* -745.2 -588.7 -234.1
Gf°, kJ/mol* -667.9 -533.1 -225.7
S° J/K* 77.7 47.2 62.8
Density, g/cm3 (21°C) 4.692 6.29
mp, °C 801 >1600
bp, °C 1155
Sublimation, °C ca. 700 ca. 1100
Solubility, g/L
Water 28°C 0.49 insoluble
* Standard enthalpy and free energy of formation from the elements and entropy at
298.15 K


While copper keeps its chemical pure form until its end use as alloying element or final product in dif-
ferent applications, molybdenum is mainly won as molybdenite (MoS2) which is the purified concen-
trate that results from beneficiation. Molybdenite is the starting point for the production of technical-
grade molybdenum trioxide and ferromolybdenum, which are used as alloying materials in the steel
production.
4.1 Copper
Before end-use the pure metal produced in refineries or re-melting plants is manufactured into semi
fabricated products, such as shapes and ingots (known as “semis”). These manufacturing processes are
subdivided into hot working, cold working and, if necessary, process annealing. These processes are
dealt with in Part IXX - Metal Processing.
Few of the semis are ready for use as such. They more typically travel to first-stage manufacturers,
who make metal rods, wire, and casting, and then to component manufacturers who produce such
things as motors, valves, and insulated cables (Fig. 4.1). The components in turn are incorporated by
final-stage manufacturers into a variety of finished products – automobiles, commercial buildings, re-
sidential housing, and so forth. These uses are shown in Tab. 4.1, they account for about 79% of the
world-wide consumption of copper. The figures in Tab. 4.1 are useful guides to the relative use magni-
tudes, albeit these copper containing products are the basis for copper re-use. The relations between
end-use, production and in- and export for Germany are given in Fig. 4.2.
Generally it can be said that copper is losing some traditional markets to other materials. One of them
is for telecommunications cables, mainly due to increasing use of glass fibres and wireless devices for
local distribution – the so-called “last mile”. Other traditional copper markets, such as copper roofing,
copper water pipe, and brass hardware of all kinds are under pressure from cheaper materials, includ-
ing aluminium, galvanised iron and plastics. On the other hand, copper is virtually irreplaceable for
local distribution of electric power, household wiring, motor-generator windings, electronic circuitry
and (as alloy) for some kinds of heat exchangers. Further substitution of copper by aluminium, in par-
ticular, seems unlikely.
Fig. 4.1 Copper “semis” production. An illustration of the transformation of copper from pure metal fabrication
flows through component manufacture (intermediate products) to a selection of final products. I and E indi-
cate import and export flows, respectively (Graedel et al. (2002))

Tab. 4.1 Principal end-uses of copper in 1990. RT is the average residence time of copper in the respective use; S/W
is the estimated distribution between scrap (recovering) and waste (disposal/incineration) at the end of use
(Graedel et al. (2002)).
# Category Use (%) RT (years) S/W

1 Building wire 14 45 50/50
2 Tube 12 60 45/55
3 Alloy rod 11 20 10/90
4 Magnet wire 9 15 50/50
5 Telecommunication wire 8 50 25/75
6 Power cable 8 40 60/40
7 Copper sheet and strip 8 50 60/40
8 Alloy sheet and strip 7 25 20/80
9 Casting alloys 6 30 50/50
10 Motor vehicle wire 4 10 80/20
11 Appliance wire 4 20 50/50
12 Bare wire 3 10 0/100
13 Copper rod 2 40 60/40
14 Alloy tube 2 35 95/5
15 Wire (other) 1 5 0/100
16 Alloy wire 1 5 5/95
17 Chemical and powder <1 1 5/95
Fig. 4.2 Copper foundry production, imports, exports & apparent consumption in Germany, 1913 - 1998 (3 year mov-
ing averages) (Ayres et al. (2002)).
4.2 Molybdenite concentrate

Flotation remains the preferred and almost exclusive method of upgrading the molybdenite mineral to
an industrial-grade molybdenite concentrate containing 90 – 95 % MoS2 [1317-33-5], the balance be-
ing gangue.

Some commercial-grade molybdenite concentrate is subjected to additional grinding and flotation to

produce lubricant-grade molybdenum disulfide (99 % MoS2). The purified product is added to grease
and oils as a solid lubricant. Molybdenite is converted to technical-grade MoO3.
4.3 Molybdenum trioxide and Ferromolybdenum

Molybdenum is mainly used as an alloying element in steel, cast iron, and superalloys to increase har-
denability, strength, toughness, and corrosion resistance. Initially, molybdenum was used in the lamp
industry. Nowadays it is an increasingly important material in a wide range of applications (Tab. 4.2).
In high-temperature applications, molybdenum competes with iron-, nickel- and cobalt-base superal-
loys, ceramics, and other high-melting metals (tungsten, tantalum, and niobium).
Molybdenite concentrate is converted to technical-grade MoO3 by roasting it in air in a multiple-hearth
furnace. The principal commercial molybdenum product is technical-grade molybdenum trioxide
(MoO3) [1313-27-5]. In addition to being the major compound for adding molybdenum to steel, it is
also the starting compound for all other molybdenum products including ammonium dimolybdate,
pure molybdenum trioxide, sodium molybdate, molybdenum metal, and finally ferromolybdenum.
In the steel and foundry industry molybdenum is added to melts, either as technical oxide or in the
form of ferromolybdenum [11121-95-2]. Compared to pure molybdenum, ferromolybdenum dissolves
much more easily in the steel melt and is cheaper to produce.
The principal raw materials used for ferromolybdenum production are molybdenum concentrates and
technical-grade molybdenum trioxide. Technical-grade molybdenum trioxide is commonly used as the
molybdenum raw material. Two principal processes are used for molybdenum oxide reduction: car-
bothermic reduction in a submerged arc furnace and metallothermic reduction. The most commonly
applied method is the silicothermic reduction, a type of metallothermic reduction (Sebenik et al.
(1997)).
Tab. 4.2 Uses of molybdenum and molybdenum alloys (Sebenik et al. (1997)).
Area of application Examples of uses

Lamp and lighting industries support wires, sealing ribbons, dimming caps
Electronic and semiconductor industries semiconductor base plates, heat sinks, contact pins, sputter-
ing targets, control grids, klystron and travelling-wave tube
components
High-temperature and vacuum furnace heating elements, thermal radiation shields, furnace ware
construction
Glass and ceramic industries glass-melting electrodes, installations in glass production
tanks, drawing dies, crucibles for producing sapphire single
crystals
Casting technology and metalworking forging dies for isothermal forging, extrusion dies, die-casting
moulds, hot-galvanizing equipment
Coating spray wire, spray powder, evaporation boats, sputtering tar-
gets, components of chemical vapour deposition equipment
Nuclear technology furnace parts and charging equipment for sintering UO2; first
wall, diverter, and limiter components for experimental fusion
reactors
Medicine rotating X-ray anodes, collimators

In this study the production of copper and molybdenum is modelled as coupled production system. In
a first step the ore – jointly containing copper and molybdenum – is exploited. This mining and ben-
eficiation step is formally a multi-output process, yielding copper concentrate and molybdenite con-
centrate. Depending on whether the raw material was an ore from primary molybdenum porphyry or
from another sulphide deposit, either copper is the co-product or molybdenite. Since the coupled pro-
duced copper from primary molybdenum production is of no further importance, the coupled produced
molybdenite from primary copper production accounts to 55% of the molybdenum production world-
wide. The global molybdenite mix is thus formed together with the 40% “pure” molybdenite produc-
tion, accounting for 95% of the overall molybdenum production (Fig. 5.1).
exploitation copper-ore,
copper,
copper, primary, at
 concentrate, at 
refinery, RNA
beneficiation, RNA
RNA
molybdenum,
molybdenite
concentrate,
coupleproduction Cu
copper,
copper, primary, at
refinery, RLA
beneficiation, RLA
RLA
molybdenum,
molybdenite
concentrate,
coupleproduction Cu
copper,
copper, primary, at
refinery, RER
beneficiation, RER
RER
molybdenum,
molybdenite copper, primary, at
concentrate, refinery, GLO
coupleproduction Cu
molybdenite,
copper,
copper, primary, at concentrate,
refinery, RAS coupleproduction Cu,
beneficiation, RAS
GLO
RAS
molybdenum,
molybdenite
concentrate,
coupleproduction Cu
58%
copper,
copper, primary, at molybdenite, at plant,
refinery, ID GLO
beneficiation, ID
ID
molybdenum,
molybdenite
concentrate, 42%
coupleproduction Cu
exploitation molybdenite
molybdenite,
concentrate, GLO
ore, GLO
copper, concentrate,
coupleproduction Mo,
GLO
Fig. 5.1 Schematic representation of the coupled production of copper and molybdenite in ecoinvent. RNA, RLA,
RER, RAS, ID refer to the regions North America, South / Middle America, Europe, Russia / Asia, Indonesia /
Australia; GLO is the global average.

The inventory representing the consumption of copper in Europe consists of a mix of imported pri-
mary metal from different parts of the world, imported copper concentrate which is refined in Europe
itself, and secondary metal, which is partly recycled into foundries and partly into prior metallurgical
steps (Fig. 5.2). As a proxy for Europe, the situation for Germany in 1994 is chosen.
The meta information on the modules described in this chapter are summarised in Tab. 5.1.
exploitation copper-ore, RLA
copper,
concentrate, at copper, primary, at

beneficiation,

refinery, RLA 7.5% Primary copper, RLA
RLA
molybdenum,
molybdenite concentrate,
coupleproduction Cu
exploitation copper-ore, ID
copper,
copper, primary, at
refinery, ID 2.2% Primary copper, ID
beneficiation, ID
molybdenum,
coupleproduction Cu
copper,


beneficiation,

refinery, other 9.4% Primary copper, Other
other
other
molybdenum,
coupleproduction Cu
exploitation copper-ore, RER
molybdenum,
coupleproduction Cu
copper,

beneficiation,

refinery, RER 23.9% Primary copper, RER
RER
copper, blister copper,

at primary smelter, RER
635'000 t
10%
Secondary Copper (old

copper, at regional storage, RER
copper, secondary,
at reffinery 22.0%
scrap), RER
Copper scrap, 325'000 t
from collection
point
Secondary Copper
22.0%
325'000 t (prompt scrap), RER
copper,
Transportation to RER
concentrate, at 
beneficiation, RLA
(652'883 t)
copper, copper, from imported

Primary copper, from
concentrate, at   concentrates, at refinery, 13.1% imported concentrate,
beneficiation, OCE RER
193'000 t RER
copper,
concentrate, at 
beneficiation, RER
Fig. 5.2 Schematic representation of the consumer mix of copper in RER in ecoinvent. As a proxy for Europe, the
consumption pattern for Germany in 1994 is chosen.

Tab. 5.1 Ecoinvent meta information for the processes for the production of primary copper and primary molybde-
num.
Name mining copper ore copper, primary, at refinery

Location GLO, RNA, RLA, RER, RAS, ID GLO, RNA, RLA, RER, RAS, ID
Unit kg kg
The module includes the pre-treatment
The module includes a mining and a benefici- of the ore, the reduction and the refining
Included Processes ation step with the mining infrastructure and copper to cathode copper, including the
disposal of overburden and tailings (see part country-specific share in hydrometalur-
"Auxiliary Processes"). gically won copper (SX-EW)
Amount 1 1
Local Name Abbau Kupfererz Kupfer, primär, ab Raffinerie
Copper concentrate // Kupfererz konzentrat //
Synonyms
Molybdenite // Molybdänit
The multi-output-process "mining copper-ore"
delivers the co-products "copper concentrate,
at beneficiation" and "molybdenum, molyb-
denite concentrate, couple production Cu". The
product "copper concentrate" is intermediary
required in the following copper production
General comment to
(5.2.5 Winning of primary copper in ecoinvent).
reference function
The module "molybdenite concentrate" is de-
signed for the use in Ecoinvent as starting ma-
terial for the manufacturing of molybdenum tri- The product "copper, primary, at refin-
oxide and ferromolybdenum, both materials ery" is used as pure metal or as alloying
which used in the manufacturing of high quality element in various technical applica-
steels and alloys. tions.
CAS Number 7440-50-8
End Date 2003 2003
Period 1 1
Other Period Text
This module is valid for the whole correspond-
ing region. Some agents used like lime or so- This module is valid for the whole corre-
dium cyanide originate from data sets valid for sponding region. Some agents used like
Geography text
Switzerland or Europe. Processes like blasting lime or sodium cyanide originate from
and transport originate from data sets valid for data sets valid for Switzerland or
Switzerland or Europe. Europe.
The ore is pre-treated, reduced and re-
Mining is done 70% open pit and 30% under- fined according to the country specific
ground, followed by joint beneficiation of cop- mix of process alternatives: reverbera-
Technology text per and molybdenite through flotation, where tory furnace 23.7%; flash smelting fur-
considerable amounts of agents are added. naces 60.7%; other 6.2%.; SX-EW
Overburden and tailings are disposed near the 9.4%. An overall abatement for sulphur
mining site. dioxide of 45.4% was estimated.
Production Volume 9.5 million tonnes in Cu in 1994 9.5 million tonnes in Cu in 1994
Sampling Procedure literature literature
Extrapolations see Geography and Technology see Geography and Technology

num. (continued)
copper, blister-copper, at primary

Name copper, SX-EW, at refinery
smelter
Location RER GLO
Unit kg kg
The module includes the mining of the
ore including the mining infrastructure,
Included Processes the leaching and extraction, the electro
The module includes the same steps as the winning and the disposal of the leaching
primary copper production without the refining. residues.
Amount 1 1
Local Name Kupfer, Blisterkupfer, ab Verhüttung Kupfer, SX-EW, ab Raffinerie
Synonyms
The product "blister copper", i.e. raw copper
after converting with an average Cu content of
98%, is an intermediary product from primary
General comment to
copper production used uniquely in the secon- The product "copper, SX-EW, at refin-
reference function
dary copper production. It makes 90% of the ery" is used in the production mix repre-
feed for secondary copper, whereas copper sented in module "copper, primary, at
scrap makes the remaining 10%. refinery" of the respective region.
End Date 2003 2003
Period 1 1
Other Period Text
This module is valid for the whole correspond-
ing region. Some agents used like lime or so-
Geography text
dium cyanide originate from data sets valid for This module is valid for an average
Switzerland or Europe. world-wide process.
Heap leaching was assumed as average
The ore is pre-treated and reduced according process. The ore is mined in open cut
to the country specific mix of process alterna- technique, leached and extracted with
Technology text tives: reverberatory furnace 6.2%; flash smelt- sulphuric acid and reduced by electro
ing furnaces 76%; other 17.8%. An overall winning. For electricity mix a mix of 60%
abatement for sulphur dioxide of 95% was es- hydropower and 40% electricity at grid
timated. UCTE was chosen.
2.3 million tonnes in Europe in 1994 , SX-EW
Production Volume
copper excluded 870 thousand tonnes in Cu in 1994

num. (continued)
copper, from imported concentrates,

Name copper, secondary, at refinery
at refinery
Location RER DE
Unit kg kg
The module includes the collection and The module includes the pre-treatment of
handling of the copper scrap, the refining the ore, reduction and refining of copper
Included Processes
and the processes for waste water treat- cathodes in RER. It does NOT include the
ment. mining and beneficiation step.
Amount 1 1
Kupfer, aus importierten Konzentraten, ab
Local Name
Kupfer, sekundär, ab Raffinerie Raffinerie
Synonyms
The module describes the production of
secondary copper in Germany with an out-
put of 176'000 t copper cathodes in 1994. It The product "copper, from imported con-
is designed for the use of the metal as part centrates, at refinery" is an auxiliary module
General comment to
of the supply mix "copper, at regional stor- used uniquely in the consumption mix in
reference function
age". It is not to be used if the impact of the order to separate concentrate production
secondary copper is considerable in a LCA and their transportation from different coun-
study. In such cases, a more detailed as- tries and the pyrolytical winning in RER. It
sessment has to be conducted. does not contain any copper bearing inputs.
End Date 2003 2003
Period 1 1
Other Period Text
This module is valid for the whole corre-
This module is valid for one single very sponding region. Some agents used like
Geography text
large plant, which is taken as proxy for pro- lime or sodium cyanide originate from data
duction in RER. sets valid for Switzerland or Europe.
The ore is pre-treated, reduced and refined
In three different stages, different types of according to the country specific mix of
copper scrap and 10% of the feed of blister process alternatives: reverberatory furnace
Technology text
copper are refined to copper cathodes. 6.2%; flash smelting furnaces 76%; other
Waste water is led to a communal treat- 17.8%. An overall abatement for sulphur
ment plant. dioxide of 95% was estimated.
Representativeness [%] 15.3 100
Env. 50% of the 2.3 million tonnes = 1.2
Production Volume
million tonnes in STAF Europe in 1994. 193000 tonnes in 1994 from this source

num. (continued)
Name mining molybdenite molybdenite, at plant

Location GLO GLO
Unit kg kg
The module includes a mining and a ben-
eficiation step with the mining infrastructure All relevant processes in mining, benefici-
Included Processes
and disposal of overburden and tailings ation, metallurgy and disposal of the tailings
(see part "Auxiliary Processes"). are included.
Amount 1 1
Local Name Abbau Molybdänit Molybdänit, ab Fabrik
Synonyms Molybdenite // Molybdänit
The module describes the mining of molyb-
denite as main product. It is designed solely The module describes the production mix
as part of the molybdenum production mix. composed of molybdenite from coupled
The multi-output-process "mining of molyb- production in copper industry and molyb-
General comment to denite" delivers the co-products "molybdenite as main product. It is designed for
reference function denite concentrate, main product" and the use in ecoinvent as starting material for
"copper, concentrate, couple production the manufacturing of molybdenum trioxide
Mo". Molybdenite from coupled production and ferromolybdenum, both materials finally
module is designed for its use in the world- used in the manufacturing of high quality
wide production-mix. steels and alloys.
End Date 2003 2003
Period 1 1
Other Period Text
Mining is done 70% open pit and 30% un-
derground, followed by joint beneficiation of
copper and molybdenite trough flotation,
Geography text
where considerable amounts of agents are
added. Overburden and tailings are dis-
posed near the mining site.
This module is valid for the whole corre-
sponding region. Some agents used like
lime or sodium cyanide originate from data
Technology text sets valid for Switzerland or Europe. Proc-
esses like blasting and transport originate
from data sets valid for Switzerland or This production mix is fully based on the
Europe. related mining of copper ore.
Production Volume 52000 t Mo in Molybdenite 130000 t Mo in Molybdenite
Extrapolations see Geography and Technology see Technology

num. (continued)
Name copper, at regional storage molybdenum, at regional storage

Location RER RER
Unit Kg kg
Transport of primary metal to Europe from The module includes the production of high
the countries importing to Europe is in- grade primary molybdenum by pyro- and
Included Processes cluded, as the import of concentrate, which hydrometallurgical processes, the disposal
is processed in RER. As import pattern the of the slag, and the treatment of the waste
situation of Germany in 1994 was chosen. water.
Amount 1 1
Local Name Kupfer, ab Regionallager Molybdän, ab Regionallager
Synonyms
Molybdenum is used e.g. in catalytic con-
verters. It is not used for alloying steel, for
which molybdenum trioxide or molybdenite
are used. For lack of specific data, the
General comment to metallurgical process of the zinc production
reference function The module characterises the copper used is used as a proxy. Only the input of the
in Germany in 1994, reflecting its origin and concentrate is adapted. Although zinc and
the fraction of secondary metal. It is de- molybdenum concentrate are both sulfidic
signed for the use of the metal in various and the metal content is almost the same,
technical applications such as alloys and this is a crude assumption and the data
construction material. quality therefore is poor.
CAS Number
End Date 2003 2003
Period 1 1
Other Period Text
In this module the consumption pattern of
Geography text Germany in 1994 is used as proxy for the
situation in RER.
This consumption mix represents the tech-
nology used of the countries importing to A mix of 80% hydrometallurgical and 20%
Technology text Germany in 1994 and differentiates be- pyrometallurgical production of zinc is cho-
tween pyrolytical processes, hydrolytical sen to represent the molybdenum produc-
processes and secondary copper. tion.
Production Volume 1.5 million tonnes in Cu in 1994 unknown
Extrapolations see Geography and Technology see Technology

5.1 Exploitation of copper ore

Copper ore is being mined for thousands of years. In the past decades the grades of worked copper
ores continuously fell and dropped recently clearly below 1% world-wide (Fig. 2.1 for U.S. deposits).
Nevertheless, mining and beneficiation techniques remain basically the same with open cut mining
and flotation beneficiation. The hydrometallurgical process skips the beneficiation step in favour of a
leaching step, which is followed by electro-winning.
5.1.1 Production process

Mining. Sulphidic copper ores are mined only 30% underground, the major part is mined in large o-
pen cut operations. According to (Ayres et al. (2002)) the ore mined in 1900 in the U.S. had a high
content of 3.4% and was mined entirely underground. Open pit mining permits the use of very large
equipment. Resulting economies of scale enable the exploitation of lower grade disseminated (por-
phyry) ores – the ores now mainly mined.
Beneficiation. After mining, the ore is first ground. In a next step it is subjected to gravity concentra-
tion to separate the metal-bearing particles from the unwanted minerals. After this first concentration
step, flotation is carried out to remove the gangue from the sulphidic minerals. For neutralisation lime
is added. In the flotation several organic chemicals are used as collector, frother, activator, depressor
and flocculant. Sometimes cyanide is used as depressant for pyrite. Tailings usually are led to tailings
heaps or ponds. As a result, copper concentrates containing around 30% Cu are produced. Molyb-
denite concentrate are further ground and purified. It leaves the process as co-product with a concen-
tration of 90 – 95 % Molybdenum disulphide.
5.1.2 Emissions & Waste

Mining
The major emissions are due to mineral born pollutants in the effluents. Open cut mining generates
large quantities of dust, which contains elevated contents of metals and sulphur. Rain percolates
through overburden and accounts to metal emissions to groundwater. Overburden is deposed close to
the mine.
Beneficiation
Ore handling and processing produce large amounts of dust, containing PM10 and several metals from
the ore itself. Flotation produces effluents containing several organic agents 3 used. Some of these
chemicals evaporate and account for VOC emissions to air. Namely xanthates decompose hydrolyti-
cally to release carbon disulphide. Tailings effluent contains additional sulphuric acid from acid rock
drainage.
Tailings are deposed as piles and in ponds. In the sulphidic tailings occurs acid rock drainage (ARD)
over a long period of time. These effects are handled in the section “Non Ferrous Metal Winning –
Auxiliary Processes”.
3
Floating agents consist of collectors (xanthate or aerofloat) and frothing reagent (eg. Methyl Isobutyl Carbinol). Xanthanes
are a broad class of organic salts formed by treatment of an alcohol with carbon disulfide in the presence of an alkali, potas-
sium ethyl xanthate (KC3H5S2O, CAS No: 140-89-6). Aerofloat is a salt of carbonyldithiophosphoric acids. A widely used
frothing reagent is e.g. methyl isobutyl carbinol, CAS: 108-11-2, syn. Methyl Amyl Alcohol.

5.1.3 Exploitation of copper-ore in ecoinvent

The modules describe the coupled production of copper- and molybdenum-concentrate in 1994 for dif-
ferent regions of the world and the world-wide production mix. Please note that in the database dis-
tinct datasets for the regions Northern America (RNA), South America (RLA), Europe (RER), Asia
and Russia (RAS), Indonesia & Australia (ID) and the world-wide average (GLO) exist. However, in
this report only the global (GLO) module is discussed in detail, the regional modules do not differ in
their structure but only in their values. This module yields copper concentrate, which is used in py-
rometallurgical processes. Hydrometallurgical copper winning – e. g. SX-EW – is based on different
beneficiation techniques. Following modules in ecoinvent are covered in this chapter:
- mining copper ore (GLO)

with the outputs:
- copper, concentrate, at beneficiation (GLO)
- molybdenum, molybdenite concentrate, coupleproduction Cu (GLO)
- mining copper ore (RNA)

with the outputs:
- copper, concentrate, at beneficiation (RNA)
- molybdenum, molybdenite concentrate, coupleproduction Cu (RNA)
- mining copper ore (RLA)

with the outputs:
- copper, concentrate, at beneficiation (RLA)
- molybdenum, molybdenite concentrate, coupleproduction Cu (RLA)
- mining copper ore (RER)

with the outputs:
- copper, concentrate, at beneficiation (RER)
- molybdenum, molybdenite concentrate, coupleproduction Cu (RER)
- mining copper ore (RAS)

with the outputs:
- copper, concentrate, at beneficiation (RAS)
- molybdenum, molybdenite concentrate, coupleproduction Cu (RAS)
- mining copper ore (ID)

with the outputs:
- copper, concentrate, at beneficiation (ID)
- molybdenum, molybdenite concentrate, coupleproduction Cu (ID)
- copper, concentrate, at beneficiation (RNA)
- copper, concentrate, at beneficiation (RLA)
- copper, concentrate, at beneficiation (RER)
- copper, concentrate, at beneficiation (RAS)
- copper, concentrate, at beneficiation (ID)
The layout of the modules with general flow information, remarks, sources, values and uncertainty in-
formation are shown in Fig. 5.3 and Fig. 5.4. Ecoinvent meta information for this process are summa-
rised in Tab. 5.1 on page 17. The following paragraphs describe calculations, sources and assumptions
chosen in this study.

Allocation
The ecological burden of the copper ore mining is distributed among the coupled products copper con-
centrate and molybdenite concentrate. As allocation procedure the revenue is chosen. The molybde-
num price is quoted as US$ per kilogram molybdenum content in molybdenite concentrate. This price
is stable since 1983 with a value of around US$ 5.75 per kg Mo (USGS (2003)), this gives a revenue
of US$ 3190 per tonne molybdenite concentrate with an assumed composition of 92.5% MoS2. Copper
concentrate is traded in large quantities. The price of copper concentrate could not yet be assessed; it is
estimated with the final bids of three auctions held in 2000 on an online web-auction 4 resulting in an
average price of 15.3 US cents per pound copper payable for refining and smelting. With a copper
price of 0.95 US$/kg a price of 522 US$ per tonne of concentrate with 29.7% copper content is calcu-
lated. These assumptions lead to an average revenue of 17 US$ and 1816 US$ per tonne molybdenite
and copper, respectively, contained in the ore – with the metallurgical yield included – and thus to an
allocation factor of 1% and 99% for molybdenite and copper concentrate, respectively.
Infrastructure and land use

The production infrastructure used directly for exploitation is inventoried within this module, whereas
the infrastructure like buildings and land-use for the operation is inventoried within auxiliary modules,
which comprises both mining and beneficiation facilities. These datasets are treated in the part auxil-
iary processes of this report. The mobile equipment used for exploration like drilling machines, scoop
trams and underground trucks is inventoried with the diesel consumption of 69.2 kJ per tonne mined
raw ore. The used module “diesel, burned in building machine” includes the infrastructure of the ma-
chines. Conveyor belts are further assumed with a capacity of 2000 t/h which yields 7.6E-5 m con-
veyor belt required per tonne mined raw ore (Tab. 5.2).
According to Part XX, Non Ferrous Metal Winning – Auxiliary Processes, land use was assessed with
the value for open pit mining (0.0025 m2/t) and the respective share in the world copper (70 %) pro-
duction, yielding a 0.00174 m2/t and 0.0522 m2a/t respectively. For underground mining no direct land
use is estimated. A lifetime of 30 years is assumed.
Tab. 5.2 Assumptions for the infrastructure used in underground mining exploitation. As lifetime ten years are esti-
mated, and 5000 working hours yearly.
action share machine throughput unit infrastructure

t / lifetime Length (m) m / t ore mined
Open pit 70% Conveyor belts 100'000'000 2000 1E-5
Underground 30% Conveyor belts 100'000'000 2000 2E-5
Mix 7.6E-5
Mining
In the considered process the ore is mined mainly in open cut operations. According to Martens et al.
(2002) 69.6 % of the copper world production originates from open pit mines, but no further differen-
tiation in regions or countries was available. Hence the world average is used for all regions. The
overburden is disposed in piles near the mine, the standard module for disposal of non-sulphidic over-
4
http://www.copperconcentrate.com/default.asp, accessed on 01.03.2003

burden was chosen, which includes the land use and transformation through the disposal. No overbur-
den is refilled according to Krauss et al. (1999), the amount is documented for each of the five world
regions. Also water use is documented, but as overall value, which is comprised in the following ben-
eficiation step. Concerning electricity supply mix, Hilbrans & Hinrichs (1999) assume for the nickel-
production a share of 60% of hydropower due to the operations high and constant energy demand.
This assumption is used also in this inventory.
Beneficiation
Once the ore is mined, it is ground and conditioned for further concentration. Considerable amounts of
lime and flotation agents are added. The separated gangue is disposed in tailings ponds; the concen-
trated ore is fed to the metallurgy, which is assumed to be on-site. The respective standard module
“disposal, sulphidic tailings, off-site” is chosen. (see “Non Ferrous Metal Winning – Auxiliary Proc-
esses”, Chapter “Disposal of Tailings and Overburden”).
The infrastructure of beneficiation is included in the mining infrastructure. Dust emissions comprise
additionally the dust emissions due to mining activities. Like in the mining step, a share of 60% hy-
droelectricity was chosen according to Hilbrans & Hinrichs (1999).
Lime: Lime input was documented in Krauss et al. (1999) for each process alternative with a range of
1200–1330g / t ore.
Steel: The steel input for milling (substitution of components for abrasion) was documented and
ranges from 690 to 920 g/t with an average of 765 g/t ground ore assumed herewith. The mill work
consists of special austenitic manganese steel of Hadfield grade, which is inventoried with chromium
steel as a proxy.
Reagents: No exact information on mass and nature of the floating agents used had been available,
therefore general values were assessed in Krauss et al. (1999) based on values in year 1978. Newer
values reported in IPPC (2001) have been taken in this study (Tab. 5.4). For copper mining a range of
25 – 300 g collector and 25 – 250 g frothing reagent per tonne sulphidic ore was recommended in
(Krauss et al. (1999)). This is confirmed by the values in Tab. 5.4. Collector and frother are invento-
ried for ecoinvent as unspecified organic chemicals. According to NPI (2001), xanthates hydrolyti-
cally decompose to 0.53 mg carbon disulphide per mg xanthate. With 189 g xanthates per tonne ore
from Tab. 5.4 this corresponds to 100 g carbon disulphide per t mined ore (Tab. 5.5).
Activator and Depressants – excluding sodium cyanate – are modelled as unspecific inorganic chemi-
cals; their emission to water is covered by the emissions reported from mining sites, which are shown
in Tab. 5.3.
For pH adjustment and leaching, the value for sulphuric acid was considered with 3.74 kg/t ore mined
(Tab. 5.5). As sulphuric acid is a by-product from the metallurgy, its production is regarded as burden
free. Requirement of other acids is neglected. Acid emissions to water due to the use of sulphuric acid
are not accounted.
Since composition and environmental fate of activator, depressant and sulphuric acid is uncertain, re-
ported data for total emissions shown in Tab. 5.4 are taken. Note, that some theoretical emissions due
to the use of the reagents are higher than the reported ones in Tab. 5.3.
Sodium cyanide is widely used for extracting gold and other metals such as silver, copper, and zinc. It
is used as leachate and as depressant for certain sulphidic minerals. A cyanide recovery of 80 – 90% is
reported due to abatement and recovery in the Golden Cross mine in New Zealand (Anonymous
(1998a)). Here a value of 80% is chosen for cyanide-abatement technology. With a reported use of
35 g NaCN per tonne ore mined (Tab. 5.4), this means a direct emission of 3.88 g HCN per tonne ore
mined. An overview of the reagents accounted for is presented in Tab. 5.5.
Water use & Effluents: Water use was documented in Krauss et al. (1999) and has been inventoried
in this study as originating from river without pumping activities. Additional to direct emissions due to

the use of chemicals, total emissions to water are chosen according to Tab. 5.3. These values are from
base metal sites, and are thus only an approximation for the real effluent concentrations in the copper
production. It is assumed here that the effluents also include acid rock drainage (ARD). ARD after
mine closure is modelled in the module “disposal, sulphidic tailings, off-site”. Effluent volume was
not documented in Krauss et al. (1999); for this study it has been assumed that the water requirements
equal the effluents.
Dust: A case study in Anonymous (1998b) on a projected open cut gold-copper mining and process-
ing facility in Australia predicts for dust emissions per tonne of ore mined values between 0.18 kg/t
and 0.46 kg/t. The prediction was made using the Emission Estimation Technique Manual for Mining
(NPI (2001)). The maximum value of 0.46 kg/t is chosen for total dust emissions. According to the
IAEA 5 the dust composition is approximated with values of a specific sulphidic nickel ore in Tab. 5.6.
For PM10 values, the ratio TDS : PM10 : PM2.5 was approximated with values from copper ore mining
101.72 : 50 : 5 g / t ore taken from the CEMIP-database 6 , i.e. one tonne emitted dust corresponds to
491.5 kg PM10 and 49.2 kg PM2.5 respectively. The airborne emissions resulting from mining and ben-
eficiation activities are summed up in Tab. 5.6. In ecoinvent, for particulates differential classes are
used adding up to the total dust emission, i. e. 49.2 kg PM<2.5, 442 kg PM2.5-10 and 508 kg PM>10.
Data quality
Mining; Data for major mass fluxes are satisfactory, while the emissions into water may be overesti-
mated due to different system boundaries of reported effluent volume from (Krauss et al. (1999)) and
total effluent emissions reported in (IPPC (2002)). High uncertainties are reflected by a high standard
deviation (stdev95) of 10.
Beneficiation; Like in mining, the emissions to water can be overestimated. The emissions from tail-
ings are assessed in an external module, which itself contains rough assumptions of a global overbur-
den and tailings composition and a simplified disposal scenario. Emissions though can vary considera-
bly from case to case.
5
The IAEA Database of Natural Matrix Reference Materials, http://www-naweb.iaea.org/nahu/external/e4/nmrm/, accessed on
01.03.2003
6
http://www.air.sk/tno/cepmeip/em_factors.php, accessed 01.03.2003

Tab. 5.3 Composition of effluents from mining activities; values are taken from total emissions to water from base
metal sites in (IPPC (2002)).
Compound Unit Range This study Number of

(geometric Samples
mean)
Discharge Mio. m3 / a 0.08 - 22.9 3.65 6
Ca g / m3 686 686 1
SO4 g / m3 1750 - 3175 2357 2
COD g / m3 2.24 - 48.5 10.4 2
Solids g / m3 2.38 - 11.3 5.17 4
Al mg / m3 69.3 - 108 86.3 2
As mg / m3 0.264 - 14.1 2.95 3
Cd mg / m3 0.0901 - 1.02 0.316 4
Co mg / m3 0.742 - 0.823 0.781 2
Cr mg / m3 0.0311 - 9.62 0.547 2
Cu mg / m3 1.24 - 44.9 7.91 5
Fe mg / m3 61.7 - 1326 291 3
3
Mn mg / m 24.7 24.7 1
Hg mg / m3 0.0155 - 0.12 0.0377 3
Ni mg / m3 0.792 - 1338 24.3 3
Pb mg / m3 0.0155 - 20 2.80 4
Zn mg / m3 5.37 – 225 75.9 5
N g / m3 2.50 - 1786 22.8 3
Tab. 5.4 Consumption of reagents in base metal mineral processing plants. Data taken from (IPPC (2002)).
Reagents Unit Range This study Number of

(geometric samples
mean)
Collectors – Xanthate g/t 110 - 300 188.6 6
Frothers - Sylvapine, g/t 9.7 - 150 37.2 5
MIBC, Dowfroth
Sodium cyanide g/t 4 - 310 35 2
Depressants – Zinc sul- g/t 35 - 432 231.5 5
phate, Iron sulphate, Ace-
tic acid, Sodiumchromate,
Dithiophosphate
Activator - Copper sul- g/t 433 - 876 538 4
phate
Flotation agents g/t 19.0 19.0 1
Flocculants - CMC g/t 1 - 100 11.1 3
Sulphuric acid (pH) g/t 387 - 12000 3739 4
Sodium hydroxide (pH) g/t 30 - 490 121 2
Nitric acid (pH) g/t 150 150 1
Hydrochloric acid (pH) g/t 1 1 1

Tab. 5.5 Emissions and input due to chemical use in beneficiation in g per tonne treated ore. The values are the ones
chosen in this study. For further details read in text.
Reagent Type of burden Unit Chosen value

Collector input technosphere g/t 189
Carbon disulfide air emission g/t 100
Frother input technosphere g/t 37
Lime input technosphere g/t 2000
Carbon dioxide air emission g/t inventoried in metallurgical step
Activator/Depressant input technosphere g/t 828
Sulphuric acid input technosphere g/t 3739
Sodium Cyanide input technosphere g/t 35.2
Cyanide water emission g/t 3.88
Tab. 5.6 Emissions to air resulting from mining and beneficiation. Values for total dust emission are taken from NPI
2001
Emission to air Unit Emission per tonne mined ore

Antimony Sb mg / mg dust 0.0000002
Arsenic As mg / mg dust 0.0000015
Beryllium Be mg / mg dust 0.0000026
Boron Be mg / mg dust 0.00001
Cadmium Cd mg / mg dust 0.00000011
Chromium Cr mg / mg dust 0.0001
Cobalt Co mg / mg dust 0.00002
Copper Cu mg / mg dust 0.00005
Fluorine F mg / mg dust 0.00095
Lead Pb mg / mg dust 0.000014
Manganese Mn mg / mg dust 0.00095
Mercury Hg mg / mg dust 0.00000005
Nickel Ni mg / mg dust 0.00008
Selenium Se mg / mg dust 0.00000005
Zinc Zn mg / mg dust 0.00019
PM<2.5 mg / mg dust 0.0492
PM2.5-10 mg / mg dust 0.442
PM>10 mg / mg dust 0.508

General Flow Information Representation in ecoinvent Uncertainty Informations
mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Remarks
Location
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
Copper, 0.99% in sulfide, Cu
83.5% yield, 62.1 kg overburden, 35.9 kg ore,
Copper in ground  resource in ground 0.36% and Mo 8.2E-3% in crude 3.56E-01 kg (Krauss, 1999) 1 1.13 (2,2,2,1,1,4,12)
35.1 kg tailings per kg Cu concentrate
ore, in ground
80% yield. Only 36% of Mo mined together wit Cu Molybdenum, 0.022% in

mining copper ore, GLO; mining
Molybdene in ground  is used --> 64% of resource use is lost and resource in ground sulfide, Mo 8.2E-3% and Cu 7.84E-03 kg own assumptions 1 1.13 (2,2,2,1,1,4,12)
allocated to Cu production 0.36% in crude ore, in ground
Land transformation
 resource land Transformation, from unknown 6.25E-05 m2 (Martens 2002) 1 2.01 (2,2,2,1,1,2,8)
from nature
Land transrormation to Transformation, to mineral
 resource land 6.25E-05 m2 (Martens 2002) 1 2.01 (2,2,2,1,1,2,8)
mine extraction site
Occupation, mineral extraction
Land use for mining  resource land 1.88E-03 m2a (Martens 2002) 1 1.51 (2,2,2,1,1,2,7)
site
construction
explosive  civil engineering No RER blasting 3.59E-02 kg (Krauss, 1999) 1 1.13 (2,2,2,1,1,4,4)
processes
construction diesel, burned in building
diesel fuel  machinery No GLO 2.83E-03 MJ (Krauss, 1999) 1 1.13 (2,2,2,1,1,4,4)
processes machine
electricity from grid, electricity, medium voltage,
 Assumption: 40% electricity from grid electricity production mix No UCTE 4.22E-02 kWh (Krauss, 1999) 1 1.14 (2,3,2,1,1,4,2)
mining production UCTE, at grid
electricity, hydropower, at run-
hydropower, mining  Assumption: 60% hydropower hydro power power plants No RER 6.33E-02 kWh (Krauss, 1999) 1 1.14 (2,3,2,1,1,4,2)
of-river power plant
exploitation
construction
infrastructrue,  exploration infrastructure, 2000 m belts assumed machinery Yes RER conveyor belt, at plant 2.72E-06 m own assumptions 1 3.36 (5,5,1,1,3,5,9)
processes
conveyor belts
infrastructure of mining 30% underground; life time output of mine 2E10 non-ferrous metal, mine,
 metals extraction Yes GLO 1.57E-10 unit own assumptions 1 3.36 (5,5,1,1,3,5,9)
(underground) kg) underground
infrastructure of mining non-ferrous metal, mine,
 70% open cut; life time output of mine 2E10 kg) metals extraction Yes GLO 3.59E-10 unit own assumptions 1 3.36 (5,5,1,1,3,5,9)
(surface) surface
waste residual material disposal, sulfidic tailings, off-
 overburden, disposed No GLO 6.21E+01 kg (Krauss, 1999) 1 1.13 (2,2,2,1,1,4,6)
management landfill site
 overburden, refilled no overburden refilled according to (Krauss 1999) 0.00E+00 kg (Krauss, 1999) 1 1.13 (2,2,2,1,1,4,6)
included in beneficiation effluents, not evident if
 effluents, mining 0.00E+00 m3 (Krauss, 1999) 1 1.13 (2,2,2,1,1,4,6)
accounted in (Krauss 1999) or not.
Fig. 5.3 Flows for “Mining of copper-ore, GLO – part mining” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg copper ore.

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Remarks
Location
name in
General
Process
Output
Name
Modul
Input
Infra-
Type
Sub-
Unit
electricity, from grid, electricity, medium voltage,
beneficiation production UCTE, at grid
hydropower, electricity, hydropower, at run-
 Assumption: 60% hydropower hydro power power plants No RER 3.76E-01 kWh (Krauss, 1999) 1 1.14 (2,3,2,1,1,4,2)
beneficiation of-river power plant
fresh and processwater
 resource in water Water, river 3.59E-02 m3 (Krauss, 1999) 1 1.13 (2,2,1,1,1,4,12)
(beneficiation)
construction limestone, milled, packed, at
lime  others No CH 4.28E-02 kg (Krauss, 1999) 1 1.04 reported values
materials plant
steel  Abrasion masticator metals extraction No RER chromium steel 18/8, at plant 2.67E-02 kg (Krauss, 1999) 1 1.20 reported values
floating agents, organic  chemicals organics No GLO chemicals organic, at plant 8.11E-03 kg (IPPC, 2002) 1 2.00 reported values
mining copper ore, GLO; beneficiation
floating agents,
 chemicals inorganics No GLO chemicals inorganic, at plant 2.76E-02 kg (IPPC, 2002) 1 2.00 reported values
inorganic
Sodium cyanide  chemicals inorganics No RER sodium cyanide, at plant 1.26E-03 kg (IPPC, 2002) 1 2.00 reported values
Sulphuric Acid  burden-free, since byproduct from metallurgy. 1.34E-01 kg (IPPC, 2002) 1 2.00 reported values
AlOH-production plant taken as proxy for

infrastructure
 benefication. A factor of 1.4 is applied to relate metals extraction Yes RER aluminium hydroxide, plant 3.02E-10 unit own assumptions 1 3.07 (3,2,2,1,3,4,9)
beneficiation
to bauxite input rather then AlOH output.
not clear, wether net disposal or partly re-filled in waste residual material disposal, sulfidic tailings, off-
 tailings, disposed No GLO 3.51E+01 kg (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,6)
mine. Here assumed as net disposal management landfill site
 tailings, refilled no tailings refilled according to (Krauss 1999) 0.00E+00 kg (Krauss, 1999) 1 1.13 (2,2,2,1,1,4,6)
composition in Tab. 5.3, high variability assumed.
 effluents, beneficiation 3.59E-02 m3 (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,6)
No treatment of waste water.
(IPPC, 2002),
 Emission Cyanide From sodium cyanide, 80% control water river Cyanide 1.39E-04 kg 1 1.84 (4,2,2,3,4,4,33)
(Anonymous 1998)
0.53 kg carbondisulfide out of 1 kg xanthate low population
 emission carbondisulfide air Carbon disulfide 3.59E-03 kg (NPI, 2001) 1 2.31 (4,2,2,5,4,4,23)
through hydrolytical decomposition density
low population
 emission carbon dioxide 0.44 kg CO2 per kg lime air Carbon dioxide, fossil 1.35E-02 kg (NPI, 2001) 1 2.30 (5,3,3,6,5,5,24)
density
including emissions from mining activities. Dust is
the sum of particulate emissions represented in
 Dust 1.65E-02 kg (NPI, 2001) 1 1.59 (3,2,1,1,3,4,25)
fractions PM<2.5, PM2.5-10, PM>10, including
metals and salts (Tab. 5.6)
including emissions from mining activities. Sum up low population (NPI, 2001), (naweb,
 PM<2.5 air Particulates, > 10 um 8.12E-04 kg 1 2.69 (3,2,1,1,5,4,19)
together with other PM fraction to total of dust density 2003), cemip
including emissions from mining activities. Sum up low population Particulates, > 2.5 um, and < (NPI, 2001), (naweb,
 PM2.5-10 air 7.31E-03 kg 1 3.69 (3,2,1,1,5,4,20)
together with other PM fraction to total of dust density 10um 2003), cemip
including emissions from mining activities. Sum up low population (NPI, 2001), (naweb,
 PM>10 air Particulates, < 2.5 um 8.40E-03 kg 1 2.69 (3,2,1,1,5,4,23)
together with other PM fraction to total of dust density 2003), cemip
low population
 waste heat air Heat, waste 2.64E+00 MJ 1 1.14 (2,3,2,1,1,4,13)
density
copper, concentrate, at
 Copper Concentrate 29.7% Cu metals extraction No GLO 1.00E+00 kg (Krauss, 1999)
beneficiation
molybdenum, molybdenite
55% Mo in concentrate (92.5% MoS2 in
 Molybdenite concentrate metals extraction No GLO concentrate, couple 4.11E-03 kg own assumptions
molybdenite)
production Cu
Fig. 5.4 Flows for “Mining of copper-ore, GLO – part mining” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg copper ore.

5.2 Winning primary copper

The standard process for the pyrometallurgical production of primary copper in use since the late 19th
century involves three major stages:
1. ore mining and beneficiation,
2. smelting and desulphurisation and
3. final refining.
The newer hydrometallurgical process, known as solvent extraction-electro-winning (SX-EW), is in a
rapid growth stage and it is to some extent displacing the older pyrometallurgical approach. The py-
rometallurgical approach is a net producer of by-product sulphuric acid, whereas the hydro-
metallurgical process is a net consumer of the acid. Thus, in the long run it is likely that the two will
be increasingly combined.
In this study primary copper production is modelled as an aggregation of the pyrometallurgical and
SX-EW production routes using the region-specific shares along with (Krauss et al. (1999)).
Tab. 5.7 Overview of the shares of different production processes in different regions, in million tonnes, and the
share of SX-EW (Krauss et al. (1999)). Note that the two regions North America and South America are not
distinguished and therefore concerning sulphur dioxide emissions are treated as one in this study.
Production in MMT North South Euro CIS 7 / Indonesia / World Com-

(106 metric tonnes) America America pe Asia Australia ment
Region in ecoinvent RNA RLA RER RAS ID GLO
Total copper prod. 3840 1185 2915 295 8760
Reverberatory furnace 895 70 595 90 2010 A
Electric furnace 30 80 170 425 B
Outokumpu 995 775 1320 40 3160 B
Inco 350 90 440 B
Kivcet 40 40 B
Mitsubishi 100 200 300 B
Noranda 180 40 220 B
Teniente 415 415 B
Vanyukov 155 155 B
Other 200 200 40 90 530 A
SX-EW 675 (17.6%) 25 (8.8%) 800 (9.4%) C
total considered 3840 1125 2610 285 8495
Abatement SO2 50% 50% 95% 20% 20% 45%
A furnace type with high SO2-emission (2.5 t / t) (Ayres et al. (2002))
B furnace type with moderate SO2-emission (0.15 t / t) (Ayres et al. (2002))
C furnace type with little SO2-emission (0.015 t / t) (Ayres et al. (2002))

There are many different process possibilities to win the metal. The chosen process depends on the
composition of the ore, the local costs of energy supply, and the local legislation. Basically two differ-
7
Commonwealth of Independent States

ent types can be distinguished: the hydrometallurgical and the pyrometallurgical process with a share
of 90.6 % and 9.4 % of the world-wide copper produced in 2004. The hydrometallurgical process is
described in detail later in Section 5.3 Hydrolytic winning of primary copper.
Pyrometallurgy; The pyrometallurgical treatment of copper concentrates includes three types of unit
operation: drying & roasting (pre-treatment), smelting & converting (reduction), and finally refining.
In the roasting step sulphur is driven off as sulphur dioxide and part of the iron is oxidised. In smelt-
ing, the roaster product is melted with a siliceous flux which combines with the oxidised iron to pro-
duce two immiscible phases, a liquid silicate slag which can be discarded, and a solution of molten
sulphides which contains the metal values. In the converting operation on the sulphide melt, more sul-
phur is driven off as sulphur dioxide, and the remaining iron is oxidised and fluxed for removal as sili-
cate slag, leaving a high-grade copper sulphide matte. In several modern operations the roasting step
has been eliminated, and the copper sulphide concentrate is treated directly in the smelter. The con-
verted copper is referred to as blister copper with a purity of 98%.
Refining; Conventional refining comprises three stages: (1) pyrometallurgical or fire refining, (2) e-
lectrolytic refining, and (3) remelting of cathodes and casting of shapes. Fire refining is applied to
crude copper from converters and furnaces, cement copper from hydrometallurgy, anode scrap from
electrolytic refining, and high-grade copper scrap, chiefly unalloyed wire scrap. About 80 % of the
world copper production is refined by electrolysis. This treatment yields copper with high electrical
conductivity and provides for separation of valuable impurities, especially precious metals. Copper
cathodes must be remelted and cast to shapes because the structure of copper formed by electro-
crystallization is not suitable for working to semi finished products. Cathodes are remelted in several
types of special furnaces that perform the tasks of melting, post refining (if necessary), holding, and
casting.
5.2.2 Emission
Metallurgy
Sulphur dioxide; In every process steps sulphur dioxide is emitted to air. Recovery of sulphur dioxide
is only economic for high concentrated off-gas. SO2 emissions are by far the most important in the
metallurgy.
Carbon dioxide; In the beneficiation step, considerable amounts of lime is added to the ore for pH-
stabilisation. Lime forms later flux in the metallurgical step. Lime decomposes into CO2 to form cal-
cite.
Dust; Dust carries over from the roasting, smelting and converting processes. Particulate emissions to
air consist of metals and thus are often returned to the leaching process after treatment.
Emissions to water; Metals and their compounds and substances in suspension are the main pollut-
ants emitted to water. The metals concerned are Cu, Ni, Co, As, and Cr. Other significant substances
are chlorides and sulphates. Wastewater from wet gas cleaning (if used) of the different metallurgical
stages is the most important source. The leaching stages are usually operated on a closed circuit and
drainage systems, and are therefore regarded as minor sources.
Refining
Also in this step the combustion of sulphur leads to emissions of SO2. Nitrogen oxides are produced in
significant amounts during acid digestion using nitric acid. Chlorine and HCl can be formed during a
number of digestion, electrolytic, and purification processes. Dust and metals are generally emitted
from incinerators and furnaces. Organic compounds, namely dioxins, can be emitted from smelting
stages resulting from the poor combustion of oil and plastic in the feed material. All these emissions
are subject to abatement technologies and controlling. Large quantities of effluents contain substantial
amounts of metals and organic substances.

5.2.3 Waste
Metallurgy & Separation from Non Ferrous Metals
The production of metals is related to the generation of several co-products, residues and wastes,
which are listed in the European Waste Catalogue. Some of the process specific residues can be reused
or recovered in preliminary process steps (e. g. dross, filter dust) or construction (e. g. cleaned slag).
Residues also arise from the treatment of liquid effluents, the main residue being gypsum waste and
metal hydroxides from the wastewater neutralisation plant. These residuals have to be disposed, usu-
ally in lined ponds.
Refining
Like in the pyrometallurgical step, considerable quantities of solid residuals are generated, which are
mostly recycled within the process or sent to other specialists to recover any precious metals. Final
residues generally comprise hydroxide filter cakes (iron hydroxide, 60% water, cat I industrial waste).
5.2.4 Copper, blister copper, at primary smelter in ecoinvent

This module is uniquely used in the secondary copper production where, besides copper scrap, 10% of
the feed consists of blister copper, i.e. raw copper after converting with an average content of 98%.
The inventory of this module is identical to that of the primary production of copper in RER, with the
only difference that the refining step is omitted and with 98% of the in- and outputs, since the metal
content of blister copper is only 98%. The inventory has an output of 1 kg blister copper.
5.2.5 Winning of primary copper in ecoinvent

This part is mainly based on Krauss et al. (1999). The general process description was taken from
Fabian (1997), whereas the process specific emissions are from IPPC (2001) and EEA (1999). In
Krauss et al. (1999) data for the different process alternatives in copper production was collected in
detail. The distinction there was made between procedures depending on different furnaces: Inco,
Teniente, Outokumpu, Mitsubishi, Noranda, Vanyukov, reverberatory furnace, electric furnace,
Kivcet. According to the share of the different procedures in the respective countries an aggregated
production mix for a country was calculated (Tab. 5.7). Where applicable, the respective share in SX-
EW – copper is included into the production mix.
Please note that in the database distinct datasets for the regions Northern America (RNA), South A-
merica (RLA), Europe (RER), Asia and Russia (RAS), Indonesia / Australia (ID), and the world-wide
average (GLO) exist. However, in this report only the global (GLO) module is discussed, the regional
modules do not differ in their structure but in their values according to Tab. 5.8.
The module describes the production mix for refined copper in 1994. It is designed for the use of the
metal as raw material for the semis manufacturing or as alloying element. The modules with general
flow information, remarks, sources, values and uncertainty information are shown in Fig. 5.5, Fig. 5.6
and Fig. 5.7. The meta information for this process in ecoinvent are summarised in Tab. 5.1 on page
17 ff. The following paragraphs describe calculations, sources and assumptions chosen in this study.
The functional unit of this process is one kilogram of cathode copper.
Metallurgy
In this study all nine process alternatives from Tab. 5.7 have been considered according to their pro-
duction share given in Krauss et al. (1999). These alternatives cover 84 % of the world production of
primary copper. The inventory already is partly aggregated and reflects the mix of the different proc-
ess alternatives within a certain process step, e.g. electricity consumption in the roasting step. For de-

tails on single process alternatives, e. g. the Outokumpu – process, please consult Krauss et al. (1999)
or IPPC (2001).
Pre-treatment; Flux demand (quartz sand) is documented in Krauss et al. (1999). Additional 250 kg
of limestone per tonne of copper has been assumed in this study according to Ayres et al. (2002). Dry-
ing is done by gas. Of the electricity 40% is assumed to originate from the grid while run-off plants
deliver 60%. Sulphuric acid produced from off-gas is not inventoried (cut-off). All process specific
emissions are inventoried in the smelting step.
Smelting and Reduction; 40% electricity is from grid, 60% hydropower assumed. A production of
steam is reported which is not inventoried as by-product. Dross and slag are reported to be recycled,
for this study however they are assumed to be disposed of in a landfill. In this step the major part of
the sulphur dioxide is emitted, which represent the main impact besides the energy use. The values for
emissions vary in a wide range from 4 kg to 2000 kg SO2 per tonne metal produced (Anonymous
(1998c)), but single values of much higher magnitude are also reported, e. g. nickel-copper smelting in
Norils’k (RUS) with 5300 kg SO2 / t metal produced (Hochfeld (1997)). The target for “best available
technique” in Europe is set to 6 – 16 kg SO2 / t Cu (IPPC (2001)), the World Bank mentions 25 kg
SO2 / t Cu for “modern plants using good industrial practices”. SO2 emissions depend basically on the
possibility of economical recovering. Recovering of sulphur dioxide as sulphuric acid is economic if
the SO2 content in the off-gas is sufficient and therefore is dependent from the production process and
the control-factor (abatement). In this study following emission factors are chosen relating to Tab. 5.7:
reverbatroy furnace and undefined processes (“other”): 2500 kg / t Cu; all other alternatives: 150 kg / t
Cu. The control-factors chosen are shown in Tab. 5.7, they are based on Ayres et al. (2002). For this
calculation the regions North and South America are treated equally due to missing specific data. The
resulting SO2 emissions are shown in Tab. 5.8; they range from 36 kg / t Cu (RER) to 1296 kg / t Cu
(ID) with a world average of 458 kg / t Cu. The calcination of previously added lime (pre-treatment
and beneficiation, see Section 5.1.3 Exploitation of copper-ore in ecoinvent) releases 0.44 kg CO2 per
tonne lime. Several emissions to air are documented for European conditions in EEA (1999); the re-
sulting emissions are chosen for this study (Tab. 5.9). Elaborate abatement technology – values of col-
umn “impact control” in Tab. 5.8 – are applied only for RNA and RER, whereas for the other regions
values corresponding to limited control are chosen. VOC emissions are mentioned in EEA (1999) with
a value of about 0.015 kg / t (unabated, including roasting and smelting). Dust emissions according to
Anonymous (1998c) range between 0.1 kg / t Cu and 20 kg / t Cu, whereas IPPC (2001) mentions for
modern European plants a range of 0.16 kg / t Cu to 1 kg / t Cu. In this study 0.4 kg / t Cu is chosen,
which is assumed to apply for all covered processes. The distribution of particulate sizes is taken from
CEPMEIP 8 (Tab. 5.10). For water use the value for primary nickel production of 1.4 m3 per tonne
concentrate worked is chosen (Hochfeld (1997)). The volume of the effluent is assessed with the value
for water use and respective emissions into water are calculated according to the content in Tab. 5.11.
Refining; No further wastes are considered from this step. Region’s specific production mix is com-
piled with the specific pyrolytical technologies and the specific share in SX-EW – copper.
8
http://www.air.sk/tno/cepmeip/em_factors.php, accessed on 01.03.2003

Tab. 5.8 Pyrometallurgy of copper. Background data per tonne of copper produced for the different regions covered
in this study after (Krauss et al. (1999)).
Step / Description Unit RNA RLA RER RAS ID GLO

Input output
Pre-treatment
Copper concentrate1) t 3.60 3.18 4.14 4.06 3.16 3.48
Compressed Air Nm3 1230 1370 1300 1600 1320 1450
electricity kWh 45 91 38 70 94 65
electricity, grid kWh 18 36.4 15.2 28 37.6 26
hydropower kWh 27 54.6 22.8 42 56.4 39
fossil energy (gas) MJ 1234 218 1379 1162 545 980
off-gas Nm3 1400 1580 1450 1900 1860 1650
sulphuric acid kg 140 380 130 220 380 220
Reduction
electricity kWh 145 52 200 205 86 155
electricity, grid kWh 58 20.8 80 82 34.4 62
hydropower kWh 87 31.2 120 123 51.6 93
fossil energy (oil MJ 4779 9641 4532 6592 9146 6592
heavy)
fossil energy (gas) MJ 72.6 72.6 36.3
fossil energy (coke) kg 1 2 1
Compressed air Nm3 3600 3640 4400 4400 3500 3900
Electrode graphite kg 1 0.1 1 1 0.3 1
Oxygen Nm3 363 152 300 310 208 290
silica sand kg 790 750 750 930 803 831
sulphuric acid t 3.6 1.95 3.6 2.7 1.94 2.52
slag t 1.78 1.755 1.85 1.9 1.65 180
off-gas Nm3 6500 8300 7000 7800 8300 7600
sulphur dioxide kg 397 397 36 577 1297 458
Refining
electricity kWh 309 324 309 340 324 324
electricity, grid kWh 124 130 124 136 130 130
hydropower kWh 185 194 185 204 194 194
fossil energy (gas) MJ 11217 2468 2360 2541 2468 2468
off gas MJ 300 315 300 330 315 315
dross (dry) t 10 10 10 10 10 10
Compressed air t 300 315 300 330 315 315
1) Krauss et al. (1999) give different values for copper content in concentrates and for concentrate consumption (per t of
copper) in the chapters for mining and for metallurgy. For consistency within the data, the values given in the mining
chapter are used in ecoinvent.

Tab. 5.9 Metal emission factors to air for primary copper production (EEA (1999)). Chosen value is the average of the
given range. Missing ranges – tagged with * – have been extrapolated from limited / impact control with the
average factor of 20. Values for “impact control” apply only to RNA and RER.
Compound impact control (g / t Cu) limited control (g / t Cu)

Arsenic 32.5 (15-50) 1000 (500-1500)
Chromium 0.05 1
Cadmium 6.5 (3-10) 350 (200-500)
Copper 250 (200-300) 2750 (1500-4000)
Indium 0.125 (0.05-0.2)* 2.5 (1-4)
Manganese 15 (5-25)* 300 (100-500)
Nickel 55 (10-100)* 1950 (900-3000)
Lead 150 (50-250)* 2500 (1000-4000)
Antimony 5.5 (1-10) 125 (50-200)
Selenium 5.5 (1-10) 100 (50-150)
Tin 6.25 (2.5-10) 125 (50-200)
Vanadium 0.375 (0.25-0.5) 7.5 (5-10)
Zinc 150 (100-200) 750 (500-1000)
Bismuth 7.5 (5-10)* 150 (100-200)
Mercury 0.1 (0.1-0.1) 2 (2-2)*
Tab. 5.10 Emissions to air per tonne copper. Data was taken from similar processes in nickel (A) and copper (B) met-
allurgy (IPPC (2001)). The distribution of the particle size is taken from CEPMEIP (C), values for VOC from
EEA (1999) (D).
Substance Amount in kg / t Source

(this study)
Dust 0.41 A, B
PM<2.5 0.0005 C
PM2.5-10 0.3 C
PM>10 0.10 C
SO2 See Tab. 5.8 estimation
CO 0.03 A
VOC 0.015 D
TCDD 1.00E-08 B

Tab. 5.11 Composition of different wastewater effluents after treatment. A simple average is chosen. Data from Euro-
pean copper operations (IPPC (2001)).
Sub- Process Surface run Direct cool- This study

stance Water off ing water (average)
3 3 3
[g / m ] [g / m ] [g / m ] [g / m3]
Cu 0.045 0.063 0.05 0.0527
(0.01-0.2) (0.01-0.4) (0.01-0.25)
Pb 0.006 0.032 0.01 0.0160
(0.001-0.04) (0.005-0.2) (0.001-0.1)
As 0.032 0.014 0.01 0.0187
(0.01-0.1) (0.003-0.07) (0.001-0.1)
Ni 0.024 0.028 0.011 0.0212
(0.004-0.15) (0.002-0.4) (0.002-0.06)
Cd 0.003 0.004 0.001 0.00273
(0.0001-0.1) (0.0002-0.1) (0.0001-0.003)
Zn 0.045 0.11 0.1 0.0848
(0.01-0.2) (0.03-0.4) (0.02-0.5)
Cr 0.029
Hg 0.00029
Sn 0.029
Data quality
Process data is satisfactory and valid for over 84 % of the primary copper production. The main emis-
sions dominating possibly the whole process is sulphur dioxide, which itself is not reported in Krauss
et al. (1999). Reported values range in two orders of magnitude. Poor data is available on composition
and fate of solid waste, for which the reuse rate was not documented. High uncertainties in volume and
composition of the effluents are assumed.

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Remarks
Location
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
not inventoried in Ecoinvent, energy consumption
Compressed Air  1.31E+00 Nm3 (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,4)
copper, primary, at
comprised in chosen value

refinery, GLO; pre-
90.6% of copper input (rest by copper from SX- copper, concentrate, at

Copper concentrate  metals extraction No GLO 3.15E+00 kg (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,4)
EW); 96.8% yield, concentrate at 29.7% Cu beneficiation
electricity, high voltage,
electricity, grid  40% electricity from grid electricity production mix No UCTE 2.36E-02 kWh (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,2)
treatment
production UCTE, at grid

hydropower  60% electricity from hydropower hydro power power plants No RER 3.53E-02 kWh (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,2)
natural gas, burned in
fossile energy (gas)  natural gas heating systems No RER 8.88E-01 MJ (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,1)
industrial furnace >100kW
limestone  others No CH 2.26E-01 kg (Ayres, 2002) 1 1.51 (5,2,2,3,1,3,4)
materials plant
Infrastructure
 metals extraction Yes GLO non-ferrous metal, smelter 9.57E-12 unit own assumptions 1 3.07 (3,2,2,1,3,4,9)
metallurgy
 off-gas not inventoried, no data 1.49E+00 Nm3 (Krauss, 1999) 1
not inventoried, assumed to be burden free (cut
 sulfuric acid 1.99E-01 kg (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,4)
off)
Fig. 5.5 Flows for “copper, primary, at refinery, GLO – part pre-treatment” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg copper.

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Remarks
Location
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
construction
silica sand  additives No DE silica sand, at plant 7.53E-01 kg (Krauss, 1999) 1 1.51 (5,2,2,3,1,3,4)
materials
copper, primary, at refinery, GLO; reduction
heavy fuel oil, burned in

fossile energy (oil
 oil heating systems No RER industrial furnace 1MW, non- 5.97E+00 MJ (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,1)
heavy)
modulating
fossile energy (gas)  natural gas heating systems No RER 3.29E-02 MJ (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,1)
hard coal, burned in industrial
fossile energy (coke)  hard coal heating systems No RER 2.59E-02 MJ (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,2)
furnace 1-10MW
Compressed air  3.53E+00 Nm3 (Krauss, 1999) 1 1.48 (4,4,4,5,3,5,3)
Electrode graphite  metals extraction No RER anode, aluminium electrolysis 9.06E-04 kg (Krauss, 1999) 1 2.07 (5,5,5,6,4,6,4)
Oxygen  chemicals inorganics No RER oxygen, liquid, at plant 2.63E-01 kg (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,4)
water use  for overall process resource in water Water, river 4.41E-03 m3 (Krauss, 1999) 1 1.13 (2,2,1,1,1,4,12)
Inventoried as wastewater treatment w/o treatment, sewage,
waste wastewater
 Effluent pollution. For overall process, waterborn emission No CH unpolluted, to wastewater 4.41E-03 m3 (Hochfeld, 1999) 1 1.13 (2,2,1,1,1,4,13)
management treatment
see Tab. 5.11 treatment, class 3
disposal, nickel smelter slag,
50% recycling assumed --> disposal 50%. Proxy for waste residual material
 slag No CH 0% water, to residual material 8.15E-01 kg (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,6)
Ni smelter slag used management landfill
landfill
 off-gas not inventoried, no data 6.88E+00 Nm3 (Krauss, 1999)
low population (Krauss, 1999),
 sulphur dioxide in off-gas comprise emissions from drying and roasting air Sulfur dioxide 4.15E-01 kg 1 1.64 (5,4,1,3,3,5,15)
density (Anonymous 2000)
low population
 carbon dioxide (lime) 0.44 kg / kg lime air Carbon dioxide, fossil 9.96E-02 kg (Krauss, 1999) 1 1.89 (5,4,1,3,3,5,7)
density
 Total dust composition in Tab 5.9 and Tab 5.10 9.53E-03 kg (IPPC, 2002)
low population
 PM<2.5 see Tab 5.9 and Tab 5.10 air Particulates, < 2.5 um 4.60E-07 kg (IPPC, 2002) 1 3.07 (3,2,1,1,3,4,27)
density
low population Particulates, > 2.5 um, and <
 PM2.5-10 see Tab 5.9 and Tab 5.10 air 2.76E-04 kg (IPPC, 2002) 1 1.34 (3,2,1,1,3,4,28)
density 10um
low population
 PM>10 see Tab 5.9 and Tab 5.10 air Particulates, > 10 um 9.21E-05 kg (IPPC, 2002) 1 1.50 (3,2,1,1,3,4,29)
density
low population
 Carbon monoxide see Tab 5.9 and Tab 5.10 air Carbon monoxide, fossil 2.72E-05 kg (IPPC, 2002) 1 5.08 (3,2,1,1,3,4,22)
density
NMVOC, non-methane volatile
low population
 VOC see Tab 5.9 and Tab 5.10 air organic compounds, 1.36E-05 kg (IPPC, 2002) 1 1.59 (3,2,1,1,3,4,16)
density
unspecified origin
low population Dioxins, measured as 2,3,7,8-
 TCDD see Tab 5.9 and Tab 5.10 air 1.81E-12 kg (IPPC, 2002) 1 3.07 (3,2,1,1,3,4,21)
density tetrachlorodibenzo-p-dioxin
Fig. 5.6 Flows for “copper, primary, at refinery, GLO – part reduction” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg copper.

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Remarks
Location
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
copper, primary, at

refinery, GLO;
hydropower 60% electricity from hydropower hydro power power plants No RER 1.76E-01 kWh (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,2)
fossile energy (gas)  natural gas heating systems No RER 2.24E+00 MJ (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,1)
refining
 off gas not inventoried, no data 2.85E-01 MJ (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,4)
not inventoried, assumed to be burden free (cut
 sulfuric acid 2.28E+00 kg (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,4)
off)
 dross (dry) internal recovery, not inventoried 9.06E-03 kg (Krauss, 1999)
low population
 waste heat (over all) air Heat, waste 1.77E+00 MJ 1 1.14 (2,3,2,1,1,4,13)
density
copper from SX-EW  metals extraction No GLO copper, SX-EW, at refinery 9.42E-02 kg (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,1)
 Copper Cathode metals extraction No GLO copper, primary, at refinery 1.00E+00 kg (Krauss, 1999) 1
Fig. 5.7 Flows for “copper, primary, at refinery, GLO – part refining” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg copper.

5.3 Hydrolytic winning of primary copper

This part is mainly based on (Krauss et al. (1999)). The general process description was taken from
(Fabian (1997)). Process specific emissions derived from (IPPC (2002)) and (IPPC (2001))

Oxide ores and supergene sulphide ores (i.e. ores not containing iron) can be recovered most easily by
hydro-metallurgical techniques. The general steps of mining and refining are identical to those de-
scribed in Section 5.1 Exploitation of copper ore. The difference in this type of production is in the
beneficiation and the metallurgical step. Hydrometallurgical process is only applicable to oxidic ore,
sulphides have to be roasted prior to leaching.
Several hydrometallurgical processes are in commercial operation for the treatment of copper mattes
to copper cathodes. The dominant process in the copper production is the solvent extraction-electro-
winning (SX-EW), which essentially by-passes the beneficiation and smelting stages, substituting a
leaching stage followed by cementation or electro-winning.
Hydrometallurgical processes can generally be subdivided into the following steps (Fabian (1997)):
1. Pre-treatment, physical (e.g., comminution, grinding) or chemical (e.g., roasting)
2. Leaching, chiefly with sulphuric acid or ammoniacal solutions, sometimes under oxidising condi-
tions. In copper production, leaching is used in a number of ways:
- In situ leaching, applied in loose ore bodies. In situ leaching recovers copper from small oxide
ore bodies (too small to mine economically) or worked-out underground mines. Dump leach-
ing is used to recover copper from large quantities (millions of tonnes) of strip mine waste
with a very low grade. Dilute sulphuric acid is trickled through the material. Once the process
starts it continues naturally if water and air are circulated through the heap. The time required
is typically measured in years. Sulphur dioxide is emitted during such operations. Soluble
copper is then recovered from drainage tunnels and ponds. Copper recovery rates vary from
30% to 70%.
- Dump or heap leaching of lumpy or ground material. Heap leaching is essentially the same as
dump leaching, except that it is applied to oxide ores, which are previously ground or crushed
and heaped on a prepared surface.
- Vat or percolation leaching of small-sized material. In vat leaching, oxide ores are crushed and
placed in large concrete vats (up to 18,000 tonnes capacity), then flooded by concentrated sul-
phuric acid, which converts most of the copper to copper sulphate. Extraction of one tonne of
copper from ore with a grade of 1% requires 4400 litres of 96% sulphuric acid. Vat leaching
typically recovers 75% of the copper. The process is fairly rapid (hours to days).
- Agitation or slime leaching of fine-grade material without pressure
- Pressure leaching of fine-grade material with or without oxygen

In practice, only a few proven chemicals, lixiviants, are used: Dilute sulphuric acid for basic
carbonates and sulphates, and silicates; Solutions of iron (III) salts (sulphate) as oxidising
agents in sulphuric acid for sulphidic ore; Ammoniacal solutions of ammonium salts (carbon-
ate) for oxidised ores. Recently, the bacteria-assisted leaching of low-grade copper ores and
industrial wastes has been developed. Certain autotrophic bacteria accelerate the oxidation of
copper and iron sulphides. Typical species are Thiobacillus thiooxidans, the reaction takes
place slowly in atmospheric oxygen (Acid Rock Drainage), but the bacteria catalyse it.

3. Solution cleaning, e.g. by precipitation of impurities and filtration or selective enrichment of cop-
per by solvent extraction or ion exchange. The solvent extraction process comprises two steps: se-
lective extraction of copper from an aqueous leach solution into an organic phase (extraction cir-
cuit) and the re-extraction or stripping of the copper into dilute sulphuric acid to give a solution
suitable for electro winning (stripping circuit).
4. Separation, precipitation of copper metal or copper compounds such as Cu2O, CuS, CuCl, CuI,
CuCN, or CuSO4 · 5 H2O (crystallisation)
5. Post-treatment, e.g., fire-refining of crude cement copper or thermal reduction of copper com-
pounds
However, because solvent extraction-electro-winning (SX-EW) requires ten times as much electrical
energy as classical electrolytic refining (2000 – 2600 kWh t–1), and because established solvent extrac-
tion techniques do not work well on copper sulphide ores, electro-winning techniques have remained
complementary to smelting. Today about 10 – 15% of the world copper production is from SX-EW of
copper oxide ores (Graedel et al. (2002) and Krauss et al. (1999)).
5.3.2 Emissions and wastes

Several emissions to air and water occur in the mining step. These emissions are identical to those in
the conventional mining process and are shortly discussed there.
Pre-treatment; In the eventual roasting step to remove sulphides, considerable amounts of sulphur
dioxide may be emitted into the air, since the concentration of sulphur dioxide in the off-gas probably
is too small for recovery as sulphidic acid. Contrary to the beneficiation step of the pyrometallurgical
process, no tailings develop additionally to the mining.
Leaching; For in-situ and heap leaching considerable amounts of sulphuric acid and leaching agents
emit into water and air. No figures are currently available on the dimension of such emissions. After
the solvent-solvent extraction, considerable amounts of leaching residues remain, which consist of un-
dissolved minerals and the remainders of leaching chemicals.
Electro-winning; Most of the electrolyte and the sulphuric acid is recovered into the leaching step.
Presumably no considerable wastes and emissions occur, but there are no available data.
5.3.3 Hydrolytic winning of primary copper in ecoinvent.

This part is mainly based on (Krauss et al. (1999)). The general process description was taken from
(Fabian (1997)), but no process specific emissions were documented there.
The module describes the production of copper with the SX-EW process. It is designed for the use of
the metal as part of the production mix “copper, primary, at plant”.
The modules with general flow information, remarks, sources, values and uncertainty information are
shown in Fig. 5.8 and Fig. 5.9. Ecoinvent meta information for this process is summarised in Tab. 5.1
on page 17 ff. The following paragraphs describe calculations, sources and assumptions made in this
study. The functional unit of this process is one kilogram of cathode copper.
Mining; Mining is similar to what is described in Section 5.1 Exploitation of copper-ore, but no ben-
eficiation occurs. No exact information on the mining step is available in Krauss et al. (1999), there-
fore a mining process described in the previous section has been chosen as proxy. This is done with
specific values for diesel use, explosives, and energy according to open pit mining. For further details
please consult the Section mentioned above. Data on ore input, overburden, and gangue are docu-
mented in Krauss et al. (1999). Since no further information is available on presumably used crushing

and grinding operations of the material, they have been estimated with additional energy and mechani-
cal wear (milling equipment) from the conventional copper-beneficiation in RLA. No use of condi-
tioning agents is to be expected.
Pre-treatment: Leaching and extraction; Following Krauss et al. (1999), heap leaching is consid-
ered as leaching process. Of the total electricity use of 600 kW /t Cu, 250 kWh/t Cu and 350 kWh/t Cu
are required for leaching (sprinkling) and extracting (pumping), respectively. Krauss et al. (1999) state
a loss of 171 m3 / t Cu leachate through infiltration and evaporation, and 1.7 t / t Cu sulphuric acid and
276 t / t Cu leaching residues but their fate is not evident. Since no information on the magnitude of
leachate loss into the ground exists, no emissions to water are considered. Thus additional emissions
of heavy metals, and sulphate into the ground are to be expected. Leaching agents were inventoried
with the inventory of ammonia production as proxy.
Electro-winning and Refining; Use of sulphuric acid was in no case inventoried, since it is treated in
this study as by-product of pyrolytical winning of metals from sulphidic ores. No further emissions or
wastes are considered important, since the electrolyte is fully recycled to the leaching stage. Losses in
electrolyte and leachate are inventoried in the pre-treatment step.
Data quality
There is not sufficient information for declaring this data set of good data quality. In particular, uncer-
tainty for energy consumption in a likely additional crushing / milling step (not modelled herewith)
and probable leaching to ground may lead to an underestimation of the environmental impacts. Since
the energy consumption considered for electro-winning is much higher than in the pyrolytical proc-
esses, this will probably dominate the environmental impacts; thus, the disregard of beneficiation and
leaching impacts may be compensated.

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Remarks
Location
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
70.0% yield, 554.0 kg overburden, 277.0 kg ore,
copper, SX-EW, at refinery, GLO; mining
Copper in ground  resource in ground 0.36% and Mo 8.2E-3% in crude 1.43E+00 kg (Krauss, 1999) 1 1.13 (2,2,2,1,1,4,12)
0.0 kg tailings per kg Cu concentrate
ore, in ground
Land transformation
 resource land Transformation, from unknown 4.82E-04 m2 (Martens 2002) 1 2.06 (3,2,2,1,3,2,8)
from nature
Land transrormation to Transformation, to mineral
 resource land 4.82E-04 m2 (Martens 2002) 1 2.06 (3,2,2,1,3,2,8)
Land use for mining  resource land 1.45E-02 m2a (Martens 2002) 1 1.58 (3,2,2,1,3,2,7)
site
construction
explosive  civil engineering No RER blasting 2.77E-01 kg (Krauss, 1999) 1 1.25 (2,2,2,1,3,4,4)
processes
diesel fuel (energy)  machinery No GLO 8.31E+00 MJ (Krauss, 1999) 1 1.25 (2,2,2,1,3,4,4)
and leaching
processes machine
electricity from grid, electricity, medium voltage,
hydropower, mining  Assumption: 60% hydropower hydro power power plants No RER 9.13E-01 kWh (Krauss, 1999) 1 1.25 (2,3,2,1,3,4,2)
electricity, from grid, Assumption: 40% electricity from grid, of total electricity, medium voltage,
 electricity production mix No UCTE 1.67E+00 kWh assumption 1 1.58 (4,2,2,3,4,4,2)
beneficiation 15.1 kWh / t raw ore production UCTE, at grid
hydropower, Assumption: 60% electricity from grid, of total electricity, hydropower, at run-
 hydro power power plants No RER 2.51E+00 kWh assumption 1 1.58 (4,2,2,3,4,4,2)
beneficiation 15.1 kWh / t raw ore of-river power plant
steel  Abrasion masticator metals extraction No RER chromium steel 18/8, at plant 1.91E-01 kg assumption 1 1.20 (4,2,2,3,4,4,4)
exploitation
construction
infrastructrue,  exploration infrastructure, 2000 m belts assumed machinery Yes RER conveyor belt, at plant 2.10E-05 m own assumptions 1 1.58 (4,2,2,3,4,4,4)
processes
conveyor belts
infrastructure of mining non-ferrous metal, mine,
 including crushing / grinding facilities metals extraction Yes GLO 1.39E-08 unit own assumptions 1 3.07 (3,2,2,1,3,4,9)
(surface) surface
waste residual material disposal, sulfidic tailings, off-
 overburden, disposed No GLO 5.54E+02 kg (Krauss, 1999) 1 1.13 (2,2,2,1,1,4,6)
management landfill site
 overburden, refilled no overburden refilled according to (Krauss 1999) 0.00E+00 kg (Krauss, 1999) 1 1.13 (2,2,2,1,1,4,6)
 effluents, mining no data, effluents probable, not inventoried 0.00E+00 m3 1 1.13 (2,2,2,1,1,4,6)
Fig. 5.8 Flows for “copper, SX-EW, at refinery – part mining” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg copper.

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Remarks
Location
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
treatment and reduction
refinery, GLO; pre-

copper, SX-EW, at
not inventoried, assumed to be by-product from

sulfuric acid (leaching)  1.70E+00 kg
pyrolytical winning
water use  resource in water Water, river 1.70E-01 m3 (Krauss, 1999) 1 1.13 (2,2,1,1,1,4,12)
ammonia, liquid, at regional
leaching agents  approximated with ammonia inventory chemicals inorganics No RER 1.70E-03 kg (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,4)
storehouse
infrastructure leaching  chemical plant chosen as proxy chemicals organics Yes RER chemical plant, organics 4.00E-10 unit own assumptions 1 3.07 (3,2,2,1,3,4,9)
 leaching residues not inventoried, no data 2.76E+02 kg (Krauss, 1999) 1

hydropower  60% electricity from hydropower hydro power power plants No RER 1.37E+00 kWh (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,2)
low population
 waste heat (over all) air Heat, waste 3.09E+01 MJ 1 1.14 (2,3,2,1,1,4,13)
density
 copper, SX-EW, at refinery metals extraction No GLO copper, SX-EW, at refinery 1.00E+00 kg (Krauss, 1999)
Fig. 5.9 Flows for “copper, SX-EW, at refinery – part metallurgy” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg copper.

5.4 Winning of secondary copper

Krauss et al. (1999) give a picture of imports of copper and copper concentrate to Germany. This mix
is used in this study as proxy for Europe.

Secondary copper consists of various types of scrap. Prompt scrap is directly reused in foundries and
is not further processed. Old scrap has to be treated in a secondary copper smelter, where a variety of
metal values are recuperated. Depending on the chemical composition, the raw materials of a secon-
dary copper smelter are processed in different types of furnaces, including:
- blast furnaces (up to 30% of Cu in the average charge),
- converters (about 75% Cu), and
- anode furnaces (about 95% Cu).

A scheme of the process considered is given in Fig. 5.10. The blast furnace metal (“black copper”) is
treated in a converter; then, the converter metal is refined in an anode furnace. In each step additional
raw material with corresponding copper content is added. In the blast furnace, a mixture of raw mate-
rials, iron scrap, limestone and sand as well as coke is charged at the top. Air that can be enriched with
oxygen is blown through the tuyeres. The coke is burnt and the charge materials are smelted under re-
ducing conditions. Black copper and slag are discharged from tapholes (EEA (1999)).
The converters used in primary copper smelting, working on mattes containing iron sulphide, generate
surplus heat and additions of scrap copper are often used to control the temperature. The converter
provides a convenient and cheap form of scrap treatment, but often with only moderately efficient gas
cleaning. Alternatively, hydrometallurgical treatment of scrap, using ammonia leaching, yields to solu-
tions which can be reduced by hydrogen to obtain copper powder (EEA (1999)).
Alternatively, these solutions can be treated by solvent extraction to produce feed to a copper-winning
cell.
Converter copper is charged together with copper raw materials in anode furnace operation. For smelt-
ing the charge, oil or coal dust is used, mainly in reverberatory furnaces. After smelting, air is blown
on the bath to oxidise the remaining impurities.
Leaded brasses, containing as much as 3% of lead, are widely used in various applications and recy-
cling of their scrap waste is an important activity. Such scrap contains usually much swarf and turn-
ings coated with lubricant and cutting oils. Copper-containing cables and motors contain plastic or
rubber insulants, varnishes, and lacquers. In such cases, scrap needs pre-treatment to remove these
non-metallic materials. The smaller sizes of scrap can be pre-treated thermally in a rotary kiln pro-
vided with an after-burner to consume smoke and oil vapours (so-called Intal process). There are also
various techniques available to remove rubber and plastic insulations of electric cables.
5.4.2 Emissions and waste

The stages used for secondary copper production are generally similar to those for primary production
but the raw material is usually oxidic or metallic and process conditions are therefore varied. Other
than in the primary copper production elevated levels of halogenated organic compounds may arise,
such as TCDD. Slags are usually used in construction.

5.4.3 Winning of secondary copper in ecoinvent

This part is mainly based on Rentz et al. (1999), who described one single large plant. These values
are added with Bruch et al. (1995).
The module describes the production of secondary copper in Germany as proxy for Europe. It is de-
signed for the use of the metal as part of the supply mix “copper, at regional storage”. It is not to be
used if the impact of the secondary copper is expected to be considerable in a LCA study. In such
cases, a more detailed analysis has to be conducted.
shown in Fig. 5.11 and Fig. 5.12. Ecoinvent meta information for this process is summarised in Tab.
5.1 on page 17 ff.
The values for material input and output was taken from Rentz et al. (1999), which describe the worlds
biggest plant for secondary production (Hüttenwerke Kayser AG, GER). No allocation to other by-
products like copper sulphate has been considered, but the whole ecological burden is attributed to the
main product secondary copper cathodes. Electricity demand for electrolysis was taken from Bruch et
al. (1995), whose fossil energy demand for smelting (9.21 GJ / t Cu) is similar to this used in this stu-
dy (overall 9.1 GJ / t Cu). Copper scrap was modelled with a proxy module for the collection and han-
dling of iron scrap. Since scrap does not count as resource according to ecoinvent cut-off rule – i. e. no
inventory entry –, the different scrap origin (iron instead of copper) does not play a role. About 10 %
of blister copper is used from primary copper production into secondary production.
Emissions were taken from Rentz et al. (1999) and added with values reported in EEA (1999) (Tab.
5.12 and Tab. 5.13). Effluent volume was mentioned in Bruch et al. (1995) with 1 m3 / t Cu. This vol-
ume is represented in the dataset as “treatment of clean waste water”, thus no emissions to water are
inventoried. The emissions into water – i. e. after waste water treatment – are accounted for in the this
dataset as direct emissions. The specific emission factor for TCDD (50 μg / t Cu) was reported in EC
(2002).
Data quality
The overall quality is sufficient for the use proposed. But poor data exist for collection and benefici-
ation of old scrap and no data on leaching agents and emissions from Hydrolytic processes were avail-
able. However, the bias should be negligible because of high level of throughputs and energy demand.

Fig. 5.10 Secondary copper production. Process scheme after Rentz et al. (1999). Scheme relates to the plant “Hüt-
tenwerke Kayser AG”, Germany.
Tab. 5.12 Secondary copper production. Emissions to air from two different sources (EEA (1999) and Rentz et al.
(1999)); the last column shows the value chosen in this study.
Emission to air (g / t Cu) (Rentz 1999) (EEA 1999) This study

Particulate matter 470 470
PM<2.5 282 282
PM2.5-10 94 94
PM>10 94 94
SO2 3000 3000
NOx 1000 1000
CO 2000 2000
Arsenic 1 2 2
Antimony 3 3
Cadmium 3 3
Copper 50 85 85
Lead 50 90 90
Nickel 1 1
Zinc 100 375 375

Tab. 5.13 Secondary copper production. Emissions to water according to Rentz et al. (1999) for one big plant (Hüt-
tenwerke Kayser AG).
Emission to water after treatment. [g / t ]

Copper 1
Lead 0.5
Arsenic 0.1
Nickel 0.3
Cadmium 0.01
Zinc 0.2

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Remarks
Location
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
Copper scrap  proxy for collecting scrap metals extraction No RER iron scrap, at plant 7.39E-01 kg (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,4)
copper, secondary, at refinery,
copper, blister-copper, at
Blister copper  metals extraction No RER 1.14E-01 kg (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,4)
primary smelter
Copper alloy scrap  proxy for collecting scrap metals extraction No RER iron scrap, at plant 1.99E-01 kg (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,4)
Ext. Intermeiate
 proxy for collecting scrap metals extraction No RER iron scrap, at plant 2.27E-01 kg (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,4)
materials
Copper-iron material  proxy for collecting scrap metals extraction No RER iron scrap, at plant 1.42E-01 kg (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,4)
RER; inputs

limestone  others No CH 7.39E-02 kg (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,4)
materials plant
construction
silica  additives No DE silica sand, at plant 6.25E-02 kg (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,4)
materials
coke  Hu=28.6 MJ/kg -> 4063 MJ / t Cu hard coal heating systems No RER 4.06E+00 MJ (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,1)
furnace 1-10MW
coal  Hu=28.9 MJ/kg -> 2463 MJ / t Cu hard coal heating systems No RER 2.46E+00 MJ (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,1)
furnace 1-10MW
Fuel (oil)  Hu=41.2 MJ/kg -> 2575 MJ / t Cu oil heating systems No RER industrial furnace 1MW, non- 2.58E+00 MJ (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,1)
modulating
11.34 GJ / t Cu primary energy in electricity, electricity, high voltage,
Electricity, grid  electricity production mix No UCTE 1.10E+00 kWh (Bruch 1995) 1 1.13 (2,4,2,3,1,nA,2)
effectiveness 0.35 = 110 kWh / t Cu production UCTE, at grid
infrastructure
 metals extraction Yes GLO non-ferrous metal, smelter 9.57E-12 unit own assumptions 1 3.07 (3,2,2,1,3,4,9)
pyrometallurgy
infrastructure leaching  chemical plant chosen as proxy chemicals organics Yes RER chemical plant, organics 4.00E-10 unit own assumptions 1 3.07 (3,2,2,1,3,4,9)
Fig. 5.11 Flows for “copper, secondary, at refinery” (inputs) and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg copper.

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Remarks
Location
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
 Slag internal use in building, not inventoried. 4.55E-01 kg (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,6)
waste residual material
 Anode slime No CH 0% water, to residual material 5.68E-03 kg (Rentz 1999) 1 1.13 (2,4,2,3,1,nA,6)
refinery, RER; outputs
copper, secondary, at
management landfill
landfill
Conists of different fractions (PM2.5, PM10, TSP)
 Particulate matter 4.70E-04 kg (Rentz 1999) 1 1.52 (2,4,2,3,1,nA,25)
and heavymetals. Composition see Tab. 5.12
 TCDD air 5.00E-11 kg (EC 2002) 1 10.00 (2,4,2,3,1,nA,21)
Inventoried as wastewater treatment w/o
treatment, sewage,
pollution. Pollution after treatment is inventoried waste wastewater
 Effluents No CH unpolluted, to wastewater 1.00E-03 m3 (Bruch 1995) 1 1.13 (2,4,2,3,1,nA,6)
extra according Tab 5.13. 1 m3 / t Cu effluents management treatment
treatment, class 3
assumed.
 Lead-tin alloys by-product, no burden 2.10E-02 kg (Rentz 1999)
 Zinc oxides by-product, no burden 5.11E-02 kg (Rentz 1999)
 Nickel sulphate by-product, no burden 1.36E-02 kg (Rentz 1999)
 Copper sulphate by-product, no burden 1.25E-02 kg (Rentz 1999)
 waste heat from electricity, 100% to air 3.97E+00 kg calculation 1 1.13 (2,4,2,3,1,nA,13)
 Copper cathodes metals extraction No RER copper, secondary, at refinery 1.00E+00 kg (Rentz 1999)
Fig. 5.12 Flows for “copper, secondary, at refinery” (outputs) and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg copper.

5.5 Copper, at regional storage
5.5.1 Copper, from imported concentrates, at refinery in ecoinvent

This module is identical to “Copper, primary, at refinery, RER” but without neither any input of cop-
per concentrate, nor any input of SX-EW copper. In this way it can be used to be combined with im-
ported concentrates, whose production – mining and beneficiation – otherwise would be counted dou-
ble.
5.5.2 Copper, at regional storage in ecoinvent

This module is based on information on the origin and type of imported copper for Germany in 1994
cited in Krauss et al. (1999). The layout relates to Fig. 5.2.
The module characterises the copper supply in Germany in 1994. It is designed for the use of the metal
in various technical applications such as alloys and construction material in European production
processes.
formation are shown in Fig. 5.13. Ecoinvent meta information for this process is summarised in Tab.
5.1 on page 17 ff.
Data on the amount and origin of the imported primary copper and concentrate is given in Tab. 5.14
and Tab. 5.15, from which the transport service in tkm results. Krauss et al. (1999) estimate the secon-
dary copper use with 650'000 t in 1994, where 50% went directly into the foundries and 50% into the
secondary copper production.
For copper and concentrate supplied by countries with no specification of the origin (“other”), the re-
spective modules for South America is used, but for transport distance the resulting mean distance
from RER, RLA, and ID has been chosen.
For prompt scrap, which is used directly in the foundries, the respective module for collection and
handling of iron scrap has been chosen as proxy.
Data quality
The overall quality of this module is fair, depending on accuracy of the consumption values. The sys-
tem of copper production in its different pathways is far more complex than pictured in this study. A
more comprehensive approach as shown in Bertram et al. (2002) may enhance the quality of the data.
Tab. 5.14 Import of copper concentrate to Germany in 1994. In the last column values for transport service in tonnes
and kilometres (tkm) are given (Krauss et al. (1999)). Per kg “copper, from imported concentrates” (=13.1%
in the supplyer mix “copper, at regional storage”) 3.38 kg concentrates in average are required (=0.441 kg
concentrates per kg “copper, at regional storage”).
Origin RER RLA ID other total Per kg concentrate

concentrate 197669 t 206235 t 164316 t 84612 t 652832 t 1 kg
ship sea 2750 km 14268 km 18800 km 11939 km 7.59E+09 tkm 11.62 tkm
ship river 412 km 412 km 1.20E+08 tkm 0.184 tkm
train 190 km 190 km 5.36E+07 tkm 0.082 tkm
lorry 45 km 310 km 10 km 122 km 8.48E+07 tkm 0.13 tkm

Tab. 5.15 Import of primary copper (=56% in “copper, at regional storage”) to Germany in 1994. In the last column val-
ues for transport service in tonnes and kilometres (tkm) are given (Krauss et al. (1999)).
Origin RER RLA ID other imported total Per kg Per kg

Primary mix
copper 353281 t 111078 t 32584 t 138979 t 193000 t 635922 t 1 kg 0.56 kg
ship sea 3071 km 14100 km 20000 km 12390 km 5.02E+09 tkm 6.06 tkm 3.40 tkm
ship river 87 km 87 km 4.28E+07 tkm 0.052 tkm 0.029 tkm
train 2000 km 550 km 1050 km 1200 km 200 km 1.01E+09 tkm 1.22 tkm 0.681 tkm
lorry 15 km 200 km 108 km 100 km 6.18E+07 tkm 0.075 tkm 0.042 tkm
Tab. 5.16 Copper supply in the European region and the resulting fractions in the supply dataset “copper, at regional
storage, RER” (column “% of total”). Copper supply from imported concentrates means that the copper is
produced within Europe (Germany being reference region) from concentrates that are extracted elsewhere
(see Tab. 5.17).
Copper supply primary Copper secondary copper % of primary % of total

RER 353281 t 43% 23.9%
RLA 111078 t 13% 7.51%
ID 32584 t 4% 2.20%
Other 138979 t 17% 9.40%
from imported concentrates 193000 t 23% 13.1%
total primary copper 828922 t 100% 56.0%
Prompt scrap 325000 t 22.0%

Old scrap 325000 t 22.0%
total secondary 650000 t 44.0%
total copper 1478922 t 100%
Tab. 5.17 Concentrate imported for the production of primary copper within Europe (Germany being reference region).
The resulting concentrate requirements in the dataset “copper, at regional storage” are displayed in the last
column “kg / kg ‘copper, at regional storage’”.
Concen- Amount % from imported kg / kg imported con- kg / kg “copper, at

trate concentrates centrates regional storage”
RER 197669 t 30% 1.02 0.134
RLA 206235 t 32% 1.07 0.139
ID 164316 t 25% 0.85 0.111
other 84612 t 13% 0.44 0.057
total 652832 t 100% 3.38 0.441
copper pro- 193000 t 13.1%

duced from
imported
conc.
average grade 29.6%

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Remarks
Location
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
Primary copper
 metals extraction No RLA copper, primary, at refinery 7.51E-02 kg (Krauss, 1999) 1 1.07 (2,1,1,3,1,1,4)
from RLA
primary copper
 metals extraction No ID copper, primary, at refinery 2.20E-02 kg (Krauss, 1999) 1 1.07 (2,1,1,3,1,1,4)
from ID
primary copper
 metals extraction No RER copper, primary, at refinery 2.39E-01 kg (Krauss, 1999) 1 1.07 (2,1,1,3,1,1,4)
from RER
primary copper No information on origin. Proxy
 metals extraction No RLA copper, primary, at refinery 9.40E-02 kg (Krauss, 1999) 1 1.07 (2,1,1,3,1,1,4)
other countries production in RLA chosen
import primary transport, transoceanic freight
 transport systems ship No OCE 3.40E+00 tkm (Krauss, 1999) 1 2.00 (2,1,1,3,1,1,5)
copper (sea) ship
import primary
 transport systems ship No RER transport, barge tanker 2.90E-02 tkm (Krauss, 1999) 1 2.00 (2,1,1,3,1,1,5)
copper (river)
copper, at regional storage
import primary
 transport systems train No RER transport, freight, rail 6.81E-01 tkm (Krauss, 1999) 1 2.00 (2,1,1,3,1,1,5)
copper (train)
import primary transport, lorry >16t, fleet
 transport systems road No RER 4.18E-02 tkm (Krauss, 1999) 1 2.00 (2,1,1,3,1,1,5)
copper (lorry) average
copper copper concentrate, at
 metals extraction No RLA 1.39E-01 kg (Krauss, 1999) 1 1.07 (2,1,1,3,1,1,4)
concentrate RLA beneficiation
 metals extraction No ID 1.11E-01 kg (Krauss, 1999) 1 1.07 (2,1,1,3,1,1,4)
concentrate ID beneficiation
metals extraction No RER 1.34E-01 kg (Krauss, 1999) 1 1.07 (2,1,1,3,1,1,4)
concentrate RER beneficiation
copper
No information on origin. Proxy copper concentrate, at
concentrate other  metals extraction No RLA 5.72E-02 kg (Krauss, 1999) 1 1.07 (2,1,1,3,1,1,4)
production in RLA chosen beneficiation
countries
import copper transport, transoceanic freight
 transport systems ship No OCE 5.13E+00 tkm (Krauss, 1999) 1 2.00 (2,1,1,3,1,1,5)
concentrate (sea) ship
import copper
 transport systems ship No RER transport, barge tanker 8.10E-02 tkm (Krauss, 1999) 1 2.00 (2,1,1,3,1,1,5)
concentrate (river)
import copper
 transport systems train No RER transport, freight, rail 3.63E-02 tkm (Krauss, 1999) 1 2.00 (2,1,1,3,1,1,5)
concentrate (train)
import copper transport, lorry >16t, fleet

 transport systems road No RER 5.73E-02 tkm (Krauss, 1999) 1 2.00 (2,1,1,3,1,1,5)
concentrate (lorry) average
primary production copper, from imported
 metals extraction No DE 1.31E-01 kg (Krauss, 1999) 1 1.07 (2,1,1,3,1,1,4)
DE concentrates, at refinery
prompt scrap from as proxy collection and handling
 metals extraction No RER iron scrap, at plant 2.20E-01 kg (Krauss, 1999) 1 1.07 (2,1,1,3,1,1,4)
RER of iron scrap used.
secondary copper
 metals extraction No RER copper, secondary, at refinery 2.20E-01 kg (Krauss, 1999) 1 1.07 (2,1,1,3,1,1,4)
from RER
copper, primary, at
 metals extraction No RER copper, at regional storage 1.00E+00 kg (Krauss, 1999)
regional storage
Fig. 5.13 Flows for “copper, at regional storage” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg copper.

5.6 Copper alloys: brass bronze and brazing solders

From the multitude of different copper alloys the two most important are modelled as brass and
bronze:
- Brass in ecoinvent is cartridge brass (70/30 brass, C26000) composed of 70% (w/w) copper and
30% (w/w) zinc.
- Bronze in ecoinvent is CuSn4 (C51000) composed of 94.97% (w/w) Cu, 5% (w/w) Sn and 0.03%
(w/w) P.
From the multitude of brazing solders the general purpose cadmium-free solders are often used. In this
class, the CuZn40 is an important material that is chosen as "typical" brazing solder in this study.
Production of copper alloys is basically done by melting the different metals and casting of ingots. No
specific data are available for the processes. It is assumed that due to the high vapour pressure of zinc
and tin, parts of these inputs arise as air emissions. It is further assumed that filters are commonly used
to abate these emissions. The filter dust containing mainly zinc or tin is assumed to be recycled.
The casting of brass and bronze to the final form needs the moulds and energy for melting the alloy.
Moulds are either made of metal (steel) or of sand.
5.6.1 Copper alloy production in ecoinvent

The three modules "brass, at plant", "bronze, at plant", and "brazing solder, cadmium free, at plant"
include the material and energy inputs for melting the metals. The energy is calculated using the heat
capacity and the enthalpy of fusion of the constituents (see Tab. 5.18). A mix of 50% natural gas, 40%
heavy fuel oil, and 10% electricity is assumed for the heat production. A metal loss of 1% is assumed.
This metal is recycled.
Tin and zinc emissions to air (after abatement) are assumed 0.1% (w/w) of the respective input.
The standard distances according to Frischknecht et al. (2003) for the transportation of the metal in-
puts are used.
The infrastructure is approximated by the infrastructure for aluminium casting. Thereby it is assumed
that the amount of infrastructure corresponds to the volume of material throughput and thus is recalcu-
lated according to the different densities of aluminium and copper alloys.
The ecoinvent meta-information for the copper alloy production is given in Tab. 5.19 and the in- and
output flows for "brass, at plant", "bronze, at plant", and "brazing solder, cadmium free, at plant" are
given in Fig. 5.14, Fig. 5.15 and Fig. 5.16 respectively. Those for "casting, brass" and "casting,
bronze" are given in Fig. 5.17 and Fig. 5.18 respectively.

Tab. 5.18 Calculation of energy consumption for brass and bronze production
Copper Zinc Tin Brass Bronze Solder Unit Source

thermal capacity 383 385 227 J/(kg*K)
Binz et al.
heat of fusion (*105) 2.05 1.11 0.596 J/kg
(1984)
melting point 1083 419.5 231.9 950 1050 900 °C
production temp. 1200 1200 1200 1100 1100 1100 °C estimate
energy input (theoretical) 0.665 0.573 0.332 0.643 0.654 0.634 MJ/kg
calculation
energy input (for ecoinvent)1 0.731 0.630 0.365 0.708 0.720 0.698 MJ/kg
fraction Cu2 1 0 0 0.707 0.9595 0.606 -
fraction Zn2 0 1 0 0.303 0 0.404 - definition
fraction Sn2 0 0 1 0 0.0505 0 -
1 10% heat loss during heating and melting assumed
2 including 10% material loss for brass, bronze and solder
5.6.2 Data quality

All data for the brass and bronze production are assumptions based on qualitative information and
theoretical models. Thus uncertainty is rather high. These data are of insufficient quality for LCA
where these metals play an important role.

Tab. 5.19 Meta information for the production processes of the modelled copper alloys
brazing solder, cad-

Name brass, at plant bronze, at plant
mium free, at plant
Location CH CH RER
Unit kg kg kg
copper and zinc including copper and tin including copper and tin including
Included Processes their melting and casting of their melting and casting of their melting and casting of
brass ingots bronze ingots solder
Amount 1 1 1
Hartlot, cadmiumfrei, ab
Local Name
Messing, ab Werk Bronze, ab Werk Werk
Synonyms
Stands for brass with 70% Stands for bronze with Stands for solder with 60%
Cu and 30% Zn. All data 95% Cu and 5% Sn. All Cu and 40% Zn. All data
General Comment to ref- are calculated based on data are calculated based are calculated based on
erence function assumptions and theoreti- on assumptions and theo- assumptions and theoreti-
cal models. Their overall retical models. Their over- cal models. Their overall
quality is poor. all quality is poor. quality is poor.
CAS Number
Start Date 2000 2000 2000
End Date 2000 2000 2000
riod 1 1 1
Other Period Text
Production data relate to Production data relate to Production data relate to
the European average. the European average. the European average.
Geography text Transports of inputs relate Transports of inputs relate Transports of inputs relate
to the consumption in to the consumption in to the consumption in
Switzerland Switzerland Europe
Assumed energy mix for Assumed energy mix for Assumed energy mix for
Technology text melting. Abatement of air melting. Abatement of air melting. Abatement of air
emissions assumed emissions assumed emissions assumed
Production Volume
based on literature and based on literature and based on literature and
Sampling Procedure
internet internet internet
Extrapolations none none none

Tab. 5.19 Meta information for the production processes (cont.)
Name casting, brass casting, bronze

Location CH CH
Unit kg kg
Melting of copper and tin
Melting of copper and zinc and casting of bronze
Included Processes
and casting of brass parts. parts. Metal input is not
Metal input is not included. included.
Amount 1 1
Local Name Giessen, Messing Giessen, Bronze
Synonyms
All data are calculated All data are calculated
General Comment to ref- based on assumptions and based on assumptions and
erence function theoretical models. Their theoretical models. Their
overall quality is poor. overall quality is poor.
CAS Number
End Date 2003 2003
riod 1 1
Other Period Text
Data relate to the Euro- Data relate to the Euro-
Geography text
pean average. pean average.
Assumed energy mix for Assumed energy mix for
Technology text melting. Abatement of air melting. Abatement of air
emissions assumed emissions assumed
Production Volume
based on literature and based on literature and
Sampling Procedure
internet internet
Extrapolations none none

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Location
Remarks
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
copper  metals extraction No RER copper, at regional storage 7.07E-01 kg calculation 1 1.70 (4,5,3,5,4,5,3)
zinc for coating, at regional

zinc  metals extraction No RER 3.03E-01 kg calculation 1 1.70 (4,5,3,5,4,5,3)
storage
heat from natural heat, natural gas, at industrial

 natural gas heating systems No RER 3.54E-01 MJ calculation / estimation 1 1.70 (4,5,3,5,4,5,1)
gas furnace >100kW
heat from heavy heat, heavy fuel oil, at

 oil heating systems No RER 2.83E-01 MJ calculation / estimation 1 1.70 (4,5,3,5,4,5,1)
fuel oil industrial furnace 1MW
brass, at plant
electricity, medium voltage,

electricity  electricity production mix No UCTE 1.97E-02 kWh calculation / estimation 1 1.70 (4,5,3,5,4,5,2)
(4,5,n.A.,n.A.,n.A.,n.A.,
transport rail  600 km transport systems train No RER transport, freight, rail 6.06E-01 tkm estimation 1 2.09
5)
(4,5,n.A.,n.A.,n.A.,n.A.,
transport road  50 km transport systems road No CH transport, lorry 28t 5.05E-02 tkm estimation 1 1.30
6)
Infrastructure for aluminium casing, corrected by

infrastructure  metals extraction Yes RER aluminium casting, plant 4.90E-11 unit calculation / estimation 1 3.39 (4,5,3,5,4,5,9)
density (Al 2700 kg/m3, Brass 8470 kg/m3)
 waste heat from electricity use air unspecified Heat, waste 7.08E-02 MJ calculation 1 1.70 (4,5,3,5,4,5,13)
 Zn 0.1% of Zn input to air air unspecified Zinc 3.03E-04 kg calculation / estimation 1 1.95 (4,5,3,5,4,5,31)
 waste dust from filters to recycling --> not inventoried 9.80E-03 calculation / estimation
 brass metals extraction No CH brass, at plant 1.00E+00 kg
Fig. 5.14 Flows for "brass, at plant" and its representation in the ecoinvent database

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Location
Remarks
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
tin  metals extraction No RER tin, at regional storage 5.05E-02 kg calculation 1 1.70 (4,5,3,5,4,5,3)

gas furnace >100kW

bronze, at plant

(4,5,n.A.,n.A.,n.A.,n.A.,
5)
(4,5,n.A.,n.A.,n.A.,n.A.,
transport road  50 km transport systems road No CH transport, lorry 28t 5.05E-02 tkm estimation 1 1.30
6)

 Sn 0.1% of Sn input to air air unspecified Tin 5.05E-05 kg calculation / estimation 1 1.95 (4,5,3,5,4,5,31)
 bronze metals extraction No CH bronze, at plant 1.00E+00 kg
Fig. 5.15 Flows for "bronze, at plant" and its representation in the ecoinvent database

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Location
Remarks
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
zinc  brazing solder, cadmium free, at plant metals extraction No RER

zinc for coating, at regional
storage
4.04E-01 kg calculation 1 1.70 (4,5,3,5,4,5,3)

gas furnace >100kW


(4,5,n.A.,n.A.,n.A.,n.A.,
5)
(4,5,n.A.,n.A.,n.A.,n.A.,
transport road  100 km transport systems road No RER transport, lorry 32t 1.01E-01 tkm estimation 1 1.30
6)

 Zn 0.1% of Zn input to air air unspecified Zinc 4.04E-04 kg calculation / estimation 1 1.95 (4,5,3,5,4,5,31)
brazing solder, cadmium free,

 brazing solder metals extraction No RER 1.00E+00 kg
at plant
Fig. 5.16 Flows for "brazing solder, cadmium free, at plant" and its representation in the ecoinvent database

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Location
Remarks
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
gas furnace >100kW


casting, brass

 Zn 0.1% of zn input to air air unspecified Zinc 3.03E-04 kg calculation / estimation 1 1.95 (4,5,3,5,4,5,31)
 casting of brass metals processing No CH casting, brass 1.00E+00 kg
Fig. 5.17 Flows for "casting, brass" and its representation in the ecoinvent database

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Location
Remarks
name in
General
Process
Output
Modul
Name
Input
Infra-
Type
Sub-
Unit
gas furnace >100kW

casting, bronze


 Sn 0.1% of Sn input to air air unspecified Tin 5.05E-05 kg calculation / estimation 1 1.95 (4,5,3,5,4,5,31)
 casting of bronze metals processing No CH casting, bronze 1.00E+00 kg
Fig. 5.18 Flows for "casting, bronze" and its representation in the ecoinvent database

5.7 Telluride and Silver containing slime as by-product

The chief source of tellurium is a by-product of electrolytic copper refining, where it concentrates into
the anode slimes. This source accounts for over 90% of the supply, although tellurium is also recov-
ered from the flue dusts and gases generated during the smelting of copper, bismuth and lead ores, and
from lead refinery skimmings 9 .
Like tellurium, also silver accumulates in the anode slimes. Different sources report that the share of
silver from copper extraction declined from around 30% in 1996 to 25% today (JMB (2002); Klapwijk
et al. (2006); Renner et al. (2002)).
Since the composition of the arising slimes chiefly depends on the composition of the copper anodes,
the content of silver and tellurium varies considerably (Tab. 5.20).
Tab. 5.20 Characterisation of different anode slimes. Numbers are in %wt; span relates to reported minimal – maximal
values; the mean value in Cooper (1990) is the median of the reported operations (n=18).
Operations Ag Te Se Cu Source
Canadian smelter 0.2 - 3.5 4 - 25 4 - 15 16 - 24 Chen & Dutrizac (2004)
Noranda slimes 0.52 10.9 21.3 20.3 Knockaert (2002)
Canadian copper re- 21.3 3.19 10.9 20.3 Knockaert (2002)
finers
Nippon mining 20.58 3.64 15.23 4.73 Knockaert (2002)
Inco 6 2 7 17 Knockaert (2002)
18 operations world- 10.8 1.1 7.9 21.3 Cooper (1990)
wide (4.1 - 62) (0 - 3) (0.6 - 46) (2.0 - 53)
5.7.1 Anode slime generation in ecoinvent

The multi-output-process "anode slime generation, primary copper production" delivers the co-
products "anode slime, silver and tellurium containing, primary copper production" and "cathode,
copper, primary copper production". The product "cathode, …" is an intermediary in the production of
refined copper and is not to be used by the LCA practitioner. The flow "anode slime, …" is a by-
product of this process that receives part of the burdens allocated.. (Fig. 5.19).
shown in Fig. 5.20. The ecoinvent meta information for this process is summarised in Tab. 5.21.
Copper production Silver production

anode slime processing of
generation, anode slime, silver, from copper
anode slime, silver and tellurium
primary copper primary copper production, at refinery
containing, primary copper production
production production
cathode, copper, primary copper production copper-telluride cement, from copper production
Fig. 5.19 Process layout of the anode slime generation and its relation to other processes in ecoinvent.
9
http://www.mmta.co.uk/economicsFacts/miningJournalReview.aspx, “Tellurium” in the Mining Journal Review, The Minor
Metals Trade Association, accessed in July 2007.

Tab. 5.21 Meta information for the dataset anode slime generation, primary copper production.
Name anode slime generation, primary copper production

Location GLO
Unit kg
Data Set Version 1
The module includes the upstream production of primary copper represented by the pri-
Included Processes mary copper input and the resource depletion of silver and telluride found jointly with
copper.
Amount 1
Local Name Anodensumpf Erzeugung, primäre Kupferproduktion
The multi-output-process "anode slime generation, primary copper production" delivers
the two co-products "anode slime, silver and tellurium containing, primary copper pro-
General comment to duction" and "cathode, copper, primary copper production". The product "anode slime,
reference function …" is an intermediary in the production of secondary gold from electric scrap and is not
to be used by the LCA practitioner. The flow "cathode, …" is a by-product of this process
that receives part of the burdens allocated.
Start Date 1993
End Date 2003
Period 1
Geography text Data refers to global copper production and hence reflects a global average.
Technology text Electrorefining of copper anode.
Sampling Procedure literature
Extrapolations see Geography and Technology
Both reference flows are intended to be used as raw material in the production of the respective met-
als. For further information on process and inventory of the refined metals refer to the respective Parts
within this report.
Based on anode slime compositions listed in Tab. 5.20, the anode slime is assumed to contain 20% re-
sidual copper, 20% silver and 3% tellurium. The nominal generation of this anode slime is harmonised
with the world production average over the last three years (Tab. 2.4). Considering that 25% of silver
(4’875 t/a average from 2004-2006) and 95% of tellurium (107.6 t/a average from 2004-2006) were
supplied by the copper industry, 326 g refined silver and 7 g refined tellurium are produced per tonne
copper that is extracted (Tab. 5.22). Given an extraction efficiency of 100% for silver (closed cycle)
the amount of 1.63 kg anode slime (20% silver) per tonne cathode copper produced is calculated. Tak-
ing the Te content of 3% in the slime, 49 g Te per tonne copper extracted would then be co-extracted
on a global average. However, this 49 g/t contrast with the 7 g/t Te that are estimated with a top-down
calculation basing on mine production figures. This mismatch between bottom-up and top-down cal-
culation can be explained by the fact, that in many cases the rarer but jointly extracted metals such as
Tellurium are completely recovered on a global average. Many operations do not recover Tellurium at
all. A second reason may be found in the limited extraction efficiency when leaching tellurium.
However, the allocation of the upstream burden from the copper production – which determines this
inventory – is allocated by means of the effective produced amount of silver and tellurium from this
source (top-down figures). Therefore the uncertainty about slime composition and completeness of the
recovery of the resources in the ground do not affect this inventory model but the amount of silver and
tellurium resources required (input flows from nature).
The upstream burden is reflected by the production of one kg copper on a global average; the associ-
ated burdens are allocated to the reference flows by their value. The elementary flows that represent

the depletion of the resource itself for tellurium and silver are introduced using the same extraction
yields as for copper, since the metals are incorporated in the same mineral particles.
Tab. 5.22 Production factors for the by-products silver and tellurium from copper production, three years average
(USGS (2007)).
Year Copper mine Silver from Tellurium from Silver per cop- Tellurium per
production, tpa copper, tpa copper, tpa per, t / t copper, t / t
2004 14'600'000 4'925 88.065 3.37E-04 6.03E-06
2005 15'000'000 4'825 113.05 3.22E-04 7.54E-06
2006 15'300'000 4'875 121.6 3.19E-04 7.95E-06
mean 14'966'667 4'875 107.6 3.26E-04 7.17E-06
5.7.2 Allocation and resource correction

The allocation split is done by the value represented by the reference flows “cathode copper” and “an-
ode slime”. Since the anode slime has no market price as such, it is assessed by the marginal profit that
is assumed to be in the order of the overall operations profit. A value of 10% of the price of the refined
commodity (telluride and silver) has been chosen.
As explained in the preceding section the value of the anode slime is measured in the effective amount
of silver and tellurium that is produced on a global average, including the extraction losses from the
point where silver or tellurium is being extracted, i.e. CuTe precipitate. Therefore, and in accordance
with Tab. 5.22, it is assumed that 326 g Ag and 7.97 g Te per tonne produced copper cathode are ef-
fectively extracted, assuming a closed-loop situation for silver (recycling the electrolyte to the process)
and 90% for the preparation of CuTe as precipitate. The proceeds for refined silver and tellurium are at
269 $/kg and 151 $/kg (Tab. 5.23).
With this approach an internal value of 5.45 $ per kg anode slime and 4.48 $ per kg cathode copper
and a respective allocation split of 0.2% of the burden to the 1.63 g of anode slime and the remaining
99.8% of the burden to the 1 kg of cathode copper.
Part of the copper resource that is represented by the upstream inventory for copper production is allo-
cated by the economic procedure to the anode slime and has to be corrected.
The depletion of silver (326 g) and tellurium (49 g) per tonne of copper extracted is not accounted for
in the upstream processes represented by the copper production. They therefore have to be introduced
and are allocated fully to the anode slime.
Per kg copper produced, 1.12 kg copper resource is depleted. According to the allocation split 0.20%
of this amount or 2.22E-3 kg is attributed to the anode slime and has to be subtracted and re-allocated
to the cathode copper produced.

Tab. 5.23 Average market price for copper (LME), silver and tellurium in $ per tonne. Data compiled after USGS (2003)
through 2004 and USGS (2007) for 2005/6.
Silver Copper Tellurium

$/t $/t $/t
1996 166'900 2'404 46'300
1997 157'200 2'358 41'900
1998 178'000 1'734 39'700
1999 169'000 1'674 33'100
2000 161'000 1'944 30'900
2001 140'000 1'694 33'100
2002 148'000 1'671 35'300
2003 157'000 1'879 37'500
2004 207'000 2'910 22'000
2005 238'880 3'704 211'644
2006 360'088 6'812 220'462
This study 268'656 4'475 151'369
(mean 2004-2006)
5.7.3 Data quality

This inventory is mainly based on theoretical considerations to reflect the upstream burden and re-
sources depletion in order to allocate them to the reference flow of anode slime. High uncertainties in
the amount and allocation split are assumed.

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
structure
Category
Standard
category
Remarks
Location
name in
General
Process
Output
Name
Modul
Input
Infra-
Type
Sub-
Unit
Copper input  by alloc split metals extraction No GLO copper, primary, at refinery 1.00E+00 kg 1
primary copper production
Tellurium, 0.5ppm in sulfide,
anode slime generation,
Tellurium in
 to slime resource in ground Te 0.2ppm, Cu and Ag, in 4.89E-05 kg 1 1.09 (2,2,1,2,1,3,3)
copper resource
crude ore, in ground
Silver, 3.2ppm in sulfide, Ag
Silver in copper
 to slime resource in ground 1.2ppm, Cu and Te, in crude 3.26E-04 kg 1 1.09 (2,2,1,2,1,3,3)
resource
ore, in ground
Necessary to correct the
allocation of copper resource to
resource correction, CuMo,
Correction Copper  to slime the slime by economic metals extraction No GLO 2.22E-03 kg 1 1.09 (2,2,1,2,1,3,3)
copper, negative
allocation split. Subtract 0.20%
of a total 1.12 kg required.
The subtracted amount is

resource correction, CuMo,
Correction Copper  to copper allocated to the cathode copper metals extraction No GLO 2.22E-03 kg 1 1.09 (2,2,1,2,1,3,3)
copper, positive
flow.
cathode, copper, primary

 cathode, copper 99.8% metals extraction No GLO 1.00E+00 kg 1 1.09 (2,2,1,2,1,3,3)
copper production
anode slime, silver and
Anode slime, from
 0.2% metals extraction No GLO tellurium containing, primary 1.63E-03 kg 1 1.09 (2,2,1,2,1,3,3)
copper refining
copper production
Fig. 5.20 Flows for “anode slime generation, primary copper production” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg copper.

5.8 Molybdenum
Molybdenite is the starting material for molybdenum trioxide and ferromolybdenum, which is the raw
material for alloying steels with molybdenum. Since molybdenite is almost exclusively won from por-
phyry deposits, its production process is related to the exploitation of copper ores. Secondary porphyry
deposits bear lower concentrations of molybdenum, in this case it is won only as co-product in the
copper production (see Section 2.2 Primary Molybdenum on page 7). Molybdenum from co-
production is modelled as co-product in Section 5.2 Winning primary copper. The ecological burden is
distributed between molybdenite and copper concentrate according to their revenue (price times pro-
duction).
According to the scheme in Fig. 5.1 a separate module for the exploitation of molybdenite as main
product is necessary as well as a production mix module for molybdenite. These two modules are dis-
cussed in the following sections.
5.8.1 Exploitation of molybdenite in ecoinvent

The module describes the exploitation of molybdenite as main product. It is based entirely on the re-
lated process mining of copper ore. It is designed solely as part of a production mix. The multioutput-
process “mining of molybdenite” delivers the two co-products “molybdenite concentrate, main prod-
uct” and “copper, concentrate, coupleproduction Mo”.
The ecoinvent meta information for this process is summarised in Tab. 5.1 on page 17 ff.
The primary porphyry deposits are richest in molybdenum and are exploited for its molybdenum val-
ues. Hence the attribution of the ecological burden is different, since copper concentrate now is the by-
product. Beside of the different resource demanded and the consequently different output of molyb-
denite – relating to the higher Mo content (0.041% in crude ore) – no other differences with respect to
the Module "mining copper ore" (GLO) exist. Details are discussed in chapter 5.1 of this study.
Allocation
The ecological burden of the molybdenite mining is distributed among the coupled products copper
concentrate and molybdenite concentrate. As allocation procedure the revenue is chosen. With a price
of US$ 5.75 per kg Mo (USGS (2003)) this gives a revenue of US$ 3190 per tonne molybdenite con-
centrate with an assumed composition of 92.5% MoS2. Copper concentrate is traded in large quanti-
ties. However, the price of copper concentrate could not yet be assessed; it is estimated with the final
bids of three auctions held in 2000 on an online web-auction 10 resulting in an average price of 15.3 US
cents payable for refining and smelting per pound copper. With a copper price of 0.95 US$/kg, a price
of 522 US$ per tonne of concentrate with 29.7% copper content is calculated These assumptions and a
production ratio of 74 kg molybdenite concentrate per 3.5 t copper concentrate lead to an average
revenue of 236 US$ and 1816 US$ per tonne molybdenite and copper contained in the ore, respec-
tively, – with the metallurgical yield included – and to an allocation factor of 11% and 89% for mo-
lybdenite and copper concentrate, respectively.
Data quality
The same constraints apply like for the module “mining of copper ore”. The assessment will highly
vary with the price of copper concentrate and molybdenite. The yield of molybdenite is not evident; it
is merely estimated based on the content of molybdenum in the ore.
10
http://www.copperconcentrate.com/default.asp, accessed at 01. 03. 2003

5.8.2 Molybdenite, at plant in ecoinvent

The module describes the production mix composed of molybdenite from coupled production in cop-
per industry and molybdenite as main product.
Ecoinvent meta information for this process are summarised in Tab. 5.1 on page 17 ff.
Within the world-wide production of molybdenite a share of 58% originates from co-production in the
copper industry, 42% originate from primary deposits. The molybdenite from scheelite leaching in the
tungsten production is neglected in this study. Since molybdenite is a intermediary product, it is mod-
elled as production mix, i.e. no transportation and energy use for the distribution is included in this
module. Molybdenite from co-production is inventoried as “Molybdenum, molybdenite concentrate,
couple-production Cu”, the main product molybdenite as “Molybdenum, molybdenite concentrate,
main product”.
For the subsequent production of molybdenum oxide and later ferromolybdenum, transportation to the
manufacturing plants and transport of the finished product to the user has to be considered.
Data quality
The main constraints in this module are missing information on the production process. The choice of
copper production as proxy is a strong simplification. The overall assessment will highly vary with
share – molybdenum as main- or by-product – and price of copper concentrate and molybdenite.
5.8.3 Molybdenum, at regional storage in ecoinvent

Molybdenum is produced from molybdenite by roasting to MoO3 and subsequent metallurgical steps.
No process specific data for these steps have been found. Thus the roasting and metallurgical proc-
esses from the zinc production are used as a proxy. The concentrate of zinc and molybdenite are both
sulphidic and the metal concentration is similar. Thus the inventory for the production of 1 kg "zinc, at
regional storage" is used directly for describing the production of 1 kg molybdenum. To adjust the
concentrate input, the amount of zinc concentrate that is needed in the production of zinc is subtracted
and the molybdenite concentrate therefore inventoried. The amount of molybdenite concentrate is cal-
culated assuming the same yield as for zinc concentrate to zinc. It is slightly lower than the amount of
zinc concentrate, because the metal concentration is slightly higher (55% Mo vs. 53% Zn).
formation are shown in Fig. 5.21. The ecoinvent meta information for this process is summarised in
Tab. 5.1 (page 17 ff).
Data quality
The very crude approximation of the previously described production of molybdenum with data from
another metal implies a poor data quality. The data set must only be used if the relevance of molybde-
num is low in a LCA study. If the focus of such a study would lie on molybdenum, the process would
need to be further investigated.

Source for
structure
Comment
ecoinvent
Deviation
Standard
Category
Remarks
category
Location
name in
General
Output
Process
Name
value
Modul
Infra-
mean
Input
value
Mean
Type
Sub-
Unit
M olybdenite
molybdenit
e, at plant
molybdenum concent rate,

from coupled  met als ext raction No GLO 5.79E-01 kg (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,4)
couple production Cu
production
M olybdenite
M olybdenum concentrate,
from main  met als ext raction No GLO 4.21E-01 kg (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,2)
main product
production
 M olybdenite, mix met als ext raction No GLO molybdenite, at plant 1.00E+00 kg (Krauss, 1999) 1 1.10 (2,2,2,3,1,3,2)
M olybdenite  concentration 55% met als ext raction No GLO molybdenite, at plant 1.83E+00 kg calculated 1 2.15 (4,5,1,2,5,5,12)
zinc, primary, at regional
zinc met als ext raction No RER 1.00E+00 kg definition 1 2.15 (4,5,1,2,5,5,4)
regional storage

molybdenum, at
storage
exact amount
zinc concentrate, at
zinc concentrate  concentration 53% met als ext raction No GLO -1.90E+00 kg calculated 1 1.00 inventoried in zinc
beneficiat ion
production
resource exact amount
0.013004 kg Cd ore in resource correct ion, PbZn,
correction Cd in  met als ext raction No GLO 3.95E-02 kg calculated 1 1.00 inventoried in zinc
0.62565 kg Zn concentrate cadmium, positive
zinc concentrate production
Ressource exact amount
0.00021673 kg In ore in resource correct ion, PbZn,
correction In in  met als ext raction No GLO 6.59E-04 kg calculated 1 1.00 inventoried in zinc
0.62565 kg Zn concentrate indium, positive
zinc concentrate production
molybdenum, at regional
 M olybdenun met als ext raction No RER 1.00E+00 kg
storage
Fig. 5.21 Flows for “molybdenite, at plant” and “molybdenum, at regional storage” and their representation in ecoinvent.

6. Literature
6 Literature
www.ecoinvent.ch.
Anonymous (1998a) Anonymous (1998a) Cyanide Management. In: Best Practice Environmental
Management in Mining. Environment Australia, Online-Version under:
http://www.ea.gov.au/industry/sustainable/mining/booklets/cyanide/.
Anonymous (1998b) Anonymous (1998b) Dust Control. In: Best Practice Environmental Manage-
ment in Mining. Environment Australia, Online-Version under: http://www.
ea.gov.au/industry/sustainable/mining/booklets/dust/.
Anonymous (1998c) Anonymous (1998c) Pollution Prevention and Abatement Handbook. The World
Bank Group, Online-Version under: http://lnweb18.worldbank.org/essd/
essdext.nsf/51ByDocName/PollutionPreventionandAbatementHandbook.
Ayres et al. (2002) Ayres R. U., Ayres L. W. and Råde I. (2002) The Life Cycle of Copper, its Co-
Products and Byproducts. MMSD - Mining, Minerals and Sustainable Develop-
ment, Online-Version under: http://www.iied.org/mmsd/mmsd_pdfs/ayres_
lca_main.pdf.
Bertram et al. (2002) Bertram M., Graedel T. E., Rechberger H. and Spatari S. (2002) The contempo-
rary European copper cycle: waste management subsystem. In: Ecological Eco-
nomics(42), pp. 43-57.
Binz et al. (1984) Binz J., Friedrich M., Gebauer P., Höhn W., Ineichen R., Kappus H., Marzetta
B., Schilt H., Schnyder A., Thöni W., Bächtiger P., Ess H., Etter K., Florin R.,
Joho P., Karrer H., Muehry H., Plüss A., Rosenthaler E., Rüetschi R., et al.
(1984) Formeln und Tafeln. 3. Edition. Orell Füssli Verlag, Zürich.
Bruch et al. (1995) Bruch K. H., Gohlke D., Kögler C., Krüger J., Reuter M., Röpenack I. v., Rom-
bach E., Rombach G. and Winkler P. (1995) Sachbilanz einer Ökobilanz der
Kupfererzeugung und -verarbeitung, Teil 1. In: Metall, 49(4), pp. 252-257.
Chen & Dutrizac (2004) Chen T. T. and Dutrizac J. E. (2004) Gold in the electrorefining of copper and
the decopperizing of copper anode slimes. In: Jom, 56(8), pp. 48-52.
Cooper (1990) Cooper W. C. (1990) The Treatment of Copper Refinery Anode Slimes. In: Jom-
Journal of the Minerals Metals & Materials Society, 42(8), pp. 45-49.
EC (2002) EC (2002) European Dioxin Inventory - Secondary copper production. European
Commission. Retrieved from http://europa.eu.int/comm/environment/dioxin/
stage1/seccopper.pdf. Date of last revision: 22.08.2002.
EEA (1999) EEA (ed.) (1999) Primary Copper Production (B336). In: EMEP / CORINAIR
Emission Inventory Guidebook. 3rd edition October 2002 Update Edition. Euro-
pean Enviornmental Agency, Online-Version under: http://reports.eea.eu.int/
technical_report_2001_3/en/en/tab_content_RLR.
Fabian (1997) Fabian H. (1997) Copper. In: Ullmann's encyclopedia of industrial chemistry
(ed. Anonymous). 5th edition on CD-ROM Edition. Wiley & Sons, London.

6. Literature
Graedel et al. (2002) Graedel T. E., Bertram M., Fuse K., Gordon R. B., Lifset R., Rechberger H. and
Spatari S. (2002) The contemporary European copper cycle: The characteriza-
tion of technological copper cycles. In: Ecological Economics(42), pp. 9-26.
Hilbrans & Hinrichs (1999) Hilbrans H. and Hinrichs W. (1999) Stoffmengenflüsse und Energiebedarf bei
der Gewinnung ausgewählter mineralischer Rohstoffe; Teilstudie Nickel. In:
Hochfeld (1997) Hochfeld C. (1997) Bilanzierung der Umweltauswirkungen bei der Gewinnung
von Platingruppen-Metallen für PKW-Abgaskatalysatoren. In: Werkstattreihe,
Vol. 101. Öko-Institut, Freiburg i. Breisgau.
IPPC (2001) IPPC (2001) Integrated Pollution Prevention and Control (IPPC); Reference
Document on Best Available Techniques in the Non Ferrous Metals Industries.
European Commission. Retrieved from http://www.jrc.es/pub/english.cgi/0/
733169
IPPC (2002) IPPC (2002) Integrated Pollution Prevention and Control (IPPC); Draft Refer-
ence Document on Best Available Techniques for Management of Tailings and
Waste-Rock in Mining Activities. European Commission. Retrieved at
01.03.2003 from http://www.jrc.es/pub/english.cgi/0/733169
Isbell (1997) Isbell C. A. (1997) Copper Alloys. In: Ullmann's encyclopedia of industrial
chemistry (ed. Anonymous). 5th edition on CD-ROM Edition. Wiley & Sons,
London.
JMB (2002) JMB (2002) Spezialausgabe: Silber - November 2002. In: JMB News. Johnson
Matthey & Brandenberger AG, Zürich, Online-Version under: http://www.
johnson-matthey.ch/download/2002JMBNewsSilver241002.pdf.
Klapwijk et al. (2006) Klapwijk P., Walker P., Ryan P., Newman P. and others (2006) World Silver
Survey 2006 - A Summary. GFMS & The Silver Institute, Online-Version un-
der: http://www.silverinstitute.org/publications/.
Knockaert (2002) Knockaert G. (2002) Tellurium. In: Ullmann's Encyclopedia of Industrial
Chemistry. Online version, posting date: September 15, 2000 Edition. Wiley-
Interscience, Online-Version under: http://dx.doi.org/10.1002/14356007.a26_
177.
Krauss et al. (1999) Krauss U., Wagner H. and Mori G. (1999) Stoffmengenflüsse und Energiebedarf
bei der Gewinnung ausgewählter mineralischer Rohstoffe; Teilstudie Kupfer. In:
Martens et al. (2002) Martens P. N., Ruhrberg M. and Mistry M. (2002) Flächeninanspruchnahme des
Kupfererzberbaus. In: Erzmetall, 55(5/6), pp. 287-293.
NPI (2001) NPI (2001) Emission Estimation Technique Manual for Mining. In: National
Pollution Inventory. Environment Australia, Online-Version under:
http://www.npi.gov.au/.
Renner et al. (2002) Renner H., Schlamp G., Hollmann D., Lüschow H. M., Rothaut J., Knödler A.,
Hecht C., Schlott M., Drieselmann R., Peter C. and Schiele R. (2002) Silver, Sil-
ver Compounds, and Silver Alloys. In: Ullmann's Encyclopedia of Industrial

6. Literature
Interscience, Online-Version under: http://dx.doi.org/10.1002/14356007.a24

_107.
Rentz et al. (1999) Rentz O., Krippner M., Hähre S. and Schultmann F. (1999) Report on Best
Available Techniques (BAT) in Copper Production. Deutsch-Französisches In-
stitut für Umweltforschung (DFIU) and University of Karlsruhe (TH).
Sebenik et al. (1997) Sebenik R. F., Burkin A. R. and Dorfler R. R. (1997) Molybdenum and Molyb-
denum Compounds. In: Ullmann's encyclopedia of industrial chemistry (ed.
Anonymous). 5th edition on CD-ROM Edition. Wiley & Sons, London.
USGS (2003) USGS (2003) Mineral Commodity Summaries. U.S. Department of the Interior,
U.S. Geological Survey. Retrieved at 01.03.2003 from http://minerals.
usgs.gov/minerals/pubs/mcs/. Date of last revision: 26-Feb-2003@09:36.
USGS (2007) USGS (2007) Mineral Commodity Summaries. US. Geological Survey. Re-
trieved from http://minerals.usgs.gov/minerals/pubs/mcs/

Part IV
Nickel
Data v2.1 (2009)
Authors: Mischa Classen, EMPA Dübendorf

Hans-Jörg Althaus, EMPA Dübendorf
Summary

ported in Frischknecht et al. (2006). The changes were implemented with v1.3 and affect mainly the amount of
added lime and the herewith related emissions of carbon dioxide in the metallurgical step.
The amount of limestone input in the Ferronickel-Dataset was increased by factor 4. The CO2 emission related
to limestone decomposition in MO-Process (#35) is decreased a factor 16. The allocation factors applied on this
emission (Values v1.2: 8.46% on Copper / 91.54% on Nickel) are corrected to 14% on Copper / 86% on Nickel.
Further more. The amount of sulphidic tailings disposed in the MO-Process increased by 2/3.
In Version 2.1, the representation of part of the electricity input into the Ferronickel-dataset was corrected from
UCTE mix to hydropower.
Chapter “6 Cumulative Results and Interpretation” of the original report has been deleted for the current update.
Summary
The primary nickel and ferronickel production is modelled for the global situation.
Nickel is co-produced with Copper and the model is made accordingly. Allocation is made by revenue.
For a lack of data the secondary nickel production could not be inventoried. However, since most of the nickel is
used in alloys it is not recycled separately but with the alloyed steel.
Mining, beneficiation and the different metallurgical processes are inventoried vertically aggregated because of
difficulties to clearly allocate some of the flows to one single step and because the intermediate steps would not
make sense in another context. The Figures in this report however contain the available information to the single
steps.

Table of Contents
Table of Contents
SUMMARY ......................................................................................................................I
1 INTRODUCTION ....................................................................................................... 1
2 RESERVES AND RESOURCES OF NICKEL ................................................................... 2

2.1 Sulfidic Ores.......................................................................................................................... 2
2.2 Laterite Ores: Limonite and Silicate...................................................................................... 3

3.1 Class I Nickel ........................................................................................................................ 5
3.2 Ferronickel ............................................................................................................................ 6

4.1 Nickel (class I)....................................................................................................................... 8
4.2 Ferronickel ............................................................................................................................ 8
4.3 Nickel Alloys......................................................................................................................... 8

5.1 Winning of Class I Nickel ................................................................................................... 13
5.1.2 Emission.............................................................................................................................. 14
5.1.3 Waste................................................................................................................................... 16
5.1.4 Class I Nickel in ecoinvent.................................................................................................. 16
5.2 Ferronickel .......................................................................................................................... 32
5.2.2 Emission.............................................................................................................................. 32
5.2.3 Waste................................................................................................................................... 33
5.2.4 Ferronickel in ecoinvent...................................................................................................... 33
6 LITERATURE ......................................................................................................... 39

1. Introduction
1 Introduction
This chapter describes the production of Nickel from cradle to consumer in Europe. Inventory data is
taken mainly from (Hilbrans & Hinrichs (1999)) together with background data from IPPC (2002) and
IPPC (2001). Market data on nickel and copper are provided by the London Metal Exchange 1 .
Since nickel always is associated with other metals, the production yields a fraction of other metals
than nickel, mainly copper (Cu), cobalt (Co) and platinum gruop metals. The process of primary pro-
duction described in this chapter is formally a multioutput-process with the coupled products Ni and
Cu. This makes it necessary to allocate the inventory to these products. The demand for nickel-
containing ore is allocated by mass. The rest of the inventory is allocated by revenue. The coupled
production of cobalt is neglected because due to a lack of data no meaningful allocation factors could
be established.
Known annual world production capacity of primary nickel products in the early 1990s was approxi-
mately 1240 kilotonnes, 485 of which were in former Soviet and East Bloc countries. In recent years,
annual world production has averaged in excess of 900 kilotonnes. Primary nickel products fall
broadly into two categories - Class I and Class II. Class I products contain at least 99.8% nickel whilst
all other primary nickel products are Class II. Unlike Class I products, Class II products range widely
in nickel content.
1
London Metal Exchange, daily and monthly stocks and prices under http://www.lme.co.uk/data_prices/ home.html, accessed
at 01.03.2003

2. Reserves and Resources of Nickel
2 Reserves and Resources of Nickel

cle. When recovered the properties are theoretically identical. The metal can be re-used in another
product life cycle, provided that its concentration makes a recovery profitable. The recovered metal is
referred to as secondary metal, whereas the metal originating directly from the ore is referred to as
primary metal. In LCIs these two types – primary and secondary metal – are often differentiated be-
cause they bear a different ecological burden.
Since nickel is used in ecoinvent as intermediate product in different metal manufacturing industries,
no secondary production is covered in this report. Secondary nickel is consumed directly in the form
of re-melted scrap and other recycled products generally in the production of ferro-nickel and stainless
steel. Other secondary materials such as catalysts and precipitator dusts are recovered in the primary
smelting processes, usually in the slag furnace.
Nickel ores can be classified in a sulphidic type and a oxidic type. An overview over the different
nickel bearing ores is given in Tab. 2.1, the composition of some ores in Tab. 2.2. Although 80 % of
the world's known nickel resources on land are in oxide ore bodies, only 40 % of nickel production
comes from this source. The greater part of the nickel produced is recovered from sulphide ores. The
nickel content of sulphide ores can be concentrated relatively easily and cheaply by physical methods,
while oxide ores must be treated in bulk. The potential for recovering the more numerous and valuable
co-products also favours the exploitation of sulphide ore. However, the grades of sulphide ores are
continually declining as the richer ores are consumed, and the cost of underground mining, particu-
larly labour costs, is continually increasing. Inevitably therefore, oxide ore bodies, which are all rela-
tively uniform in grade and can be surface mined, must in future support a larger share of nickel pro-
duction (Kerfoot (1997)).
A detailed overview over the global factory production and a statistic on US use and production of
nickel is available in the online version of the USGS “Mineral Commodity Summary” (USGS (2003)).
The world-wide mine production by country in 2002 and the estimated reserves is listed in Tab. 2.3.
2.1 Sulfidic Ores

Sulphide nickel ores consist principally of nickeliferous pyrrhotite (Fe7S8) , pentlandite (Ni, Fe)9S8,
and chalcopyrite (CuFeS2). Pentlandite, the commonest of the sulphide minerals, probably accounts
for 60 % of world nickel production. Nickeliferous pyrrhotite is normally the most abundant phase in a
nickel ore. It contains nickel in solid solution (0.2 – 0.5 % Ni) in addition to very finely divided pent-
landite inclusions. Chalcopyrite is the most important copper mineral, although cubanite is often pre-
sent in lesser amounts. Sulphide ores typically grade 0.4 – 2.0 % nickel, 0.2 – 2.0 % copper, 10 –30 %
iron, and 5 – 20 % sulphur. The balance consists of silica, magnesia, alumina, and calcium oxide.
Although the major minerals are the same in all large sulphide ore bodies, the relative amounts of pyr-
rhotite, pentlandite, and chalcopyrite vary widely. The relative amounts of the three minerals largely
determine the choice of processing route. In addition to nickel production, a large number of valuable
co-products can be recovered from sulphide ores, including copper, cobalt, the platinum group metals,
gold, silver, selenium, tellurium, sulphur, and iron. Cobalt occurs principally in solid solution in pent-
landite, at concentrations ranging from 1 to 5 % of the nickel content.
Major sulphide ore bodies occur in Canada, the Soviet Union, the Republic of South Africa, Australia,
Zimbabwe, and Finland. According to (Hilbrans & Hinrichs (1999)) 62 % of the world nickel produc-
tion originates from sulfidic ore in 1994.

2.2 Laterite Ores: Limonite and Silicate

The oxidic ores of nickel are formed by a chemical concentration process that occurs as a result of the
lateritic weathering of peridotite rock. The dissolved nickel and magnesium, and the colloidal silica,
percolate downwards in the laterite deposit, where they precipitate as hydrated silicates.
Some of the nickel may in fact remain in the upper layer, which is thus enriched in iron and nickel, but
depleted in magnesium and silica. Such material, consisting mainly of ferric oxide minerals, is de-
scribed as limonitic. This type of ore is typically relatively rich in cobalt and chromium. By contrast,
in the silicate minerals, the separation of iron and nickel is more complete, and the nickel is present in
silicate minerals with high magnesia content. Both types of material are almost always found in a lat-
erite deposit, but in widely differing proportions.
Oxide nickel mineral deposits are currently exploited in Australia, Brazil, Cuba, the Dominican Re-
public, Greece, Indonesia, the Philippines, New Caledonia, Yugoslavia, the United States, and the So-
viet Union. The laterite deposits in Oregon and California were formed when the climate was sub-
tropical in past geologic epochs.
Tab. 2.1 Principal nickel minerals (Kerfoot (1997)).
Mineral Ideal formula Nickel

content,%
Sulfides
Pentlandite (Ni,Fe)9S8 34.22
Millerite NiS 64.67
Heazlewoodite Ni3S2 73.30
Polydymite Ni3S4 57.86
Siegenite (Co,Ni)3S4 28.89
Violarite Ni2FeS4 38.94
Arsenides
Niccolite NiAs 43.92
Rammelsbergite NiAs2 28.15
Gersdorffite NiAsS 35.42
Antimonides
Breithauptite NiSb 32.53
Silicates and oxides
Garnierite (Ni,Mg)6Si4O10(OH)8 47%
Nickeliferous limonite (Fe,Ni)O(OH) · nH2O low

Tab. 2.2 Composition of some nickel ores after (IPPC (2001)).
Ore Type Ni% Cu% Co%

Murrin Murrin Laterite 1.25 0.08
Cawse Laterite 1 0.07
Cerro Matoso Laterite 2.89
Selebi-Phikwe Sulphide 0.65 0.75 0.06
Falconbridge, Sudbury Sulphide 1.7 1.8 0.05
Falconbridge, Raglan Sulphide 3.2 0.9 0.05
INCO, Copper Cliff Sulphide 1.55 2 0.04
Outokumpu, Silver Sulphide 9.8 0.11
Swan
Mount Keith Sulphide 0.6 0 0.01
Tab. 2.3 Mine production and reserves of nickel in 2001. “Reserves” is referred to as “the part of the reserve base
which cold be economically extracted or produced at the time of determination”, while “Reserve base” de-
and production practices” (USGS (2003)).

2001 [kT Ni / a] 2002 [kT Ni / a] [kT Ni] [kT Ni]
(estimated)
Australia 197 186 22000 27000
Botswana 26.2 23.2 490 920
Brazil 45.4 44.9 670 6000
Canada 193.361 188 5200 15000
China 51.5 55.7 3600 7600
Colombia 52.962 55.4 900 1100
Cuba 70.662 73.1 5600 23000
Dominican Republic 31 24.3 690 1000
Greece 20.83 22.4 490 900
Indonesia 102 105 3200 13000
New Caledonia 117.554 98.2 4400 12000
Philippines 27.359 31.8 940 5200
Russia 325 328 6600 9200
South Africa 36.443 38 3700 12000
Venezuela 13.6 20.6 610 610
Zimbabwe 8.145 9.69 15 260
Other countries 9.24 13.7 1300 5100
World total (rounded) 1330 1320 61000 140000


Nickel is rarely used without further processing. Nickel is marketed solely on the basis of its chemical
composition and according to its chemical purity as either a Class I or a Class II product. Typical
compositions for Class I and II products are given in Tab. 3.2.
Class I includes the high-purity products, electrolytic cathode, carbonyl refined granules and powder,
and hydrogen-reduced nickel in powder or briquette form.
Class II includes the various grades of ferronickel, nickel oxide, and metallised nickel oxide. In this
study only ferronickel is of significance because of its use in the manufacturing of alloy steel and
stainless steel.
However, Class I and Class II are intermediate market products. Most of it is used for alloying other
metals. Stainless steels are particularly valued for their hygienic properties in hospitals and kitchens.
In some applications, such as jet engines, nickel alloys are essential and cannot be substituted with
other materials.
In ecoinvent only the intermediate nickel products “Class I Nickel” and “Ferronickel” are of impor-
tance. They are asked in the steel manufacturing and production of other alloys.
The reported effects of nickel and its compounds on humans are wide ranging, comprising effects that
are both beneficial - the probable essentiality of nickel in humans - as well as harmful - skin allergy
and, in certain circumstances, respiratory cancer (NIDIS (2003)).
Health Effects include
- Allergy of the skin to nickel - resulting in dermatitis. The following image shows a case of nickel
eczema (dermatitis).
- Allergy of the lung to nickel - resulting in asthma
- In certain circumstances of exposure in refining, there may be a risk of cancer of the paranasal si-
nuses (close to the nose) and of cancer of the lung. This risk has now probably been completely
removed in most modern plants.
- Nickel tetracarbonyl - an intermediate in the Mond process for refining nickel is very toxic to the
lungs and heart.
3.1 Class I Nickel

Commercial nickel of this class (CAS 7440-02-0) is normally 99.7 % pure or better. Synonyms are
“unwrought nickel, not alloyed”. Nickel is a silver-white metal. Factors that make nickel and its alloys
commercially valuable commodities include strength, corrosion resistance, high ductility, good ther-
mal and electrical conductivity, magnetic characteristics (below 340 °C) and catalytic properties.
Some physical properties of pure nickel are given in Tab. 3.1. Nickel is not tarnished by air and is re-
sistant to fluorine (F2) but is attacked by acids (except conc. HNO3). Class I nickel is marketed as
cathodes, granules, powder, briquettes and rounds.

Tab. 3.1 Some physical properties of pure nickel (Kerfoot (1997)).
property value
Melting Point (mp) 1455°C (1728 K)
Boiling Point (bp) 2730°C (3003 K)
Relative density (25°C) 8.9
Volume increase on melting 4.5%
Heat of fusion (mp) 302 J/g
Heat of sublimation (25°C) 7317 J/g
Heat of vaporization (Tcrit) 6375 J/g
Standard entropy 29.81 J/K
Thermal conductivity (0-100°C) 88.5 W m–1 K–1
Heat capacity (0-100°C) 0.452 J g–1 K–1
Electrical resistivity (20°C) 6.9 /cm
Temperature coefficient of electrical resis- 6.8×10–3 K–1
tivity(0-100°C)
Thermal expansion coefficient (0-100°C) 13.3×10–6 K–1
Modulus of elasticity 199.5 GPa
Brinell hardness 85
3.2 Ferronickel
Ferronickel (CAS 11133-76-9) is a market product of the nickel industry. It consists of an alloy of iron
with nickel, with a nickel content of 20 - 50 % (Tab. 3.2). It is used in the manufacturing of stainless
steels.
Tab. 3.2 Commercial forms of nickel (Kerfoot (1997)).
Type Composition, %
Ni Co Cu Fe C S O
Class I:
Cathodes >99.90 0.005 0.002 0.01 0.001
–5
Pellets >99.97 5×10 0.001 0.0015 <0.10 0.0003
Powder 99.74 <0.10 <0.10 <0.010 <0.15
Briquettes 99.90 0.03 0.001 0.01 0.01 0.0035
Rondelles 99.25 0.37 0.046 0.022 0.022 0.004 0.042
Class II:
Ferronickel 20-50 1-2 rest 1.5-1.8 <0.3
Nickel oxide 76.0 1.0 0.75 0.30 0.006 rest


Unlike other non-ferrous metals, nickel is rarely used by itself but is commonly mixed with other met-
als to produce alloys. There are many different alloys containing nickel, each providing a particular
combination of technical properties (corrosion resistance, mechanical properties and service life). Thus
the two forms of nickel – class I nickel and ferronickel – are to high degree intermediate products for
the metal industry.
Tab. 4.1 Production of different nickel products (USGS (2003)).
2000
Ferronickel 220000 20%
Metal 638000 57%
Oxide sinter 98200 9%
Chemicals 18000 2%
Unspecified 143000 13%
Total 1117200
More than 80 %of primary nickel is consumed in some 3’000 different alloys, grouped generally as
stainless steels, alloy steels, and non-ferrous alloys. About 65 % of primary nickel is used in stainless
steels alone, as shown in Fig. 4.1.
2
Fig. 4.1 Uses of nickel in different industries. Source: NiDI , 1996
Electroplating accounts for roughly 10 % of primary nickel consumption, with the balance consumed
by foundries and a host of other uses including chemicals, catalysts, batteries, welding rods, coinage,
pigments for enamels/glasses/ceramics, electronics, and printing inks.
Most of the plating, foundry and "other" applications are "end-uses" of nickel; that is to say, the prod-
ucts are used directly by the customer or "end-user". Steels and other nickel alloys, on the other hand,
are "intermediate" products that must be further processed or "transformed" into end-use commercial
products in a number of industrial applications. These applications are shown in Fig. 4.2.
The major end-use sectors for nickel shown in Fig. 4.2 reflect various applications for the many of
known nickel alloys as well as the direct use of primary nickel for electroplated and other products
noted above. The use therefore can be diverted into an intermediate use in the form of metallic nickel
(class I) and Ferronickel and into the end-use as nickel alloys.
2
Nickel Development Institute, http://www.nidi.org/, accessed at 01. 03. 2003.

3
Fig. 4.2 Application of nickel Source: NiDI , 1996
4.1 Nickel (class I)

Primary nickel metal powder is used in many applications including battery manufacture, coin produc-
tion, welding products, fuel filters, brake linings, catalysts, cutting tools, paints, plastics, electronics,
implants, and surface treatments by plasma spraying.
4.2 Ferronickel
Ferro-alloys are mainly used as master alloys in the iron and steel industry, because it is the most eco-
nomic way to introduce an alloying element in the steel melt. Ferronickel in this case therefore is used
as master alloy in the production of stainless steels. Further more it is used as starting materials in the
preparation of chemicals and pure metal; as reducing agents (e.g., the use of ferrosilicon to reduce rich
slag); as alloying elements in nonferrous alloys and as starting materials for special products such as
amorphous metals.
4.3 Nickel Alloys

Nickel-bearing, "austenitic" stainless steels contain variable amounts of nickel, between 5 to 25 per-
cent, and average about 10 % nickel. Commercial uses include chemical and food processing equip-
ment, transportation equipment, structural claddings and other architectural applications, durable con-
sumer goods such as cookware, and hospital and medical applications.
Nickel alloys can be hot or cold worked. Other processes include annealing, drawing, forging, spin-
ning, and joining by soldering, brazing, gas welding, and resistance welding. Machining of most
nickel alloys requires high-speed tools and flooding with cutting compounds.
3
Nickel Development Institute, http://www.nidi.org/, accessed at 01. 03. 2003.

Tab. 4.2 Some of the more common nickel alloys.
Alloy Constituents
Monel metals combinations of nickel and copper
Inconel metals nickel alloys containing chromium and iron
Hastelloy metals combinations of chromium, molybdenum and iron
Illium nickels nickel alloys with chromium, molybdenum and copper
Nickel alloy steels are used in heavy machinery manufacturing, armoured plate, tools, and high-
temperature equipment including gas turbines and scrubbers. Non-ferrous alloys include nickel-copper
alloys, typically containing about 65% nickel, and copper-nickel alloys, typically containing about
10% nickel, used extensively in desalination plants for converting seawater into fresh water. Ni-
Cu/Cu-Ni alloys are also employed in marine equipment and inorganic acid and alkali handling
equipment. Nickel-chromium alloys, containing approximately 40-70% nickel, are used in corrosive
chemical environments and high-temperature applications such as heating elements and gas turbine
aero-engine components. Nickel-copper-zinc alloys, including nickel silver with less than 18% nickel,
are used for decorative purposes, including jewellery.
Different alloy types are used in specific applications. The constitution of some common nickel alloys
is given in Tab. 4.2.
- Corrosion-resistant alloys: nickel, monel, hastelloy, illium, nirolium for use in chemical and proc-
ess plant, plating coinage, food processing and water treatment.
- High temperature alloys: inconel, incoloy, nimonic, nimocast used for furnace components, boiler
components, heat-treatment equipment, power generation, heat exchangers.
- Electrical Alloys: nickel, brightray, nichrome, constantan for use in ferry electronics, heating ele-
ments, industrial furnace elements, resistors, thermocouples and spark plugs.
- Magnetic Alloys: nickel, permalloy, mumetal used in magnetostrictive devices, magnetic shields,
HF transformation, loading cells, transformer and rotor laminations, temperature compensators.
- Controlled-expansion and constant modulus alloys: invar, nile, ni-Span, elinvar, bimetals, thermo-
stats, glass-sealing and weighing machines.
- Miscellaneous alloys: nitinol, hy-stor employed in shape-memory devices, hydrogen storage and
brazing alloys.

The winning of nickel can be classified in two main process chains as shown in Fig. 5.1. The first
process chain represents the winning of nickel from sulphidic ore, yielding Class I Nickel as market
product. This class of nickel is used in a wide range of applications. The other path represents the win-
ning of ferronickel from lateritic ore.
The aim of this study is to inventory the two intermediate products “Nickel Class I” and “Ferronickel”
as raw materials for nickel alloys. These two modules represent the most important application of
nickel. The system boundaries are set at the production gate, both modules represent the production
mix, no transport to the desired region of the application is included.
The meta information of the modules are summarised in Tab. 5.1.
Market
Mining Benefication Metallurgy Applications
Products
Nickel Class I
cathodes other 4%
pyro-
54% granules coins 3%
metallurgy
60% sulfidic concentrate powder batteries 3%
hydro-
6% briquettes plating 6%
metallurgy
raw ore rondelles alloys 14%
30%
pyro-
ferronickel metallurgy 6%
metallurgy
hydro- stainless
40% lateritic nickel oxide 64%
metallurgy steels
10%
Fig. 5.1 Winning of nickel from ore to application. Two main process chains can be identified, the sulphidic path
yielding Class I nickel and the lateritic path yielding ferronickel. Ferronickel is mainly used in the production
of stainless steel, whereas Class I nickel is used in a wide range of applications.

Tab. 5.1 Ecoinvent meta information for the processes for the production of Class I nickel and ferronickel.
Name nickel production, sulphidic ore, primary

Location GLO
Unit kg
The module includes a mining and a beneficiation step with the mining and benefi-
ciation infrastructure and disposal of overburden and tailings. Subsequently it in-
cludes the metallurgy step with the disposal of slag, the infrastructure and the sepa-
Included Processes ration of the co-product copper, and the refining step yielding the desired Class I
nickel inclusively the metallurgical infrastructure. Production, application and emis-
sions of most agents and additives used in beneficiation and metallurgy are also
included.
Amount 1
Local Name Nickel-Produktion, sulfidisches Erz, primär
Synonyms
The module describes the global production mix for Class I nickel in 1994 with a
coupled production of a certain amount of copper. The multioutput-process “nickel
production, sulphidic ore, primary” delivers the co-products “nickel, 99.5%, at plant”
and “copper, primary, coupleproduction nickel”. It is designed for the use of the
metal as intermediate product or as raw material in the plating industry or the manu-
facturing of stainless steels and alloys. It is not to be used if the impact of the Class
ence function
I nickel is considered to be high. In such cases, a more detailed analysis depending
on scope and allocation procedures has to be conducted. The data used is mainly
based on a study of the energy and material streams resulting from the production
of class I nickel. Lacking data mainly on direct process specific emissions were
taken form similar processes for copper winning.
CAS Number
Start Date 1994
End Date 2003
Other Period Text
This module represents a global production mix. Some processes like blasting, lime
Geography text and energy production and transport originate from data sets valid for Switzerland
or Europe.
Mining is done underground, followed by beneficiation trough flotation where con-
siderable amounts of agents and lime are added. Overburden and tailings are dis-
posed on the mining site and partly re-filled. This module covers the five major pro-
Technology text
duction routes, aggregated according to their market share in 1994. Sulphur dioxide
in the off-gas is recovered producing sulphuric acid. For electricity mix a mix of 60%
hydropower and 40% electricity at grid was chosen.
Production Volume 547'684 t nickel in sulfidic raw ore in 1994

Tab. 5.1 Ecoinvent meta information for the processes for the production of Class I nickel and ferronickel. (cont.)
Name ferronickel, 25% Ni, at plant

Location GLO
Unit kg
The module includes a mining and a beneficiation step with the mining infrastruc-
ture. Subsequently it includes the metallurgy step with the disposal of slag and the
Included Processes
metallurgical infrastructure. Production, application and emissions of most agents
and additives used in beneficiation and metallurgy are also included.
Amount 1
Local Name Ferronickel, 25% Ni, ab Werk
Synonyms ferro-alloy//Eisenlegierung
The module describes the global production mix for ferronickel in 1994. It is de-
signed for the use of the metal as raw material in the manufacturing of stainless
steels and alloys. It is not to be used if the impact of ferronickel is considered to be
General Comment to ref-
high. In such cases, a more detailed analysis has to be conducted. The data used
erence function
is mainly based on a study of the energy and material streams resulting from the
production of class I nickel. Lacking data mainly on direct process specific emis-
sions were taken form similar processes for copper winning.
CAS Number
Start Date 1994
End Date 2003
riod 1
Other Period Text
This module represents a global production mix. Some processes like blasting, lime
Geography text and energy production and transport originate from data sets valid for Switzerland
or Europe.
Mining is done in open cuts, followed by beneficiation with a simple mechanical
separation. Overburden and tailings are disposed inside the mining panels.This
Technology text
module covers the major production route for ferronickel. For electricity mix a mix of
60% hydropower and 40% electricity at grid was chosen.
Production Volume 329'478 t nickel in lateritic raw ore in 1993

5.1 Winning of Class I Nickel

This part is mainly based on (Hilbrans & Hinrichs (1999)). General process description was taken
from Kerfoot (1997). Allocation is done by revenue between the coupled products Nickel and Copper
using the values of 6647$ / t Nickel and 1986$ / t Copper.

Mining. 95% of sulphidic nickel ores are mined underground in depths between 200m and 1800m, the
ore is transferred to the beneficiation. Widening of the tunnels is mainly done by blasting. The over-
burden – material, which does not contain PGM-bearing ore – is deposed off-site and is partially re-
filled into the tunnels.
Beneficiation. After mining, the ore is first ground. In a next step it is subjected to gravity concentra-
tion to separate the metallic particles from the PGM-bearing minerals. After this first concentration
step, flotation is carried out to remove the gangue from the sulphidic minerals. For neutralisation lime
is added. In the flotation several organic chemicals are used as collector, frother, activator, depressor
and flocculant. Sometimes cyanide is used as depressant for pyrite. Tailings usually are led to tailing
heaps or ponds. As a result, nickel concentrates containing 7 - 25% Ni are produced.
Metallurgy. There are many different process possibilities to win the metal. The chosen process de-
pends on the composition of the ore, the local costs of energy carrier and the local legislation. Basi-
cally two different types can be distinguished: the hydrometallurgical and the pyrometallurgical proc-
ess with a share of 6 % and 54 % of the worldwide nickel produced. An overview is given in Fig. 5.2.
Fig. 5.2 Overview over some process possibilities for nickel winning. The chosen process depends on the composi-
tion of the ore, the local costs of energy carrier and the local legislation. Scheme from (IPPC (2001))
Pyrometallurgy. The pyrometallurgical treatment of nickel concentrates includes three types of unit
operation: roasting, smelting, and converting. In the roasting step sulphur is driven off as sulphur diox-
ide and part of the iron is oxidised. In smelting, the roaster product is melted with a siliceous flux
which combines with the oxidised iron to produce two immiscible phases, a liquid silicate slag which
can be discarded, and a solution of molten sulphides which contains the metal values. In the convert-
ing operation on the sulphide melt, more sulphur is driven off as sulphur dioxide, and the remaining
iron is oxidised and fluxed for removal as silicate slag, leaving a high-grade nickel – copper sulphide
matte. In several modern operations the roasting step has been eliminated, and the nickel sulphide con-
centrate is treated directly in the smelter.
Hydrometallurgy. Several hydrometallurgical processes are in commercial operation for the treat-
ment of nickel – copper mattes to produce separate nickel and copper products. In addition, the hy-
drometallurgical process developed by Sherritt Gordon in the early 1950s for the direct treatment of
nickel sulphide concentrates, as an alternative to smelting, is still commercially viable and competi-
tive, despite very significant improvements in the economics and energy efficiency of nickel smelting

technology. In a typical hydrometallurgical process, the concentrate or matte is first leached in a sul-
phate or chloride solution to dissolve nickel, cobalt, and some of the copper, while the sulphide is oxi-
dised to insoluble elemental sulphur or soluble sulphate. Frequently, leaching is carried out in a two-
stage countercurrent system so that the matte can be used to partially purify the solution, for example,
by precipitating copper by cementation. In this way a nickel – copper matte can be treated in a two-
stage leach process to produce a copper-free nickel sulphate or nickel chloride solution, and a leach
residue enriched in copper.
Refining. In many applications, high-purity nickel is essential and Class I nickel products, which in-
clude electrolytic cathode, carbonyl powder, and hydrogen-reduced powder, are made by a variety of
refining processes.
The carbonyl refining process uses the property of nickel to form volatile nickel-carbonyl compounds
from which elemental nickel subsides to form granules. Electrolytic nickel refineries treat cast raw
nickel anodes in a electrolyte. Under current the anode dissolves and pure nickel deposits on the cath-
ode. This electrorefining process is obsolete because of high energy demand and the necessity of
building the crude nickel anode by reduction with coke. It is still practised in Russia. Most refineries
recover electrolytic nickel by direct electrowinning from purified solutions produced by the leaching
of nickel or nickel – copper mattes. Some companies recover refined nickel powder from purified
ammoniacal solution by reduction with hydrogen.
The multitude of processing possibilities was reduced in (Hilbrans & Hinrichs (1999)) to five major
process routes for nickel winning from sulphidic ores. Their layout is given in Tab. 5.2. This classifi-
cation will be used further.
Tab. 5.2 Overview over the major process routes in the nickel winning from sulphidic ores. After (Hilbrans & Hinrichs
(1999)).
Process Ore Roasting Primary Matte Separa- Primary Nickel Refi-

Smelting conver- tion co- Refining ning
ting product
fluidized Electric Fur- Matte refining
alternative 1 Nickel- Converter
bed furnace nace Smelting: (Electrorefining)
concen-
trate Precipitation
alternative 2 Flash-Smelting Converter
with hydrogen
fluidized Electric Fur- Reductive elec-
alternative 3 Converter Wet
bed furnace nace Smelting: trolysis
Nickel- Electrolytic Re-
Roasting,
alternative 4 Copper- Flash-Smelting Converter Dry fining (Electrow-
reduction
Concen- inning)
trate Reverberatory
Multiple Roasting, Carbonyl Proc-
alternative 5 Furnace smelt- Converter Dry
hearth reduction ess
ing
5.1.2 Emission
Mining
The major emissions are due to mineral born pollutants in the effluents. The underground mining op-
erations generate roughly 80 % of the dust emissions from open pit operations, since the major dust
sources do not take place underground. Rain percolate through overburden and accounts to metal
emissions to groundwater.

Beneficiation
the ore itself. Flotation produce effluents containing several organic agents 4 used. Some of these
cally to release carbon disulphide. Tailings effluent contains additional sulphuric acid from acid rock
drainage.
Sulphur dioxide. In all of the process steps sulphur dioxide is emitted to air. Recovery of sulphur di-
oxide is only economic for high concentrated off-gas. SO2 emissions are by far the most important in
the metallurgy.
Carbon dioxide. In the beneficiation step, considerable amounts of lime is added to the ore for pH-
stabilisation, lime forms later flux in the metallurgical step. Lime decomposes into CO2 to form cal-
cite.
Dust. Dust carry over from the roasting, smelting and converting processes. Particulate emissions to
the air consist of metals and thus are often returned to the leaching process after treatment.
Chlorine. Chlorine is used in some leaching stages and is produced during the subsequent electrolysis
of chloride solution. The chlorine evolved is collected and re-used in the leach stage. The presence of
chlorine in wastewater can lead to the formation of organic chlorine compounds (AOX) if solvents etc.
are also present in a mixed wastewater.
VOC. VOCs can be emitted from the solvent extraction stages. A variety of solvents are used an they
contain various complexing agents to form complexes with the desired metal that are soluble in the or-
ganic layer.
Emissions to water. Metals and their compounds and substances in suspension are the main pollutants
emitted to water. The metals concerned are Cu, Ni, Co, As and Cr. Other significant substances are
chlorides and sulphates. Wastewater from wet gas cleaning (if used) of the different metallurgical
stages are the most important sources. The leaching stages are usually operated on a closed circuit and
Refining
number of digestion, electrolytic and purification processes. Chlorine is used extensively in the Miller
process and in the dissolution stages using hydrochloric acid and chlorine mixtrues respectively. Dust
and metals are generally emitted from incinerators and furnaces. VOC can be emitted from solvent ex-
traction processes, while organic compounds, namely dioxins, can be emitted from smelting stages re-
sulting from the poor combustion of oil and plastic in the feed material. All these emissions are subject
to abatement technologies and controlling. Large quantities of effluents contain amounts of metals and
organic substances.
4

5.1.3 Waste
Mining and Beneficiation
Overburden is deposed close to the mine. Tailings are deposed as piles and in ponds. In both the over-
burden and the tailings, acid rock drainage occurs over a long period of time. These effects are han-
dled in section IX “Non Ferrous Metal Winning – Auxiliary Processes”, chapter “Disposal of Tailings
and Overburden”.
Refining
Like in the pyro- and hydrometallurgical step quantities of solid residuals are generated, which are
residues generally comprise hydroxide filter cakes (ironhydroxide, 60% water, cat I industrial waste).
5.1.4 Class I Nickel in ecoinvent

In Hilbrans & Hinrichs (1999) data for the different process alternatives shown in Tab. 5.2 was col-
lected in detail. However, in this study an average was calculated based on the values from alternative
1 to 5, weighted with their share to the world-wide production. The resulting module thus represents
an aggregated value for the world-wide production of Class I nickel.
The module describes the global production mix for Class I nickel in 1994 with a coupled production
of a certain amount of copper. The multioutput-process “nickel production, sulphidic ore, primary” de-
livers the co-products “nickel, 99.5%, at plant” and “copper, primary, coupleproduction nickel”. It is
designed for the use of the metal as intermediate product or as raw material in the plating industry or
the manufacturing of stainless steels and alloys. The layout of the modules with general flow informa-
tion, remarks, sources, values and uncertainty information are shown in Fig. 5.3, Fig. 5.4, Fig. 5.5, Fig.
5.6, Fig. 5.7, Fig. 5.8 and Fig. 5.9. The following paragraphs describe calculations, sources and as-
sumptions chosen in this study. The functional unit of this process is one kilogram of Class I nickel.

The mobile equipment used for exploitation like drilling machines, scoop trams and underground
trucks is inventoried with the diesel consumption stated in Hilbrans & Hinrichs (1999). The used
module “diesel, burned in building machine” includes the infrastructure of the machines. 2000 m con-
veyor belts are further assumed with a capacity of 2000 t / h which gives 2E-5 m conveyor belt re-
quired per tonne mined raw ore. Since mining is underground and no heap leaching is done, no land
use for exploitation is assumed.
Mining facilities and smeltery are inventoried with respective modules, beneficiation plant is approxi-
mated according to chapter II “Non Ferrous Metal Winning – Auxiliary Processes” with the module
“aluminium hydroxide, plant”.

Tab. 5.3 Infrastructure in underground mining, assumptions used. A lifetime of ten years are estimated, and 5000
working hours yearly.
action Machine througput unit Infrastructure

Transport Conveyor belts 100'000'000 2000 2E-5
Mining.
In the considered processes the ore is mined underground. The overburden is disposed in piles near the
mine, the standard module for disposal of non-sulphidic overburden was chosen, which includes the
land use and transformation through the disposal. Refilled overburden is assumed to be burden free,
but cement and stowing material (assumption: sand) has to be added as stated in Hilbrans & Hinrichs
(1999). The degree of the refilling is documented for each alternative in Hilbrans & Hinrichs (1999).
Also water use and effluent from mining is documented. The use of freshwater was inventoried as
river water, where as the pit water was assumed to be groundwater, which has to be pumped. The
composition of the effluent was taken from IPPC (2002), its values are shown in Tab. 5.4. Concerning
electricity mix Hilbrans & Hinrichs (1999) assumes a share of 60% of hydropower due to the opera-
tions high and constant energy demand. This assumption is used also in this inventory. Cement, explo-
sives and sand are assumed to be transported by lorry over 50 km.
Beneficiation
lime and flotation agents are added. The separated gangue is disposed in tailings ponds, the concen-
trated ore is fed to the metallurgy, which is assumed to be on-site.
Dust emissions comprise additionally the dust emissions due to mining activities. Like in the mining
step, a share of 60% hydroelectricity was chosen according to Hilbrans & Hinrichs (1999).
Lime. Lime input was documented in Hilbrans & Hinrichs (1999) for each process alternative. The
specific lime input ranges from 2.0 to 8.5 kg per t ore treated with a weighted mean of 5.5 kg / t ore.
Reagents. No information on mass and nature of the floating agents used are available. For copper
mining a range of 25 – 300 g collector and 25 – 250 g frothing reagent per ton sulphidic ore is recom-
mended (Krauss et al. (1999)). This is confirmed by the values in Tab. 5.5. Collector and frother are
balanced in ecoinvent in their production as unspecified organic chemical. According to NPI (2001)
xanthates hydrolytically decompose to 0.53 mg carbon disulphide per mg xanthate. With 189 mg xan-
thates from Tab. 5.5 this is 100 mg carbon disulphide per t mined ore (Tab. 5.6).
cals, their emission to water is covered by reported emissions from mining sites shown in Tab. 5.4.
For pH adjustment and leaching the value for sulphuric acid was considered with 3.74 kg / t ore mined
(Tab. 5.6). As sulphuric acid is a by-product from the metallurgy, it is regarded as burden free. Other
acids are neglected.
ported data for total emissions shown in Tab. 5.6 are taken. Note that some theoretical emissions due
to the use of the reagents are higher than the reported ones in Tab. 5.4.
Sodium cyanide is widely used for extracting gold and other metals such as silver, copper and zinc. It
is used as leachate and as depressant for certain sulphidic minerals. The environmental fate of cyanide
compounds is complex and could not further be queried. A cyanide recovery of 80 – 90% is reported
due to abatement and recovery in the Golden Cross mine in New Zealand (Anonymous (1998a)). Here
a value of 80% is chosen for cyanide-abatement technology. With a reported use of 35 g NaCN per
tonne ore mined (Tab. 5.5), this means a direct emission of 3.75 g CN- per tonne ore mined.

Effluents. Additional to direct emissions due to the use of chemicals, total emissions to water are cho-
sen according to Tab. 5.4. These values are from base metal sites, thus only an approximation for the
real effluent concentrations in the nickel production. It is assumed here that the effluents also include
acid rock drainage (ARD). ARD after mine closure is modelled in the module “disposal, sulphidic tail-
ings, off-site”.
Dust. A case study in Anonymous (1998b) about a projected open cut gold-copper mining and proc-
essing facility in Australia predict annual dust emissions between 0.18 kg/t and 0.46 kg/t of dust pro-
duced per tonne of ore mined. The prediction was made using the Emission Estimation Technique
Manual for Mining (NPI (2001)). The maximum value of 0.46 kg/t is chosen for total dust emissions.
Underground mining activities are assumed to generate 80 % of the dust emissions compared to open
pit mining (Tab. 5.7) yielding a 0.37 kg dust per tonne mined ore. According to the IAEA 5 the dust
composition is approximated with values of a specific sulphidic nickel ore in Tab. 5.8. For PM10 val-
ues, the ratio TDS : PM10 : PM2.5 was approximated with values from copper ore mining 101.72 : 50 :
5 g / t ore taken from the CEMIP-database 6 , i. e. one tonne emitted dust corresponds to 491.5 kg PM10
and 49.2 kg PM2.5 respectively. The dust emissions resulting from underground activities are summed
up in Tab. 5.8. In Ecoinvent, for particulates differential classes are used adding up to the total dust
emission, i. e. 49.2 kg PM<2.5, 442 kg PM2.5-10 and 508 kg PM>10 per tonne dust
Tab. 5.4 Mining & Beneficiation, emissions to water. Composition of effluents from mining activities, values are
taken from total emissions to water from base metal sites in (IPPC (2002)).
Com- Unit Range This study Number of

pound (geometric Samples
mean)
Discharge Mio. m3 / a 0.08 - 22.9 3.65 6
Ca g / m3 686 686 1
SO4 g / m3 1750 - 3175 2357 2
COD g / m3 2.24 - 48.5 10.4 2
Solids g / m3 2.38 - 11.3 5.17 4
Al mg / m3 69.3 - 108 86.3 2
As mg / m3 0.264 - 14.1 2.95 3
Cd mg / m3 0.0901 - 1.02 0.316 4
Co mg / m3 0.742 - 0.823 0.781 2
Cr mg / m3 0.0311 - 9.62 0.547 2
Cu mg / m3 1.24 - 44.9 7.91 5
Fe mg / m3 61.7 - 1326 291 3
Mn mg / m3 24.7 24.7 1
Hg mg / m3 0.0155 - 0.12 0.0377 3
Ni mg / m3 0.792 - 1338 24.3 3
Pb mg / m3 0.0155 - 20 2.80 4
Zn mg / m3 5.37 – 225 75.9 5
N g / m3 2.50 - 1786 22.8 3
5
The IAEA Database of Natural Matrix Reference Materials, http://www-naweb.iaea.org/nahu/external/e4/nmrm/, accessed
01.03.2003
6

Tab. 5.5 Consumption of reagents in base metal mineral processing plants. Data taken from IPPC (2002).

(geometric samples
mean)
MIBC, Dowfroth
Sodium cyanide g/t 4-310 35 2
Dithiophosphate
phate
Flocculants - CMC g/t 1-100 11.1 3

Sulphuric acid (pH) g/t 387-12000 3739 4
Sodium hydroxide (pH) g/t 30-490 121 2

Lime input technosphere kg/t 2.0 – 8.5
Carbon dioxide air emission g/t inventoried in metallurgical step

Tab. 5.7 Dust emissions from typical coal strip-mining operations (Anonymous (1998b)). After this table underground
mining cause roughly 80% of the dust emissions of open pit mining.
Sources of dust Truck and Dragline Occurrence in

shovel op- operation underground
eration mining
% of total dust emission
Dragline n.a 27 Yes
Haul roads
Overburden 35 - Yes
Coal 7 42 Yes
Loading
Overburden 12 - Yes
Coal 4 5 Yes
Drilling
(Overburden and coal) 1 2 No
Blasting
Truck dumping 7 5 Yes
Topsoil removal 18 10 No
Exposed areas 6 7 No
Haul-road repairs 5 1 Yes

Tab. 5.8 Emissions to air resulting from mining activities per tonne mined ore. Values for total dust emission are
7
taken from Anonymous (1998b), partition of fractions of particulate matter from the CEMIP-database , the
dust composition correspond to an specific nickel sulphide with values taken from the IEAE-database (see
text).
Emission to Unit Emission per tonne

air mined ore
Cobalt mg / t ore 151
Copper mg / t ore 3680
Lead mg / t ore 36.8
Nickel mg / t ore 4526
Silver mg / t ore 1.58
Palladium mg / t ore 9.94E-03
Platinum mg / t ore 0.151
Aluminium mg / t ore 18400
Gold mg / t ore 0.0736
Calcium mg / t ore 12880
Magnesium mg / t ore 11040
Sulphur mg / t ore 36800
PM<2.5 g / t ore 18.1

PM2.5-10 g / t ore 162.8
PM>10 g / t ore 187.1
total dust g / t ore 368
Metallurgy and Separation of Copper as Co-product.

As shown in Tab. 5.2 there exist a variety of different production routes in the nickel metallurgy. In
this study all five process alternatives were considered according to their production share given in
Hilbrans & Hinrichs (1999). These alternatives cover at least 79 % of the world production of Class I
nickel. The values thus already are aggregated and reflect the mix of the different alternatives within a
certain process step, e. g. the electricity consumption in the roasting step. Consequently allocation fac-
tors for nickel differ from step to step.
Drying (alternative 2 &4). Flux is assumed to consist of 50% quartzsand and 50% lime, whose com-
bustion releases 0.44 kg CO2 per tonne lime, inventoried later in “overall process”. Drying is done by
gas. According to Hilbrans & Hinrichs (1999), 40% of the electricity consumed is from grid, 60% is
hydropower.
Roasting (alternative 1, 3 & 5). Flux is assumed to consist of 50% quartzsand and 50% lime, whose
combustion releases 0.44 kg CO2 per tonne lime, inventoried later in “overall process”. Drying is done
by oil. According to Hilbrans & Hinrichs (1999), 40% of the electricity consumed is from grid, 60% is
hydropower. A production of steam is reported in (Hilbrans & Hinrichs (1999)), which can not be ac-
counted for as benefit.
Smelting (all alternatives). Smelting is mainly done by oil (alternative 5, fluidized bed kiln) and in al-
ternative 4 high amounts of coke are required. 40% electricity are from grid, 60% hydropower as-
sumed. A production of steam is reported, which can not be accounted for as benefit. Dust emissions
from smelting are inventoried later under “overall process”. Here the major part of the sulphur dioxide
is emitted. The value of 1,5 t SO2 per tonne metal produced includes also the sulphur dioxide emis-
sions from drying and roasting. This value is reported at INCO’s environmental report for the sud-
7

bury’s operation 8 . It was chosen because it probably depicts the global situation more accurate than
the much lower values reported in (IPPC (2001)) reflecting the European situation with 0.164 t SO2 / t,
or the much higher value from the Russian operation at Noril’sk (5.3 t SO2/ t) reported in (Hochfeld
(1997)).
Converting (all alternatives). Converting is done by gas. According to Hilbrans & Hinrichs (1999),
40% of the electricity consumed is from grid, 60% is hydropower. A production of steam is reported
in (Hilbrans & Hinrichs (1999)), which can not be accounted for as benefit. Reported dust emissions
are not fugitive, they are caught in the wet scrubber and form a residue, which has to be disposed. Fu-
gitive dust emissions are inventoried later under “overall process”.
Sulphuric acid production (all alternatives). For production 40% electricity from grid, 60% hydro-
power assumed. The acid formed is not inventoried because it is assumed to be a by-product.
Other processes (all alternatives). Different electricity use from facilities etc. 40% of the electricity
from grid, 60% hydropower assumed.
Leaching / copper separation (alternatives 3, 4 & 5). Pyrometallurgical separation is done by gas.
40% electricity from grid, 60% hydropower assumed. Leaching agents reported in (Hilbrans &
Hinrichs (1999)) are approximated with the inventory for ammoniac production but no process emis-
sions were considered due to lack of data. Ammonium sulphate as by-product from alternative 2 was
not inventoried and assumed to be burden free (cut off).
Overall process. The emissions to air are taken from different sources as shown in Tab. 5.9. Effluents
were approximated with the value of 1.4 m3 per tonne concentrate worked from PGM-operations re-
ported in (Hochfeld (1997)) (Tab. 5.10). The emissions due to the effluent are taken from (IPPC
(2001)). Carbon dioxide from the calcination of lime added in several steps before is released at a rate
of 0.44 t CO2 per tonne lime added.
Purification. Until this step all inventory was allocated according the revenue obtained by co-
production of copper in alternatives 3, 4 & 5 (The only exception is the resource that is allocated by
mass (34% to copper)). In this purification step the inventory is fully allocated to nickel since no cop-
per is present any more. Each alternative differs from each other. An overview over the different con-
tributions is given in Fig. 5.7. No data on chlorine input and related emissions is available. For the re-
duction with hydrogen (alternative 2) a simple stoechiometry Ni:H of 59:1 yielding and a reaction ef-
ficiency of 90% was assumed, yielding a necessary input of 19.1 kg hydrogen per tonne nickel re-
duced. For all electricity use a share of 40% electricity from grid and 60% hydropower was assumed.
8
globally operating nickel producer, http://www.inco.com

Tab. 5.9 Emissions to air per tonne produced metal. Data was taken from similar processes in nickel (A) and copper
(B) metallurgy and multiplied with the sum of produced metals (PGM, Ni & Cu). Data is taken from (IPPC
9
(2001)), the distribution of the particle size is taken from CEPMEIP (C). VOC emission from leaching (D). In
this study values for SO2 emissions are estimated roughly with 1.5 t / t Ni as stated in the environmental re-
port of the sudbury operation.

Pb 0.03 B
Zn 0.10 B
Sn 0.007 B
Cu 0.04 B
As 0.006 B
Total dust 0.41 A
PM<2.5 0.0005 C
PM2.5-10 0.3 C
PM>10 0.10 C
SO2 1500 estimation
Ni 0.03 A
Co 0.03 A
VOC 0.20 A, D
TCDD 1.00E-08 B
Tab. 5.10 Composition of different wastewater effluents after treatment. A simple average from different reported val-
ues of European copper operations is chosen (IPPC (2001)). Values in the columns “process water”, “sur-
face run off” and “direct cooling water” represent geometric means of the range reported. Values for Cr, Hg
and Sn correspond to a copper semis production site, reported in (IPPC (2001)).
Sub- Process Surface run Direct coo- This study

stance Water off ling water (arithmetic
mean)
[g / m3] [g / m3] [g / m3] [g / m3]
Cu 0.045 0.063 0.05 0.0527
(0.01-0.2) (0.01-0.4) (0.01-0.25)
Pb 0.006 0.032 0.01 0.0160
(0.001-0.04) (0.005-0.2) (0.001-0.1)
As 0.032 0.014 0.01 0.0187
(0.01-0.1) (0.003-0.07) (0.001-0.1)
Ni 0.024 0.028 0.011 0.0212
(0.004-0.15) (0.002-0.4) (0.002-0.06)
Cd 0.003 0.004 0.001 0.00273
(0.0001-0.1) (0.0002-0.1) (0.0001-0.003)
Zn 0.045 0.11 0.1 0.0848
(0.01-0.2) (0.03-0.4) (0.02-0.5)
Cr 0.029
Hg 0.00029
Sn 0.029
9

Data quality
Mining. Data for major mass fluxes are satisfactory, while the emissions in to water may be overesti-
mated due to different system boundaries of reported effluent volume from Hilbrans & Hinrichs
(1999) and total effluent emissions reported in IPPC (2002).
Beneficiation. Like in mining, the emissions to water can be overestimated. The emissions from tail-
ings are assessed in an extern module, which itself contains rough assumptions of a global overburden
and tailings composition and a simplified disposal scenario. Emissions tough can vary considerably
from case to case.
Metallurgy. Process data is satisfactory and valid for over 90 % of the Class I nickel production. The
main emissions dominating possibly the whole process is sulphur dioxide, which is not reported in
Hilbrans & Hinrichs (1999). The chosen value of a single operation is consistent with other reported
values in IPPC (2001) and Hochfeld (1997). Poor data are available on composition and fate of solid
waste, the reuse rate was documented.
Separation of copper and further refining. Main data on energy and mass fluxes is satisfactory,
while data on direct emissions due to hydrometallurgical processes is poor. Namely chlorine emissions
can be considerable, generating both HCl emissions to air and AOX emissions to water – emissions
which are not assessed in this work. An other emission, which is not inventoried due to poor data, is
carbondisulfide used in alternative 5 in considerable amounts.

in ecoinvent
Modul name
mean value
Mean value
coefficient
Source for
Allocation
Comment
Deviation
structure
Standard
Category
category
Location
Remarks
General
Process
Output
Name
Infra-
Input
Type
Sub-
Unit
51 kg overburden and 110 kg crude ore per kg Class
I Nickel mined (resulting in 97.2 kg tailings and Nickel, 1.13% in sulfide, Ni
Ni in ground  100% to Ni; mass 12.8 kg concentrate). 82% yield of resource in ground 0.76% and Cu 0.76% in crude 1.26E+00 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,12)
mining/beneficiation, 97% yield of metallurgy --> ore, in ground
nickel production, sulphidic ore, primary, part I

79.5% overall yield.
Cu in ground  100% to Cu; mass resource in ground 0.76% and Ni 0.76% in crude 6.48E-01 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,12)
ore, in ground
to Cu; construction
explosive  27% civil engineering No RER blasting 1.66E-01 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,4)
revenue processes
to Cu; construction diesel, burned in building
diesel fuel (energy)  27% machinery No GLO 1.12E+01 MJ (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,4)
revenue processes machine
electricity from grid, to Cu; electricity, medium voltage,
27% Assumption: 40% electricity from grid electricity production mix No UCTE 1.74E+00 kWh (Hilbrans, 1999) 1 1.14 (2,3,2,1,1,4,2)
mining revenue production UCTE, at grid
to Cu; electricity, hydropower, at run
hydropower, mining  27% Assumption: 60% hydropower hydro power power plants No RER 2.62E+00 kWh (Hilbrans, 1999) 1 1.14 (2,3,2,1,1,4,2)
revenue of-river power plant
fresh and unspecified - because of the relative abundance of
to Cu;
processwater  27% pumped ground water in the mining process, resource in water Water, river 3.82E-02 m3 (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,12)
revenue
(mining) surface water is chosen here.
stowing material, to Cu; construction
 27% additional material, no excavation included additives No CH sand, at mine 4.56E+01 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,4)
sand revenue materials
mining
to Cu; construction portland calcareous cement,

cement  27% binder No CH 3.62E+00 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,4)
revenue materials at plant
to Cu;
pit water  27% resource in water Water, well, in ground 6.79E-02 m3 (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,12)
revenue
calculations
transports, lorry, to Cu; transportation of cement, explosives and sand; transport transport, lorry >16t, fleet
 27% road No RER 2.47E+00 tkm based on 1 2.11 (3,2,2,1,3,5,5)
mining revenue 50km assumed. systems average
(Hilbrans, 1999)
exploration
to Cu; construction
infrastructure,  27% exploration infrastructure, 2000 m belts assumed machinery Yes RER conveyor belt, at plant 4.20E-06 m own assumptions 1 1.26 (3,2,2,1,3,4,4)
revenue processes
conveyor belts
infrastructure of
to Cu; non-ferrous metal mine,
underground mining  27% metals extraction Yes GLO 5.50E-09 unit own assumptions 1 3.07 (3,2,2,1,3,4,9)
revenue underground
operation
overburden, to Cu; waste residual disposal, sulfidic tailings, off-
 27% No GLO 3.49E+01 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,6)
disposed revenue management material landfill site
to Cu;
 overburden, refilled 27% not inventoried, no ecological burden 1.58E+01 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,6)
revenue
to Cu;
 off-gas from blasting 27% included in module blasting (2941)
revenue
off-gas from to Cu; included in module diesel, burned in building
 27%
transports revenue machine (1350)
to Cu; Detailed composistion see Tab. 5.4. The emissions
 effluents, mining 27% 6.79E-02 m3 (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,6)
revenue are inventoried.
Fig. 5.3 Flows for the multi-output process “Nickel production” (part mining) and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg Class
I nickel. Allocation values indicate the amount of the flow that is allocated to the co-product copper.

in ecoinvent
Modul name
mean value
Mean value
coefficient
Source for
Allocation
Comment
Deviation
structure
Category
Standard
category
Location
Remarks
General
Process
Output
Name
Infra-
Input
Type
Sub-
Unit
electricity, to Cu;
 27% included in mining
benefication revenue
nickel production, sulphidic ore, primary,

fresh and
to Cu;
processwater  27% resource in water Water, well, in ground 1.52E-01 m3 (Hilbrans, 1999) 1 1.13 (2,2,1,1,1,4,12)
revenue
(benefication)
to Cu; construction limestone, milled, packed, at
lime  27% others No CH 4.96E-02 kg (Krauss, 1999) 1 2.00 reported values
revenue materials plant
transports, lorry, to Cu; transportation of lime, floating agents and sodium transport transport, lorry >16t, fleet (3,2,1,5,3,2,5);
 27% road No RER 8.15E-03 tkm calculation 1 2.07
benefication revenue cyanide; 50 km assumed. systems average calculation
infrastructure of to Cu;
 27% AlOH-production plant chosen as proxy metals extraction Yes RER aluminium hydroxide, plant 9.24E-10 unit own assumptions 1 3.07 (3,2,2,1,3,4,9)
benefication revenue
floating agents, to Cu;
 27% chemicals organics No GLO chemicals organic, at plant 2.48E-02 kg (IPPC, 2002) 1 2.00 reported values
organic revenue
floating agents, to Cu;
 27% chemicals inorganics No GLO chemicals inorganic, at plant 8.47E-02 kg (IPPC, 2002) 1 2.00 reported values
beneficiation
inorganic revenue
to Cu;
Sodium cyanide  27% assumed to be produced on site chemicals organics No RER hydrogen cyanide, at plant 3.85E-03 kg (IPPC, 2002) 1 2.00 reported values
part II
revenue
to Cu;
Sulphuric Acid  27% burden-free, since byproduct from metallurgy 2.54E-01 kg (IPPC, 2002) 1 2.00 reported values
revenue
to Cu; Tailings that are disposed of in ponds or other waste residual disposal, sulfidic tailings, off- (2,2,2,3,1,3,6);
 tailings, disposed 27% No GLO 6.50E+01 kg (Hilbrans, 1999) 1 1.10
revenue facilities. management material landfill site reported values
to Cu;
 tailings, refilled 27% not inventoried, no ecological burden 3.33E+01 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,6)
revenue
effluents, to Cu; composition in Tab. 5.4. The single emissions are (2,2,2,3,1,3,6);
 27% 1.52E-01 m3 (Hilbrans, 1999) 1 1.10
benefication revenue inventoried with a variability of 10 reported values
(IPPC, 2002),
to Cu; From sodium cyanide (35g NaCN per t crude ore,
 Emission Cyanide 27% water river Cyanide 4.12E-04 kg (Anonymous 1 1.84 (4,2,2,3,4,4,33)
revenue 80% recovered --> 3.75 g CN- per t crude ore)
1998)
emission to Cu; 0.53 kg carbondisulfide out of 1 kg xanthate low population
 27% air Carbon disulfide 1.10E-02 kg (NPI, 2001) 1 2.31 (4,2,2,5,4,4,23)
carbondisulfide revenue through hydrolytical decomposition density
to Cu; the sum of particulate emissions represented in
 Dust 27% 4.05E-02 kg (NPI, 2001) 1 1.59 (3,2,1,1,3,4,25)
revenue fractions PM<2.5, PM2.5-10, PM>10, including
metals and salts. (Tab. 5.8)
Fig. 5.4 Flows for the multi-output process “Nickel production” (part beneficiation) and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg
Class I nickel. Allocation values indicate the amount of the flow that is allocated to the co-product copper.

in ecoinvent
Modul name
mean value
Mean value
coefficient
Source for
Allocation
Comment
Deviation
structure
Category
Standard
category
Location
Remarks
General
Process
Output
Name
Infra-
Input
Type
Sub-
Unit

nickel production, sulphidic

limestone 17% flux, 50% limestone assumed others No CH 5.40E-01 kg (Hilbrans 1999) 1 1.51 (5,2,2,3,1,3,4)
to Cu; construction
silica sand  17% flux, 50% quartzsand assumed additives No DE silica sand, at plant 5.40E-01 kg (Hilbrans 1999) 1 1.51 (5,2,2,3,1,3,4)
revenue materials
drying
ore, primary, part III

electricity drying, to Cu; electricity, high voltage,
 11% 40% electricity from grid electricity production mix No UCTE 2.83E-02 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
grid revenue production UCTE, at grid
hydropower drying  11% 60% electricity from hydropower hydro power power plants No RER 4.24E-02 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
fossile energy drying to Cu; natural gas, burned in
 11% natural gas heating systems No RER 2.44E+00 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
(gas) revenue industrial furnace >100kW
limestone  13% flux, 50% limestone assumed others No CH 1.66E+00 kg (Hilbrans 1999) 1 1.51 (5,2,2,3,1,3,4)
to Cu; construction
silica sand  13% flux, 50% quartzsand assumed additives No DE silica sand, at plant 1.66E+00 kg (Hilbrans 1999) 1 1.51 (5,2,2,3,1,3,4)
revenue materials
roasting
electricity roasting, to Cu; electricity, high voltage,

hydropower roasting  11% 60% electricity from hydropower hydro power power plants No RER 1.30E-01 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
to Cu;
fossile energy (oil)  20% oil heating systems No RER industrial furnace 1MW, non- 1.80E+00 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
revenue
modulating
to Cu;
 steam, credit 4% not inventoried, no credits assumed -7.81E-01 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,4)
revenue
Fig. 5.5 Flows for the multi-output process “Nickel production” (part drying/roasting) and its representation in the ecoinvent database. Values correspond to the functional unit of 1
kg Class I nickel. Allocation values indicate the amount of the flow that is allocated to the co-product copper.

in ecoinvent
Modul name
mean value
Mean value
coefficient
Source for
Allocation
Comment
Deviation
structure
Category
Standard
category
Location
Remarks
General
Process
Output
Name
Infra-
Input
Type
Sub-
Unit
electricity melting, to Cu; electricity, high voltage,

 7% 40% electricity from grid electricity production mix No UCTE 1.16E+00 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
hydropower melting  7% 60% electricity from hydropower hydro power power plants No RER 1.74E+00 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
to Cu; natural gas, burned in
fossile energy (gas)  9% natural gas heating systems No RER 2.80E+00 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
revenue industrial furnace >100kW
to Cu; hard coal, burned in industrial
fossile energy (coke)  15% hard coal heating systems No RER 2.20E+00 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
revenue furnace 1-10MW
fossile energy to Cu;
smelting
 20% oil heating systems No RER industrial furnace 1MW, non- 2.48E+01 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
melting (oil) revenue
modulating
infrastructure to Cu;
 27% including infrastructure for separation of copper metals extraction Yes GLO non-ferrous metal smelter 4.61E-11 unit assumption 1 3.30 (4,2,1,5,4,4,9)
smeltery revenue
to Cu;
 steam, credit 15% not inventoried, no credits assumed -1.31E+01 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,4)
revenue
part IV

to Cu; waste residual
 slag 13% disposal No CH 0% water, to residual material 1.05E+01 kg (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,6)
revenue management material landfill
landfill
 dust inventoried as overall emissions
sulphur dioxide in to Cu; low population

 10% comprise emissions from drying an roasting air Sulfur dioxide 1.50E+00 kg (Inco 2002) 1 1.64 (5,4,1,3,3,5,15)
off-gas revenue density
to Cu; electricity, high voltage,
electricity, grid  15% 40% electricity from grid electricity production mix No UCTE 5.66E-01 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
revenue production UCTE, at grid
converting
hydropower  15% 60% electricity from hydropower hydro power power plants No RER 8.49E-01 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
to Cu; natural gas, burned in
fossile energy (gas)  12% energy carrier assumed to be natural gas natural gas heating systems No RER 3.05E+00 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
revenue industrial furnace >100kW
to Cu;
 steam, credit 13% not inventoried, no credits assumed -2.86E+00 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,4)
revenue
to Cu; residues in wet scrubber, approximated with slag waste residual
 airborn dust 12% No CH 0% water, to residual material 3.97E-01 kg (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,6)
revenue disposal - module management material landfill
landfill
Fig. 5.6 Flows for the multi-output process “Nickel production” (part smelting/converting) and its representation in the ecoinvent database. Values correspond to the functional unit of
1 kg Class I nickel. Allocation values indicate the amount of the flow that is allocated to the co-product copper.

in ecoinvent
Modul name
mean value
Mean value
coefficient
Source for
Allocation
Comment
Deviation
structure
Category
Standard
category
Location
Remarks
General
Process
Output
Name
Infra-
Input
Type
Sub-
Unit
sulf. Acid
nickel production, sulphidic ore, primary, part V

hydropower  12% 60% electricity from hydropower hydro power power plants No RER 7.49E-01 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
to Cu; not inventoried, assumed to be burden free (cut
 sulfuric acid 11% 1.13E+01 kg (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,4)
revenue off)
other

hydropower  17% 60% electricity from hydropower hydro power power plants No RER 1.11E+00 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
to Cu; low population
 waste heat 0.03% from total electricity air Heat, waste 5.27E+01 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
revenue density
electricity, grid to Cu; electricity, high voltage,
(copper removal) revenue production UCTE, at grid
leaching, copper separation
hydropower (copper to Cu; electricity, hydropower, at run

 4% 60% electricity from hydropower hydro power power plants No RER 4.37E-01 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
removal) revenue of-river power plant
fossile energy to Cu; natural gas, burned in
 2% energy carrier assumed to be natural gas natural gas heating systems No RER 1.76E+00 MJ (Hilbrans 1999) 1 1.25 (3,2,2,3,3,3,1)
(copper removal) revenue industrial furnace >100kW
to Cu; ammonia, liquid, at regional
leaching agents  0% approximated with ammonia inventory chemicals inorganics No RER 8.37E-02 kg (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,4)
revenue storehouse
to Cu;
ammoniac  0% no data available n.a
revenue
triisooctylamine to Cu;
 8% no data available n.a
(TIOA) revenue
to Cu; by-product from alternative (2), considered as
 ammonium sulfat 0% 2.99E-01 kg (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,6)
revenue burden free
to Cu;
 precipitates 0% internal recovery, not inventoried 1.67E-01 kg (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,6)
revenue
to Cu;
 dross (dry) 0% internal recovery, not inventoried 3.11E-02 kg (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,6)
revenue
copper as co- copper, primary, couple
 100% to Cu; mass co-product, allocation by revenue metals extraction No GLO 5.15E-01 kg (Hilbrans 1999)
product production nickel
1.4m3 per t concentrate worked as reported from
to Cu;
 effluent, general 9% PGM operations. Single emissions inventoried 1.26E-02 m3 (Hilbrans 1999) 1 2.15 (4,5,1,3,5,5,6)
revenue
overall
according to composition in Tab. 5.10.

Dust emission for to Cu; Single emissions inventoried according to
 9% 4.07E-04 kg (IPPC, 2002) 1 1.59 (3,2,1,1,3,4,25)
overall process revenue composition in Tab 5.9.
(4,2,1,1,4,4,14);
carbondioxide, from to Cu; lime addition from benefication, roasting and
 14% air unspecified Carbon dioxide, fossil 9.92E-01 kg (IPPC, 2002) 1 1.58 excluding stdev
lime addition revenue drying
of lime addition
Fig. 5.7 Flows for the multi-output process “Nickel production” (part sulphuric acid / copper separation) and its representation in the ecoinvent database. Values correspond to the
functional unit of 1 kg Class I nickel. Allocation values indicate the amount of the flow that is allocated to the co-product copper.

in ecoinvent
Modul name
mean value
Mean value
coefficient
Source for
Allocation
Comment
Deviation
structure
Standard
Category
category
Location
Remarks
General
Process
Output
Name
Infra-
Input
Type
Sub-
Unit
reduction

electricity, grid  0% mass 40% electricity from grid electricity production mix No UCTE 1.60E-02 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
4/5
 steam, credit 0% mass -2.40E-01 MJ (Hilbrans 1999)
electricity, hydropower, at run

hydropower  0% mass 60% electricity from hydropower hydro power power plants No RER 2.40E-02 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
fossile energy  0% mass natural gas heating systems No RER 3.87E+00 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
electrolysis 1/3/4

part VI
fossile energy (oil)  0% mass natural gas heating systems No RER 9.93E-02 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
chlorine  0% mass not available
chlorine emission
 0% mass not available
(HCl)
waste residual
 anode slime 0% mass approximated with slag disposal - module No CH 0% water, to residual material 7.48E-02 kg (Hilbrans 1999) 1 2.15 (4,5,1,3,5,5,6)
landfill
 spoiled anodes 0% mass internal use, not inventoried 1.82E-01 kg (Hilbrans 1999) 1
purification of

leachate (4)

fossile energy (oil)  0% mass oil heating systems No RER industrial furnace 1MW, non- 1.07E-04 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
modulating
 dross from leaching 0% mass internal recovery, not inventoried 1.07E-03 kg (Hilbrans 1999) 1 (2,2,2,3,1,3,6)
Fig. 5.8 Flows for the multi-output process “Nickel production” (part nickel reduction / purification step) and its representation in the ecoinvent database. Values correspond to the
functional unit of 1 kg Class I nickel. Allocation values indicate the amount of the flow that is allocated to the co-product copper.

in ecoinvent
Modul name
mean value
Mean value
coefficient
Source for
Allocation
Comment
Deviation
structure
Standard
Category
category
Location
Remarks
General
Process
Output
Name
Infra-
Input
Type
Sub-
Unit
reduction (2)
sulphidic ore, primary,

nickel production,
fossile energy (gas)  0% mass natural gas heating systems No RER 1.37E+00 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
part VII
hydrogen  0% mass chemicals inorganics No RER hydrogen, liquid, at plant 4.56E-03 kg (Hilbrans 1999) 1 1.51 (5,2,2,3,1,3,4)

carbonyl (5)
fossile energy (oil)  0% mass oil heating systems No RER industrial furnace 1MW, non- 1.21E+00 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
modulating
dross waste residual
 0% mass approximated with slag disposal - module No CH 0% water, to residual material 6.91E-06 kg (Hilbrans 1999) 1 2.15 (4,5,1,3,5,5,6)
carbonyleprocess management material landfill
landfill
 nickel, class I 0% metals extraction No GLO nickel, 99.5%, at plant 1.00E+00 kg (Hilbrans 1999)
Fig. 5.9 Flows for the multi-output process “Nickel production” (part carbonyl and refining) and its representation in the ecoinvent database. Values correspond to the functional unit
of 1 kg Class I nickel. Allocation values indicate the amount of the flow that is allocated to the co-product copper.

5.2 Ferronickel
This part is mainly based on Hilbrans & Hinrichs (1999). General process description was taken from
Kerfoot (1997).

Mining. Nickel oxide ores occur in two types: a limonitic type which occurs in the upper zone of a
lateritic deposit, in which ferric oxide materials are predominant, and a silicate, occurring at greater
depth in the deposit, which has lower iron content, but is usually richer in nickel than the limonite, and
contains high levels of magnesia and silica. Most deposits contain both types of ore. Usually it is pos-
sible to separate the ore by selective mining and screening into an iron-containing limonite fraction
and a magnesium silicate fraction enriched in nickel.
Beneficiation. Two processes – the Nippon Yakin Oheyama and the MINPRO–PAMCO process –
have been developed in which the nickel is first converted by high-temperature treatment to a metallic
phase, which can then be separated from the bulk of the ore by standard mineral beneficiation tech-
niques.
Metallurgy. The metallurgy of nickel oxide ores differs from that of sulphides in that oxide ores are
not amenable to most of the standard mineral beneficiation methods due to the chemical dissemination
of the nickel in the oxide minerals. Nickel laterite ores have high moisture contents (typically up to 45
%) as well as chemically bound water in the hydroxide form. For this, the ore has to be dried, usually
direct-fired rotary units operating at about 250 °C, to 15 – 20 % moisture. Calcining to dehydrate the
ore and prereduction prior to electric furnace smelting are generally carried out. Chemically bound
water is released above ca. 400 °C, reduction of the oxides to metal starts at 500 – 600 °C. Dehydra-
tion and pre-reduction of the ore in the kiln prior to smelting, optimises the utilisation of energy avail-
able from the reductants and fuel, thus reducing energy consumption in the smelting operation.
Class II nickel products such as nickel oxide, metallised nickel oxide, and ferronickel, are produced
directly by smelting and roasting. They are sufficiently pure without refining for many nickel applica-
tions such as stainless steel production. The rotary kiln – electric furnace smelting process is now used
almost universally for the production of ferronickel from oxide ores. Also ores with a low Ni/Fe ratio
can be smelted to yield an acceptable ferronickel grade.
Ferronickel refining. Crude ferronickel produced by the conventional rotary kiln – electric furnace
process usually contains high levels of carbon and sulphur. These ferronickels are first treated under
reducing conditions to remove sulphur and are then refined sequentially under oxidising conditions
with suitable fluxes to remove carbon, silicon, and phosphorus. A variety of equipment is used for the
refining of ferronickel, including electric arc furnaces, shaking ladles, and low-frequency induction
furnaces for desulphurisation, and oxygen-blown converters for silicon, carbon, and phosphorus re-
moval. Sulphur is generally removed under reducing conditions by adding soda ash, lime, or calcium
carbide to the molten ferronickel. Good agitation of the melt is essential to ensure effective mixing of
the reagent with the metal phase
5.2.2 Emission
Mining
The major emissions are due to mineral born pollutants in the effluents. Mining operations generate
large amounts of dust. Rain percolate through overburden and accounts to metal emissions to ground-
water.

Beneficiation
Beneficiation is done by classification, no emissions arise from this step.
Carbon dioxide. In the ore pre-treatment, considerable amounts of lime are added. Lime decomposes
into CO2 to form calcite.
the air consist of metals and thus are often returned to the leaching process after treatment. Dust con-
sists to a considerable part of heavy metals.
Other emissions to air. The most important pollutants from the production of ferro-alloys in general
besides dust are SO2, NOx, CO, CO2, HF, PAH and VOC. Only few SO2 coming from metallurgical
coke is formed.
emitted to water. The metals concerned are Zn, Fe, Co, Pb and Hg. Other significant substances are
chlorides and sulphates. Wastewaters from wet gas cleaning (if used) of the different metallurgical
stages are the most important sources. The leaching stages are usually operated on closed circuit and
5.2.3 Waste
Overburden is deposed close inside the mine panel. Since no further beneficiation takes place in the
ferronickel production, no beneficiation tailings arise.
The production of metals is related to the generation of several residues and wastes, which are listed in
the European Waste Catalogue. The most specific residues are filter dusts, sludge from wet scrubbers,
slag from the smelting process, used furnace linings and packaging material like drums or big-gabs.
These residues are partly sold as by-products, recycled to the process or in cases of wastes without
economic utility transported to a deposit or a landfill.
Some of the process specific residues can be reused or recovered in preliminary process steps (e. g.
dross, filter dust) or construction (e. g. cleaned slag). Residues also arise from the treatment of liquid
effluents, the main residue being gypsum waste and metal hydroxides from the wastewater neutralisa-
tion plant. These residuals have to be disposed, usually in lined ponds.
5.2.4 Ferronickel in ecoinvent

The module describes the global ferronickel production in 1994. It is designed for the use of the alloy
as raw material in the manufacturing of stainless steels and alloys. The layout of the modules with
general flow information, remarks, sources, values and uncertainty information are shown in Fig. 5.10,
Fig. 5.11 and Fig. 5.12. The following paragraphs describe calculations, sources and assumptions cho-
sen in this study. The functional unit of this process is one kg of ferronickel. A general description of
the module and a summary of its meta information can be found in Tab. 5.1.
The mobile equipment used for exploitation like drilling machines, scoop trams and underground
trucks is inventoried with the diesel consumption stated in Hilbrans & Hinrichs (1999). The used
module “diesel, burned in building machine” includes the infrastructure of the machines. 10’000m
conveyor belts are further assumed with a capacity of 2000 t / h which yields 2E-5 m conveyor belt
required per tonne mined raw ore.

Transformation and occupation of land through direct exploitation are assumed with the standard
value for non-ferrous metals without heap leaching processes of 0.0025 m2/t and 0.075 m2a/t ore
mined, that is 0.055 m2/t and 1.65 m2a per tonne Nickel produced.
Mining facilities and smeltery are inventoried with respective modules according to chapter II “Non
Ferrous Metal Winning – Auxiliary Processes”.
Mining.
In the considered processes the ore is mined in open cuts. The overburden is disposed in piles inside
the mining panel, therefore no disposal of overburden is assumed (Hilbrans & Hinrichs (1999)). Also
water use and effluent from mining is documented in Hilbrans & Hinrichs (1999). The use of freshwa-
ter was inventoried as river water, where as the pit water was assumed to be groundwater, which has to
be pumped. The composition of the effluent was taken from IPPC (2002), its values are shown in Tab.
5.4. Concerning electricity mix Hilbrans & Hinrichs (1999) assumes a share of 60% of hydropower
due to the operations high and constant energy demand. This assumption is used also in this inventory.
For laterites no further beneficiation is necessary and possible except classification. Therefore no tail-
ings arise.
Metallurgy
Energy. In Hilbrans & Hinrichs (1999) the energy demand for drying and roasting of the wet ore is
stated, which contributes to a major extent to the overall energy demand in the ferronickel production
(Fig. 5.11).
Flux. The input of limestone is estimated roughly with values reported from the related manufacturing
processes for ferro-chrome and calcium silicate in IPPC (2001). Choosing a geometric mean of the
values a value of 470 kg limestone per tonne ferro-alloy is required. Consequentially 0.44 kg CO2 per
tonne flux is emitted after the formula CaCO3  CaO + CO2.
Dust. The emission of dust is estimated roughly considering values from ferro-chrome production.
There overall values ranging from 0.5 to 1 kg per tonne ferro-alloy is reported in IPPC (2001). In this
study a value of 1 kg per tonne is chosen using a PM-partition reported in CEPMEIP 10 . Dust composi-
tion is approximated with the specific values from class I nickel production in this section, resulting in
emission factors in Tab. 5.11.
Effluent. Specific value of 6.9 m3 per tonne ferro-alloy chosen from IPPC (2001).
Air emissions. No VOC-emissions were assumed since no hydrometallurgical processes are consid-
ered. For TCDD the same values as in class I nickel metallurgy are chosen (IPPC (2001)).
Data quality
mated due to different system boundaries of reported effluent volume from Hilbrans & Hinrichs
(1999) and total effluent emissions reported in IPPC (2002).
Metallurgy. Process data is satisfactory. Poor data are available on composition and fate of solid
waste, which has to be seen as major impact within the overall process due to its big mass.
10

Tab. 5.11 Emissions to air per tonne produced ferronickel. Data was taken as proxy from similar processes in ferro-
chrome (A) and copper metallurgy (B) reported in IPPC (2001), the distribution of the particle size is taken
11
from CEPMEIP (C).

Pb 0.077 B
Zn 0.25 B
Sn 0.016 B
Cu 0.10 B
As 0.014 B
Total dust 1.0 A
PM<2.5 0.0013 C
PM2.5-10 0.75 C
PM>10 0.25 C
Ni 0.03 A
Co 0.03 A
TCDD 1.00E-08 B
11

in ecoinvent
Modul name
mean value
Mean value
coefficient
Source for
Allocation
Comment
Deviation
structure
Standard
Category
category
Location
Remarks
General
Process
Output
Name
Infra-
Input
Type
Sub-
Unit
Overall 57.5% yield assumed. 18 kg overburden and Nickel, 1.98% in silicates,
Ni in ground  resource in ground 1.74E+00 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,12)
22 kg ore required for 1 kg ferronickel 1.04% in crude ore, in ground
ferronickel, 25% Ni, at plant, part I

construction
explosive  civil engineering No RER blasting 1.20E-03 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,4)
processes
diesel fuel (energy)  machinery No GLO 1.91E+00 MJ (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,4)
processes machine
electricity from grid, electricity, high voltage,
 40% electricity from grid electricity production mix No UCTE 9.29E-02 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
hydropower, mining  60% electricity from hydropower hydro power power plants No RER 1.39E-01 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
exploration
construction
infrastructure,  exploration infrastructure, 10'000 m belts assumed machinery Yes RER conveyor belt, at plant 8.00E-07 m own assumptions 1 1.26 (3,2,2,1,3,4,4)
processes
conveyor belts
mining
non-ferrous metal mine,

infrastructure mining  metals extraction Yes GLO 2.00E-09 unit own assumptions 1 3.07 (3,2,2,1,3,4,9)
surface
Land trans-
 0.0025 m2/t ore mined resource land Transformation, from unknown 5.49E-05 m2 (IPPC, 2002) 1 3.07 (3,2,2,1,3,4,9)
formation, ground
Land trans-ormation, Transformation, to mineral
 0.0025 m2/t ore mined resource land 5.49E-05 m2 (IPPC, 2002) 1 2.08 (3,2,2,1,3,4,8)
Occupation, mineral
Land use for mining  average lifetime of 30 a assumed resource land 1.65E-03 m2a (IPPC, 2002) 1 1.59 (3,2,2,1,3,4,7)
extraction site
overburden,
 no overburden disposed, but refilled in mining pit 0.00E+00 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,6)
disposed
 overburden, refilled not inventoried, no ecological burden 1.80E+01 kg (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,6)
Detailed composistion see Tab. 5.4, same
 effluents, mining 8.00E-03 m3 (Hilbrans, 1999) 1 1.13 (2,2,2,1,1,4,6)
composition as underground mining effluents
the sum of particulate emissions represented in
 Dust 1.01E-02 kg (NPI, 2001) 1 1.59 (3,2,1,1,3,4,25)
fractions PM<2.5, PM2.5-10, PM>10, including
metals and salts.
Fig. 5.10 Flows for “Ferronickel, part mining” and its representation in the ecoinvent database. The values correspond to the functional unit of the production of 1 kg ferronickel

in ecoinvent
Modul name
mean value
Mean value
coefficient
Source for
Allocation
Comment
Deviation
structure
Standard
Category
category
Location
Remarks
General
Process
Output
Name
Infra-
Input
Type
Sub-
Unit
electricity drying, electricity, high voltage,
grid production UCTE, at grid
drying

hydropower drying  60% electricity from hydropower hydro power power plants No RER 3.90E-01 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
fossile energy drying natural gas, burned in
 natural gas heating systems No RER 2.88E+01 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
ferronickel, 25% Ni, at plant, part II

(gas) industrial furnace >100kW
electricity roasting, electricity, high voltage,
roasting

hydropower roasting  60% electricity from hydropower hydro power power plants No RER 4.80E-01 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
fossile energy hard coal, burned in industrial
 hard coal heating systems No RER 4.00E+01 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
roasting (coke) furnace 1-10MW
electricity melting, electricity, high voltage,
 40% electricity from grid electricity production mix No UCTE 2.72E+00 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
hydropower melting  60% electricity from hydropower hydro power power plants No RER 4.08E+00 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
fossile energy
 oil heating systems No RER industrial furnace 1MW, non- 3.75E+00 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
melting (oil)
modulating
flux, ferronickel  flux, limestone assumed others No CH 4.69E-01 kg (IPPC 2001) 1 1.51 (5,2,2,3,1,3,4)
materials plant
infrastructure
 metals extraction Yes GLO non-ferrous metal smelter 6.48E-11 unit own assumptions 1 3.07 (3,2,2,1,3,4,9)
smeltery
melting

waste residual
 slag approximated with slag disposal - module No CH 0% water, to residual material 1.24E+01 kg (Hilbrans 1999) 1 2.15 (4,5,1,3,5,5,6)
landfill
in scrubber, approximated with slag waste residual
 dust No CH 0% water, to residual material 3.00E-01 kg (Hilbrans 1999) 1 2.15 (4,5,1,3,5,5,6)
disposal - module management material landfill
landfill
(4,2,1,1,4,4,14);
carbondioxide, from calculations,
 lime addition as melting flux air unspecified Carbon dioxide, fossil 2.06E-01 kg 1 1.58 excluding stdev
lime addition (IPPC, 2002)
of lime addition
Composition as in Class I Nickel
 effluent, general production (Tab 5.10). The emissions are 6.90E-03 m3 (IPPC 2001) 1 2.15 (4,5,1,3,5,5,6)
inventoried.
dust, from overall Composition in Tab 5.11. The emissions
 1.00E-03 kg (IPPC, 2002) 1 1.59 (3,2,1,1,3,4,25)
process are inventoried.
Fig. 5.11 Flows for “Ferronickel, part drying, roasting and melting” and its representation in the ecoinvent database. The values correspond to the functional unit of the production of
1 kg ferronickel

in ecoinvent
Modul name
mean value
Mean value
coefficient
Source for
Allocation
Comment
Deviation
structure
Category
Standard
category
Location
Remarks
General
Process
Output
Name
Infra-
Input
Type
Sub-
Unit
low population

other/transp
 waste heat from total electricity air Heat, waste 3.34E+01 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
density
electricity other, electricity, high voltage,
plant, part III


hydropower other  60% electricity from hydropower hydro power power plants No RER 1.80E-01 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)

electricity, grid  40% electricity from grid electricity production mix No UCTE 2.00E-01 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
refining

hydropower  60% electricity from hydropower hydro power power plants No RER 3.00E-01 kWh (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,2)
fossile energy (gas)  natural gas heating systems No RER 1.00E+00 MJ (Hilbrans 1999) 1 1.10 (2,2,2,3,1,3,1)
 Ferronickel metals extraction No GLO ferronickel, 25% Ni, at plant 1.00E+00 kg (Hilbrans 1999)
Fig. 5.12 Flows for “Ferronickel, parts others/transports and refining” and its representation in the ecoinvent database. The values correspond to the functional unit of the production
of 1 kg ferronickel

6. Literature
6 Literature
Althaus et al. (2004) Althaus H.-J., Chudacoff M., Hischier R., Jungbluth N., Osses M. and Primas A.
(2004) Life Cycle Inventories of Chemicals. Final report ecoinvent 2000 No. 8.
EMPA Dübendorf, Swiss Centre for Life Cycle Inventories, Dübendorf, CH,
Online-Version under: www.ecoinvent.ch.
ment in Mining. Environment Australia, Online-Version under:
http://www.ea.gov.au/industry/sustainable/mining/booklets/dust/.
European Commission. Retrieved from
http://www.jrc.es/pub/english.cgi/0/733169
Kerfoot (1997) Kerfoot D. G. E. (1997) Nickel. In: Ullmann's encyclopedia of industrial chem-
istry (ed. Anonymous). 5th edition on CD-ROM Edition. Wiley & Sons, Lon-
don.
NIDIS (2003) NIDIS (2003) The Nickel Page. Nickel Producers Environmental Research As-
sociation. Retrieved at 01.03.2003 from http://www.nipera.org
U.S. Geological Survey. Retrieved at 01.03.2003 from
http://minerals.usgs.gov/minerals/pubs/mcs/. Date of last revision: 26-Feb-
2003@09:36.

6. Literature

Part V
Platinum Group Metals (PGM)

Platinum, Palladium, Rhodium
Data v2.1 (2009)

Final report of updated Swiss National Life Cycle Inventory Database "ecoinvent", version 2.1.
Summary

ported in Frischknecht et al. (2006). The changes were implemented with v1.3 affect the co-product dataset of
nickel due to an incorrect nickel price underlying the allocation factors.
The correction caused an increase of attributed exchanges for nickel and a decrease for the others, especially in
the case of PGM-production in Russia, since there the co-product nickel has a considerable share in the output.
No changes were made in version 2.1 compared to version 2.0.
Summary
The primary production of platinum group metals (PGM) is modelled for South Africa and Russia. Together
these two countries produce more than 90% of the world’s PGM.
For the secondary production of PGM the recycling process of automotive catalysts is modelled.
The consumer mixes for PGM in Europe are finally modelled from the outputs of these three modules.
steps.

Table of Contents
Table of Contents
SUMMARY ......................................................................................................................I
1 INTRODUCTION ....................................................................................................... 1
2 RESERVES AND RESOURCES OF PGM...................................................................... 2

2.1 Primary PGM ........................................................................................................................ 2
2.2 Secondary PGM .................................................................................................................... 5
3 CHARACTERISATION OF THE PGMS ......................................................................... 6

3.1 Platinum ................................................................................................................................ 6
3.2 Palladium............................................................................................................................... 6
3.3 Rhodium ................................................................................................................................ 6
4 USE / APPLICATION OF PGM ................................................................................... 8

5.1 Allocation ............................................................................................................................ 11
5.1.1 Primary Production ............................................................................................................. 11
5.1.2 Secondary Production ......................................................................................................... 12
5.2 PGM consumer mix............................................................................................................. 15
5.2.1 PGM, at regional storage, in Ecoinvent .............................................................................. 16
5.3 Winning of Primary PGM ................................................................................................... 21
5.3.2 Emissions ............................................................................................................................ 23
5.3.3 Waste................................................................................................................................... 24
5.3.4 Primary PGM in ecoinvent.................................................................................................. 24
5.4 Winning of Secondary PGM ............................................................................................... 41
5.4.2 Emissions and Waste........................................................................................................... 41
5.4.3 Secondary PGM in ecoinvent.............................................................................................. 42
6 LITERATURE ......................................................................................................... 45

1 Introduction
1 Introduction
This chapter describes the production of the platinum-group metals Platinum (Pt); Palladium (Pd) and
Rhodium (Rh) from cradle to consumer in Europe. Inventory data is taken mainly from (Hochfeld
(1997)) with background data from (IPPC (2002)) and (IPPC (2001)). Market data on PGM and non
ferrous metals are provided by Johnson Matthey 1 and the London Metal Exchange respectively 2 .
Since the PGM are associated with other metals, their production always yields a fraction of other
metals than PGM, mainly Nickel (Ni) and Copper (Cu). Hence the process of primary production de-
scribed in this chapter is formally a multioutput-process with the coupled products Pt, Pd, Rh, Ni and
Cu. This makes it necessary to allocate the inventory to these products. The demand of PGM-
containing ore is allocated by mass. The rest of the inventory is allocated by revenue.
PGM are highly valuable, their occurrence is scarce. PGM-containing ores are mined only in few sites.
The deposits in Noril’sk (RUS) and Bushveld (RSA) contribute over 90% to the global PGM-
production. The data on these two PGM winning operations therefore represents the global primary
production of PGM according to their market share.
The great value of the PGM makes recycling profitable to some extent. This secondary production is
approximated with the declared amount of metal from the recycling of automotive catalysts. For
Europe this assumption is reasonable, since automotive catalysts account for over 90% of the metal
use.
The market for PGM is of great importance. On the one hand, market prices determine the profitably
of the PGM winning, both for primary and secondary production. On the other hand the inventory is
allocated by revenue to the different metals. An ecological assessment of PGM therefore is always de-
pendent on the actual market situation. Further more, the producing countries hold stocks of PGM in
order to influence market price and to protect their own PGM-demanding industries. As precious
metal, Platinum in particular is dealt as investment.
1
Johnson Matthey’s Website “Platinum Today”, http://www.platinum.matthey.com, accessed at 01.03.2003
2
London Metal Exchange, daily and monthly stocks and prices under http://www.lme.co.uk/data_prices/ home.html, accessed
at 01.03.2003
2 Reserves and Resources of PGM

cle. When recovered the properties are theoretically identical. The metal can be re-used in another
product life cycle, provided that its concentration makes a recovery profitable. The recovered metal is
referred to as secondary metal, whereas the metal originating directly from the ore is referred to as
primary metal. In LCIs these two types – primary and secondary metal – are often differentiated be-
cause they bear a different ecological burden.
2.1 Primary PGM

The PGMs in primary deposits have been transferred from the earth's interior. Tectonic movements of
the earth's crust, followed by the eruption of magma, have led to their presence in regions close to the
surface. Almost invariably, the platinum group elements have separated from ultrabasic magmas.
Norite (Mg – Fe – Ca – Al silicate) contains mainly sulphidic intrusions, whereas intrusions in dunite
(Mg – Fe silicate) are mainly sulphide free.
When primary deposits are washed and weathered, secondary deposits are formed, also known as
placers or alluvial deposits. Mechanical concentration of the heavy constituents by flowing water takes
place, together with chemical dissolution and precipitation of the platinum group metals. Alluvial de-
posits usually originate from dunite. For easily washable sand, a PGM content of 0.05 ppm is eco-
nomic.
The PGMs occur in a large number of minerals. Workable ore deposits contain mainly sperrylite
(PtAs2), cooperite (PtS), stibiopalladinite (Pd3Sb), laurite (RuS2), ferroplatinum (Fe – Pt), polyxene
(Fe – Pt – other platinum metals), osmiridium (Os – Ir), and iridium platinum (Ir – Pt). These minerals
are associated with particular carrier materials, e.g., iron pyrites, nickel iron pyrites, or chrome iron ore
(Renner (1997)).
The ratio and concentration of the PGM and their accompanying metals vary from site to site (s. Tab.
2.1). PGM deposits can be classified into the types oxidic-silicatic associations (e. g. UG2) and sul-
fidic associations, which can further be divided into PGM-dominated (e. g. Merensky, Stillwater) and
Ni-Cu-dominated (Sudbury, Norsil’sk) (Hochfeld (1997)). Of the estimated 112’900 t PGM-resources
55.6% are supposed to be found in South Africa, 15.9% in Russia, 12.3% in Finland, 8.9% in Zim-
babwe and 6.0% in the USA (Hochfeld (1997)). However, the PGM-reserves – the profitable workable
deposits – are mainly located in South Africa and Russia (s. Fig. 2.1).

Tab. 2.1 Characteristic metal concentrations in the most important PGM-mining sites. Data from Hochfeld (1997)
Element Merensky UG2 Stillwater Sudbury Noril’sk

RSA RSA USA CAN RUS
Ni % 0.28 0.16 <0.1 1.4 – 1.5 up to 3
Cu % 0.17 0.03 <0.1 1.1 – 1.3 up to 4
Pt ppm 4.82 3.22 6.0 0.38 2.5
Pd ppm 2.04 3.24 20.0 0.39 7.3
Rh ppm 0.24 0.54 0.21 0.03 0.2
2
Ru ppm 0.44 0.72 2 0.36 2 0.04 2 n. a. 2
2
Ir ppm 0.06 0.11 2 0.21 2 0.01 2 n. a. 2
2
Os ppm 0.04 0.07 2 -2 0.01 2 n. a. 2
2
Au ppm 0.26 0.07 0.21 - 0.25
2
Values for Ru, Ir and Os are taken from Hagelücken (2001b).
Fig. 2.1 PGM-reserves by region. Reserves are defined as at the present time profitably workable deposits. The
global PGM-resources are estimated to be 112'900 t. “Multiple of annual demand” indicates the years until
the actual resources are depleted assuming a constant demand.
Today worked ores are from primary deposits, whereas earlier mainly placers in Russia (Ural) were
exploited. Production mainly depends on the market price of the PGMs and the accompanying metals
like Ni or Cu. As seen from geographical distribution of the main resources, major producers are lo-
cated in South Africa (RSA) on the Bushveld Complex (Merensky-Reef and UG2) and in the Russian
Federation (RUS) at Noril’sk-Talnakh. These two major sites account for over 90 % of the PGM pro-
duction world-wide. The deposits in RSA are mined by seven big companies in RUS the company
Norilsk Nickel is responsible for the whole exploitation. Fig. 2.2, Fig. 2.3 and Fig. 2.4 show the de-
velopment of the production of platinum, palladium and rhodium respectively split by country. Other
PGM winning countries are the USA (Stillwater), Canada (Sudbury) and Zimbabwe. They contribute
only little to the global supply even though the capacity of the US palladium deposit has been greatly
increased in recent years.

Platinum, Supply by Country

100% 200 t
90% 180 t
80% 160 t
70% 140 t
60% 120 t
50% 100 t
40% Others 80 t
30% North America 60 t
20% Russia 40 t
South Africa
10% 20 t
Total Supply (tonnes)
0% t
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Fig. 2.2 Platinum production by country. Bold line represents total supply in tonnes. Grey areas indicate the coun-
tries share in the global supply. Data taken from JohnsonMatthey (2003), grey areas correspond to the axis
on the left (percentage of total supply), whereas the total supply correspond to the axis on the right (supply
in tonnes).
Palladium, Supply by Country

100% 300 t
90%
80% 250 t
70%
200 t
60%
50% 150 t
40%
Others
30% North America
100 t
20% Russia
South Africa
50 t
10%
0% t
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Fig. 2.3 Palladium production by country. Bold line represents total supply in tonnes. Grey areas indicate the coun-
tries share in the global supply. Data taken from (JohnsonMatthey (2003)), grey areas correspond to the axis
in tonnes).
Rhodium, Supply by Country

100% 30 t
90%
80%
25 t
70%
20 t
60%
50% 15 t
40%
Others
30%
10 t
North America
20% Russia
5t
10% South Africa
0% t
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Fig. 2.4 Rhodium production by country. Bold line represents total supply in tonnes. Grey areas indicate the coun-
tries share in the global supply. Data taken from JohnsonMatthey (2003), grey areas correspond to the axis
in tonnes).

2.2 Secondary PGM

Another important source is the recovery of secondary PGM metals. Large amounts of PGMs are used
in automotive catalysts. Due to the high value of the metals, automotive catalysts are recovered. They
are dismantled, the ceramic carrier then is dissolved and the metals regained through hydrometallurgi-
cal processes. This source is still small compared to the total demand world-wide, but it is rising. In
the USA already 11.5 t of total 40.1 t platinum used in 2001 originated from automotive catalysts.
This value is higher than in Europe because of the earlier introduction of automotive catalysts. The
pool of PGM in US automotive catalysts is stable. In Europe and Japan this pool still rises until
steady-state is reached with a recovery of expected 52% of all used automotive catalysts for Germany
(Hagelücken (2001b)). This future recovery quota is uncertain, because the PGM recovery depends
mainly on the market price. Further development probably will lead to a higher demand and thus to a
higher recovery quota. The actual recovery in Europe amounts 2.8 t, 1.4t and 1.1t for platinum, palla-
dium and rhodium respectively (for calculations and comments see. 5.1.2 Secondary Production).
Large quantities of secondary materials are present in the form of used Pt-Rh gauze, used for catalytic
formation of nitric acid. At the present, the pool of such gauze amounts about 50 t world-wide. Also
the glass industry generates large quantities of defective components for recovery. Defective labora-
tory equipment – crucibles and dishes – makes a considerable contribution. Also large quantities of Pt-
and Rh-containing thermocouple components are provided by the electrical industry (Renner (1997)).
In addition, an increasing proportion of the cars currently being scrapped are fitted with catalysts – it
was only in 1993 that all new gasoline cars in Europe were equipped with catalytic converters. This
will lead consequently to a higher amount of PGM’s recovered.
Often, the recovery operation is included in the sales contract. Materials are sent directly for recovery
and do not appear on the raw materials market. Because of this, no data on the recovery of the existing
PGM pool is available.

3 Characterisation of the PGMs

This chapter is mainly based on information provided by the International Platinum Association’s
website 3 . Some physical and chemical properties of the PGMs are summed up in Tab. 3.1.
3.1 Platinum
Platinum is a silvery-white metal and is probably best known for its use in jewellery. However, jewel-
lery actually represents only about 38% of overall platinum use, an equivalent volume of platinum is
used in catalytic converters, with the remaining 20% being used in other industrial applications.
Platinum is hard and extremely dense. Platinum and its relatives, iridium and osmium, are the most
dense metals known (platinum is nearly twice as dense as lead and 11% more dense than gold). It has
a high melting point and high temperature stability and corrosion resistance; and it is a good oxidation
catalyst, conductive and oxidation resistant. It’s biological compatibility makes it an important com-
ponent in medical applications.
3.2 Palladium
Like platinum, palladium is silvery-white in colour. Its melting point is the lowest of all PMG, but it is
still high compared to other popular metals. It is also the least dense of the PMGs. It has also high
temperature stability and corrosion resistance. The rarest of all PGMs apart from iridium, palladium is
also a good oxidation catalyst, conductive, oxidation resistant and ductile when annealed.
Palladium has the ability to absorb big amounts of hydrogen. This makes it an efficient and safe hy-
drogen storage medium and purifier. It is also used in chemical processes that require hydrogen ex-
change between two reactants, such as that which produces butadiene and cyclohexane, the raw mate-
rials for synthetic rubber and nylon.
Palladium's catalytic qualities make it necessary in catalytic converters and air purification equipment.
Its chemical stability and electrical conductivity lead to more effective and durable plating than using
gold in electronic components.
3.3 Rhodium
Its hardness makes it an alloying agent to harden platinum. With this hardness, along with its high
melting point, high temperature stability and corrosion resistance, rhodium is used in many industrial
processes such as glass, glass fibre and nitric acid production. Like other PGMs, Rhodium has cata-
lytic qualities, an thus is used in catalytic converters.
3
http://www.platinuminfo.net. accessed at 01.03.2003
Tab. 3.1 Physical and chemical properties of the platinum group metals.
Property unit Rh Pd Pt
Atomic number 45 46 78
Atomic weight g / mol 102.91 106.42 195.08
-4
Rarity in lithosphere ppm 2 10 6 10-4 0.001
3
Density g/cm 12.41 12.02 21.45
10 cm Ω
6 -1 -
Electrical conductivity 0.211 0.095 0.0966
1
Electrical resistivity µ Ω cm 4.33 9.93 9

-1 -1
Thermal conductivity at 0°C W m °C 150 76 73
Melting point °C 1960 1554 1772
-2
Hardness (Brinell value) MN m 1100 37.3 392
Tensile strength, annealed con- kg/mm2 71 17 14
dition

4 Use / Application of PGM

Common to all of the PGMs is their use in electronics and as an element in catalytic converters. Cata-
lytic converters are mainly used in automobiles, in the petroleum industry and hydrogen fuel cells. The
use in automotive catalysts covers a good part of the overall PGM demand, in the case of Rhodium it
amounts over 80%. The resource needed for automotive catalysts is covered partly from recycled auto-
motive catalysts. This share differs greatly from country to country.
Second largest segment is the use of the metals in jewellery, namely in the case of Platinum. Another
important field of application is the use in dental alloys.
Magnetic alloys containing platinum (mainly with cobalt) are components in the hard disks of 90% of
the computers manufactured 2001. It is the largest single electronic application for platinum with rap-
idly increasing demand.
PGM catalysts for nitric acid production take the form of gauze made out of fine wire. Rhodium is
added for strength and to reduce the amount of platinum loss during the conversion of the gas.
PGMs are widely used in medicine because they are inert and do not corrode inside the body. Allergic
reactions to PGM are extremely rare. They have good electrical conductivity, which makes them an
ideal electrode material.
Tab. 4.1 resumes the use of different PGMs, Fig. 4.1, Fig. 4.2 and Fig. 4.3 show the PGM demand by
application in 2001.
The demand of the different PGM varies greatly depending on stocks and price. Overall, the platinum
used equals in magnitude the demand of palladium (198 t Pt or 6’370 troz 4 and 4’880 troz or 152 t Pd
respectively in 2002), whereas the demand of Rhodium of 18 t or 568 troz is markedly smaller
(JohnsonMatthey (2003)).
5
Tab. 4.1 Main applications of PGMs. Table taken from Johnson Matthey
Application PGMs used

Automotive control of vehicle pollution Pt, Pd, Rh
catalyst
Chemical catalysts for chemical synthesis Pt, Pd, Rh, Ru
electrode coating Ir, Ru
Dental alloys for dental restorations Pt, Pd
Electronics computer hard disks Pt
multi-layer ceramic capacitors Pd
crucibles to grow single crystals Ir
Fuel Cells catalysts for power generation Pt
Glass glassmaking equipment Pt, Rh
Investment coins and bars Pt
Jewellery wedding rings, fashion jewellery Pt
Medical anti-cancer drugs, implants Pt
Petroleum catalysts for gasoline refining Pt
Sensors temperature & gas sensing Pt, Rh
Other spark plugs, pollution control Pt, Ir
4
troz = troy ounce, 1'000 troz correspond to 31.1 kg.
5
Johnson Matthey’s website “Platinum Today”, http://www.platinum.matthey.com, accessed at 01.03.2003
Fig. 4.1 Platinum, demand by application in 2001. Values in ounces (1000 troy ounces correspond to 31 kg), data
6
taken from International Platinum Association
Fig. 4.2 Palladium, demand by application in 2001. Values in ounces (1000 troy ounces correspond to 31 kg), data
6
Fig. 4.3 Rhodium, demand by application in 2001. Values in ounces (1000 troy ounces correspond to 31 kg), data
6
6
International Platinum Association‘s website, http://www.platinuminfo.net/, accessed at 01.03.2003
Platinum group metal, which is consumed in Europe, consists of a certain fraction of secondary mate-
rials originating from scrap. This fraction of secondary material is joined before consumption with
primary material, originating from ore. Two single plants are responsible for around 90% of the pri-
mary platinum, palladium and rhodium production – these are the operations in Norsil’sk in Russia
and in the Bushveld Complex in South Africa. The PGM are however not the only noteworthy prod-
ucts of these enterprises. Namely Noril’sk Nikkel production is characterised by the production of
nickel and copper, PGM being a co-product. This fact makes it obvious that the ecological burden aris-
ing from the primary production has to be allocated to the primary products nickel, copper, platinum,
palladium and rhodium. The same is true for the secondary production, where the metal values are re-
covered jointly. Fig. 5.1 gives an overview over the process scheme and the corresponding modules in
ecoinvent.

Russia (RU) Platinum group Platinum group Europe (RER)

South Africa
(ZA) metal production, metal production,
primary secondary
mining collection
beneficiation benefication
metallurgy metallurgy
separation of co-
products
copper, primary, nickel, primary,
from platinum from platinum
group metal group metal
refining refining
RU Platinum, RU Palladium, RU Rhodium, Production Mix:

primary, at primary, at primary, at Russia (RU), South
Africa (ZA)
refineryZA refinery ZA refinery ZA
Transport to Europe
Rhodium, at
Platinum, at Palladium, at Consumer Mix in
regional Europe (RER)
regional storage regional storage
storage
Fig. 5.1 Scheme of the PGM production in ecoinvent.
5.1 Allocation
The PGM production in ecoinvent is described formally as multi output process, yielding the coupled
products nickel, copper and PGM as a mixture. The next chapter explains the allocation procedure in
detail.
5.1.1 Primary Production

The exact ratio of the different PGMs present in this mixture is not reported in the literature used. It
was assumed that the ratio of the PGM produced corresponds to the corresponding ratio in the raw ore
(see Tab. 2.1).
For the coupled production of Ni, Cu and PGM the demand of ore is allocated according to the pro-
duction output by mass Hochfeld (1997). The ore is handled in ecoinvent as a mixed ore with 4.7E-4%
Pt, 3.1E-4% Pd, 0.2E-4% Rh in crude ore. The copper and nickel content in the ore depends on the
deposit. For RU it is 3.2% and 2.3% respectively and for ZA the copper content is 0.052% ant the
nickel content is 0.037%. As copper and nickel are leaving the PGM production before the final refin-
ery step, the burdens of refining are allocated only to PGM while the burdens of the other steps are al-
located to copper, nickel and the PGMs. Thus different overall allocation factors for the different in-
ventory flows are established depending on the flows occurrence and magnitude in the different pro-
duction steps.
For all inventories other than ore an allocation by revenue is made, because this reflects economic re-
ality of metal mining in a higher degree than an allocation by mass. Exploitation of an ore is always
dependent on sufficient revenue expected. If a specific ore lacks sufficient revenue expected, it is not
mined. Already mined ore with insufficient metal content usually is put aside for further beneficiation
at a later time, at which market prices allow a profitable production of the metal. As ores contain dif-
ferent metals, the overall accounting of all metals produced is decisive. Even if the winning of a cer-
tain single metal is not profitable, it may be through cross-subsidising by another more profitable co-
product. Thus the standard way chosen to allocate ecological burdens – except the demand of ore – is
the allocation by revenue.
As PGM-mining companies have long term decision horizons, the average price of the last ten year
period was chosen, this is for Pt, Pd and Rh US$ 14’000, US$ 9’420 and US$ 30’700 per kilogram re-
spectively 7 . The prices for Ni and Cu in this time period were US$ 6.65 and US$ 1.99 per kilogram
respectively 8 . The fluctuation in price is documented in Fig. 5.2, Fig. 5.3, Fig. 5.4 and Fig. 5.5. Allo-
cation and environmental burden may change depending on the period the price average is based on.
Tab. 5.1 gives a summary of the values chosen to calculate the allocation factors in the primary pro-
duction. The resulting factors are summed up in Tab. 5.2
5.1.2 Secondary Production

Similar considerations are made for the allocation of secondary PGMs that originate from recovered
automotive catalysts. It is assumed that all the secondary metal comes from automotive catalysts.
Other sources of scrap were neglected. As in primary production, no data on the ratio of single PGM
produced was available. This ratio differs greatly from scrap to scrap. Instead of the ore composition
like in primary production, here the amount of annually recovered Pt, Pd and Rh respectively (in
Europe) are chosen as proxy for the scrap composition. This composition ratio was used for allocation
by mass. Multiplication with the market price gives the ratio used for the allocation by revenue.
Since no data on European rhodium recovery was available, it was approximated with the figure for
global recovery rate of 15% 9 and a total annual Rh demand in Europe of 6.4 t. For the year 2002 this
gives an amount of 1.11 t/a recovered Rh in Europe. The figures for Pt and Pd recovered in Europe are
documented and amount 2.80 t/a Pt and 1.40 t/a Pd in the year 2002 (JohnsonMatthey (2003)).
This ratio of 1.11t/a : 1.40t/a : 2.80t/a for Rhodium : Palladium : Platinum corresponds to allocation
factors summed up in Tab. 5.2.
7
Data taken from Johnson Matthey’s Website “Platinum Today”, http://www.platinum.matthey.com, accessed at 01.03.2003
8
Data taken from “London Metal Exchange”, http://www.lme.co.uk/data_prices/ home.html, accessed at 01.03.2003
9
Total global rhodium demand of 20.7 t and an amount of 3.08 t recovered from automotive catalysts world-wide gives secon-
dary rhodium a share of 15 % in the global demand.
Fig. 5.2 Price development of platinum and palladium in the past ten years. Values in US$ per troy ounce.
Fig. 5.3 Price development of rhodium in the past ten years. Values in US$ per troy ounce.

Nickel, Price in US$ per Tonne

$25'000
27-Month
15-Month
$20'000
3-Month
Cash
$15'000
$10'000
$5'000
$-
01.89 01.91 01.93 01.95 01.97 01.99 01.01 01.03
Fig. 5.4 Price development of nickel in the past ten years. Values in US$ per tonne.
Copper, Price in US$ per Tonne

$4'000
27-Month
$3'500 15-Month
3-Month
$3'000
Cash
$2'500
$2'000
$1'500
$1'000
$500
$-
01.89 01.91 01.93 01.95 01.97 01.99 01.01 01.03
Fig. 5.5 Price development of copper in the past ten years. Values in US$ per tonne.
Tab. 5.1 Assumptions made for allocation in the primary production. Details in the text above.
Unit Primary PGM- Primary PGM- global

production (ZA) production (RU)
functional unit, 1 t PGM t / t PGM 1 1
Platinum in PGM-mix % share in PGM 68% 25%
Palladium in PGM-mix % share in PGM 29% 73%
Rhodium in PGM-mix % share in PGM 3% 2%
Nickel as co-product t / t PGM 147 2’310
Copper as co-product t / t PGM 104 3’200
Platinum, price on 10 year average US$ / tonne 14'017’726
Palladium, price on 10 year average US$ / tonne 9’420’169
Rhodium, price on 10 year average US$ / tonne 30'703’965
Nickel, price on 10 year average US$ / tonne 6’647
Copper, price on 10 year average US$ / tonne 1’986

Tab. 5.2 Resulting factors for allocation based on values shown in Tab. 5.1. These factors allocate the ecological
burden to the respective products. Allocation by mass reflects the PGM composition in the raw material
(ore, scrap). Allocation by revenue is calculated by multiplication of the PGM content in the raw material
with the respective market price. The allocation by mass is solely used for the allocation of PGM-ore to the
respective products.
Primary Production Secondary Production

Products Allocation by Allocation by Allocation by Allocation by Allocation per Allocation per
mass (ZA) mass (RU) revenue (ZA) revenue (RU) mass (RER) revenue (RER)
Pt 0.27% 0.005% 66% 11% 53% 45%
Pd 0.11% 0.013% 19% 21% 26% 15%
Rh 0.01% 0.0004% 7% 2% 21% 39%
Ni 41.27% 41.88% 7% 47% - -
Cu 58.33% 58.10% 1% 19% - -
5.2 PGM consumer mix

In ecoinvent, following modules represent the use of platinum, palladium and rhodium in Europe
(consumer mix):
- “platinum, at regional storage ”
- “palladium, at regional storage ”
- “rhodium, at regional storage ”

The consumer mix is an assembly of the primary PGM transported from ZA and RU to Europe and
secondary PGM, which are recovered and consumed in Europe (Fig. 5.6). The modules for primary
and secondary production are described in the sections 5.3 and 5.4 respectively.
The layout of the modules with general flow information, remarks and values are shown in Fig. 5.7,
Fig. 5.8 and Fig. 5.9. An overview over the module’s general information and metadata is given in
Tab. 5.3
Russia (RU) Platinum group Platinum group Europe (RER)

South Africa
(ZA) metal production, metal production,
primary secondary
RU Platinum, RU Palladium, RU Rhodium, Production Mix:

primary, at primary, at primary, at Russia (RU), South
Africa (ZA)
Transport to Europe
Rhodium, at
Platinum, at Palladium, at Consumer Mix in
regional Europe (RER)
regional storage regional storage
storage
Fig. 5.6 Scheme of the PGM, at regional storage, in ecoinvent.

Tab. 5.3 Ecoinvent meta information for the modules of consumption of platinum, palladium and rhodium.
Name palladium, at regional platinum, at regional rhodium, at regional

storage storage storage
InfrastructureProcess 0 0 0
Unit kg kg kg
Dataset Version 2.0 2.0 2.0
IncludedProcesses Primary PGM transported Primary PGM transported Primary PGM transported
from ZA and RU for con- from ZA and RU for con- from ZA and RU for con-
sumption in Europe and sumption in Europe and sumption in Europe and
secondary PGM recov- secondary PGM recov- secondary PGM recov-
ered and consumed in ered and consumed in ered and consumed in
RER. RER. RER.
Amount 1 1 1
LocalName Palladium, ab Regional- Rhodium, ab Regional-
lager Platin, ab Regionallager lager
Synonyms Platingruppenmet- Platingruppenmet- Platingruppenmet-
all//platinumgroup metal all//platinumgroup metal all//platinumgroup metal
GeneralComment The consumer mix is an The consumer mix is an The consumer mix is an
assembly of the primary assembly of the primary assembly of the primary
palladium transported platinum transported from rhodium transported from
from ZA and RUS for con- ZA and RU for consump- ZA and RU for consump-
sumption in Europe and tion in Europe and secon- tion in Europe and secon-
secondary palladium re- dary platinum recovered dary rhodium recovered
covered and consumed in and consumed in RER and consumed in RER
RER according the global according to the global according to the global
market share in 2002. market share in 2002. market share in 2002.
StartDate 2002 2002 2002
EndDate 2002 2002 2002
Data valid for entire period 1 1 1
Geography Text Global supply data for Global supply data for Global supply data for
Europe used, no import Europe used, no import Europe used, no import
statistic available. Supply statistic available. Supply statistic available. Supply
per region is assumed to per region is assumed to per region is assumed to
correspond to the import correspond to the import correspond to the import
to Europe to Europe to Europe
Technology Text European consumer mix European consumer mix European consumer mix
Representativeness 92% 92% 98%
ProductionVolume 54.9 t / a 52.4 t / a 7.49 t / a
SamplingProcedure Literature Literature Literature
Extrapolations see Geography and Tech- see Geography and Tech- see Geography and Tech-
nology nology nology
UncertaintyAdjustments none none none
5.2.1 PGM, at regional storage, in Ecoinvent

These modules are valid for the consumption of the platinum group metals (PGM) Platinum, Palla-
dium and Rhodium in Europe.
The secondary production is reflected by the recovery of automotive catalysts in Germany according
to the actual recycling rate for Europe (Tab. 5.4). It is assumed to be valid for whole Europe.

The primary-production is reflected by the production processes of RSA and RUS according to their
market-share. These two countries produce over 90 % of the PGMs. The mix of consumed PGM is as-
sumed to correspond to the production mix shown in chapter 2.1 An average of the share in the global
production over the last ten years was chosen (Tab. 5.5). Due to the importance of PGM, transport is
assumed to take place by plane, the metal being protected with a case of 50% of its mass.
Data quality
Allocations will shift with changing market situation. The recycling rate – i. e. the ratio of secondary
platinum – may be underestimated, as most of the recovered material does not appear on the market.
Tab. 5.4 Demand of secondary and primary PGM. Assessment of the share of the secondary production in the world
wide demand. The recovery from automotive catalysts has to be added to the total demand to obtain the net
demand. For Rh, no region specific figures are available, a world-wide recovery rate is chosen. Values are
taken from JohnsonMatthey (2003). Note that the palladium demand in the US is exceptionally small in 2002,
resulting in an overestimated ratio of secondary production, which was stable at around 5% of the total
consumption until 2001. This is due to the use of stocks by US auto manufacturers in lieu of metal pur-
chases.
Values in t for 2002 Pt Pd Rh

total global demand (primary) 198 152 17.7
total global recovery (secondary) 17.7 11.5 3.1
global net demand (total) 216 163 20.7
global ratio secondary 8% 7% 15%
total European demand (primary) 52.1 51.0 6.4
total European recovery (secondary) 2.8 1.4 1.1
European net demand (total) 54.9 52.4 7.5
European ratio secondary 5% 3% 15%
21% (2002)
US ratio secondary 28%
5% (1993 – 2001)
Japanese ratio secondary 4.2% 3.0%
Tab. 5.5 Production of primary metal in South Africa and the Russian Federation. Ten year averages chosen from
data in JohnsonMatthey (2003).
Share in global primary production Pt Pd Rh

(ten year average)
South Africa (ZA) 72.2% 26.3% 72.9%
Russian Federation (RU) 20.1% 63.2% 23.3%
Represented global market volume 92.3% 89.5% 96.2%
South Africa (ZA), relating to total 78.2% 29.4% 75.7%
Russian Federation (RU), relating to total 21.8% 70.6% 24.3%

Deviation 95%
Modul name in
Process Name
Subcategory
Infrastructure
mean value
Source for
ecoinvent
Comment
Standard
Category
Location
Remarks
Output
General
Input
Type
Unit
M e an value
A ccording to market share: 72% platinum, primary, at (JohnsonMatthey

Platinum, RSA  metals extraction No ZA 0.74 kg 1 1.12 (1,2,1,1,1,4,3)
of global primary production ref inery , 2003)
platinum, at regional
A ccording to market share: 20% platinum, primary, at (JohnsonMatthey

Platinum, RUS  metals extraction No RU 0.21 kg 1 1.12 (1,2,1,1,1,4,3)
of global primary production ref inery , 2003)
storage
Recycling solely f rom recovered

Platinum, platinum, secondary, (JohnsonMatthey
 autocatalysts: 5% of metals extraction No RER 0.05 kg 1 1.12 (1,2,1,1,1,4,3)
recycing at ref inery , 2003)
consumption
Only aircargo considered,
packaging is roughly estimed to
transport transport, aircraf t,
Transportation  be 50% of the metal mass. airplane No RER 17.49 tk m def ault distances 1 1.12 (1,2,1,1,1,4,3)
systems f reight
Distance to RER: f rom ZA
15000km, RU 2500 km
Platinum, at platinum, at regional
 metals extraction No RER 1.00 kg
consumer storage
Fig. 5.7 Flows for “Platinum, at regional storage” and its representation in the ecoinvent database

Deviation 95%
Modul name in
Process Name
Subcategory
Infrastructure
mean value
Source for
ecoinvent
Comment
Standard
Category
Location
Remarks
Output
General
Input
Type
Unit
M e an value
Palladium, palladium, primary, at (JohnsonMatthey

 metals extraction No ZA 0.29 kg 1 1.12 (1,2,1,1,1,4,3)
palladium, at regional
RSA ref inery , 2003)

Palladium, palladium, primary, at (JohnsonMatthey
 metals extraction No RU 0.68 kg 1 1.12 (1,2,1,1,1,4,3)
RUS ref inery , 2003)
Recycling solely f rom recovered palladium,
Palladium, (JohnsonMatthey
storage
 autocatalysts: 3% of metals extraction No RER secondary, at 0.03 kg 1 1.12 (1,2,1,1,1,4,3)

recycing , 2003)
consumption ref inery
only aircargo considered,
transport transport, aircraf t,
Transportation  be 50% of the metal mass. airplane No RER 8.98 tk m def ault distances 1 1.12 (1,2,1,1,1,4,3)
systems f reight
15000km, RU 2500 km
Palladium, at palladium, at regional
consumer storage
Fig. 5.8 Flows for “Palladium, at regional storage” and its representation in the ecoinvent database

Deviation 95%
Modul name in
Process Name
Subcategory
Infrastructure
mean value
Source for
ecoinvent
Comment
Standard
Category
Location
Remarks
Output
General
Input
Type
Unit
M e an value
Rhodium, A ccording to market share: 73% rhodium, primary, at (JohnsonMatthey

 metals extraction No ZA 0.64 kg 1 1.12 (1,2,1,1,1,4,3)
RSA of global primary production ref inery , 2003)
rhodium, at regional
Rhodium, A ccording to market share: 23% rhodium, primary, at (JohnsonMatthey

 metals extraction No RU 0.21 kg 1 1.12 (1,2,1,1,1,4,3)
RUS of global primary production ref inery , 2003)
storage
Recycling solely f rom recovered

Rhodium, rhodium, secondary, (JohnsonMatthey
 autocatalysts: 15% of metals extraction No RER 0.15 kg 1 1.12 (1,2,1,1,1,4,3)
recycing at ref inery , 2003)
consumption
only aircargo considered,
Transportatio transport transport, aircraf t,
 be 50% of the metal mass. airplane No RER 15.26 tk m def ault distances 1 1.12 (1,2,1,1,1,4,3)
n systems f reight
15000km, RU 2500 km
Rhodium, at rhodium, at regional
consumer storage
Fig. 5.9 Flows for “Rhodium, at regional storage” and its representation in the ecoinvent database

5.3 Winning of Primary PGM

This part is mainly based on Hochfeld (1997). Lacking data – mainly on direct emissions to air and to
water – was taken from different sources where stated. Because of the big variance between the differ-
ent production processes due to different ore composition, primary PGM production is represented by
the two major sites in Bushveld (ZA) and Noril’sk (RU). The producing companies in these two de-
posits determine the PGM production within their countries and world-wide.
While the companies in RSA are mainly PGM-dominated, Norilsk Nickel in the Russian Federation
earns a large share through Nickel and Copper. Also the ratio of the PGM produced differ between the
production in RSA and the one in RUS (see chapter 5.1 above).
For this reasons, the production process in RSA and the one in RUS are treated separately. Their prod-
ucts – the pure metals Pt, Pd, Rh, Ni and Cu – are assembled later according to their market share.

The production process described is valid for both the operations in RSA and in RUS. Differences are
stated.
Mining
The PGM containing ore is mined underground and transferred to the beneficiation. Widening of the
tunnels is mainly done by blasting. The overburden – material, which does not contain PGM-bearing
ore – is deposed off-site and is partly refilled into the tunnels.
Beneficiation
After mining, the ore is first ground. In a next step it is subjected to gravity concentration to separate
the metallic particles from the PGM-bearing minerals. After this first concentration step, flotation is
carried out to remove the gangue from the sulfidic minerals. For neutralisation lime is added. In the
flotation several organic chemicals are used as collector, frother, activator, depressor and flocculant.
Sometimes cyanide is used as depressant for pyrite. Tailings usually are led to tailing heaps or ponds.
Metallurgy
The pyrometallurgical step may be divided in three substeps, which can be effected in combination
depending on the procedure used. The process is basically the same as in the nickel or copper produc-
tion. Emitted sulphur dioxide usually is recovered from the off-gas stream producing sulphuric acid.
Roasting. The sulphidic ore is roasted at 600 – 700 °C. This oxidises the iron present, dries and pre-
heats the ore. Since iron has got a bigger affinity to oxygen than other metals, it gets oxidised to slag
releasing the sulphur as sulphur dioxide (Formula 5.1). This reaction is strongly exothermic. Other
metals associate again with sulphur to sulphur compounds.
3 Fe7S8 + 38 O2 → 7 Fe3O4 + 24 SO2 (5.1)
Smelting. In this step, the bigger part of the gangue and the iron sulphide is eliminated. The valuable
metals concentrate in the matte. Smelting is done at temperatures of 1400 °C. Main reaction is the
slagging of the prior formed magnetite (Fe3O4) with the flux (calcite, silicate), according to the formu-
las 5.2 and 5.3.
3 Fe3O4 + FeS → FeO + SO2 (5.2)
10 FeO + 5 SiO2 → 5 (2FeO·SiO) (5.3)
The melt is parted from magnetite and now consists of a sulphidic and a oxidic phase. The matte is
tapped at 1400 °C and processed in the converting step.

Converting. In the converting step, iron sulphide is removed from the low-grade matte by oxidation
and slagging. The slag, which contains high levels of metals, is returned to the smelting step for recov-
ery. The final product is high-grade, low-iron nickel – copper matte, which typically contains 20 % S,
less than 1 % Fe, and also some precious metals. Converting is a batch operation. Air or oxygen-
enriched air is blown through the molten matte to form iron oxides and remove sulphur as sulphur di-
oxide. The iron oxides combine with added silica flux to form an iron silicate slag (formulas 5.4 and
5.5).
2 FeS + 3 O2 → 2FeO + 2SO2 (5.4)
2 FeO + SiO2 → 2FeO·SiO2 (5.5)
In the processes regarded in this study Pierce-Smith-Converters are used.
Separation from Non Ferrous Metals

The main accompanying metals in the PGM winning are Nickel and Copper. In the case of the Russian
process the winning of these two metals is the main objective. The separation is done in a hydrometal-
lurgical process. In a typical hydrometallurgical process, the concentrate or matte is first leached in a
sulphate or chloride solution to dissolve the metals, while the sulphide is oxidised to insoluble elemen-
tal sulphur or soluble sulphate. In this way a nickel – copper matte can be treated in a two-stage leach
process to produce a copper-free nickel sulphate or nickel chloride solution, and a leach residue en-
riched in copper. The copper in the copper-rich residue is solubilised by repeated leaching and recov-
ered from solution electrolytically as cathode copper. Nickel is recovered from the purified nickel sul-
phate or chloride solution either electrolytically as a pure nickel cathode or by chemical reduction with
hydrogen to give pure nickel powder. The remaining leach residue contains all PGM, which are ex-
tracted in the refining step.
Refining
The usual starting point for separating platinum group metals or producing compounds and catalysts is
an aqueous solution of PGM. Most raw materials can be dissolved in oxidising acids. Once dissolved,
there are namely two processes for refining.
Selective precipitation. This is the classical process. Separation is based on the principles of analyti-
cal chemistry of PGM. In the crystallisation process, the solubility of chloro complexes of the PGM
are very important, as is the possibility of altering this solubility. The temperature dependence of this
solubility is generally quite large and can be utilised for precipitation. An even more useful phenome-
non in separation technology is the effect on solubility of the addition of a common ion, which is util-
ised in precipitation crystallisation. Conversion of the precipitated salts to their metals is carried out
mainly by thermal decomposition at 1000 °C.
Solvent extraction. This process is based on the ability of the metals to form together with organic
molecules stable complexes. After total dissolution the liquid is treated with a cascade of different
specific solvents. Liquid – liquid extraction is characterised by distribution coefficients that vary
greatly from element to element. These can usually be further modified by chemical methods. Al-
though separations are generally better than those effected by precipitation, coarse separation by sol-
vent extraction must be further purified.
The principal raw materials are concentrates produced from ores, mattes and slimes from nickel and
copper operations. Secondary materials such as spent catalyst or electronic component scrap are also
significant sources. PGM refining is complex and individual process stages may have to be repeated to
achieve the required purity. Precious metal refineries are complex networks of main and subsidiary
processes. It is therefore not possible to identify single process steps and their contribution to emis-
sions and consumptions. The product of this step is the pure platinum group metal.

5.3.2 Emissions
Mining
Major emissions are due to mineral born pollutants present in effluents. Being underground, the min-
ing has less emissions to air than open pit mining. The major dust sources are on the surface therefore
the dust emissions are in the same magnitude like in open pit mining. Rain percolate through overbur-
den and account to metal emissions to groundwater.
Beneficiation
the ore itself. Flotation produce effluents containing several organic agents 10 used. Some of these
cally to release carbon disulphide. Tailings effluent contain additional sulphuric acid from acid rock
drainage.
Sulphur dioxide. In all of the process steps sulphur dioxide is emitted to air. Recovery of sulphur di-
oxide is only economic for high concentrated off-gas. Therefore recovery is not effected in all process
alternatives. SO2 emissions are by far the most important in the metallurgy.
Carbon dioxide. In the beneficiation step, considerable amounts of lime is added to the ore for pH-
stabilisation, lime forms later flux in the metallurgical step. Lime decomposes into CO2 to form cal-
cite.
the air consist of metals and thus are often returned to the leaching process after treatment.
Chlorine. Chlorine is used in some leaching stages and is produced during the subsequent electrolysis
of chloride solution. The chlorine evolved is collected and re-used in the leach stage. The presence of
chlorine in wastewater can lead to the formation of organic chlorine compounds (AOX) if solvents etc.
are also present in a mixed wastewater.
VOC. VOCs can be emitted from the solvent extraction stages. A variety of solvents are used an they
contain various complexing agents to form complexes with the desired metal that are soluble in the or-
ganic layer.
emitted to water. The metals concerned are Cu, Ni, Co, As and Cr. Other significant substances are
chlorides and sulphates. Wastewater from wet gas cleaning (if used) of the different metallurgical
stages are the most important sources. The leaching stages are usually operated on a closed circuit and
Refining
number of digestion, electrolytic and purification processes. Chlorine is used extensively in the Miller
process and in the dissolution stages using hydrochloric acid and chlorine mixtrues respectively. Dust
10
and metals are generally emitted from incinerators and furnaces. VOC can be emitted from solvent ex-
traction processes, while organic compounds, namely dioxins, can be emitted from smelting stages re-
sulting from the poor combustion of oil and plastic in the feed material. All these emissions are subject
to abatement technologies and controlling. Large quantities of effluents contain amounts of metals and
organic substances.
5.3.3 Waste
Overburden are deposed close to the mine. Tailings are deposed as piles and in ponds. In both the
overburden and the tailings occur acid rock drainage over a long period of time. These effects are han-
dled in the section “Non Ferrous Metal Winning – Auxiliary Processes”, chapter “Disposal of Tailings
and Overburden”.
Refining
Like in the pyro- and hydrometallurgical step quantities of solid residuals are generated, which are
residues generally comprise hydroxide filter cakes (ironhydroxide, 60% water, cat I industrial waste).
5.3.4 Primary PGM in ecoinvent

This module is modelled in two different specifications: “PGM-Production, primary, RSA” for the
South African operations and “ PGM-Production, primary, RUS” for the Russian operations. Both
modules are similarly modelled, exceptions are mentioned in the text below. The scheme of this multi-
output process is shown in Fig. 5.10.
formation are shown in Fig. 5.11 - Fig. 5.15. An overview over the module’s general information and
metadata is given in Tab. 5.6. The following paragraphs describe calculations, sources and assump-
tions chosen in this study.

Russia (RU) Platinum group

South Africa
(ZA) metal production,
primary
mining
beneficiation
metallurgy
separation of co-
products
copper, primary, nickel, primary,
from platinum from platinum
group metal group metal
refining
RU Platinum, RU Palladium, RU Rhodium,

primary, at primary, at primary, at
Fig. 5.10 Scheme of the multi-output process “platinum group metal production, primary” in Ecoinvent. Its jointly
produced co-products are “platinum, primary, at refinery”, “palladium, primary, at refinery”, “rhodium, pri-
mary, at refinery”, “nickel, primary, couple-production, platinum group metal production” and “copper, pri-
mary, couple-production, platinum group metal production”.

Tab. 5.6 Ecoinvent meta information for the multi-output processes in the primary production of PGM in Russia and
South Africa.
Name platinum group metal production, primary platinum group metal production, primary
Location ZA RU
Infrastructure Proc-
ess 0 0
Unit kg kg
Dataset Version 2.0 2.0
Included Processes The module includes a mining and a benefici- The module includes a mining and a benefici-
ation step with the mining infrastructure and ation step with the mining infrastructure and
disposal of overburden and tailings. Subse- disposal of overburden and tailings. Subse-
quently it includes the metallurgy step with the quently it includes the metallurgy step with the
disposal of slag, the infrastructure and the disposal of slag, the infrastructure and the
separation of the co-products nickel and cop- separation of the co-products nickel and cop-
per, and the refining step yielding the desired per, and the refining step yielding the desired
PGM-mix inclusively the refining infrastruc- PGM-mix inclusively the refining infrastruc-
ture. Production, application and emissions of ture. Production, application and emissions of
most agents used in beneficiation and metal- most agents used in beneficiation and metal-
lurgy are also included. lurgy are also included.
Amount 1 1
Local Name Platingruppenmetalle-Produktion, primär Platingruppenmetalle-Produktion, primär
Synonyms Platingruppenmetall//platinumgroup metal Platingruppenmetall//platinumgroup metal
General Comment The multioutput-process “PGM-Production,
The multioutput-process “PGM-Production, primary” delivers the co-products “platinum,
primary” delivers the co-products “platinum, primary, at refinery”, “palladium, primary, at
primary, at refinery”, “palladium, primary, at refinery”, “rhodium, primary, at refinery”,
refinery”, “rhodium, primary, at refinery”, “nickel, primary, from platinum group metal
“nickel, primary, from platinum group metal production” and “copper, primary, from plati-
production” and “copper, primary, from platinum group metal production” in the Russian
num group metal production” in South Africa Federation (RU). The module is designed for
(ZA). The module is designed for the use of the use of the metal in technical systems,
the metal in technical systems, where it plays where it plays a minor role like the use in
a minor role like the use in manufacturing of manufacturing of electronic or technical chem-
electronic or technical chemistry using certain istry using certain catalysts. It is not to be
catalysts. It is not to be used if the impact of used if the impact of the PGM within the mod-
the PGM within the modelled process in elled process in scope is considered to be
scope is considered to be high. In such cases, high. In such cases, a more detailed analysis
a more detailed analysis depending on scope depending on scope and allocation proce-
and allocation procedures has to be con- dures has to be conducted. The data used is
ducted. The data used is mainly based on a mainly based on a LCA study for autocata-
LCA study for autocatalysts in Germany. lysts in Germany.
End Date 2002 2002
Data valid for entire
period 1 1
Geography Text Some agents used like lime or sodium cya- Some agents used like lime or sodium cya-
nide originate from data sets valid for Switzer- nide originate from data sets valid for Switzer-
land or Europe. Processes like blasting and land or Europe. Processes like blasting and
transport originate from data sets valid for transport originate from data sets valid for
Switzerland or Europe. Switzerland or Europe.

Tab. 5.6 Ecoinvent meta information for the multi-output processes in the primary production of PGM in Russia and
South Africa. (cont.)
Technology Text Mining is done underground, followed by ben-

eficiation trough flotation where considerable Mining is done underground, followed by ben-
amounts of agents are added. Overburden eficiation trough flotation where considerable
and tailings are disposed on the mining site amounts of agents are added. Overburden
and partly refilled. In ZA a electric arc furnace and tailings are disposed on the mining site
with a Søderberg electrode system and a and partly refilled. In RU a Outokumpu flash
Pierce-Smith-Converter is used. sulphur diox- furnace followed by a Pierce-Smith-Converter
ide in the off-gas is recovered producing sul- is used. The resulting nickel matte is treated
phuric acid. The separation of non ferrous by elektrorefining. No recovery of sulphur di-
metals is done hydrometallurgically, the refin- oxide in the off gas. The refining is done by
ing by selective precipitation. Allthough elec- selective precipitation, using data from the ZA
tricity production in ZA is mainly coal based, process as proxy. For electricity mix the
the UCTE production mix is inventoried. UCTE production mix is inventoried
Representativeness 63% 23%
Production Volume 221 t/a 81 t/a
Sampling Procedure Literature Literature
Uncertainty Adjust-
ments none none

The production infrastructure used directly for exploration is inventoried within this module, whereas
the infrastructure like buildings and land-use for the operation is inventoried in an extern module. The
mobile equipment used for exploration like drilling machines, scoop trams and underground trucks is
inventoried assessing a standard value for diesel, burned in building machine, i. e. 69.2 kJ per tonne
mined raw ore. 2000 m conveyor belts are further assumed with a capacity of 2000 t / h which yields
2*10-5 m conveyor belt required per tonne mined raw ore. Mining facilities are inventoried as unspe-
cific module according to part II “Non Ferrous Metal Winning – Auxiliary Processes”.
Transformation and occupation of land through direct exploration are assumed with the standard value
for non-ferrous metals without heap leaching processes of 0.0025 m2/t and 0.075 m2a/t ore mined.
Mining
mine (see. Part II “Non Ferrous Metal Winning – Auxiliary Processes”). Based on IPPC (2002) a
value of 50 % refilled overburden is assumed. Demand for water was not available for the process in
RUS, it was approximated with the value for RSA (of 0.21m3 / t mined ore). Two third of the water
resource use is assumed to be ground water, the rest is surface water. No volume of mining effluents
was available, values for water demand were taken. Composition of mining effluents is approximated
with values taken from total emissions to water from base metal sites (s. Tab. 5.10). Transport from
mine to beneficiation were neglected, it is assumed to take place in direct vicinity of the mine. For die-
sel use the value of 69.2 kJ per tonne mined raw ore for underground mining given in Hilbrans &
Hinrichs (1999) was chosen.
Dust emissions from mining activities are comprised in the beneficiation step below.
Infrastructure of the mine is modelled with the unspecific module “mining infrastructure” from part II
“Non Ferrous Metal Winning – Auxiliary Processes”, chapter “Disposal of Tailings and Overburden”.
The electricity mix is approximated with the UCTE production mix. For the ore demand an overall

yield of 90 % is assumed, based on the 91 % primary yield in the nickel winning process reported in
Hilbrans & Hinrichs (1999).
Beneficiation
lime and flotation agents are added. The separated gangue is disposed in tailings ponds, the concen-
trated ore is fed to the metallurgy, which is assumed to be on site.
The infrastructure of beneficiation is included in mining infrastructure. Dust emissions comprise addi-
tionally the dust emissions due to mining activities. By approximation again electricity production mix
for the UCTE is chosen. 50 % of the reported energy consumption is assumed to be used to dry the
material, using natural gas, burned in industrial furnace >100kW.
Lime. The value for lime – geometric mean – is taken from copper mining where a range from 1 to 4
kg per tonne sulphidic ore is recommended (Krauss et al. (1999)). This value is confirmed by IPPC
(2002), Tab. 5.11.
Reagents. No information on mass and nature of the floating agents used is available. For copper min-
ing a range of 25 – 300 g collector and 25 – 250 g frothing reagent per ton sulphidic ore is recom-
mended (Krauss et al. (1999)). This is confirmed by the values in Tab. 5.11. Collector and frother are
balanced in ecoinvent in their production as unspecified organic chemical. According to NPI (2001)
xanthates hydrolytically decompose to 0.53 mg carbon disulphide per mg xanthate. With 189 mg xan-
thates from Tab. 5.11 this is 102.4 mg carbon disulphide per tonne mined ore.
cals, their emission to water is covered by reported emissions from mining sites shown in Tab. 5.10.
For pH adjustment and leaching the value for sulphuric acid was considered with 3.74 kg / t ore mined
(Tab. 5.11). As sulphuric acid is an by-product from the metallurgy, it is regarded as burden free.
Other acids are neglected.
ported data for total emissions shown in Tab. 5.10 are taken.
is used as leachate and as depressant for certain sulphidic minerals. The environmental fate of cyanide
compounds is complex and could not further be quarried. A cyanide recovery of 80 – 90% is reported
due to abatement and recovery in the Golden Cross mine in New Zealand (Anonymous (1998a)).
Here, a value of 80% is chosen for cyanide-abatement technology. With a reported use of 35 g NaCN
per tonne ore mined (Tab. 5.11), this means a direct emission of 3.88 g HCN per tonne ore mined. An
overview of the reagents accounted for is presented in Tab. 5.12.
Effluents. For the beneficiation effluents of the Russian production no data was available, the same
value like RSA are taken (0.35 m3 / t mined ore). Additional to direct emissions due to the use of
chemicals, total emissions to water are chosen according to Tab. 5.10. These values are from base
metal sites, thus only an approximation for the real effluent concentrations in the PGM production. It
is assumed here that the effluents also include acid rock drainage (ARD). ARD after mine closure is
modelled in the module “disposal, sulphidic tailings, off-site”.
Dust. A case study in (Anonymous (1998b)) about a projected open cut gold-copper mining and proc-
essing facility in Australia predict annual dust emissions between 0.18 kg/t and 0.46 kg/t of dust pro-
duced per tonne of ore mined. The prediction was made using the Emission Estimation Technique
Manual for Mining (NPI (2001)). The maximum value of 0.46 kg/t is chosen for total dust emissions.
Underground mining activities are assumed to generate 80 % of the dust emissions compared to open
pit mining (Tab. 5.8) yielding a 0.37 kg dust per t mined ore. According to NPI (2001) the dust com-

position is approximated with values in Tab. 5.7 the average earth’s crust. For PM10 values, the ratio
TDS : PM10 : PM2.5 was approximated with values from copper ore mining 101.72 : 50 : 5 g / t ore
taken from the CEMIP-database 11 , i. e. one tonne emitted dust corresponds to 491.5 kg PM10 and 49.2
kg PM2.5 respectively. The dust emissions resulting from underground activities are summed up in
Tab. 5.9. In Ecoinvent, for particulates differential classes are used adding up to the total dust emis-
sion, i. e. 49.2 kg PM<2.5, 442 kg PM2.5-10 and 508 kg PM>10 per tonne dust.
Tab. 5.7 Mineral born dust composition depending on rock. Data taken from (NPI (2001)). Values for earth’s crust are
Element Unit Symbol Basalt Granite This study

(Earth’s Crust)
Antimony g/t Sb 0.69 0.2 0.2
Arsenic g/t As 1.5 1.5 1.5
Beryllium g/t Be 0.3 5 2.6
Boron g/t Be 8 12 10
Cadmium g/t Cd 0.13 0.09 0.11
Chromium g/t Cr 200 4 100
Cobalt g/t Co 35 1 20
Copper g/t Cu 90 13 50
Fluorine g/t F 510 1400 950
Lead g/t Pb 3 24 14
Manganese g/t Mn 1500 400 950
Mercury g/t Hg 0.12 0.08 0.05
Nickel g/t Ni 150 0.5 80
Selenium g/t Se 0.05 0.05 0.05
Zinc g/t Zn 150 240 190
11
Tab. 5.8 Dust emissions from typical coal strip-mining operations (Anonymous (1998b)). After this table underground
mining cause roughly 80% of the dust emissions of open pit mining.
Sources of dust Truck and Dragline Occurrence in

shovel op- operation underground
eration mining
% of total dust emission
Dragline n.a 27 Yes
Haul roads
Overburden 35 - Yes
Coal 7 42 Yes
Loading
Overburden 12 - Yes
Coal 4 5 Yes
Drilling
Blasting
Truck dumping 7 5 Yes
Topsoil removal 18 10 No
Exposed areas 6 7 No
Haul-road repairs 5 1 Yes
Tab. 5.9 Emissions to air resulting from mining and beneficiation activities per tonne mined ore.
Unit Emission per tonne mined ore

Antimony mg / t 0.0736
Arsenic mg / t 0.552
Beryllium mg / t 0.96
Boron mg / t 3.68
Cadmium mg / t 0.04048
Chromium mg / t 36.8
Cobalt mg / t 7.36
Copper mg / t 18.4
Fluorine mg / t 349.6
Lead mg / t 5.152
Manganese mg / t 349.6
Mercury mg / t 0.0184
Nickel mg / t 29.44
Selenium mg / t 0.0184
Zinc mg / t 69.92
Particulate g/t 181
Matter <10µm
Particulate g/t 18.1
Matter <2.5µm

Tab. 5.10 Composition of effluents from mining activities, values are taken from total emissions to water from differ-
ent base metal sites in (IPPC (2002)).
Com- Unit Range This study Number of

pound (geometric Samples
mean)
Discharge Mio. m3 / a 0.08 - 22.9 3.65 6
Ca g / m3 686 686 1
SO4 g / m3 1750 - 3175 2357 2
COD g / m3 2.24 - 48.5 10.4 2
Solids g / m3 2.38 - 11.3 5.17 4
Al mg / m3 69.3 - 108 86.3 2
As mg / m3 0.264 - 14.1 2.95 3
Cd mg / m3 0.0901 - 1.02 0.316 4
Co mg / m3 0.742 - 0.823 0.781 2
Cr mg / m3 0.0311 - 9.62 0.547 2
Cu mg / m3 1.24 - 44.9 7.91 5
Fe mg / m3 61.7 - 1326 291 3
Mn mg / m3 24.7 24.7 1
Hg mg / m3 0.0155 - 0.12 0.0377 3
Ni mg / m3 0.792 - 1338 24.3 3
Pb mg / m3 0.0155 - 20 2.80 4
Zn mg / m3 5.37 – 225 75.9 5
N g / m3 2.50 - 1786 22.8 3
Tab. 5.11 Consumption of reagents in base metal mineral processing plants. Data taken from (IPPC (2002)).
Reagents Unit Range This stucy Number of

(geometric samples
mean)
MIBC, Dowfroth
Dithiophosphate
phate
Lime g/t 350 - 43685 2000 1) 6
1)
Value taken from (Krauss et al. (1999)), which correspond to the geometric mean of 2304 g / t


Carbon dioxide air emission inventoried in metallurgical step
Metallurgy and Separation of the non-ferrous metals

The ore concentrate is worked in the typical process steps. In the RSA the sulphidic matte is separated
from the slag with an electric arc furnace with a Søderberg electrode system. The ore is poorer in sul-
phur, resulting in a higher oxygen demand. Air and Oxygen is blown through the matte in a Pierce-
Smith-Converter. The cooled matte is ground and magnetically separated. After the separation the
PGM-containing part is leached and treated hydrometallurgically. After the electrolytic precipitation,
the remaining PGM containing dross is treated in the refinery. The transport to the refinery is assumed
to be short and therefore neglected.
In Russia the ore is treated in a Outokumpu flash furnace, where it is roasted and smelted in an auto-
genous process (i.e. the heat of the exothermal oxidation of FeS is used in the process, s. 5.3.1). The
matte is converted in a Pierce-Smith-Converter. The off-gases of the conversion are not treated. The
copper-nickel matte is separated in nickel and copper concentrate, latter being processed further in the
near copper facility. The nickel matte is treated with electrorefining (roasting and reduction with coke)
and cast into anodes. The remaining PGM containing dross is transferred to the refinery in Krasno-
jarsk.
In both processes lime was added in beneficiation which releases now fossil carbon dioxide. Flux is
assumed to consist of 50% lime, which also releases carbon dioxide. 0.75 t flux per t matte is reported
for copper in Krauss et al. (1999). Burning of 1 t lime (CaCO3) yields 0.44 t CO2.
Effluent is approximated with demand on water, which was only available for RSA and extrapolated
for RUS with 1.4 m3 per t concentrate input. The effluents are treated. An overview of concentrations
in different waste waters after treatment is shown in Tab. 5.14. The values are assembled from differ-
ent processes in the nickel and copper metallurgy (IPPC (2001)). The data shows high variations, a
geometric mean was built to calculate average emissions. To cover the uncertainties, a standard devia-
tion of 10 was chosen for each emission.
The atmospheric emissions are also taken from relative processes from copper and nickel operations
observed in IPPC (2001). These emissions are stated as mass per metal produced, therefore it is multi-
plied with the total metal production of the two considered processes. The distribution of particulate
matter is taken from CEPMEIP 12 . The values for sulphur dioxide are stated in Hochfeld (1997). Emis-
sions of sulphur dioxide are one key issue of the metallurgy. In IPPC (2001) much lower values are
reported, this seems to be true for European operations. Inco Ltd. states in their environmental report
12
http://www.air.sk/tno/cepmeip/.accessed 01.03.2003
Emissions of 1’500 kg SO2 / t metal produced in the period between 1997 and 2001. The correspond-
ing values in Hochfeld (1997) are 5’300 for the Russian and 760 kg SO2 / t for the RSA operations.
In Hochfeld (1997) no specific values for the separation step are reported except for energy, water,
dross and the coupled products nickel and copper.
Tab. 5.13 Emissions to air per tonne produced metal. Data was taken from similar processes in nickel (A) and copper
(B) metallurgy and multiplied with the sum of produced metals (PGM, Ni & Cu). Data is taken from IPPC
(2001), the distribution of the particle size is taken from CEPMEIP (C). VOC emission from leaching (D). In
this study values for SO2 emissions are taken from Hochfeld (1997) (E).

(this study)
Pb 0.03 B
Zn 0.10 B
Sn 0.007 B
Cu 0.04 B
As 0.006 B
Total dust 0.407 A
PM<2.5 0.000508 C
PM2.5-10 0.304 C
PM>10 0.102 C
SO2, RSA 760 E
SO2, RUS 5300 E
Co 0.03 A
VOC 0.20 A, D
TCDD 1.00E-08 B

Tab. 5.14 Composition of different wastewater effluents after treatment. A simple average from different reported val-
ues of European copper operations is chosen (IPPC (2001)). Values in the columns “process water”, “sur-
face run off” and “direct cooling water” represent geometric means of the range reported. Values for Cr, Hg
and Sn correspond to a copper semis production site, reported in (IPPC (2001)).
Sub- Process Surface run Direct cool- This study

stance Water off ing water (average)
3 3 3
[g / m ] [g / m ] [g / m ] [g / m3]
Cu 0.045 0.063 0.05 0.0527
(0.01-0.2) (0.01-0.4) (0.01-0.25)
Pb 0.006 0.032 0.01 0.0160
(0.001-0.04) (0.005-0.2) (0.001-0.1)
As 0.032 0.014 0.01 0.0187
(0.01-0.1) (0.003-0.07) (0.001-0.1)
Ni 0.024 0.028 0.011 0.0212
(0.004-0.15) (0.002-0.4) (0.002-0.06)
Cd 0.003 0.004 0.001 0.00273
(0.0001-0.1) (0.0002-0.1) (0.0001-0.003)
Zn 0.045 0.11 0.1 0.0848
(0.01-0.2) (0.03-0.4) (0.02-0.5)
Cr 0.029
Hg 0.00029
Sn 0.029
Refining of PGM
The refining step is covered only partially in Hochfeld (1997). Only for the operations in RSA elec-
tricity and gas consumption and water demand are reported. These values are added with specific
emissions from five large precious metal refineries from IPPC (2001) in Tab. 5.15.. Values for BOD is
based on the reported value for COD which is assumed to amount 150% of BOD (Zimmermann et al.
(1996)). Values for TOC are calculated using the standard procedure approximating the organic matter
with C6H12O6 (Frischknecht et al. (2003b)). Particle distribution is assessed with values for open
hearth exhausts from steel foundries after ESP control from EPA (1998). Transports are included in
emissions to air.
Precious metal refineries recover simultaneously several precious metals from a variety of raw materi-
als. No information about the way of allocation is available, an allocation by mass is assumed, i. e.
emissions of a tonne precious metal correspond to 1 tonne PGM, notwithstanding any possible co-
product.
For the operation in RUS the same burden is assumed. The PGM-containing matte has to be trans-
ported from Norsil’sk 1500 km up Yenissey river to Krasnojarsk for refining.

Tab. 5.15 Reported emissions to air and water from 5 large refinery processes. Data taken from IPPC (2001). Annual
operation hours assumed with 6000 h / a. Values for BOC, TOC and DOC calculated
Parameter Unit This study Range No of

(geometric sam-
mean) ples
Emissions to air
Production tonnes / a 940 102 - 2500 5
Dust kg / t PGM 16.4 2 - 127 5
PM<2.5 kg / t PGM 8.1 8.1 1
PM2.5-10 kg / t PGM 5.9 5.9 1
PM>10 kg / t PGM 2.4 2.4 1
NOx kg / t PGM 35.2 7 - 154 4
SO2 kg / t PGM 18.6 3.1 - 232 3
Emissions to water
Ag g / t PGM 3.7 2.78 - 4.8 2
Pb g / t PGM 37 3.51 - 12353 5
Hg g / t PGM 0.8 0.05 - 12 4
Cu g / t PGM 141.5 8.35 - 26961 5
Ni g / t PGM 65.0 4.8 - 16078 5
COD kg / t PGM 25.8 11.1 - 60 2
BOD kg / t PGM 17.2 calc.
TOC kg / t PGM 10.1 calc.
DOC kg / t PGM 10.1 calc.
Data quality
The uncertainties are assessed using the pedigree matrix (see Frischknecht et al. (2003b)). For detailed
results compare the different flows and the ecoinvent meta information.
mated due to different system boundaries of reported effluent volume from Hochfeld (1997) and total
effluent emissions reported in IPPC (2001).
Beneficiation. Like in mining, the emissions to water can be overestimated. The emissions from tail-
ings are assessed in an extern module, which itself contains rough assumptions of a global overburden
and tailings composition and a simplified disposal scenario. Emissions tough can vary considerably
from case to case, it is not evident, whether the calculated emissions meet the reality.
Metallurgy. Process data is satisfactory and valid for over 90 % of the PGM. The main emissions
dominating possibly the whole PGM process is sulphur dioxide, which is reported for both processes
and found to be consistent with other data Krauss et al. (1999), IPPC (2001) and Hilbrans & Hinrichs
(1999). Poor data are available on composition and fate of solid waste, an overestimation may occur
due to higher recycling and reuse rates than assumed in this work.
Separation of the non-ferrous metals. Main data on energy and mass fluxes is satisfactory, while
data on direct emissions due to hydrometallurgical processes is poor. Namely chlorine emissions can
be considerable, generating both HCl emissions to air and AOX emissions to water – emissions which
are not assessed in this work due to lacking data.
Refining. No data on many specific emissions in to the air was available. Namely high-impact sol-
vents and complexants are not registered. Due to the closed-loop strategies of such refineries, only
small emissions of problematic substances are expected.

Source for mean

Modul name in
Process Name
Deviation 95%
Infrastructure
Subcategory
ecoinvent
Comment
Category
Standard
Remarks
Location
General
Output
value
Input
Type
Unit
Mean value ZA Mean value RU
Pt, Pt 4.8E-4%, Pd
90% yield assumed, RU: 89'750 kg
2.0E-4%, Rh 2.4E-5%, (Krauss, 1999),
Ore (Pt)  ore per kg PGM; ZA: 265'000 kg resource in ground 7.53E-01 7.53E-01 kg 1 1.66 reported values
Ni 3.7E-2%, Cu 5.2E- (Hilbrans, 1999)
ore per kg PGM
2% in ore, in ground
Pd, Pd 2.0E-4%, Pt
4.8E-4%, Rh 2.4E-5%, (Krauss, 1999),
Ore (Pd)  ore per kg PGM; ZA: 265'000 kg resource in ground 3.20E-01 3.20E-01 kg 1 3.90 reported values
Ni 3.7E-2%, Cu 5.2E- (Hilbrans, 1999)
ore per kg PGM
Rh, Rh 2.4E-5%, Pt
ore is allocated by mass to 90% yield assumed, RU: 89'750 kg
4.8E-4%, Pd 2.0E-4%, (Krauss, 1999),
Ore (Rh)  the coupled products Pt, Pd, ore per kg PGM; ZA: 265'000 kg resource in ground 3.77E-02 3.77E-02 kg 1 33.19 reported values
Ni 3.7E-2%, Cu 5.2E- (Hilbrans, 1999)
platinum group metal production, primary (mining)

Rh, Ni and Cu ore per kg PGM
Ni, Ni 3.7E-2%, Pt
4.8E-4%, Pd 2.0E-4%, (Krauss, 1999),
Ore (Ni)  ore per kg PGM; ZA: 265'000 kg resource in ground 1.64E+02 1.64E+02 kg 1 0.01 reported values
Rh 2.4E-5%, Cu 5.2E- (Hilbrans, 1999)
ore per kg PGM
Cu, Cu 5.2E-2%, Pt
4.8E-4%, Pd 2.0E-4%, (Krauss, 1999),
Ore (Cu)  ore per kg PGM; ZA: 265'000 kg resource in ground 1.15E+02 1.15E+02 kg 1 0.01 reported values
Rh 2.4E-5%, Ni 3.7E- (Hilbrans, 1999)
ore per kg PGM
The supply mix of the mining

The rest of the inventory is electricity, medium
Electricity, operations are assumed to be
 allocated by revenue (s. electricity production mix No UCTE voltage, production 9.81E+03 7.48E+03 kWh (Hochfeld, 1997) 1 1.26 (3,2,1,1,3,4,2)
mining similar to the electricity mix of the
chapter Allocation) as stated. UCTE, at grid
aluminium industry
Water,
mining,  resource in water Water, well, in ground 3.71E+01 1.26E+01 m3 (Hochfeld, 1997) 1 1.26 (3,2,1,1,3,4,12)
ground
Water,
 resource in water Water, river 1.86E+01 6.28E+00 m3 (Hochfeld, 1997) 1 1.26 (3,2,1,1,3,4,12)
mining, fresh
construction civil
Blasting  No RER blasting 1.33E-01 4.49E-02 kg (Hochfeld, 1997) 1 1.26 (3,2,1,1,3,4,1)
processes engineering

Diesel  machinery No GLO 1.95E+04 6.90E+03 MJ (Hochfeld, 1997) 1 2.33 (4,3,2,2,4,5,5)
infrastructure, construction
 machinery Yes RER conveyor belt, at plant 5.65E-03 1.99E-03 m (Hochfeld, 1997) 1 3.93 (5,4,3,3,5,6,9)
conveyor belt processes
Mine, 50 a lifetime; 400'000 t / a -> 5e-11 non-ferrous metal,

 metals extraction Yes GLO 4.99E-06 1.41E-05 unit rough estimate 1 3.07 (3,2,1,1,3,4,9)
Infrastructure units per kg ore mined mine, underground
not clear, wether net disposal or
Overburden, waste residual disposal, sulfidic
 partly re-filled in mine. Assumption: No GLO 8.75E+03 4.49E+03 kg (IPPC, 2002) 1 1.13 (2,2,1,1,1,4,6)
disposed management material landfill tailings, off-site
50% refilled.
not clear, wether net disposal or
Overburden,
 partly re-filled in mine. Assumption: 8.75E+03 4.49E+03 kg (IPPC, 2002) 1 1.13 (2,2,1,1,1,4,6)
refilled 50% refilled.
For effluent composition see Tab.

Effluents, 5.14. Due to high heterogenity of
 5.57E+01 1.88E+01 m3 (IPPC, 2002) 1 1.13 (2,2,1,1,1,4,6)
mining effluent, a standard deviation of 10
is assumed for concentrations
Fig. 5.11 Flows for “PGM-Production, primary – Part mining” and their representation in the ecoinvent database. Values in column “Mean value RUS” correspond to the Russian PGM
production, the values in the column “Mean value RSA” correspond to the South African PGM production. Uncertainty assumptions are identical for both modules. Values re-
late to the production of 1 kg PGM.
Genera l Flow Information Re prese nta tion in ecoinvent Uncerta inty Informations
Deviation 95%
Modul name in
Process Name
Subcategory
Infrastructure
mean value
Source for
ecoinvent
Comment
Standard
Category
Location
Remarks
Output
General
M e an value M e an value
Input
Type
Unit
ZA RU
Inventory is allocated by electricity, medium

Electricity, production
 revenue (s. chapter electricity No UCTE voltage, production 4.02E+03 1.18E+03 k Wh (Hochf eld, 1997) 1 1.13 (2,2,1,1,1,4,2)
benefication mix
Allocation) as stated. UCTE, at grid
Water, 0.35 m3 per t ore mined is
 resource in w ater Water, river 9.36E+01 3.17E+01 m3 (Hochf eld, 1997) 1 1.13 (2,2,1,1,1,4,12)
benefication assumed
construction limestone, milled,
Lime addition  others No CH 5.30E+02 1.79E+02 kg (Krauss, 1999) 1 2.00 reported values
platinum group metal production, primary

Floating agents, chemicals organic, at
 chemicals organics No GLO 5.98E+01 2.03E+01 kg (IPPC, 2002) 1 2.00 reported values
organic plant
Floating agents, chemicals inorganic,
 chemicals inorganics No GLO 2.04E+02 6.91E+01 kg (IPPC, 2002) 1 2.00 reported values
inorganic at plant
hydrogen cyanide, at
Sodium Cyanide  chemicals organics No RER 9.33E+00 3.16E+00 kg (IPPC, 2002) 1 2.00 reported values
plant
burden-free, since byproduct
(beneficiation)
Sulphuric A cid  6.11E+02 2.07E+02 kg (IPPC, 2002) 1 2.00 reported values

from metallurgy
Assumed to equal the demand of
Ef fluents,
 w ater. Composition is given in 9.36E+01 3.17E+01 m3 Rough estimation
benef ication
Tab. 5.10
(4,2,2,3,4,4,33);
Emission From sodium cyanide, 80% including stdev
 w ater river Cyanide 1.03E+00 3.48E-01 kg (NPI, 2001) 1 3.84
Cyanide control of sodium
cyanide
including emissions f rom mining
activities. Dust is the sum of
particulate emissions
Dust (air,
 represented in fractions PM<2.5, 1.04E+02 3.67E+01 kg (NPI, 2001) 1 1.59 (3,2,1,1,3,4,25)
benef ication)
PM2.5-10, PM>10, including
metals and salts. Composition is
given in Tab. 5.9.
low
due to hydrolysation of
 carbon disulf ide air population Carbon disulf ide 2.65E+01 8.97E+00 kg (NPI, 2001) 1 2.31 (4,2,2,5,4,4,23)
xanthates
density
not clear, w ether net disposal or residual
w aste disposal, sulf idic
 Tailings partly re-f illed in mine. Here material No GLO 1.28E+05 3.94E+04 kg (Hochf eld, 1997) 1 1.13 (2,2,2,3,1,4,6)
management tailings, of f -site
assumed as 50% disposal landfill
Fig. 5.12 Flows for “PGM-Production, primary – Part beneficiation” and their representation in the ecoinvent database. Values in column “Mean value RU” correspond to the Russian
PGM production, the values in the column “Mean value ZA” correspond to the South African PGM production. Uncertainty assumptions are identical for both modules. Val-
ues relate to the production of 1 kg PGM.

Deviation 95%
Modul name in
Process Name
Subcategory
Infrastructure
mean value
Source for
ecoinvent
Comment
Standard
Category
Location
Remarks
Output
General
M e an value M e an value
Input
Type
Unit
ZA RU

Electricity, production
 revenue (s. chapter electricity No UCTE voltage, production 6.00E+03 1.16E+04 k Wh (Hochfeld, 1997) 1 1.13 (2,2,2,3,1,4,2)
metallurgy mix
A llocation) UCTE, at grid
natural gas, burned
Gas, platinum group metal production, primary (metallurgy) heating
 natural gas No RER in industrial furnace 0.00E+00 1.25E+05 MJ (Hochfeld, 1997) 1 1.13 (2,2,2,3,1,4,1)
metallurgy systems
>100kW
water,
 resource in w ater Water, river 1.28E+01 7.14E+01 m3 (Hochfeld, 1997) 1 1.13 (2,2,2,3,1,4,12)
metallurgy
chemicals inorganic,
Flux  chemicals inorganics No GLO 1.88E+02 4.14E+03 kg (Hochfeld, 1997) 1 1.26 (3,2,1,1,3,4,3)
at plant
infrastructure non-ferrous metal,
 metals extraction Y es GLO 2.60E-08 1.45E-07 unit assumption 1 3.30 (4,2,1,5,4,4,9)
smeltery smelter
Carbondioxid (4,2,1,1,4,4,14);
 e, from limestone added in benef iciation air unspecified Carbon dioxide, fossil 2.33E+02 7.90E+01 kg (Hochfeld, 1997) 1 1.58 excluding stdev
limestone of lime addition
Carbondioxid
 air unspecified Carbon dioxide, fossil 4.14E+01 9.10E+02 kg (Hochfeld, 1997) 1 1.58 (4,2,2,3,4,4,14)
e, from flux
metallurgy
 composition in Tab 5.14 1.28E+01 7.14E+01 m3 (Hochfeld, 1997) 1 1.58 (4,2,2,3,4,4,6)
effluent
Sulfur dioxide low
 (air, air population Sulfur dioxide 1.91E+02 2.95E+04 kg (Hochfeld, 1997) 1 1.38 (4,2,2,1,3,5,15)
metallurgy) density
Dust (air,
 Composition in Tab 5.13 1.02E-01 2.24E+00 kg (IPPC, 2002) 1 1.59 (3,2,1,1,3,4,25)
metallurgy)
NMV OC, non-
low
VOC (air, methane volatile
 in dust, composition in Tab 5.13 air population 5.01E-02 1.10E+00 kg (IPPC, 2002) 1 1.59 (3,2,1,1,3,4,16)
metallurgy) organic compounds,
density
unspecif ied origin
low
TCDD (air, 2,3,7,8-
 in dust, composition in Tab 5.13 air population 2.51E-09 5.52E-08 kg (IPPC, 2002) 1 3.07 (3,2,1,1,3,4,21)
metallurgy) tetrachlorodibenzo-p-
density
dioxin
disposal, nickel
residual
value f or RU including dross w aste smelter slag, 0%
 slag material No CH 6.40E+02 4.14E+04 kg (Hochfeld, 1997) 1 1.13 (2,2,2,3,1,4,6)
f rom follow ing separation step. management w ater, to residual
landf ill
material landfill
Fig. 5.13 Flows for “PGM-Production, primary – Part metallurgy” and their representation in the ecoinvent database. Values in column “Mean value RU” correspond to the Russian

Source for mean

Modul name in
Process Name
Deviation 95%
Infrastructure
Subcategory
ecoinvent
Comment
Category
Standard
Remarks
Location
General
Output
value
Input
Type
Unit
Mean value
production, primary (separation

Electricity,
 revenue (s. chapter electricity production mix No UCTE voltage, production 1.59E+03 5.56E+03 kWh (Hochfeld, 1997) 1 1.13 (2,2,2,3,1,4,2)
separation Allocation) UCTE, at grid
from non-ferrous metals) natural gas, burned in

platinum group metal
heating
Gas, separation  natural gas No RER industrial furnace 3.98E+03 0.00E+00 MJ (Hochfeld, 1997) 1 1.13 (2,2,2,3,1,4,1)
systems
>100kW
water,
 resource in water Water, river 1.28E+01 8.27E+02 m3 (Hochfeld, 1997) 1 1.13 (2,2,2,3,1,4,6)
separation
disposal, nickel
waste residual smelter slag, 0%
 dross No CH 3.45E+00 6.35E+01 kg (Hochfeld, 1997) 1 1.26 (3,2,1,1,3,4,6)
material landfill
As co-product, allocation by
nickel, primary, from
nickel as co- revenue: ZA 2%, RU 47%,
 metals extraction No platinum group metal 1.47E+02 2.31E+03 kg (Hochfeld, 1997) 0 0.00 0
product geography corresponding to
production
respective origin (ZA / RU)
As co-product, allocation by
copper, primary, from
copper as co- revenue: ZA 1%, RU 19%,
 metals extraction No platinum group metal 1.04E+02 3.21E+03 kg (Hochfeld, 1997) 0 0.00 0
product geography corresponding to
production
respective origin (ZA / RU)
Fig. 5.14 Flows for “PGM-Production, primary – Part separation” and their representation in the ecoinvent database. Values in column “Mean value RU” correspond to the Russian

Source for mean

Modul name in
Process Name
Deviation 95%
Infrastructure
Subcategory
ecoinvent
Comment
Category
Standard
Remarks
Location
General
Output
value
Input
Type
Unit
Mean value
electricity, medium
Electricity,
 electricity production mix No UCTE voltage, production 4.59E+02 4.59E+02 kWh (Hochfeld, 1997) 1 1.25 (2,2,2,1,3,4,2)
refining Total allocation to PGM. UCTE, at grid
Pt:72% - Pd:20% - Rh:8% natural gas, burned in
heating
Gas, refining  natural gas
systems
No RER industrial furnace 7.76E+02 7.76E+02 MJ (Hochfeld, 1997) 1 1.25 (2,2,2,1,3,4,1)
>100kW
Transportatio only RU, 1500 km riverboat from
 platinum group metal production, primary (refining) - 6.35E+04 (Hochfeld, 1997)
n, river noril'sl to krasnojarsk
infrastructure chemical plant,
 chemicals organics Yes RER 4.00E-10 4.00E-10 unit assumption 1 3.30 (4,2,1,5,4,4,9)
refinery organics
low population (EPA, 1998),

 PM<2.5 air Particulates, > 10 um 8.09E-03 8.09E-03 kg 1 2.08 (3,2,2,5,3,2,26)
density (IPPC, 2001)
low population Particulates, > 2.5 um, (EPA, 1998),

 PM2.5-10 air 5.93E-03 5.93E-03 kg 1 3.07 (3,2,2,5,3,2,27)
density and < 10um (IPPC, 2001)
low population (EPA, 1998),

 PM>10 air Particulates, < 2.5 um 2.40E-03 2.40E-03 kg 1 1.34 (3,2,2,5,3,2,28)
density (IPPC, 2001)
low population
 Nox air Nitrogen oxides 3.52E-02 3.52E-02 kg (IPPC, 2001) 1 1.59 (3,2,1,1,3,4,16)
density
low population
 SO2 air Sulfur dioxide 1.86E-02 1.86E-02 kg (IPPC, 2001) 1 1.26 (3,2,1,1,3,4,15)
density
 Pb water river Lead 3.73E-05 3.73E-05 kg (IPPC, 2001) 1 5.08 (3,2,1,1,3,4,35)

 Hg water river Mercury 7.83E-07 7.83E-07 kg (IPPC, 2001) 1 5.08 (3,2,1,1,3,4,35)
 Cu water river Copper, ion 1.41E-04 1.41E-04 kg (IPPC, 2001) 1 5.08 (3,2,1,1,3,4,35)
 Ni water river Nickel, ion 6.50E-05 6.50E-05 kg (IPPC, 2001) 1 5.08 (3,2,1,1,3,4,35)
COD, Chemical
 COD water river 2.58E-02 2.58E-02 kg (IPPC, 2001) 1 1.59 (3,2,1,1,3,4,32)
Oxygen Demand
BOD5, Biological assumption COD
 BOD water river 1.72E-02 1.72E-02 kg 1 1.84 (4,2,2,3,4,4,32)
Oxygen Demand = 1.5 * BOD
calculated using
TOC, Total Organic
 TOC water river 1.01E-02 1.01E-02 kg standard 1 1.59 (3,2,1,1,3,4,32)
Carbon
procedure
DOC, Dissolved Assumption DOC
 DOC water river 1.01E-02 1.01E-02 kg 1 1.59 (3,2,1,1,3,4,32)
Organic Carbon = TOC
as co-product, allocation by platinum, primary, at
 Platinum metals extraction No ZA 6.78E-01 2.50E-01 kg (Hochfeld, 1997)
revenue: ZA 66%, RU 11% refinery
as co-product, allocation by palladium, primary, at
 Palladium metals extraction No ZA 2.88E-01 7.30E-01 kg (Hochfeld, 1997)
as co-product, allocation by rhodium, primary, at
 Rhodium metals extraction No ZA 3.39E-02 2.00E-02 kg (Hochfeld, 1997)
Fig. 5.15 Flows for “PGM-Production, primary – Part refining” and their representation in the ecoinvent database. Values in column “Mean value RU” correspond to the Russian PGM
production, the values in the column “Mean value ZA” correspond to the South African PGM production. Uncertainty assumptions are identical for both modules. Values re-
late to the production of 1 kg PGM.

5.4 Winning of Secondary PGM

As shown in chapter 2, considerable amounts of PGMs originate from secondary production. Since the
intended module is a consumer-mix valid for European consumers, the secondary production of PGM
represents the collection, beneficiation and refining of PGM containing scrap in the European Region.
The process in concern is mainly based on (Hochfeld (1997)) and describes the secondary production
in Germany, which will stand for the whole European region. Further more only data on recycling of
automotive catalysts are available. Since it can be assumed, that by far the most part of PGM recycling
origins from recovered automotive catalysts, this study focuses solely on recycling of automotive cata-
lysts and neglect other sources of secondary PGMs.

The production process can mainly be described in the three steps: Collection, Beneficiation and Re-
fining. Information on production process is taken from (Hagelücken (2001a)).
Collection. Spoiled automotive catalysts are bought by specialised enterprises from different origins.
Part of it originates from scrap dealer recycling end-of-life cars. Further more during the cars operat-
ing phase, defective catalysts are exchanged in garages. The third sources is the production waste, i. e.
defective catalysts which do not reach market. In most cases, there are fix agreements between the dif-
ferent supplier and the intermediary trade. Although the trade with PGM containing scrap is risky.
Motor car manufacturer built up their own internal recycling systems with their contractor garages and
gained access to exchanged catalysts.
Beneficiation. The catalysts are dismantled and then sold to refining companies. Refineries too have
long-term contracts with the intermediary trade.
Refinery. The scrap first is shredded and then hydrometallurgically or pyrometallurgical processed.
The direct hydrometallurgical path dissolves the scrap with strong acids for following precipitation.
This has the handicap of effluents and larger amounts of dross and PGM losses. Therefore it is rarely
used nowadays. In the pyrometallurgical process, the scrap is smelted and the ceramic wafer is
slagged. The PGMs are concentrated in the collector metal, usually copper. Low-content PGM scrap is
often smelted together with other non-ferrous metal matte. This is cheap, but effects a high loss in
PGM. The collector matte from the furnace then is treated hydrometallurgically by re-precipitation as
described in 5.3.1. In this step usually production scrap from other industries (glass, chemical labora-
tories) is joined. The overall PGM-yield is 98 % for platinum and 85 % for rhodium – no data on re-
covery of palladium was available.
5.4.2 Emissions and Waste

Air
Emissions during collection and beneficiation are gases from transportation and dust from dismantling.
Also in this step the combustion leads to emissions of SO2. Nitrogen oxides are produced in significant
amounts during acid digestion using nitric acid. Chlorine and HCl can be formed during a number of
digestion processes. Chlorine is used extensively in the dissolution stages using hydrochloric acid and
chlorine mixtures respectively. Dust and metals are generally emitted from incinerators and furnaces.
VOC can be emitted from solvent extraction processes, while organic compounds, namely dioxins, can
be emitted from smelting stages resulting from the poor combustion of oil and plastic in the feed mate-
rial. All these emissions are subject to abatement technologies and controlling.

Water
During collection and beneficiation no serious emissions are reported. Effluents from refining contain
considerable amounts of metals and organic substances.
Waste
Solid residues from pyrometallurgical step are usually re-used in copper facilities, final residues gen-
erally comprise hydroxide filter cakes.
5.4.3 Secondary PGM in ecoinvent

This module in ecoinvent describes the collecting of automotive catalysts in Germany, the dismantling
of the catalysts and the pyrometallurgical processing followed by a hydrometallurgical purification
step. The data is mainly taken from Hochfeld (1997) and IPPC (2001).
The scrap is assumed to be transported over 600 km by lorry, further emissions are neglected. The ori-
gin of the automotive catalysts is burden-free. For dismantling an average electricity-input of 50
kWh/t is assumed in Hochfeld (1997).
The metallurgical step consists of an arc-furnace. Values for electricity-input and lime and copper ad-
dition are reported in Hochfeld (1997) for one big German enterprise, but no direct specific emissions.
These values are taken from IPPC (2001), representing the same refining-step as in the primary pro-
duction. Data on the hydrometallurgical electrolysis still lacks and is assessed with data on refining
from the module “Primary Production of PGM”. The emissions from Tab. 5.15 used there and the re-
spective energy use per tonne PGM from Hochfeld (1997) are also chosen for secondary production.
The layout of the modules with general flow information, remarks, sources, values uncertainty infor-
mation are shown in Fig. 5.16. An overview over the meta information is given in Tab. 5.16.
Data quality
The uncertainties are assessed using the pedigree matrix (see Frischknecht et al. (2003b)). For detailed
results compare the different flows and the ecoinvent meta information.
Based on the literature no satisfactory process data was available. The hydrometallurgical step is ap-
proximated with emissions and energy use of the refining in the primary production.

Tab. 5.16 Ecoinvent meta information for the multi-output processes in the secondary production of PGM in Europe
from recycled automotive catalysts.
Name platinum group metal production, secondary

Location RER
InfrastructureProcess 0
Unit kg
Dataset Version 2.0
IncludedProcesses The module includes a collection step, with the transportation of the
scrap to the intermediary dealer, a beneficiation step and a metallurgi-
cal step. Production, application and emissions of the most important
agents are also included.
Amount 1
LocalName Platingruppenmetalle-Produktion, sekundär
Synonyms Platingruppenmetall//platinumgroup metal
GeneralComment This module describes the collecting of autocatalysts in Germany, the
dismantling of the catalysts and the pyrometallurgical processing fol-
lowed by a hydrometallurgical purification step. The multioutput-
process “PGM-Production, secondary, RER” delivers the co-products
“platinum, secondary, at refinery”, “palladium, secondary, at refinery ”
and “ rhodium, secondary, at refinery”. The smelter slag disposal is
inventoried as disposal of inert material, thus neglecting potentially im-
portant (but unknown) heavy metal emissions.
StartDate 1995
EndDate 2002
Data valid for entire pe-
riod 1
Geography Text Data for Germany are used in this module for Europe. Some agents
used like lime or transports originate from data sets that are valid only
for Switzerland.
Technology Text After collection and transportation to the intermediary dealer, the scrap
is ground in a mill. The resulting material is fed to specialised refiner-
ies. The metallurgical step consists of an arc-furnace. Same refining
process as in primary production assumed (selective precipitation)
Representativeness 4.6%
ProductionVolume 5.3 t / a
SamplingProcedure Literature
UncertaintyAdjustments none

Deviation 95%
Modul name in
Process Name
Subcategory
Infrastructure
mean value
Source for
ecoinvent
Comment
Standard
Category
Location
Remarks
Output
General
Input
Type
Unit
M e an value
Not inventoried, assumed to be

spoiled
 burden free. Average content of 400.00 kg 1 1.08 (1,2,2,1,1,3,12)
converter PGM: 0.25%
transport to transport
 assumption: 600 km transport road No RER transport, lorry 32t 240.00 tk m 1 2.01 (1,2,2,1,1,3,5)
platinum group metal production,
benefication systems
electricity, medium
Electricity production
 electricity No UCTE voltage, production 20.00 k Wh 1 1.29 (3,2,1,5,3,4,2)
(grinding) mix
UCTE, at grid
electricity, medium
Electricity production
 electricity No UCTE voltage, production 600.00 k Wh 1 1.29 (3,2,1,5,3,4,2)
(smelting) mix
UCTE, at grid
Copper, loss copper, at regional
 metals extraction No RER 1.00 kg 1 1.60 (4,2,1,5,4,4,4)
storage
secondary
to slag
construction lime, hydrated, loose,
lime  binder No CH 80.00 kg 1 1.60 (4,2,1,5,4,4,4)
materials at plant
 CO2 due to lime addition air unspecif ied Carbon dioxide, f ossil 35.20 kg 1 1.60 (4,2,1,5,4,4,14)
disposal, inert w aste,

 Slag No 5% w ater, to inert 445.80 kg (Hochfeld, 1997) 1 1.29 (3,2,1,5,3,4,6)
material landf ill
Ratio reflecting the reported

Platinum, platinum, secondary,
 average PGM recycling in RER. metals extraction No RER 0.53 kg
secondary at ref inery
Allocation of 45% of the burden.
Ratio reflecting the reported palladium,

Platinum,
 average PGM recycling in RER. metals extraction No RER secondary, at 0.26 kg
secondary Allocation of 15% of the burden. ref inery
Ratio reflecting the reported

Platinum, rhodium, secondary,
 average PGM recycling in RER. metals extraction No RER 0.21 kg
secondary at ref inery
Allocation of 39% of the burden.
Fig. 5.16 Flows for “PGM-production, secondary, RER ” and its representation in the ecoinvent database

6 Literature
6 Literature
www.ecoinvent.ch.
http://www.ea.gov.au/industry/sustainable/mining/booklets/dust/.
EPA (1998) EPA (1998) Stationary Point and Area Sources. In: Compilation of Air Pollutant
Emission Factors, AP-42, Vol. 1. Fifth Edition, Online-Version under:
http://www.epa.gov/ttn/chief/ap42/index.html.
Frischknecht et al. (2003a) Frischknecht R., Jungbluth N., Althaus H.-J., Doka G., Dones R., Hellweg S.,
Hischier R., Humbert S., Margni M., Nemecek T. and Spielmann M. (2003a)
Implementation of Life Cycle Impact Assessment Methods. Final report ecoin-
vent 2000 No. 3. Swiss Centre for Life Cycle Inventories, Dübendorf, CH,
Frischknecht et al. (2003b) Frischknecht R., Jungbluth N., Althaus H.-J., Doka G., Dones R., Hellweg S.,
Hischier R., Nemecek T., Rebitzer G. and Spielmann M. (2003b) Overview and
Hagelücken (2001a) Hagelücken C. (2001a) Der Kreislauf der Platinmetalle - Recycling von Kataly-
satoren. In: Autoabgaskatalysatoren, Vol. 612. pp. 95-115. Expert Verlag, Ren-
ningen. Online-Version under: http://www.dmc-
2.de/pmc_eng/Veroeffentlichungen_2/Der%20Kreislauf%20der%20Platinmetall
e%20-%20Recycling%20von%20Katalysatoren.pdf.
Hagelücken (2001b) Hagelücken C. (2001b) Die Märkte der Katalysatormetalle Platin, Palladium und
Rhodium. In: Autoabgaskatalysatoren, Vol. 612. pp. 95-115. Expert Verlag,
Renningen. Online-Version under: http://www.dmc-
2.de/pmc_eng/Veroeffentlichungen_2/Die%20M%C3%A4rkte%20der%20Katal
ysatormetalle%20Pt%20Pd%20Rh.pdf.
European Commission. Retrieved from
http://www.jrc.es/pub/english.cgi/0/733169

6 Literature

JohnsonMatthey (2003) JohnsonMatthey (2003) Platinum 2002 Interim Review at 01.03.2003 from
http://www.platinum.matthey.com/publications/pdf2002InterimReport/Int02-
FullDoc.pdf
Renner (1997) Renner H. (1997) Platinum Group Metals and Compounds. In: Ullmann's ency-
clopedia of industrial chemistry (ed. Anonymous). 5th edition on CD-ROM Edi-
tion. Wiley & Sons, London.
Zimmermann et al. (1996) Zimmermann P., Doka G., Huber F., Labhardt A. and Ménard M. (1996) Ökoin-
ventare für Entsorgungsprozesse; Grundlagen zur Integration der Entsorgung in
Ökobilanzen. Vol. ESU-Reihe Nr. 1/96. Eidg. Technische Hochschule (ETH),
Zürich.

Part VI
Chromium
Data v2.1 (2009)

Summary

ported in Frischknecht et al. (2006). The changes were implemented with v1.2 which corrected the amount of
resource required in the chromite dataset by factor two. Additionally to the reported errors the emissions of sol-
ids, PAH and chromium to water had been corrected by a factor 1E-3 in the dataset “ferrochromium, at plant” as
compared to the data in version 1.3.
The flow names in the tables displaying unit process data have not been updated consistently to reflect the name
changes with version 2.0.
Summary
This part describes the production of chromite, ferrochromium and metallic chromium from cradle to consumer
in Europe. Chromite is used for the production of chemicals and refractories. Ferrochromium is used as alloy for
steel and chromium is mainly used for special iron free alloys (superalloys), e.g. for high-temperature exposed
components in chemical and petrochemical industries.
steps.
Since no chromium is recycled within the chromium industry but directly to the foundries of the stainless steel
industry, no secondary chromium is considered in this study.

Table of Contents
Table of Contents
SUMMARY ......................................................................................................................I
1 INTRODUCTION ....................................................................................................... 1
2 RESERVES AND RESOURCES OF CHROMIUM ............................................................. 2

2.1 Primary Chromium................................................................................................................ 2
2.2 Secondary Chromium............................................................................................................ 2
3 CHARACTERISATION OF THE MATERIALS .................................................................. 4

3.1 Chromite ore.......................................................................................................................... 4
3.2 Chromium Metal ................................................................................................................... 4
3.3 Ferrochromium...................................................................................................................... 4

4.1 Chromite ore.......................................................................................................................... 7
4.2 Chromium Chemicals............................................................................................................ 8
4.3 Chromium Metal ................................................................................................................... 8
4.4 Ferrochromium...................................................................................................................... 9
4.5 Chromium Alloys .................................................................................................................. 9

5.1 Exploitation of chromite...................................................................................................... 15
5.1.2 Emissions & Waste ............................................................................................................. 15
5.1.3 “chromite, at beneficiation” in ecoinvent............................................................................ 15
5.2 Winning of high-carbon ferrochromium ............................................................................. 21
5.2.2 Emission.............................................................................................................................. 22
5.2.3 Waste................................................................................................................................... 22
5.2.4 Ferrochromium, high-carbon, 68% Cr, at plant in ecoinvent.............................................. 22
5.2.5 Ferrochromium, high-carbon, 68% Cr, at regional storage in ecoinvent ............................ 28
5.3 Production of metallic chromium........................................................................................ 30
5.3.1 Emissions & Wastes............................................................................................................ 30
5.3.2 “Chromium, at regional storage” in Ecoinvent ................................................................... 31
6 LITERATURE ......................................................................................................... 34

1. Introduction
1 Introduction
This part describes the production of ferrochromium and metallic chromium from cradle to consumer
in Europe. Inventory data for the ferrochromium production is taken mainly from Adelhardt & Antre-
kowitsch (1998) together with background data from IPPC (2002), IPPC (2001), Papp (1994) and
EEA (1999).
Like many metals used as alloying element, chromium is recycled from steel-scrap in foundries di-
rectly as secondary alloys. Depending on their composition, these secondary chromium containing ma-
terials are used together with ferrochromium as secondary source for chromium. In this study, only the
primary material ferrochromium and metallic chromium is considered. Secondary chromium sources –
i.e. alloys from recycled steel scrap – have to be inventoried in the respective production module, e.g.
in the stainless steel production.
The main environmental impact by producing refractory metals is metal containing dust and hard
metal powder as well as fume from the smelting processes for instance by producing chromium metal.
Dust emissions occur from storage, handling of raw materials and products and the furnace operation,
where both stack and fugitive emissions play an important role. Hydrogen fluoride, which is used in
several processes, is highly toxic and needs therefore to be handled very carefully to avoid any health
problems for the plant staff. Another environmental impact of the production of hard metals is the tox-
icity of metal compounds such as chromium.
Over the last 15 to 20 years, the world pattern of the ferro-alloy market has deeply changed:
- The consumption of the developing countries has vastly increased with the development of their
steel production;
- Their production has even more increased because they have taken a growing share of the tradi-
tional markets of the industrialised countries, where the steel production was stagnant or growing
at a slow rate.
- The ferro-alloy industry is facing a growing proportion of imports, at first from the new industrial-
ised countries and in recent years from the countries of East Europe and the CIS (former Soviet
Union).
As a consequence, EU ferro-alloy production has been submitted to a difficult competition that results
in a decreasing trend of the total amount of ferro-alloys produced. Although the last years showed
some stabilisation, even a slightly increase of the production, the European industry remains very sen-
sitive.

2. Reserves and Resources of Chromium
2 Reserves and Resources of Chromium

cle. When recovered the properties are theoretically identical to those in the raw material. The metal
can be re-used in another product life cycle, provided that its concentration makes a recovery eco-
nomic. The recovered metal is referred to as secondary metal, whereas the metal originating directly
from the ore is referred to as primary metal. In LCIs these two types – primary and secondary metal –
are often differentiated because they bear a different ecological burden.
2.1 Primary Chromium

Of the many minerals that contain chromium only the chromium spinels are of economic importance.
Chromite is found in stratiform and podiform deposits.
- Stratiform deposits (seam-like) Layers up to a meter thick; e.g. Bushveld Igneous Complex (South
Africa), the Great Dyke (Zimbabwe) and the Stillwater Complex (United States).
- Podiform deposits (shaped like sacks or tubes). Range in size; Ural Mountains, Albania, Zim-
babwe, and the Philippines.
Generally, richer lumpy Cr bearing ores have been preferred for smelting, whereas those with Cr2O3
content less than 40 % have been used in refractories. The internal texture of the chromite ore bodies
varies widely. The closest chromite crystal packing results in the formation of massive ores containing
75 - 85 vol% of chromite.
Production and resources are presented in Tab. 2.1. The largest chromite-ore-producing countries (In-
dia, Kazakhstan, and South Africa) accounted for about 76% of world production. South Africa alone
accounts for more than 45% of world production and has been the major supplier of chromium in the
form of chromite ore and ferrochromium to western industrialised countries (USGS (2003)).
2.2 Secondary Chromium

The major end use of chromium is in stainless steel, and this is the major form in which chromium is
recycled. Ferrochromium is added to iron at steel-producing plants to make the iron-chromium alloy
stainless steel.
Stainless steel scrap can substitute for ferrochromium as a source of chromium. Scrap is generated dur-
ing the manufacturing process (new scrap) and as a result of recycling obsolete equipment (old scrap).
Scrap from these sources is collected and sorted by grade (i.e., chemical composition) in scrap yards.
Scrap brokers play a role in moving material from where it is recovered to where it is consumed. The
steel industry consumes stainless steel scrap as a source of chromium and nickel. Thus chromium units
are recycled when stainless steel is reused. A study of domestic stainless steel found that its average
chromium content is about 17% (USGS (2003)). Papp (2001) reports a recycling rate of 20% for the
US in 1998 with a domestic (primary) production of 144 kt and a consumption of old (75.3 kt) and
new scrap (28.6 kt) of total 103.9 kt, imports of 263 kt and releases of stock of 37.2 kt to meet the
consumption of chromium in the fabrication of stainless steel of 522 kt and exports of 22.5 kt (all val-
ues in chromium contained).

2. Reserves and Resources of Chromium
The typical raw materials for the production of secondary ferro alloys are the following residues gen-
erated mainly within the production of stainless steel:
- Electric arc furnace (EAF) filter dust
- Converter filter dust
- All kind of scale fines
- Shot blasting dust
- Grinding dust
Since no chromium is recycled within the chromium industry but directly to the foundries of the
stainless steel industry, no secondary chromium is considered in this study.
Tab. 2.1 Mine production and reserves of chromium in 2001. “Reserves” is referred to as “the part of the reserve
base which could be economically extracted or produced at the time of determination”, while “Reserve
base” denotes “that part of an identified resource that meets specified minimum (...) criteria related to cur-
rent mining and production practices” (USGS (2003)).

2001 [kt Cr / a] 2002 [kt Cr / a] [kt Cr] [kt Cr]
(estimated)
United States — — — 7000
India 1680 1900 18000 39000
Kazakhstan 2050 2300 410000 410000
South Africa 5500 5800 790000 5500000
Other countries 2900 3000 420000 1100000
World total (may be 12100 13000 1600000 7100000
rounded)

3. Characterisation of the Materials
3 Characterisation of the Materials

Based on the raw material chromite, various chromium containing materials are produced for the use
in a broad field of applications.
3.1 Chromite ore

Chromite [53293-42-8] is a spinel FeO · Cr2O3. In nature it is a mixture described by the formula
(Fe2+, Mg)O · (Cr, Al, Fe3+)2O3. The theoretical Cr2O3 content in chromite would be 67.9% (46.5%
Cr), but in reality this value ranges between 25% and 60% Cr2O3 (17% - 40.8% Cr). The chromium
content in chromite is 2 · 52 / 152 = 68.4%, Adelhardt & Antrekowitsch (1998) calculated a world-
wide average content of 37.3% Chromite (25.5% Cr) in raw ore. Chromite ore rarely contains more
than 50 % Cr2O3; other minerals such as SiO2 can also be present. Typical composition of some impor-
tant metallurgical chromite ores are summarised in Tab. 3.1.
A high Cr : Fe ratio is advantageous to produce an alloy with high chromium content. Chromite ores
are classified as follows:
- Ores rich in chromium: > 46 % Cr2O3 , Cr : Fe > 2 : 1; for the production of ferrochromium
- Ores rich in iron: 40 – 46 % Cr2O3 , Cr : Fe < 2 : 1; for the production of charge chrome and for
the chemical industry
- Ores rich in aluminium: > 60 % (Cr2O3 + Al2O3), > 20 % Al2O3; for refractories
Metallurgical-grade chromite ores are classified as hard lumpy or friable lump types, fines, and con-
centrates. Concentrates are produced by mechanical upgrading of lean ores or fines. In the production
of high-carbon ferrochromium, generally a lumpy type of chromite ore is necessary. About 80 % of
chromite ores in the western world are fines (< 10 mm). Therefore, efforts have been made to agglom-
erate these. Fines of chromite ores can be used to produce low-carbon ferrochromium.
3.2 Chromium Metal

At room temperature Chromium [7440-47-3] is resistant to ordinary corrosive agents, which explains
its use as an electroplated, protective coating. It dissolves in non-oxidising mineral acids, such as hy-
drochloric and sulphuric acids, but not in cold aqua regia or nitric acid, which passivate the metal. At
elevated temperature it reacts with halogens, silicon, boron, nitrogen, oxygen, and carbon. Chromium
and chromium-rich alloys are brittle at room temperature (Downing et al. (1997)).
Chromium metal is a metallurgical industry product. It is produced by one of two processes: electroly-
sis or aluminothermic reduction. Chromium metal produced by these processes is essentially the same
differing only in amount of various trace chemical constituents. The largest production of chromium
metal is by the aluminothermic reduction process (ICDA).
Selected physical properties of elemental chromium are given in Tab. 3.2.
3.3 Ferrochromium
In the metallurgical industry, the principal product is ferrochromium, a mixture of chromium and iron
with varying amounts of other chemical constituents. Ferrochromium is further classified by its carbon
content as high-carbon, medium-carbon, or low-carbon ferrochromium. By far, the largest production

of ferrochromium is the high-carbon variety. It is the major source of chromium for the production of
steel alloys. 1
Ferrochromium is a master alloy of iron and chromium, containing 45 – 95 % Cr and various amounts
of iron, carbon, and other elements. The ferrochromium alloys are classified by their carbon content
and are known by their French names because basic work in this field was carried out mainly in
France:
- High-carbon ferrochromium (HC ferrochromium) with 4 – 10 % C, "ferrochrome carbure"
- Medium-carbon ferrochromium (MC ferrochromium) with 0.5 – 4 % C, "ferrochrome affiné"
- Low-carbon ferrochromium (LC ferrochromium) with 0.01 – 0.5 % C, "ferrochrome suraffiné"

The mechanical and chemical properties of steel can be improved by alloying it with ferrochromium.
Chromium combined with nickel gives stainless steel excellent chemical resistance (Downing et al.
(1997)). Some physical properties of different ferrochromium alloys are given in Tab. 3.3
Tab. 3.1 Analyses of some chromite ores, metallurgical grade (Downing et al. (1997))
Chromite ore, wt%

Compound Transvaal Zimbabwe USSR lump Turkey Albania
lump friable lump lump
Cr2O3 42.55 49.53 53.73 47.58 40.5
FeO 21.85 11.6 8.5 9.45 11.4
Fe2O3 4.85 2.2 4.3 3.1 –
MgO 9.26 17.52 17.3 18.7 23.3
Al2O3 15.5 7.3 9.4 8.8 7.3
SiO2 5.54 5.14 4.3 8.33 12.6
L.O.I.e 0.2 2.3 2.1 2.25 3.4
Cr : Fe 1.43 3.20 3.81 3.42 3.13
1
ICDA, International Chromium Development Association, http://www.chromium-asoc.com/, accessed 01.03.2003

Tab. 3.2 Some physical properties of chromium (Downing et al. (1997)).
Chromium
Ar 51.996 l
Atomic number 24
mp 1857 °C
bp 2672 °C
Density  at 20 °C 7.19 g/cm3
Crystal structure cubic, body centered
Specific heat at 25 °C 23.25 J mol–1 K–1
Molar entropy S°298 23.64 J mol–1 K–1
Heat of fusion 16.93 kJ/mol
Latent heat of vaporization at bp 344.3 kJ/mol
Linear coefficient of thermal expan- 6.2×10–6
sion at 20 °C
Resistivity at 20 °C 12.9×10–8  m
Thermal conductivity at 20 °C 67 W m–1 K–1
Tab. 3.3 Some physical properties of ferrochromium and ferrosilicochromium (Downing et al. (1997)).
Alloy Density (r), Melting range, °C

g/cm3 Liquidus Solidus
Chromium metal (electrolytic) 7.2 1900
Chromium metal (aluminothermic) 7.2 1850
Low-carbon ferrochromium (72% Cr, 0.01% C) 7.35 1690 1660
High-carbon ferrochromium (69% Cr, 4–6% C, 1% Si) 7.2 1500 1350
High-carbon ferrochromium (64% Cr, 5% C, 1% Si) 7.1 1450 1340
Ferrochromium, charge grade (63% Cr, 5.5% C, 7% Si) 6.7 1500 1400
Ferrochromium, charge grade (56% Cr, 6% C, 5% Si) 6.8 1493 1660
Ferrochromium silicon (36% Cr, 40% Si) 5.3 1388 1360


Chromium has a wide range of uses in metals, chemicals, and refractories. However, in this study only
ferrochromium is of further importance. A schematic overview is given in the following Fig. 4.1.
Fig. 4.1 The chromium life cycle process from mining to commercial, scheme taken from ICRA’s homepage at
http://www.chromium-asoc.com/
4.1 Chromite ore

Chromite ore is the central intermediary product in the chromium industry. From it the three main
product groups are made: ferrochromium, chromium chemicals and refractories. Chromite is used di-
rectly in the refractory industry to make bricks, mortar, and ramming and gunning mixes. Chromite
enhances thermal shock and slag resistance, volume stability, strength and is used mainly in the fer-
rous and non-ferrous industry, cement industry and glass manufacturing. In foundries for certain ap-
plications chromite is used as moulding sand, for which chromite sand with more specific properties is
applied.
In 1994 around 8% of the chromium produced was used in the refractory industry, while 80% went
into the production of ferrochromium and 12% into the chemical industry (Adelhardt & Antrekowitsch
(1998)).
In the refractory industry, the principal products are shaped and unshaped refractories. Refractories are
used to line metallurgical furnaces that produce steel or copper alloys. They are also used in the heat
exchange section of glass making furnaces and in cement kilns. These refractories are consumed in the
process of making steel, copper, glass, and cement, but they are not part of those products. The annual

chromite production for 2002 is given in Fig. 4.2 2 . Further details on use of refractories in the glass
industry can be found in Hischier (2003)
The consumption of chromium was reported in Papp (1994) to be 30 t Cr / Mt cement and 450 t Cr /
Mt copper in the cement and copper industry respective. Adelhardt & Antrekowitsch (1998) reports
the use of 2 and 5 kg refractories per tonne of converter and electric steel respectively.
After use, refractories are recuperated in order to remove hexavalent chromium, which is built under
high temperature and oxidising conditions (Adelhardt & Antrekowitsch (1998)).
4.2 Chromium Chemicals

In the chemical industry, the principal product is sodium dichromate. Sodium dichromate is used both
as feed material for the production of other chromium chemicals and as process material for the sur-
face treatment of metals. The use of chromite by the chemical industry is substantially smaller than
that of the metallurgical industry. Chemicals made from or along with sodium dichromate, are indus-
trial products including chromic acid, chromic oxide, and potassium dichromate. The uses of chro-
mium chemicals are quite diverse and include metal finishing, leather tanning, pigment production,
and wood preservation2.
Chromium compounds are used in numerous fields. In addition to the applications mentioned, chrom-
ates have long been used in printing as an aid in photomechanical reproduction. For some time, chro-
mium dioxide has been a component of magnetic tapes for information storage. About 12 % of the
chromium used world-wide is in the form of chromium chemicals (Adelhardt & Antrekowitsch
(1998)).
Directly or via several intermediate stages, sodium dichromate (Na2Cr2O7 · 2 H2O; CAS 7789-12-0), is
the starting material for the production of all chromium compounds and pure chromium metal. For fur-
ther details on use and production see Althaus et al. (2003)
Chromium(III) oxide (Chromium Hydroxide) (Cr2O3; CAS 1308-38-9), finds widespread application
as a green pigment resistant to atmospheric conditions and heat. In addition, it is used as a colorant in
glass products and printing inks, as a vitrifiable pigment in the ceramics industry, and as a polishing
agent because of its considerable hardness.
Chromium(IV) Oxide (Chromium Dioxide) (CrO2; CAS 12018-01-8) is used as a magnetic pigment.
So far no other uses have achieved any significance.
Chromium trioxide (CrO3; CAS 1333-82-0), also known as chromic acid anhydride or chromic acid, is
made by the reaction of sodium dichromate with sulphuric acid. Electroplating is the most important
field of application of chromium(VI) oxide. Numerous mixtures containing chromium trioxide are on
the market; these "compounds" often contain hexafluorosilicates which improve the properties of the
chromium coatings. Chromic acid solutions are also used for passivating zinc, aluminium, cadmium,
and brass. Proprietary mixtures predominantly contain additions of fluoride, nitrate, and phosphate
ions. Other uses for chromic acid are in the production of chromium dioxide and in wood preservation.
4.3 Chromium Metal

Pure chromium metal, produced by electrolytic or aluminothermic processes, is used for alloying non-
ferrous engineering materials. Chromium metal is a major source of chromium for non-ferrous alloys
that exclude iron, such as certain superalloys. The most common materials are nickel-based and co-
balt-based alloys, most of which are used as high temperature alloys (Downing et al. (1997)).
2
http://www.chromium-asoc.com/publications/frame.html

The aerospace industry is the major source of demand for chromium metal in the form of superalloys,
which are manufactured and consumed mostly in North America and Europe. World chromium metal
demand was increasing by 1’000 t/yr over the last years. Chromium metal production is estimated at
22’000 t in 2001; 21’000 t in 2000; and 20’000 t in 1999. The trend is expected to continue. Besides
aerospace industry, the major end use industries are gas turbine, and petrochemical (USGS (2003)).
4.4 Ferrochromium
Ferrochromium is used principally as a master alloy to produce chromium-containing steel and cast
iron. Chromium in steel imparts corrosion resistance because of the formation of a thin, continuous,
impervious chromium oxide film on the steel surface.
Most ferrochromium is used to manufacture stainless steel. Martensitic grades with ca. 13 % Cr are
used for applications such as knives, whereas ferritic grades with 18 – 22 % Cr are used as deep-
drawing sheets. Austenitic CrNi steels exhibit especially good corrosion (acid) resistance and are used
for equipment in the chemical industry as well as for food-processing machinery. The annual produc-
tion of ferrochromium dropped from 5'720'000 t in 2000 to 4'680'000 in 2001 (USGS (2003)).
Ferrochromium is classified usually into two major groups: low-carbon and high-carbon ferrochro-
mium. It is used as a source of chromium by the steel industry and as a source of chromium for non-
ferrous alloys that permit some iron content.
High-carbon ferrochromium is one of the chromium industry’s most important products with a share
of 91% in the production. Low-carbon ferrochromium production accounts to 9% of the total and
dropped from 792’738 t in 1988 to 413’079 t in 2001; an annual compound growth rate of 0.95 over
the time period (USGS (2003)).
Technological changes in the consuming industry permit the substitution of high-carbon ferrochro-
mium for low-carbon ferrochromium. The substitution is desirable because low-carbon ferrochromium
production is more expensive than high-carbon ferrochromium production.
The annual ferrochromium production for 2002 is given in Fig. 4.3.
4.5 Chromium Alloys

Stainless steel is composed of two broad categories – austenitic and ferritic. The names are related to
the molecular structure of the steel but also identify which grades require nickel (austenitic) and which
do not (ferritic).
The only true chromium-based alloys that have been developed and used commercially are a few
chromium – nickel alloys. These materials contain 50 – 60 % chromium; the remaining percentage is
nickel, with either niobium (columbium) or titanium specified as carbon and nitrogen scavengers.
These alloys, often referred to as superalloys, are used in the construction of high-temperature exposed
components in chemical and petrochemical industries (Downing et al. (1997)).
The most widely known chromium containing alloy is the stainless steel AISI 302 (Fe/Cr18/Ni8),
which contains 18 % Cr and 8 % Ni. A small amount of molybdenum further improves the acid resis-
tance. Heat-resistant steel (24 – 26 % Cr and 19 – 21 % Ni) is used in power stations for steam boilers
and heat exchangers; superalloys based on nickel or cobalt (ca. 30 % Cr) are used for gas turbine parts.
The annual production of stainless steels for 2002 is given in Fig. 4.4.

Tab. 4.1 Uses of chromium compounds (Anger et al. (1997))
Branch of industry Product Use

Building industry chromium(III) oxide pigment for coloring building materials
Chemical industry dichromates, chromium(VI) oxide oxidation of organic compounds, bleach-
ing of montan
waxes, manufacture of chromium complex
dyes
chromium(III) oxide catalysts
Printing industry dichromates photomechanical reproduction processes
chromium(VI) oxide chromium plating of printing cylinders
Petroleum industry chromates(VI) corrosion protection
Paints and lacquers chromates, chromium(III) oxide pigments
Refractory industry chromium(III) oxide additive for increasing slag resistance
Electroplating chromium(VI) oxide bright and hard chromium plating
Wood industry chromates, chromium(VI) oxide in mixtures of salts for protecting wood
against fungi
and insects
Leather industry basic chromium(III) sulfates tanning of smoothed skins
Metal industry chromium boride, chromium car- flame sprays
bide
chromium(III) oxide polishing agents
Metallurgy chromium(III) oxide aluminothermic extraction of pure chro-
mium metal
Textile industry dichromates dyeing with chrome dyes
basic chromium(III) acetates and mordanting of textiles
chromium(III) fluorides
Recording industry chromium(VI) oxide magnetic information storage
Pyrotechnics industry dichromates additive to igniting mixtures

Fig. 4.2 Annual chromite ore production in 2001.
Fig. 4.3 Annual ferrochromium production in 2001.
Fig. 4.4 Annual production of stainless steel in 2001, the main demand in ferrochromium.

This study focuses on the main processes in the chromium production as shown in Fig. 4.1. As shown
in chapter 4 Use / Application of Material no secondary production of chromium has to be considered,
since the recycling of chromium occurs directly into the foundries in the form of re-melted alloys from
chromium containing scrap.
281'000 t
low-coal
Ferro-
75%
chromium
3'182'400 t
2'901'000 t
Metallurgical Plant high-coal
5'000 t
electrolytic
production
20'000 t
Chromium
Metal
alumino-
15'000 t
thermic
reduction
17%
4'256'000 t
733'000 t
733'000 t
Chromite Chemical Plant Chromium Chemicals
8%
340'000 t
340'000 t
Refractory Industry refractories
Fig. 5.1 Scheme of the chromium production system considered. The study focuses on the production of high-
carbon ferrochromium, which is based on chromite. Chromite itself is used in different other modules in
ecoinvent like “portachrom, at plant” Hischier (2003), “sodium dichromate, at plant” Althaus et al. (2003) and
in the cement production. Metallic chromium is won from two different processes. Values shown are in ton-
nes of chromium contained and are based on USGS (2003) and Adelhardt & Antrekowitsch (1998).

Tab. 5.1 Ecoinvent meta information for the processes for the production of chromite and ferrochromium.
chromite, ore concentrate, at ben- ferrochromium, high-carbon, 68%

Name
eficiation Cr, at plant
Location GLO GLO
Unit kg kg
The module includes a mining and a ben- The module includes the pre-treatment of
eficiation step with the mining infrastruc- the ore and the reduction to high-carbon
ture and disposal of overburden and tail- ferrochromium. No secondary production
ings (see part II "Auxiliary Processes"). considered, since secondary chromium is
recycled to the foundries and has to be
inventoried in the respective module, i.e.
when casting the final raw material in the
Included Processes stainless steel production.
Amount 1 1
Chromit, Erz, ab Aufbereitung Ferrochrom, hochgekohlt, 68% Cr, ab
Local Name Werk
Synonyms chromeisenstein ferro-chrome//ferrochrome//chromeisen
The module describes the production of The module describes the production of
chromite in 1994 for a world-wide produc- chromite in 1994 for a world-wide produc-
tion average. It is de-signed for the use of tion average. It is designed for the use of
chromite as intermediary product in the chromite as intermediary product in the
production of ferrochromium, dichromate production of ferrochromium, dichromate
and refractories. Process data is taken and refractories. This module does not
mainly from a material flow analysis for the consider secondary sources of chromium.
consumption of chromium compounds in Process data is taken mainly from a mate-
Germany. The ore concentrate produced rial flow analysis for the con-sumption of
has a content of 30.7% Cr (44.9% Cr2O3) chromium compounds in Germany. Re-
by weight. sults are used as a proxy for the European
General Comment situation.
End Date 2003 2003
Data valid for entire pe- 1 1
riod
This module relates to the global situation. This module relates to the world wide pro-
Some agents used like lime originate from duction situation. Some Processes origi-
data sets valid for Switzerland or Europe. nate from data sets valid for Switzerland or
Processes like blasting and transport Europe.
originate from data sets valid for Switzer-
Geography Text land or Europe
Mining is done 21% open pit and 79% un- Direct reduction of chromite ores with car-
derground, followed by a beneficiation of bon in three-phase submerged arc fur-
the ore trough classification. Overburden naces. 50% ordinar technology, 50% new
and tailings are disposed near the mining processes like Outokumpu, Showa Denko,
Technology Text site. Krupp-CODIR & plasma furnace.
9.7 million tonnes in 1994 2.2 million tonnes chromium contained in
Production Volume ferrochrome in 1994

Tab. 5.1 Ecoinvent meta information for the processes for the production of chromite and ferrochromium. (cont.)
ferrochromium, high-carbon, 68% chromium, at regional storage

Name
Cr, at regional storage
Location RER RER
Unit kg kg
The module includes the transportation of Both aluminothermic production and elec-
high-carbon ferrochromium to Europe. No trolysis are aggregated in this module. It
secondary production considered, since comprises the production of Cr2O3 as in-
secondary chromium is recycled to the termediary product, the aluminothermic
foundries and has to be inventoried in the process, the transport of the raw materials
respective module, i.e. when casting the to Europe and the electrolysis. No emis-
final raw material in the stainless steel sions are taken into account. No secon-
production. dary production considered, since secon-
dary chromium is recycled to the foundries
and has to be inventoried in the respective
module, i.e. when casting the final raw ma-
Included Processes terial in the stainless steel production.
Amount 1 1
Ferrochrom, hochgekohlt, 68% Cr, ab Re- Chrom, ab Regionallager
Local Name gionallager
Synonyms ferro-chrome//ferrochrome//chromeisen chrome
The module describes the consumption of The module describes the chromium metal
chromite in Europe in 1994. It is designed consumed Europe in 1994. It is designed
for the use of chromite as intermediary solely for the use of the metal as alloying
product in the production of ferrochro- element in super-alloys (Ni-Cr). Usually
mium, dichromate and refractories. This stainless steels are made from ferro-
module does not consider secondary chrome and not from chromium metal.
sources of chromium. Process data is Chromium is considered to be made by
taken mainly from a material flow analysis aluminothermic process (75%) and elec-
for the consumption of chromium com- trolysis (25%). This module does not con-
pounds in Germany. Results are used as a sider secondary sources of chromium. The
proxy for the European situation. Disposal general date quality is not satisfying, due
of smelter slag is inventoried as disposal to a lack of information.
of inert material, thus neglecting the po-
tentially important (but quantitatively un-
General Comment known) heavy metal emissions.
End Date 2003 2003
Data valid for entire pe- 1 1
riod
This module relates to the consumption in World-wide production mix is chosen as
Europe, although the import mix for Ger- proxy for the consumption mix in Europe.
many is chosen. Production relates to a
Geography Text global average.
Most transport by ship Metallic chromium is produced by alumino-
thermic process (75%) and electroylsis of
Technology Text dissolved ferrochromium (25%)
n.a. 350'000 t in GER (1994), no info on 22'000 t prod glob (2000), RER roughly
ProductionVolume consumpt. in RER 20% (4'300 t). No data on consump.

5.1 Exploitation of chromite

Mining. Chromite ores are usually mined underground. Adelhardt & Antrekowitsch (1998) report an
estimated share of 71% underground mining activities in the mine production in 1994 considered in
their study, representing 78% of the total production. Of a surveyed part of 52% of the total world
production, 69% of the chromite originated from stratiform deposits.
Beneficiation. After mining, the ore is first crushed in several stages with jaw and / or cone crushers,
and then subsequently ground with rod and or ball mills and finally screened for classification. In a
second step the classified material is subjected to gravity concentration to separate the metal-bearing
particles from the unwanted minerals. For this drum separators and de-watering screens for lumps are
used and cone separators and a high-gradient magnetic separator for fine material. No flotation is
done. Chromite yields vary in a range from 65% to 85% (Adelhardt & Antrekowitsch (1998)).

Mining
large quantities of dust, which contains elevated contents of metals. Rain percolates through overbur-
den and leads to metal emissions to groundwater. Overburden is deposed close to the mine.
Beneficiation
the ore itself. Tailings are deposed as piles and in ponds. Since the tailings are not sulphidic, no acid
rock drainage (ARD) occurs. In the tailings material the most significant contents are Cr and Ni, which
occur as insoluble compounds and are considered not to cause any negative effects IPPC (2002).
5.1.3 “chromite, at beneficiation” in ecoinvent

The module describes the world-wide average production of chromite in 1994. It is designed for the
use of chromite as intermediary product in the production of ferrochromium, dichromate and refracto-
ries. Process data is taken mainly from Adelhardt & Antrekowitsch (1998), which conducted a mate-
rial flow analysis for the consumption of chromium compounds in Germany. Results are used as a
proxy for the European situation. Emission factors are mainly taken from IPPC (2002) and
Anonymous (1998b).
Infrastructure and resource use

The mobile equipment used for exploration like drilling machines, scoop trams and underground
trucks is inventoried with the diesel consumption reported in Adelhardt & Antrekowitsch (1998). The
used module “diesel, burned in building machine” includes the infrastructure of the machines. Use of
conveyor belt was assessed according Tab. 5.2.
The world average chromite content after beneficiation amounts to 30.7% (44.9% Cr2O3), a yield from
mining and beneficiation of 81% can be calculated (Adelhardt & Antrekowitsch (1998)).

According to part IX, chapter 2.1 of this report (Non Ferrous Metal Winning – Auxiliary Processes)
no direct land use for underground mining (71%) is inventoried. Land use for open pit mining (29%)
per tonne of ore is assessed with 0.0025m2/t. Thus the average land transformation for mining (under-
ground and open pit) is 0.000725 m2/t ore and the occupation is 0.0218 m2a/t ore (a lifetime of 30
years is assumed). Mining facilities are inventoried as unspecific module according to part IX (Non
Ferrous Metal Winning – Auxiliary Processes).
Tab. 5.2 Assumptions for infrastructure used in underground mining exploration. As lifetime ten years are estimated,
and 5000 working hours yearly.
Action Share Machine Throughput Unit Infrastructure

Open pit 29 % Conveyor belts 100'000'000 10’000 1*10-4
Underground 71 % Conveyor belts 100'000'000 2’000 2*10-5
Mix 4.32*10-5
Mining
According to Adelhardt & Antrekowitsch (1998) 71 % of the chromite world production originates
from underground mines. The overburden is disposed in piles near the mine, 1.77 kg per kg ore mined.
The standard module for disposal of non-sulphidic overburden was chosen, which includes the land
use and transformation through the disposal. No overburden is refilled according to Adelhardt &
Antrekowitsch (1998). In the world-wide average 0.35 m3/t crude ore of the pit water is used for the
beneficiation. The remaining pit water is not inventoried as output but its emissions are inventoried
together with the emissions of the beneficiation step.
Beneficiation
Once the ore is mined, it is ground and classified. No flotation is needed in beneficiation. The sepa-
rated gangue is disposed in tailings ponds, the concentrated ore is fed to the metallurgy, which is as-
sumed to be on-site. The respective standard module “disposal, non-sulphidic tailings, off-site” is cho-
sen (c.f. part IX (Non Ferrous Metal Winning – Auxiliary Processes), chapter 2.3 (Disposal of Tailings
and Overburden)).
The beneficiation infrastructure is approximated with a similar facility from the aluminium production
(c.f. part 1 (Aluminium)). Dust emissions comprise additionally the dust emissions due to mining ac-
tivities. Hilbrans & Hinrichs (1999) assume for the nickel mining a share of 60% hydroelectricity due
to the operations high and constant energy demand. The same assumption is made for chromite. The
value comprises the electricity demand of the mining, which is expected to be small compared to the
beneficiation.
Steel: The steel input (abrasion) for milling is documented in IPPC (2002) for a single Finnish chro-
mite mine (Kula) and amounts to 300 g / t ground ore. The mill (jaws, balls, etc.) consists of special
austenitic manganese steel of Hadfield grade, which is approximated with chromium steel in the in-
ventory.
Water use & effluents: Water use is documented in Adelhardt & Antrekowitsch (1998) and is inven-
toried in this study as originating from river without pumping activities. Total emissions to water are
chosen according to Tab. 5.3. These values are reported from a chromite site in Finland (Kula) and
recommended values. The reported value of 150 mg / l COD is assumed to amount to 150% of BOD
(Zimmermann et al. (1996)). Values for TOC are calculated using the standard procedure in

Frischknecht et al. (2003) approximating the unknown organic matter with the simplified formula for
organic matter C6H12O6. This gives 58.7 mg / l TOC and DOC respectively.
Dust: A case study in (Anonymous (1998a)) on a projected open cut gold-copper mining and process-
ing facility in Australia predict dust emissions between 0.18 kg/t and 0.46 kg/t of dust produced per
tonne of ore mined. The prediction was made using the Emission Estimation Technique Manual for
Mining (NPI (2001)). The maximum value of 0.46 kg/t is chosen for total dust emissions, the value for
underground mining is assessed with 80% of this amount. The dust composition is approximated with
the chromite ore composition from Tab. 3.1. For PM10 values, the ratio TSP : PM10 : PM2.5 was ap-
proximated with values from manganese ore mining 101.72 : 50 : 5 g / t ore taken from the CEMIP-
database 3 , i.e. one tonne emitted dust corresponds to 492 kg PM10 and 49 kg PM2.5 respectively. The
airborne emissions resulting from mining and beneficiation activities are summed up in Tab. 5.4. In
ecoinvent, for particulates differential classes are used adding up to the total dust emission, i.e. 49 kg
PM<2.5, 442 kg PM2.5-10 and 508 kg PM>10 per tonne of total dust.
Data quality
Mining: Data for major mass fluxes are satisfactory, while the emissions into water may be overesti-
mated due to different system boundaries of reported effluent volume from Adelhardt & Antreko-
witsch (1998) and total effluent emissions reported in IPPC (2002) and Anonymous (1998b).
Beneficiation: Like in mining, the emissions to water can be overestimated. No emissions from the
disposal of the tailings are modelled since no information about such emissions is available.
3

Tab. 5.3 Effluent levels to be achieved during operation and after mine closure according to Anonymous (1998b)
(Maximum value) and reported values from Kemi site (Finnland) after IPPC (2002) (Reported value).
Parameter Maximum Reported This Study

value (mg/l) value (mg/l) (mg/l)
Mining Type Base metal Chromite ore
pH 6-9
TSS 50 20.6 50
Oil and grease 10 10
Cyanide 1.0 - b)
Free 0.1 - b)
Weak acid dissociable (WAD) 0.5 - b)
COD 150 150
Arsenic 0.1 - c)
Cadmium 0.1 - c)
Chromium (hexavalent) 0.1 - a)
Chromium, ion 0.1 a)
c)
Chromium (in solid) 0.05 -
Copper 0.5 - c)
Iron 3.5 6.9 6.9
Lead 0.2 - c)
Mercury 0.01 - c)
Nickel 0.5 0.5
Zinc 2 - c)
Total metals 10 -
a)
Inventoried in Ecoinvent as trivalent chromium ion. Hexavalent chromium is assumed to oxidise
quickly to trivalent chromium in the course of abatement and aeration.
b)
chromite is no sulphidic ore, therefore no use of this substance and consequently no emission is as-
sumed
c)
Maximum values relates to the whole range of base metal extraction. This specific emission is not
assumed to be of importance in chromite exploitation.
taken from Anonymous (1998a), partition of fractions of particulate matter from the CEMIP-database (mining
of manganese ore), the dust composition correspond to the average chromite ore from “3.1 Chromite ore” in
this study.

air mined ore
Iron g / t ore 48.5
Silicon g / t ore 13.3
Chromium g / t ore 126
Aluminium g / t ore 20.2
Magnesium g / t ore 41.0
PM<2.5 g / t ore 19.4
PM2.5-10 g / t ore 175
PM>10 g / t ore 201

Modul name
in ecoinvent
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
81% yield assumed (19% Cr loss in
Chromium, 25.5 in
Chromium in mining / beneficiation), 1.48kg ore
 resource in ground chromite, 11.6% in 0.38 kg (Adelhardt 1998) 1 1.13 (2,2,3,1,1,2,12)
ground and 1.74kg overburcden per kg
crude ore, in ground
chromite
diesel fuel construction diesel, burned in

 machinery No GLO 0.081 MJ (Adelhardt 1998) 1 1.13 (2,2,3,1,1,2,1)
(energy) processes building machine
construction
explosive  civil engineering No RER blasting 1.5E-3 kg (Adelhardt 1998) 1 1.13 (2,2,3,1,1,2,4)
processes
exploration
exploration infrastructure, 2000 m construction (5,5,1,1,1,5,9); rough
infrastructrue,  machinery Yes RER conveyor belt, at plant 6.4E-8 m estimation 1 3.32
belts assumed processes estimate
conveyor belts
infrastructure of
50 a lifetime; 400'000 t / a -> 5e-11 non-ferrous metal (5,5,1,1,1,5,9); rough
mining  metals extraction Yes GLO 5.3E-11 unit estimation 1 3.32
mining
units per kg ore mined mine, underground estimate

(underground)
infrastructure of 51 a lifetime; 400'000 t / a -> 5e11 non-ferrous metal (5,5,1,1,1,5,9); rough

 metals extraction Yes GLO 2.2E-11 unit estimation 1 3.32
mining (surface) units per kg ore mined mine, surface estimate
Land (3,2,3,1,1,2,8);
direct land use, represents specific Transformation, from (Martens 2002),
transformation  resource land 1.1E-6 m2 1 2.03 calculation based on
share of surface mines unknown calculations
from nature reported value
Land (3,2,3,1,1,2,8);
direct land use, represents specific Transformation, to (Martens 2002),
transrormation to  resource land 1.1E-6 m2 1 2.03 calculation based on
share of surface mines mineral extraction site calculations
mine reported value
(3,2,3,1,1,2,7);
Land use for direct land use, represents specific Occupation, mineral (Martens 2002),
 resource land 3.2E-5 m2a 1 1.53 calculation based on
mining share of surface mines extraction site calculations
reported value
waste
overburden, residual material disposal, non-sulfidic
 no overburden assumed to be refilled manageme No GLO 1.77 kg (Adelhardt 1998) 1 1.16 (3,2,3,1,1,2,6)
disposed landfill overburden, off-site
nt
(1.5 - 0.25) m3/t pit water per crude ore (0.25 m3/t used in
beneficiation) without treatment. The composition is given in Tab.
 effluents, mining 1.9E-3 m3 (Adelhardt 1998) 1 1.13 (2,2,3,1,1,2,6)
5.3. The emissions are invnetoried with the emisisons of
beneficiation. The uncertainties of the single emissions are high.
Fig. 5.2 Flows for “chromite, at beneficiation – part mining” and their representation in the ecoinvent database. Values correspond to the functional unit of 1 kg of chromite.

Mean value
Source for
ecoinvent
Deviation
Comment
Standard
Category
Location
category
structure
Remarks
Output
name in
General
Process
Modul
Name
value
Input
Infra-
mean
Sub-
Type
Unit
electricity, f rom Comprises electricity demand of electricity, medium
grid   w aste heat mining. Assumption: 40% electricity electricity production mix No UCTE voltage, production 0.0150 k Wh (Adelhardt 1998) (2,2,3,1,1,2,2)
benef iciation f rom grid UCTE, at grid
Comprises electricity demand of electricity,
hydro
hydropow er   w aste heat mining. Assumption: 60% pow er plants No RER hydropow er, at run- 0.0225 k Wh (Adelhardt 1998) (3,2,3,1,1,2,2)
pow er
hydropow er of-river pow er plant
fresh and
processw ater  resource in w ater Water, river 2.7E-3 m3 (Adelhardt 1998) 1 1.13 (2,2,3,1,1,2,12)
(benef iciation)
chromium steel 18/8,
steel  Abrasion in mill (steel balls) metals extraction No RER 4.5E-4 kg (IPPC, 2002) 1 1.2 (2,3,2,3,1,3,4)
at plant
infrastrucure of Aluminiumhydroxide production as aluminium hydroxide, (5,5,1,1,1,5,9);
 metals extraction Y es RER 3.1E-15 unit estimation
beneficiation
benef iciation proxy plant rough estimate

w aste (3,2,3,1,1,2,6);
no tailings ref illed according to residual disposal, non-sulf idic
 tailings, disposed manageme No GLO 0.49 kg (Adelhardt 1998) 1 1.16 reported
(Adelhardt. 1998) material landf ill tailings, of f-site
nt values
Emissions inventoried according to

composition in Tab. 5.3, they are (2,2,3,1,1,2,6);
eff luents,
 not show n in this f igure. Emissions 2.3E-3 m3 (Adelhardt 1998) 1 1.13 reported
beneficiation
have high uncertainties. No values
treatment of w aste w ater.
Including emissions from mining

particulate emissions represented
in fractions PM<2.5, PM2.5-10,
 Dust 5.9E-4 kg (NPI, 2001) 1 1.59 (3,2,1,1,3,4,25)
PM>10, including metals. These
emissions are inventoried as
show n in Tab 5.4 but not show n in
this f igure.
chromite, ore
concentrate, 44.9% Chromium
 Chromite metals extraction No GLO concentrate, at 1 kg (Adelhardt 1998)
contained
benef iciation
Fig. 5.3 Flows for “chromite, at beneficiation – part beneficiation” and their representation in the ecoinvent database. Values correspond to the functional unit of 1 kg chromite.

5.2 Winning of high-carbon ferrochromium

As seen in Fig. 5.1, high-carbon ferrochromium accounts for the major part of chromium consump-
tion. Although the traditional type for steel manufacturing was the high-chrome low-carbon ferro-
chrome. Much more economic and hence of higher importance for the industry nowadays is high-
carbon ferrochrome. Since the introduction of the AOD process (Argon-Oxygen-Decarburisation) the
steel industry prefers high-carbon ferrochrome, which originates predominantly from stratiform depos-
its rich in iron (Adelhardt & Antrekowitsch (1998)).

High-carbon ferrochromium is produced almost exclusively by direct reduction of chromite ores with
carbon (coke, coal, or charcoal) in large, three-phase submerged arc furnaces with 10 - 50 MVA ca-
pacity, corresponding to 15 000 – 60 000 t/a of ferrochromium production.
The oxides of iron and chromium present in the chromite can be readily reduced at high temperature
with carbon. Because of the tendency of chromium to form carbides, a carbon-containing alloy is al-
ways obtained. Carbides with higher carbon content formed initially at lower temperature react at
higher temperature with Cr2O3 and form carbides with lower carbon content; finally, reduction of SiO2
– in the case of silicochromium production – starts at higher temperature. The basic reaction can be
formalised as follows:
3 (Cr, Fe)7C3 + Cr2O3  (Cr, Fe)23C6 + 3 CO

Electric arc furnaces work continuously with an input of chromite, coke, dolomite, silica and bauxite
or other Al2O3 containing materials. Metal and slag are tapped in regular intervals. The submerged arc
furnaces use Søderberg electrodes where the electrode may be formed by hot paste, briquettes, blocks
or paste cylinders.
The coke rate is calculated on the basis of the stoichiometric requirement of the oxides and on the
amount of dissolved carbon in the alloy; allowance is made for some combustion at the top of the fur-
nace and for reaction with moisture. The slag composition is important to produce metal of desired
quality and to maintain smooth furnace operation.
Because more than 80 % of chromite ore mined in the western world is finer than 10 mm, agglomera-
tion processes have been adopted to provide good burden porosity in the electric arc furnaces. In some
ferrochromium plants, fines and lumpy ores are blended in ratios between 1 : 1 and 4 : 1. (Downing et
al. (1997)).
In elder plants with quite small furnaces the off-gas containing considerable amounts of CO is burned
in the top and cleaned thereafter. Modern plants with semi-closed furnaces do not need to burn the
CO. This results in a lower off-gas volume – factor 10 to 20 – and the possibility of re-using the gas as
secondary fuel in other processes. By using a closed, sealed furnace the volume of the off-gas can be
reduced by even a factor of 50 – 75. The smelting charge can be pre-heated in a shaft or rotary kiln by
using the CO-gas from the smelting process. The yield of chromium amounts from 85% to 95%.
Triggered by enhanced demand of chromium in the metallurgical industry, bigger plants and new
processes recently evolved. These processes - Outokumpu, Showa Denko, Krupp-CODIR and plasma
- are discussed in detail in Adelhardt & Antrekowitsch (1998). They feature much lower specific en-
ergy demand and higher yield (around 90%).
Fluxing agents, e.g., quartzite or alumina (corundum or bauxite) and lime, are charged with the burden
for slag formation.
Slag that contains metal is also crushed and the metal content recovered by means of slag-metal sepa-
ration. Recovered slag-free lump metal can be sold, while slag contaminated metal and fines are gen-
erally recycled to the smelting process.

An example of a high-carbon ferrochromium production process using a closed submerged electric arc
furnace is shown in Fig. 5.4.
Fig. 5.4 Example of a High-carbon ferrochrome production process using a closed submerged electric arc furnace
(IPPC (2001)).
5.2.2 Emission
Air: The main environmental impact by producing ferro-alloys is the emission of dust and fume from
the smelting processes. Dust emissions occur as well from storage, handling and the pre-treatment of
raw materials where fugitive dust emissions play an important role. Depending on the raw material
and the process used other emissions to air are SO2, NOx, CO-gas, CO2, polycyclic aromatic hydrocar-
bon (PAH), VOCs, and volatile metals. The formation of dioxins in the combustion zone and in the
cooling part of the off-gas treatment system (de-novo synthesis) may be possible.
Water: Water emissions are produced from cooling, granulation and other processes and site related
effluents. The wastewater generated by a wet cleaning abatement system plays also an important role.
5.2.3 Waste
Process residues and by-products with a significant amount is slag, filter dust and sludge and spent re-
fractories. These materials are already recycled and reused to a large extent where it is possible. Rich
slag, that means slag with a relatively high proportion of metal oxides, is used as raw material in other
ferro-alloy productions.
5.2.4 Ferrochromium, high-carbon, 68% Cr, at plant in ecoinvent

This part is mainly based on Adelhardt & Antrekowitsch (1998). General process description was
taken from (Fabian (1997)), process specific emissions from IPPC (2001), IPCC (1996), EPA (1998)
and EC (2002).

Adelhardt & Antrekowitsch (1998) differentiates between old (submerged electric arc furnace) proc-
esses and the newer (Outokumpu, Showa Denko, Krupp-CODIR and plasma), and assumes a share of
50% each.
The module describes the world average production mix for ferrochromium in 1994 based on
Adelhardt & Antrekowitsch (1998). It is designed for the use of the metal as alloying element. The
layout of the modules with general flow information, remarks, sources, values and uncertainty infor-
mation are shown in Fig. 5.5 (inputs) and Fig. 5.6 (output). Ecoinvent meta information for this proc-
ess are summarised in Tab. 5.1 on page 13. The following paragraphs describe calculations, sources
and assumptions chosen in this study. The functional unit of this process is one kg of ferrochrome with
an average chromium content of 68%.
Infrastructure and transport: It is assumed that the bauxite used as flux is transported by ship over
20'000 km, the additives chromite, bauxite, silica, coke, slag and the disposals slag and cake by lorry
over 50 km.
Energy: Adelhardt & Antrekowitsch (1998) calculates an average gross electricity demand of 3533
kWh containing smelting (4000 kWh and 2500 kWh for old and new process respectively), de-dusting
(74.8 kWh), crushing & screening (0.68 kWh) and production of the electrode (207 kWh), assuming
the new and the old process to be equal in production volume. In this study, the production of the elec-
trodes is inventoried in a special module and has to be subtracted from this value, resulting in an aver-
age electricity consumption of 3325 kWh / t Cr. An overview over different in- and output values from
different sources is given in Tab. 5.5
Emissions to air: Air emissions are taken from various sources, they are summarised and discussed in
Tab. 5.6.
Water use and emissions: Water use for European ferrochrome production is reported in IPPC
(2001), the average of 20 m3 per tonne is chosen resulting in 10 m3 of effluent (volume of process wa-
ter). The effluent has to be treated in a communal waste-water treatment plant, a process which is in-
ventoried separately in the chosen module “treatment, sewage, unpolluted, to wastewater treatment,
class 3”. Emissions to water inventoried are based on reported values after internal process water
treatment and before waste water treatment plant (Tab. 5.7).

Tab. 5.5 Production of ferrochrome. In- and outputs for the production of 1 t ferrochrome.
Input Unit Adelhardt & IPPC (2001) This study

Antrekowitsch
(1998)
low high
Chromite kg 2.43 2.4 3 2.43
Electricity total kWh 3325 3800 4500 3325
smelting (old) kWh 4000 4000
smelting (new) kWh 2500 2500
smelting (avg) kWh 3250 3250
de-dust kWh 74.8 74.8
crushing kWh 0.68 0.68
coke kg 340 120 225 340 a)
reductant 550 700 - a)
slag t 0.80 0.80
electrode kg 19.99 7 20 20.0
silica kg 46.2 46.2
fluxes kg 100 400 204 b)
process water m3 5 15 10
cooling water m3 5 15 10
Output
Cr in ferrochromium t 0.68
ferrochromium t 1 1 1
slag t 2.65 1 1.7 2.65
dust in scrubber kg 47.6 20 80 47.6
a)
Not evident, whether what reductant comprises coke or not. In this study it is assumed to be the sum
containing coke and other fuels. Therefore the reported value for coke in Adelhardt & Antrekowitsch
(1998) is taken as mass for reductant and inventoried as coke.
b)
Derived from the average flux of 250 kg less the 46.2 kg reported silica; this amount will be inventoried
as bauxite.

Tab. 5.6 Emissions to air in the production of 1 tonne ferrochrome. Values are form different sources: BREF (IPPC
(2001)), IPCC (IPCC (1996)), EPA (EPA (1998)) and EEA (EC (2002)). Chosen values for this study are the av-
erage of the given range, for contradictory values see notes below.
Emission Unit BREF IPPC EPA EEA This study

to Air
Low High Low High Low b) High a)
NOx kg / t 0.5 1.5 0.05 11.7 5.88 c)
NMVOC kg / t 1.7 1.7
CO kg / t 1 g) 1
PM<2.5 kg / t 0.96 49 25.0
PM2.5-10 kg / t 0.24 22 11.1
PM>10 kg / t 0 7 3.5
TSP kg / t 0.5 1 1.2 78 39.6 d)
SO2 kg / t 0.2 3 1.6
f)
PCDD μg / t 1 1
Cr III kg / t 1% 15% 8% e)
Cr VI kg / t 0.10% 0.30% 0.20% e)
CN kg / t 0.02 0.05 0.035
a)
No abatement
b)
ESP facility
c)
The average of the IPPC study is preferred to the value in BREF
d)
Averages of EPA values are chosen for world wide mix.
e)
Values given as percent in dust (TSP), resulting in emissions of 3.17 kg CrIII and 79.2 g CrVI per tonne ferrochrome.
f)
SNAP-Code 040309, „Other - Installations for smelting and casting of non ferrous metals“
g)
Value taken from ferrosilicon production
Tab. 5.7 Emissions to water in the production of 1 tonne ferrochromium. The reported values are before final waste-
water treatment and after internal abatement including lamella thickener, vacuum filter and a sand and car-
bon filter. Based on the chosen values, final emissions into water are calculated.
Emission to Water Unit BREF This Study

(before treatment) Low High
TSS kg / t 5 10 7.5 a)
Cr mg / t 10 20 15 b)
PAH mg / t 300 1000 650 c)
dry cake kg / t 200 400 300 d)
b)
Zimmermann et al. (1996) reports a transfer-coefficient for chromium into water of 51.4%. This gives
an emission after treatment of 7.7 mg Cr / t
a)
An elimination of 95% TSS assumed, resulting in an emission after treatment of 375 g TSS / t
c)
Based on Zimmermann et al. (1996) transfer coefficients of 10% to water is assumed, resulting in
emissions of 65 mg PAH / t. Emissions to air as CO2 and NMVOC are neglected.
d)
Dry cake results from internal waste-water treatment. In this study it is assumed to be disposed as inert
waste to an inert material landfill.
Data quality
Process data from Adelhardt & Antrekowitsch (1998) is satisfactory and valid for over 74 % of the
world-wide primary chromium production. Process specific emissions to air and water had to be taken
from another source covering another geography and technology. Nevertheless the overall quality of
the modules is fair.

in ecoinvent
Modul name
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
chromite, ore
Chromite  metals extraction No GLO concentrate, at 2.43 kg (Adelhardt 1998) 1 1.13 (2,2,3,1,1,2,4)
ferrochromium, high-carbon, 68%
beneficiation
not inventoried, assumed to be
slag  0.802 kg (Adelhardt 1998) 1 1.13 (2,2,3,1,1,2,4)
burden free (cut off)
electricity, medium
(Adelhardt 1998), (3,2,3,1,1,2,2); estimate
electricity, grid   waste heat 40% electricity from grid electricity production mix No UCTE voltage, production 1.33 kWh 1 1.16
estimations based on reported value
UCTE, at grid
electricity, hydropower,
(Adelhardt 1998), (3,2,3,1,1,2,2); estimate
hydropower   waste heat 60% electricity from hydropower hydro power power plants No RER at run-of-river power 2.00 kWh 1 1.16
estimations based on reported value
plant
hard coal, burned in
fossile energy
Cr, at plant
 hard coal heating systems No RER industrial furnace 1- 9.724 MJ (Adelhardt 1998) 1 1.13 (2,2,3,1,1,2,1)
(coke)
10MW
construction
silica sand  additives No DE silica sand, at plant 0.0462 kg (Adelhardt 1998) 1 1.13 (2,2,3,1,1,2,4)
materials
(IPPC, 2001),
fluxes  metals extraction No GLO bauxite, at mine 0.204 kg 1 1.74 reported values
estimations
Electrode anode, aluminium
 metals extraction No RER 0.0200 kg (Adelhardt 1998) 1 1.13 (2,2,3,1,1,2,3)
graphite electrolysis
process water  resource in water Water, river 0.0100 m3 (IPPC, 2001) 1 1.11 (2,3,2,3,1,3,12)
Water, cooling,
cooling water  resource in water unspecified natural 0.0100 m3 (IPPC, 2001) 1 1.11 (2,3,2,3,1,3,2)
origin
(4,4,1,1,1,5,5);
transport transport, transoceanic
transport (sea)  20000 km for bauxite ship No OCE 4.08 tkm estimate 1 2.11 calculation based on
systems tanker
assumptions
(4,4,1,1,1,5,5);
50 km for chromite, bauxit, silica, transport transport, lorry >16t,
transport (road)  road No RER 0.341 tkm estimate 1 2.11 calculation based on
coke, slag, cake systems fleet average
assumptions
production non-ferrous metal (5,5,1,1,1,5,9); rough
 metals extraction Yes GLO 7.2E-12 unit estimate 1 3.32
infrastructure smelter estimate
Fig. 5.5 Flows for “ferrochromium – inputs” and their representation in the ecoinvent database. Values correspond to the functional unit of 1 kg ferrochromium.

in ecoinvent
Modul name
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
treatment, sewage,
Inventoried as wastewater treatment waste
wastewater unpolluted, to
ferrochromium, high-carbon, 68% Cr, at
 Effluents w/o pollution. For overall process, manageme No CH 0.010 m3 (Adelhardt 1998) 1 1.13 (2,2,3,1,1,2,6)
treatment wastewater treatment,
waterborn emission see Tab. 5.7 nt
class 3
waste disposal, inert waste,
inert material
 dry cake due to effluent, Tab 5.7 manageme No CH 5% water, to inert 0.300 kg (IPPC, 2001) 1 1.33 reported values
landfill
nt material landfill
low population
 NOx air Nitrogen oxides 5.9E-3 kg (IPPC, 2001) 1 1.99 reported values
density
NMVOC, non-methane
low population volatile organic
 NMVOC air 1.7E-3 kg (IPPC, 2001) 1 1.52 (2,3,2,3,1,3,16)
density compounds,
unspecified origin
Dust consists of PM<2.5, PM2.5-10, PM>10,
including Cr III, Cr VI and CN These specific based on reported
 Dust (TSP) 4.0E-2 kg (IPPC, 2001) 1 1.97
emissions are inventoried according to table 5.6, values
but they are not shown in this figure.
plant
low population based on reported

 SO2 air Sulfur dioxide 1.6E-3 kg (IPPC, 2001) 1 1.88
density values
low population Carbon monoxide,
 CO air 1.0E-3 kg (IPPC, 2001) 1 5.01 (2,3,2,3,1,3,17)
density fossil
low population 2,3,7,8-
 PCDD de-novo synthesis air 1.0E-12 kg (IPPC, 2001) 1 3.01 (2,3,2,3,1,3,21)
density tetrachlorodibenzo-p-
dioxin
inventoried as disposal of inert
material, thus neglecting the waste disposal, inert waste,
inert material
 slag, disposal potentially important (but manageme No CH 5% water, to inert 2.65 kg (Adelhardt 1998) 1 1.1 (2,2,2,3,1,3,6)
landfill
quantitatively unknown) heavy metal nt material landfill
emissions
waste disposal, inert waste,
inert material
 dust in scrubber from wet scrubber, deposed in landfill manageme No CH 5% water, to inert 0.048 kg (Adelhardt 1998) 1 1.11 (2,3,2,3,1,3,6)
landfill
nt material landfill
 ferrochromium metals extraction No GLO carbon, 68% Cr, at 1 kg (Adelhardt 1998) 1
plant
Fig. 5.6 Flows for “ferrochromium – outputs” and their representation in the ecoinvent database. Values correspond to the functional unit of 1 kg ferrochromium.

5.2.5 Ferrochromium, high-carbon, 68% Cr, at regional storage in ecoinvent

The module describes the ferrochromium consumed Europe in 1994. The situation for Germany de-
scribed in Adelhardt & Antrekowitsch (1998) is taken as proxy for Europe. It is designed for the use of
the metal as alloying element. The layout of the modules with general flow information, remarks,
sources, values and uncertainty information are shown in Fig. 5.7. Ecoinvent meta information for this
process are summarised in Tab. 5.1 on page 13. The following paragraphs describe calculations,
sources and assumptions chosen in this study. The functional unit of this process is one tonne of ferro-
chrome consumed in Europe with an average chromium content of 68%.
Adelhardt & Antrekowitsch (1998) examined the transport means and distances of ferrochromium
from different supplying countries to Germany (Tab. 5.8 and Tab. 5.9). Since imported chromite is de-
termined for the production of low-carbon ferrochrome, no import of chromite is considered for the
production of high-carbon ferrochrome, i. e. all of the high-carbon ferrochrome consumed in Europe is
considered to be imported. Besides transport services no further impacts – such as energy consumption
through reloading and transferring – are inventoried.
Tab. 5.8 Chromite transports to Germany for low-carbon ferrochrome production in 1994. Mass, distance and result-
ing transport intensity. Data from Adelhardt & Antrekowitsch (1998).
Unit South Africa Turkey Brasilia other total

chromite t 128880 25540 10700 7410 172530
ship sea km 15500 8900 12200 12654 tkm
train km 1200 300 300 600 985 tkm
lorry km 300 300 300 76 tkm
Tab. 5.9 Ferrochromium transports to Germany in 1994. Mass, distance and resulting transport intensity. Data from
Adelhardt & Antrekowitsch (1998).
unit South Africa Russia Norway other total

ferrochromium t 88670 85470 45040 135480 354660
a)
ship sea km 15500 3159 17900 12186 11565 tkm
train km 900 5300 c) 2800 b) 3000 3004 tkm
a)
Value consists of 2'400 km direct transport from Norway and additional 15'500 km previous trans-
portation to Norway
b)
Value consists of 2'400 km direct transport from Norway and additional 2'500 km previous trans-
portation to Norway.
c)
Value consists of 2'500 km direct transport from Russia and additional 2'500 km previous transpor-
tation to Russia.

Mean value
Source for
ecoinvent
Deviation
Comment
Standard
Category
Location
category
structure
Remarks
Output
name in
General
Process
Modul
Name
value
Input
Infra-
mean
Sub-
Type
Unit
ferrochromium ferrochromium, high-
 global average metals extraction No GLO carbon, 68% Cr, at 1 kg 1 1.13 (2,2,3,1,1,2,4)
production carbon, 68% Cr, at plant
Transport regional storage (2,2,3,1,1,2,5);

transport, calculation
ferrochromium  distances and means see tab 5.9 ship No OCE 11.6 tk m (Adelhardt 1998) 1 2.02
transoceanic tanker based on
over sea
assumptions
(2,2,3,1,1,2,5);
Transport
transport calculation
ferrochromium  distances and means see tab 5.9 train No RER transport, freight, rail 3.00 tk m (Adelhardt 1998) 1 2.02
systems based on
by train
assumptions
ferrochromium, at
 metals extraction No RER carbon, 68% Cr, at 1 kg
regional storage
regional storage
Fig. 5.7 Flows for “Ferrochromium, at regional storage” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg ferrochromium.

5.3 Production of metallic chromium

Chromium metal can either be made from chromite ore and concentrates by carbo- and metallo-
thermic reductions or by electrolysis from chromic acid. In order to get chromic acid the chromite ore
is roasted with sodium carbonate followed by a leaching process with sulphuric acid to produce so-
dium chromate that can further be transformed to chromic acid IPPC (2001).
Aluminothermic process
The usual method of reduction by carbon is inapplicable to the reduction of refractory oxides and ores.
The products commonly covered under this title are usually required to be low in carbon and they are
avid carbide formers. Also, no conventional refractories would stand up to the long reduction times at
the high temperatures required. For such products, advantage is taken of the high temperatures
achieved and the rapidity of metallothermic reactions using reactive metals, notably Al (Hall (1997)).
The thermic process uses aluminium as a reducing agent for chromium hydroxide. The charge is
weighed and loaded into a bin, which is taken to an enclosed room to mix the contents. The firing pot
is prepared by ramming refractory sand mixed with water around a central former. After ramming the
firing pot, the inner surface is coated with a weak binder solution and dried under a gas fired hood be-
fore being transferred to the firing station. The raw material mix is automatically fed at a controlled
rate into the firing pot, where the exothermic reaction takes place. When the metal has solidified fol-
lowing the reaction, the firing pot is removed and transferred by crane to a cooling conveyor. On re-
moval from the cooling conveyor (by crane), the firing pot is placed on a stripping bogie for transfer-
ral to a stripping booth. Inside the closed booth, the pot casing is hoisted off the solidified metal/slag.
The slag is separated from the Chromium metal “button” and sent to a despatch storage area. Water is
used to reduce button temperature to below 100 ºC. After cooling the metal button is transferred to
other departments on site for cleaning, breaking, crushing and grinding to achieve the desired product
size (IPPC (2001)).
Electrolysis
In the electrolytic process normally high carbon ferrochrome is used as the feed material which is then
converted into chromium alum by dissolution with sulphuric acid at temperatures at about 200 ºC. Af-
ter several process steps using crystallisation filtration ageing, a second filtration and a clarifying op-
eration the alum becomes the electrolyte for a diaphragm cell. Chromium is plated onto stainless steel
cathodes until it attains a thickness of ca. 3 mm. The process is very sensitive. The additional de-
gassing (heating at 420 °C) stage is necessary because the carbon content of the electrolytic chromium
is sometimes too high for further industrial applications. The cooled chromium metal is fragmented
with a breaker prior to crushing and drumming. The generated slag can be reused as refractory lining
or sold as abrasive or refractory material. IPPC (2001).
5.3.1 Emissions & Wastes

The general manufacturing of refractory metals normally takes place at high temperatures. The envi-
ronmental impact on air, water and land can be classified as follows.
Emissions to air: Dust and fume emissions from smelting, hard metal and carbide production; other
emissions to air are ammonia (NH3), acid fume (HCl), hydrogen fluoride (HF), VOC’s and heavy met-
als.
Emissions to water: Overflow water from wet scrubbing systems; Wastewater from slag and metal
granulation; Blow down from cooling water cycles.
Wastes: Dust, fume and sludge; Slag.

5.3.2 “Chromium, at regional storage” in Ecoinvent

The module describes the chromium metal consumed in Europe in 1994. The situation for Germany
described in Adelhardt & Antrekowitsch (1998) is taken as proxy for Europe. It is designed solely for
the use of the metal as alloying element in superalloys (Ni-Cr). The layout of the modules with general
flow information, remarks, sources, values and uncertainty information are shown in Fig. 5.8. Ecoin-
vent meta information for this process are summarised in Tab. 5.1 on page 13. The following para-
graphs describe calculations, sources and assumptions chosen in this study. The functional unit of this
process is one tonne of chromium consumed in Europe.
11'150 t chromium metal was produced 1989 by aluminothermic processes, whereas 4'350 t came from
electrolytic winning (Adelhardt & Antrekowitsch (1998)). The share in electrolytic processes is not
rising significantly; a production level of 5'000 t in 2003 is assumed. This gives a share of the pre-
ferred aluminothermic process of 75 %.
Aluminothermic process
The aluminothermic process uses chromium hydroxide (Cr2O3) as raw material and aluminium powder
as reducing agent. It is the preferred production process with an assumed share of 75 %. Hall (1997)
describes the fundamental reaction practice as follows:
Cr2O3 + 2 Al  2 Cr + Al2O3 (-H = – 548 kJ) (1)

Heat required for products at 2000 °C amounts to 540 kJ. Al2O3 (H2273-H298) is 356 kJ, 2 Cr
(H2273-H298) is 184 kJ. According to this, there is a heat balance, so there should be a satisfactory
result. However, it is found in practice that 125 kJ extra heat are required, provided by Al plus
K2Cr2O7. The extra heat reaction can be formalised as follows
K2Cr2O7 + 4 Al  2 Cr + 2 Al2O3 + K2O (-H = – 1768 kJ) (2)

The energy required to heat the additional K2O is assumed to be 200 kJ, which leads to a stoichiomet-
ric factor of (1768 – (184+ 2 · 356+200))/125 = 5.4. An addition of 5.4 · (1) + (2) gives:
5.4 Cr2O3 + 14.8 Al + K2Cr2O7  12.8 Cr + 7.4 Al2O3 + K2O (3)

which gives the final stoichiometry. Hence per kg chromium 442 g K2Cr2O7, 600 g Al and 1.23 kg
Cr2O3 is necessary. Adelhardt & Antrekowitsch (1998) gives a yield of 89 % - 92 %, an average of
90.5 % is chosen.
Assuming a Cr2O3 content of 44.9% in the chromite concentrate and a yield of 90% in the Cr2O3 pro-
duction from chromite, for the 1.23 kg Cr2O3 3.05 kg chromite concentrate are required.
CaO is not used as flux in Cr manufacture as it spoils the slag for use in refractories and calcium
chromite is a stable compound. CaF2 often is used as its low melting point of 1418 °C assists in pro-
moting a smooth reaction. The amount used is often 1 – 2 % of the total charge, never more than 5 %
(Hall (1997)). However, HF emissions are not considered in this study due to the lack of data.
For transportation the information in Adelhardt & Antrekowitsch (1998) is used. It is assumed, that the
chromium is produced in Europe and the chromite is imported in the same transportation mix as illus-
trated in Tab. 5.8.
Adelhardt & Antrekowitsch (1998) states that the energy needed for the production of Cr2O3 is 85 MJ
per tonne chromium content, 250 MJ / t-Cr for the aluminothermic winning and 25 MJ / t-Cr for the
refining. The 250 MJ / t-Cr comprise the production of the aluminium powder, which amounts to 185
MJ / t-Al according to EAA (2000). With this, a sum of 317 MJ per tonne metallic chromium is re-
quired: 153 MJ for the Cr2O3-production (1.23 kg with 68% chromium contained), 139 MJ for the
aluminothermic process and finally 25 MJ for the refining. The production of aluminium and dichro-
mate are not included in this figures and inventoried in the respective modules. The energy in the

Cr2O3-production is assumed to originate from natural gas. For the aluminothermic process and refin-
ing a share of 60% hydroelectricity and 40% UCTE-mix is assumed.
Electrolytic process
The rest of chromium (25 %) is still produced from ferrochrome in an electrolytic process. No infor-
mation on material input and output were available. The energy demand is stated in Adelhardt &
Antrekowitsch (1998) to be 18 kWh, whereas Papp (1994) reports 22.7 kWh per tonne chromium cit-
ing industry sources in 1993. In this study the higher value of 22.7 kWh (=81.8 MJ) was chosen. Con-
siderable amounts of hydrogen and ammonia ought to be used for reduction and leaching, but no de-
tailed information is available. Hence none of these emissions are taken in account.

Mean value
Source for
ecoinvent
Deviation
Comment
Standard
Category
Location
category
structure
Remarks
Output
name in
General
Process
Modul
Name
value
Input
Infra-
mean
Sub-
Type
Unit
chromite, ore (5,2,2,1,1,5,4);
chromite as raw
 metals extraction No GLO concentrate, at 2.53 kg Calculation 1 1.57 estimate based on
material
benef iciation assumptions
(5,2,2,1,1,5,4);
sodium dichromate,
Dichromate  chemicals inorganics No RER 0.366 kg Calculation 1 1.57 estimate based on
at plant
assumptions
(5,2,2,1,1,5,4);
Aluminium aluminium, primary, at
 metals extraction No RER 0.498 kg Calculation 1 1.57 estimate based on
pow der plant
chromium, at regional storage
assumptions
Energy for natural gas, burned
heating
chromium  assumed to be from gas natural gas No RER in industrial f urnace 103.0 MJ (A delhardt 1998) 1 1.13 (2,2,3,1,1,2,1)
systems
hydroxide >100kW
Hydropow er for assumed to be 60% of electricity,

hydro
aluminothermic   w aste heat aluminothermic process and pow er plants No RER hydropow er, at run- 22.6 k Wh (A delhardt 1998) 1 1.16 (3,2,3,1,1,2,2)
pow er
proc. & ref ining refining of -river pow er plant
electricity f or assumed to be 40% of electricity, medium

aluminothermic   w aste heat aluminothermic process and electricity production mix No UCTE voltage, production 15.09 k Wh (A delhardt 1998) 1 1.16 (3,2,3,1,1,2,2)
proc. & ref ining refining UCTE, at grid
Transport (3,2,3,1,1,2,5);
transport transport,
chromite over  ship No OCE 12.7 tk m (A delhardt 1998) 1 2.03 calculation based on
systems transoceanic tanker
sea reported values
Transport (3,2,3,1,1,2,5);
transport
chromite by  train No RER transport, freight, rail 0.99 tk m (A delhardt 1998) 1 2.03 calculation based on
systems
train reported values
Transport (3,2,3,1,1,2,5);
transport
chromite by  road No RER transport, lorry 32t 0.076 tk m (A delhardt 1998) 1 2.03 calculation based on
systems
road reported values
Ferrochromium
 metals extraction No GLO carbon, 68% Cr, at 0.250 kg (A delhardt 1998) 1 1.13 (2,2,3,1,1,2,4)
for electrolysis
plant
electricity, medium
Energy for
  w aste heat electricity production mix No UCTE voltage, production 5.68 k Wh (Papp 1994) 1 1.13 (2,2,3,1,1,2,2)
electrolysis
UCTE, at grid
Inf rastructure,
unspecified for smeltery taken as proxy, all mass non-ferrous metal, (5,5,1,1,1,5,9); rough
 metals extraction Yes GLO 1.08E-11 unit rough estimation 1 3.32
aluminothermic input (2.84kg) accounted for. smelter estimate
and electroylsis
ferrochromium, at chromium, at regional
 global average metals extraction No RER 1.00 kg
regional storage storage
Fig. 5.8 Flows for “Chromium, at regional storage” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg ferrochromium.

6. Literature
6 Literature
Adelhardt & Antrekowitsch (1998) Adelhardt W. and Antrekowitsch H. (1998) Stoffmengenflüsse und Energie-
bedarf bei der Gewinnung ausgewählter mineralischer Rohstoffe; Teilstudie
Chrom. In: Geologisches Jahrbuch, Vol. Sonderhefte SH 3. Bundesanstalt für
Geowissenschaften und Rohstoffe, Hannover. ISBN 3-510-95831-4.
www.ecoinvent.ch.
Anger et al. (1997) Anger G., Halstenberg J., Hochgeschwender K., Uecker G., Korallus U., Knopf
H., Schmidt P. and Ohlinger M. (1997) Chromium Compounds. In: Ullmann's
encyclopedia of industrial chemistry (ed. Anonymous). 5th edition on CD-ROM
Edition. Wiley & Sons, London.
Anonymous (1998a) Anonymous (1998a) Dust Control. In: Best Practice Environmental Manage-
http://www.ea.gov.au/ industry/sustainable/mining/booklets/dust/.
Anonymous (1998b) Anonymous (1998b) Pollution Prevention and Abatement Handbook. The World
Bank Group, Online-Version under: http://lnweb18.worldbank.org/essd/ ess-
dext.nsf/51ByDocName/PollutionPreventionandAbatementHandbook.
Downing et al. (1997) Downing J. H., Deeley P. D. and Fichte R. D. (1997) Chromium and Chromium
Alloys. In: Ullmann's encyclopedia of industrial chemistry (ed. Anonymous).
5th edition on CD-ROM Edition. Wiley & Sons, London.
EAA (2000) EAA (2000) Environmental Profile Report for the European Aluminium Indus-
try. European Aluminium Association (EAA), Brussels.
EC (2002) EC (2002) European Dioxin Inventory. European Commission. Retrieved from
http://europa.eu.int/comm/environment/dioxin/download.htm. Date of last revi-
sion: 22.08.2002.
EEA (1999) EEA (ed.) (1999) Primary Copper Production (B336). In: EMEP / CORINAIR
Fabian (1997) Fabian H. (1997) Copper. In: Ullmann's encyclopedia of industrial chemistry
(ed. Anonymous). 5th edition on CD-ROM Edition. Wiley & Sons, London.

6. Literature
Hall (1997) Hall F. W. (1997) Nickel. In: Ullmann's encyclopedia of industrial chemistry
(ed. Processes A.). 5th edition on CD-ROM Edition. Wiley & Sons, London.
Hischier (2003) Hischier R. (2003) Life Cycle Inventories of Packaging and Graphical Paper.
Final report ecoinvent 2000 No. 11. EMPA St. Gallen, Swiss Centre for Life
Cycle Inventories, Dübendorf, CH, Online-Version under: www.ecoinvent.ch.
IPCC (1996) IPCC (1996) Reference Manual. In: Revised 1996 IPCC Guidelines, Vol. 3,
Online-Version under: http://www.ipcc-nggip.iges.or.jp/public/gl/invs1.htm.
European Commission. Retrieved from http://www.jrc.es/pub/english.cgi/
0/733169
Papp (1994) Papp J. F. (1994) Chromium Life Cycle Study. IC-9411. U.S. Department of the
Interior, U.S. Geological Survey, Online-Version under: http://pubs.usgs.gov/
usbmic/ic-9411/ic-9411.pdf.
Papp (2001) Papp J. F. (2001) Chromium Recycling in the United States in 1998. IC-1196.
U.S. Department of the Interior, U.S. Geological Survey, Online-Version under:
http://pubs.usgs.gov/circ/c1196c/.
U.S. Geological Survey. Retrieved at 01.03.2003 from http://minerals.usgs.gov/
minerals/pubs/mcs/. Date of last revision: 26-Feb-2003@09:36.
Zürich.

Part VII
Manganese
Data v2.1 (2009)

Summary

This part corresponds to the original report in Althaus et al. (2004). None of the changes reported in
Frischknecht et al. (2006) regarded the datasets described in this part. Therefore no alteration of the original text
was made for version 2.0.
Summary
This chapter describes the production of ferromanganese and metallic manganese from cradle to consumer in
Europe. Inventory data for the production is taken mainly from Adelhardt & Saiger (1999) together with back-
ground data from IPPC (2002), IPPC (2001) and Wellbeloved et al. (1997).
Only the primary production is inventoried because manganese is mainly used in alloys and recycled in them.
Ferromanganese is to be used for alloying steel while manganese is used for alloying aluminium or for direct use
(e.g. sputtering)

Table of Contents
Table of Contents
SUMMARY ......................................................................................................................I
1 INTRODUCTION ....................................................................................................... 1
2 RESERVES AND RESOURCES OF MANGANESE ........................................................... 2

2.1 Primary Manganese............................................................................................................... 2
2.2 Secondary Manganese........................................................................................................... 2
3 CHARACTERISATION OF THE MATERIALS .................................................................. 4

3.1 Manganese ore....................................................................................................................... 4
3.2 Manganese Metal .................................................................................................................. 4
3.3 Ferromanganese..................................................................................................................... 4

4.1 Manganese ore....................................................................................................................... 7
4.2 Manganese Metal .................................................................................................................. 7
4.3 Ferromanganese and Silicomanganese.................................................................................. 7

5.1 Exploitation of manganese ore ............................................................................................ 13
5.1.2 Emissions & Waste ............................................................................................................. 13
5.1.3 “Manganese, concentrate, at beneficiation” in ecoinvent ................................................... 13
5.2 Winning of high carbon ferromanganese ............................................................................ 20
5.2.2 Emission.............................................................................................................................. 21
5.2.3 Waste................................................................................................................................... 21
5.2.4 Ferromanganese in ecoinvent.............................................................................................. 22
5.3 Production of metallic manganese....................................................................................... 28
5.3.1 Emissions & Wastes............................................................................................................ 28
5.3.2 “manganese, pure metal, at regional storage” in ecoinvent................................................. 29
6 LITERATURE ......................................................................................................... 31

1. Introduction
1 Introduction
This chapter describes the production of ferromanganese and metallic manganese from cradle to con-
sumer in Europe. Inventory data for the production is taken mainly from Adelhardt & Saiger (1999)
together with background data from IPPC (2002), IPPC (2001) and Wellbeloved et al. (1997).
Like many metals used as alloying element, manganese is recycled from steel-scrap in foundries di-
rectly as secondary alloys. Depending on their composition, these secondary manganese containing
materials are used together with ferromanganese – and manganese ore – as secondary source for man-
ganese. In this study, only the primary material ferromanganese and metallic manganese is considered.
Secondary manganese sources – i.e. alloys from recycled steel scrap – have to be inventoried in the
respective production module, e.g. in the steel production.
Over the last 15 to 20 years, the world pattern of the ferro-alloy market has deeply changed:
- The consumption of the developing countries has vastly increased with the development of their
steel production;
- Their production has even more increased because they have taken a growing share of the tradi-
tional markets of the industrialised countries, where the steel production was stagnant or growing
at a slow rate.
- The ferro-alloy industry is facing a growing proportion of imports, at first from the new industrial-
ised countries and in recent years from the countries of East Europe and the CIS (former Soviet
Union).
As a consequence, EU ferro-alloy production has been submitted to a difficult competition that results
in a decreasing trend of the total amount of ferro-alloys produced. Although the last years showed
some stabilisation, even a slightly increase of the production, the European industry remains very sen-
sitive.

2. Reserves and Resources of Manganese
2 Reserves and Resources of Manganese

cle. When recovered the properties are theoretically identical to those in the former raw material. The
metal can be re-used in another product life cycle, provided that its concentration makes a recovery
economic. The recovered metal is referred to as secondary metal, whereas the metal originating di-
rectly from the ore is referred to as primary metal. In LCIs these two types – primary and secondary
metal – are often differentiated because they bear a different ecological burden.
2.1 Primary Manganese

Manganese is contained in many minerals, of which approximately 250 can be regarded as true man-
ganese minerals. The major ores, generally with manganese contents above 35 %, are hydrated or an-
hydrous oxides. The silicates and carbonates occur to a lesser extent. Economically significant manga-
nese deposits are all of sedimentary origin, having been dissolved from crystalline rocks and deposited
as the oxide, hydroxide, or carbonate. The major manganese minerals are listed in Tab. 2.1
(Wellbeloved et al. (1997)). Wad is the term given to the soft, earthy, hydrous mixture of manganese
and iron oxides of variable manganese content.
The most important manganese ore deposits are located in the Republic of South Africa, the Soviet
Union, Australia, Gabon in West Africa, and Brazil. The deposits in India and Mexico are of declining
importance, and the majority of the production is consumed domestically (Tab. 2.2).
Another potentially important source of manganese are the deep-sea nodules located in certain areas of
the ocean floor. Its exploitation has not yet been proven economic.
2.2 Secondary Manganese

The major end use of manganese is in iron and steel, and this is the major form in which manganese is
recycled (ISS, iron and steel scrap). A smaller fraction ends up as aluminium beverage cans, which are
recycled too. On a weight basis, aluminium beverage cans typically consisted 75% of bodies made
from alloy 3004 and 22% of lids made from alloy 5182 in the US in 1994. The nominal manganese
content of alloy 3004 is 1.1%, and that of alloy 5182 is 0.35% (USGS (1998)).
Manganese scrap enters the production process in the furnaces of the steel industry in the form of al-
loys with a specific composition. USGS (1998) reports a recycling rate of 37% and a manganese loss
in the recycling process of 52% for the US in 1998.
Since no manganese is recycled within the primary industry but directly to the foundries of the iron
and steel industry, no secondary manganese is considered in this study.

2. Reserves and Resources of Manganese
Tab. 2.1 Manganese minerals of economic importance (Wellbeloved et al. (1997))
Mineral Formula Theoretical Mn content, % Relative density

Pyrolusite -MnO2 63.2 5
Mn Mn63+SiO12
2+
Braunite 66.6 4.7–4.9
Manganite -MnOOH 62 4.3
Psilomelane (Ba, Mn2+)3(O, OH)6Mn8O16 45–60 4.4–4.7
Cryptomelane KMn8O16 62 4.3
Hausmannite Mn2MnO4 73 4.7–5.0
Jacobsite Fe2MnO4 23.8 4.8
Bixbyite (Mn, Fe)2O3 30–40 5
Rhodonite (Mn, Fe, Ca)SiO3 42 3.4–3.6
Rhodochrosite MnCO3 47.6 3.3–3.6
Bementite Mn3Si2O5(OH)4 43.2 3.5
Tab. 2.2 Mine production and reserves of manganese in 2001. “Reserves” is referred to as “the part of the reserve
base which cold be economically extracted or produced at the time of determination”, while “Reserve base”
denotes “that part of an identified resource that meets specified minimum (...) criteria related to current min-
ing and production practices” (USGS (2003)).

e
2001 [kT Mn / a] 2002 [kT Mn / a] [kT Mn] [kT Mn]
Australia 948 890 32000 82000
Brazil 1430 e 1500 25000 52000
China 500 e 500 40000 100000
Gabon 830 e 860 20000 160000
India 600 e 630 15000 33000
Mexico 100 100 4000 9000
South Africa 1479 1300 18000 114000000
Ukraine 930 e 960 140000 520000
Other countries 750 e 860 Small Small
World total (rounded) 7600 e 7600 300000 5000000
e
estimated

3 Characterisation of the Materials

Manganese ore is the raw material for several manganese alloys and manganese chemicals and finally
manganese metal. Generally, high-carbon ferromanganese and silicomanganese are produced from a
blend of manganese-containing ores, and in the case of silicomanganese, slags and silica are added.
3.1 Manganese ore

Typically, only ores containing a minimum of 35 % Mn are regarded as manganese ores. Ores contain-
ing 10 – 35 % Mn are categorised as ferruginous manganese ores, and iron ores with 5 – 10 % Mn are
referred to as manganiferous ores (Wellbeloved et al. (1997)). Pyrolusite (a mineral form of manga-
nese dioxide), braunite (an oxysilicate), and rhodochrosite (a manganese carbonate) are among the
minerals more commonly found in manganese ores (USGS (1998)).
The three most important areas of application of manganese ores and the specific requirements, are as
follows (Wellbeloved et al. (1997)):
1. Metallurgical grade ore for the iron and steel industry, which generally contains 38 – 55 % Mn.
The phosphorus content should preferably be below 0.1 %, and the concentrations of Al2O3, SiO2,
CaO, MgO, and S are important. The manganese/iron ratio is critical; a 7.5 : 1 ratio, for example,
is required for a standard ferromanganese alloy with 78 % Mn.
2. Battery grade ore, containing 70 – 85 % MnO2 (44 – 54 % Mn). The ore should generally contain
less than 0.05 % of metals more electronegative than zinc, such as copper, nickel, cobalt, and ar-
senic. The suitability of manganese dioxide for use in batteries depends on a number of factors,
including the crystal structure, surface area, pore size distribution, particle shape and size, electri-
cal conductivity, surface conditions, chemical composition, and structure defects.
3. Chemical grade ore whose specifications vary considerably depending on the end use. Included in
this category are feed stocks for electrolytic manganese and manganese dioxide, manganese
chemicals, colorants and, in the Republic of South Africa, an oxidant in uranium extraction.
Compositions of ores of some major operations are given in Tab. 3.1 and Tab. 3.2.
3.2 Manganese Metal

Manganese is a silver-grey metal, resembling iron. It is hard and very brittle, and its primary uses in a
metallic form are as an alloying, desulphurising, and deoxidising agent for steel, cast iron, and non-
ferrous metals. Due to its high tendency to oxidise, annealing colours are often evident. Manganese
belongs to group 7 and period 4 of the periodic table with an atomic number of 25 and atomic mass of
54.94. Six isotopes have been produced, of which only 55Mn is stable. The other isotopes, 51Mn, 52Mn,
54
Mn, 56Mn, and 57Mn, have half-lives ranging from 46 min to 310 d. Manganese exists in four allo-
tropic forms: , and . It has a low vapour pressure (Wellbeloved et al. (1997)).
Selected physical properties of manganese are given in Tab. 3.3.
3.3 Ferromanganese
A number of manganese-containing ferroalloys are manufactured which are used largely in the mild
steel, foundry, and stainless steel industries. The principal manganese ferroalloys and their typical
components are high-carbon ferromanganese (78% manganese, 7% carbon, balance mostly iron) and
silicomanganese (66% manganese, 17% silicon, 2% carbon). The names and typical compositions of
these alloys are given in Tab. 3.4.

Tab. 3.1 Analyses of some manganese ores, metallurgical grade (Wellbeloved et al. (1997))
Producer Product Mn Fe P SiO2 Al2O3 CaO MgO

Samancor, RSA Mamatwan Lumpy 37–38 4–6 0.08 4–6 0.5 14–16 2.4
Samancor, RSA Mamatwan Ferruginous 32–34 4–6 0.08 4–6 0.5 14–16 3.5
Samancor, RSA Wessels High Grade 50–51 9–11 0.08 5–7 0.5 4–6 0.6
Samancor, RSA Wessels Low Grade 38–40 18–20 0.08 5–7 0.5 4–6 1.2
Groote Eylandt, Australia Metallurgical Lump 49.0 2.5 0.06 6.0 4.0 0.05 0.10
Groote Eylandt, Australia Siliceous Lump 44.0 2.5 0.06 15.0 4.0 0.05 0.10
Amapa, Brazil Bitolado 48 48–49 5.5–6.9 0.08–0.11 2–3 3.7–5.5
Amapa, Brazil Carbonate Grosso 33–37 2.5–4 0.04–0.10 10–14 2.0–3.8 3.3–5.2 2.2–4.5
CVRD, Brazil Lumpy 45.0 3.7 0.08 3.6 9.8 0.25 0.30
Comilog, Gabon Gabon Lumpy 48–51 3–5 0.11–0.14 2.5–4 6 0.11 0.07
Soviet Union Tchiaturi Fines 48.2 1.1 0.142 9.2 1.6 2.26 0.50
Tab. 3.2 Typical compositions of some important battery grade manganese ores (chemical content on % dry basis,
Wellbeloved et al. (1997)
Producer MnO2 Mn Fe Pb Cu Ni As Co
CVRD, Brazil 76.5 51.0 3.2 0.035 0.050 0.080 0.004 0.035
Comilog, Gabon 83.6 1.8 0.03 0.05 0.095

Tab. 3.3 Some physical properties of metallic manganese (Wellbeloved et al. (1997)).
Property -Mn -Mn -Mn -Mn Molten Mn

Atomic number 25
Atomic mass 54.938
Crystal structure complex cubic complex cubic fcc bcc
Lattice parameter, nm 0.8894 0.6290 0.3774 0.3080
Transition temperature, °C    liquid liquidgas
727 1095 1104 1244 2032
Standard electrode potential, V 1.134
Magnetic susceptibility, m3/kg 1.21×10–7
Density, g/cm3 7.44 7.29 7.21 7.55–1.2×10–3 t
(1244–1500°C)
Brinell hardness, MPa 3923–4119
Hardness, Mohs scale 5
Volume increase on transition,    liquid
% 3.4 0.8 0.8 2.9
Resistivity,  cm 150–260×10–6 90×10–6 40×10–6
Compressibility, m2/N 8.4×10–7
Coef. of linear expansion, K–1 22.3×10–6 24.9×10–6 14.8×10–6 41.6×10–6
Latent heat of fusion, J/g 244
Latent heat of vaporization, J/g 4020
Heat of transition, J/mol 2240 2282 1800 14655 219901
Specific heat, Jg–1K–1 0.477 0.482 0.502
Enthalpy, J/mol 4999 5112
Entropy, Jmol–1K–1 32.0 32.3 173.8
Tab. 3.4 Types of ferromanganese and their general compositions (Wellbeloved et al. (1997)).¨
Alloy Composition, %
Manganese Carbon Silicon
High-carbon ferromanganese (carburé) 72-80 7.5 <1.25
Medium-carbon ferromanganese (affiné) 75-85 <2.0
Low-carbon ferromanganese (suraffiné) 76-92 0.5-0.75
Silicomanganese 65-75 <2.5 15-25
Ferromanganese silicon 58-72 0.08 23-35
Spiegeleisen 16-28 <6.5 11-45


Metallurgical applications account for most manganese consumption, of which 85% to 90% in the US
has been going to steel-making and about 8% to the manufacture of dry cell batteries. The preponder-
ance of the manganese used for making batteries is now electrolytic manganese dioxide (EMD) be-
cause usage of natural battery ore has declined greatly. The manufacture of manganese chemicals,
such as potassium permanganate, and agricultural use of manganese in animal feed and plant fertiliser
as oxide, sulphate, and oxysulphate together account for another 5% of use. These patterns of domestic
use are typical for other industrialised countries having well-developed steel industries (Fig. 4.1).
In non-metallurgical uses, the most common valences for manganese are two and four, and oxygen is
the main element with which manganese is combined. Accordingly, the mineral and commodity chem-
istry of manganese centers on such compounds as manganous oxide (MnO), manganese dioxide
(MnO2), manganese carbonate (MnCO3), and manganese sulphate (MnSO4) (USGS (1998)). The pre-
dominant products and their production route is illustrated in Fig. 4.2.
4.1 Manganese ore

Manganese ore is the central intermediary product in the manganese industry. From it the three main
product groups are made: ferromanganese for the use in the steel industry (over 90%) and manganese
compounds for the use in the chemical industry and battery manufacture. As illustrated in Fig. 4.1, bat-
tery or chemical grade ore are processed into manganese metal or manganese oxides, whereas from
metallurgical grade ores manganese alloys are made. Silicomanganese is made predominantly from
high-silica ore.
4.2 Manganese Metal

The largest application of manganese metal in the metallurgical sector is in aluminium alloying, par-
ticularly the alloys from which items such as beverage cans and siding are made. The predominant
form in which manganese is added to molten aluminium is briquette (typically 75% manganese, 25%
aluminium) made by compacting powders of the two metals. In 1990, the amount of manganese metal
used by the steel industry was about one-third as much as by the aluminium industry. Because use of
manganese metal, as opposed to ferroalloy, is a relatively expensive way of adding manganese, the
metal is added in steel making as a “trimming addition” to adjust final composition, particularly when
an increase in carbon content is not permissible. Today the world market is in excess of 140×103 t/a.
The compositions of some alloys are given in Tab. 4.1.
4.3 Ferromanganese and Silicomanganese

Generally, high-carbon ferromanganese and silicomanganese are produced from a blend of manga-
nese-containing ores, and in the case of silicomanganese, slag and silica are added. USGS (2003) es-
timates that manganese unit consumption in steel making in the US in 2001 amounts to 5.4 kg manga-
nese per tonne raw steel. The use of manganese alloys depends strongly on the steel industry. Tab. 4.2
gives an overview over the end uses of ferromanganese in the U.S. in 2001.

Fig. 4.1 Summary of manganese products and their process routes (Wellbeloved et al. (1997)).
Fig. 4.2 Simplified diagram for the production of key manganese chemicals (Jones (1994)).

Tab. 4.1 Manganese in alloys. Composition of typical alloys containing manganese (Wellbeloved et al. (1997)).
Alloy Mn, % Cu, % Ni, % Zn, % Pb, % Al, % Fe, %

Copper manganese 10-40 56-70
Manganese bronze 5-15
Resistin 15
Hard brass 1.5 56 42
Special brass 58 Al2 2 58 3 37 0.5 1.6
Special brass 59 2.5 58 3 1.5
Special brass 64 4 65 7 4
Manganin 12 83 3-5 0.2
Tab. 4.2 End uses of ferromanganese, silicomanganese and manganese metal in the U.S in 2001 (USGS (1998)).
Ferromanganese Silico- Manganese me-

End use High car- Medium and Total Mangane- tal
bon se
low carbon
Steel:
Carbon 122’000 85’300 207’000 61’200 621
High-strength, low-alloy 16’900 3’230 20’100 2’620 2
Stainless and heat-resisting 9’680 2 9’680 4’180 1’210
Full alloy 14’300 4’080 18’400 16’900 2
Unspecified 3 998 292 1’290 678 1’120
Total 163’000 92’900 256’000 85’500 2’950
Cast irons 8’100 454 8’560 1’040 5
Superalloys W W W -- W
Alloys (excluding alloy steels and superalloys) 902 436 1’340 4 16’100 5
Miscellaneous and unspecified -- W W 4 W
Total consumption 172’000 93’800 266’000 86’500 6 19’100
Total manganese content 7 134’000 75’300 210’000 59’900 19’100
Stocks, consumers and producers 9’240 15’700 24’900 16’200 2’210
W Withheld to avoid disclosing company proprietary data; included with "Alloys (excluding alloy steels and superalloys)."
-- Zero.
1
Data are rounded to no more than three significant digits; may not add to totals shown.
2
Withheld to avoid disclosing company proprietary data; included with "Steel: Unspecified."
3
Includes electrical and tool steel, and items indicated by footnote 2.
4
Withheld to avoid disclosing company proprietary data.
5
Approximately 85% of this combined total was for consumption in aluminum alloys.
6
Internal evaluation indicates that silicomanganese consumption is considerably understated.
7
Estimated based on typical percent manganese content.

This study focuses to the main processes in the manganese production. As shown in chapter 4 “Use /
Application of Material” no secondary production of manganese has to be considered, since the recy-
cling of manganese occurs directly into the foundries in the form of re-melted iron and steel scrap.
pig iron (1%) 0.5 kg / t steel
high-coal
ferromanga 5.4 kg ferromanga-
blast furnaces (15%)
nese (75%) nese / t raw steel
medium- & steel

low coal industry
ferromangan
electric submerged ese (<2%)
(90%)
Manganese ore
furnaces (75%)
silicomangan
ese (13%) manganese metal in
copper and
aluminium alloys
electroylsis, (2% of
nonmetallurg- consumption)
manganeses metal
ical uses (2%)
(2%)
uranium extraction,
chemical Brick and ceramic
Industry (3%) coloring, Chemical
oxidiser and catalyst
batteries (4%)
Fig. 5.1 Scheme of the manganese production system considered and applications of its products after Adelhardt &
Saiger (1999). Percentages relate to the total manganese contained. In this study the processes for ore ex-
ploitation, production of high-coal ferromanganese and manganese metal are considered.
According to Fig. 5.1, the following modules cover the manganese production system for the use of
manganese in its main application areas “steel industry” as ferronickel and as manganese metal in dif-
ferent specific uses. They are summarised in Tab. 5.1.
- “manganese, concentrate, at beneficiation” Fig. 5.2, Fig. 5.3
- “ferromanganese, high-coal, 74% Mn, at regional storage” Fig. 5.4, Fig. 5.5
- “manganese, pure metal, at regional storage” Fig. 5.6

Tab. 5.1 Ecoinvent meta information for the processes for the production of ferromanganese and manganese metal.
Name manganese, concentrate, at benefici- ferromanganese, high-coal, 74.5% Mn, at manganese, pure metal, at regional stor-
ation regional storage age
Location GLO RER RER
Unit kg kg kg
Dataset Version 2.0 2.0 2.0
Included Processes The module includes a mining and a benefici- The module includes the reduction to high-carbon The module production by electrolysis from ore
ation step with the mining infrastructure and dis- ferromanganese. No secondary production con- and by electrothermic process from ferromanga-
posal of overburden and tailings (see part II sidered, since secondary manganese is recycled nese and slag. Transportation to Europe is only
"Auxiliary Processes"). to the foundries and has to be inventoried in the partially considered.
respective module, i. e. when casting the final raw
material in the steel production.
Amount 1 1 1
Local Name Mangan, Erzkonzentrat, ab Aufbereitung Ferromangan, hochgekohlt, 74.5% Mn, ab Regio- Mangan, ab Regionallager
nallager
Synonyms Manganese metal // metallisches Mangan
General Comment The module describes the production of manga- The module describes the consumption of ferro- The module describes the consumption of man-
nese ore in 1994 for a world-wide production av- manganese in Europe in 1995. It is designed for ganese metal in Europe in 1994. It is designed
erage. It is designed for the use of manganese the use of the metal as alloying element. This solely for the use of the metal in special applica-
ore as intermediary product in the production of module does not consider secondary sources of tions like sputtering or as alloying element. This
ferromanganese and manganese metal. For the manganese. Process data is taken mainly from a module is explicitly not to be used as alloying ele-
use of this module in the battery manufacture, material flow analysis for the consumption of ment in bulk iron or steel industry. It does not con-
the data quality may not be sufficient. chromium in Germany. Results are used as a sider secondary sources of manganese. Process
proxy for the European situation. This module may data is based on rough estimations; the overall
be used as proxy for the silicomanganese produc- data quality is very poor.
tion
Start Date 1994 1994 2003
End Date 2003 2003 2003
Data valid for entire 1 1 1
period
Geography Text This module relates to the global situation. This module relates to the consumption in Europe, No information were available on production or
Some modules used like lime, blasting and although the import mix for Germany is chosen. consumption mix.
transport originate from data sets valid for Swit-
zerland or Europe

Name (continued) manganese, concentrate, at benefici- ferromanganese, high-coal, 74.5% Mn, at manganese, pure metal, at regional stor-
ation regional storage age
Technology Text Mining is done 70% open pit and 30% under- The ore is processed in blast furnaces (20%), The metal is won by electrolysis (assumption:
ground, followed by a beneficiation of the ore electric arc furnaces without flux (27%), electric 25%) and electrothermic processes (assumption:
trough classification. Overburden and tailings arc furnaces with calcareous flux (53%). 75%). No detailed information available, mainly
are non-sulphidic , they are disposed near the based on rough estimates.
mining site.
Representativeness 55% 88% n. a.
Production Volume 23.5 million tonnes in 1995 3.4 million tonnes manganese contained in ferro- 140'000 t in 2000
manganese in 2001
Sampling Procedure Literature literature literature
Extrapolations see Geography and Technology see Geography and Technology see Geography and Technology
Uncertainty Adjust- none none none
ments

5.1 Exploitation of manganese ore

Mining: Manganese ores are usually mined in open pits. Adelhardt & Saiger (1999) reports an esti-
mated share of 70% surface mining activities in the mine production in 1994. The deposits are of vari-
ous nature and therefore the average manganese concentration differs greatly, from Asia with 24%wt to
Australia with 48.5%wt and a world-wide average of 35.7%wt manganese content in the ore.
Beneficiation: After mining, the ore is first crushed in several stages with jaw and / or cone crushers,
and then subsequently ground with rod and or ball mills and finally screened for classification. In a
second step the classified material it is subjected to gravity concentration to separate the metal-bearing
particles from the unwanted minerals. For this drum separators and de-watering screens for lumps are
used and cone separators and a high-gradient magnetic separator for fine material. No flotation is
done. Recovery of Manganese vary in a range from 65% to 75% (Adelhardt & Saiger (1999)). Various
intermediate treatment processes can be applied to manganese natural ores or concentrates prior to
smelting to a ferroalloy or direct addition to a steel making process. These processes are employed for
one or both of the following two reasons: 1) to convert a fine ore to a more suitable size, and 2) to cal-
cine and/or reduce the ore to a more suitable final product. Agglomeration processes include noduliz-
ing, pelletizing, and briquetting. Reduction can be carried out in a rotary kiln or sinter plant. Sintering
has become the preferred process for intermediate treatment (Wellbeloved et al. (1997)).

Mining
large quantities of dust, which contains elevated contents of metals. Rain percolate through overbur-
den and accounts to metal emissions to groundwater. Overburden is deposed close to the mine.
Beneficiation
the ore itself. Tailings are deposed as piles and in ponds. Since the tailings are not sulphidic, no acid
rock drainage (ARD) occurs.
5.1.3 “Manganese, concentrate, at beneficiation” in ecoinvent

The module describes the production of manganese ore in 1994 for a world-wide production average.
It is designed for the use of manganese ore as intermediary product in the production of ferromanga-
nese, manganese metal and for the use of the ore in the manufacturing of batteries. Process data is
taken mainly from Adelhardt & Saiger (1999), which conducted a material flow analysis for the con-
sumption of manganese compounds in Germany.
Infrastructure and resource use

The mobile equipment used for exploration like drilling machines, scoop trams and underground
trucks is inventoried with the diesel consumption, which is reported in Adelhardt & Saiger (1999). The

used module “diesel, burned in building machine” includes the infrastructure of these machines. Use
of conveyor belt was assessed according Tab. 5.2.
The world average manganese content after beneficiation amounts to 42.4%, a yield from mining and
beneficiation of 69.2% can be calculated (Adelhardt & Saiger (1999)).
According to part “Non Ferrous Metal Winning – Auxiliary Processes” no direct land use for under-
ground mining (30%) is inventoried, land use for open pit mining (70%) is assessed with 0.0025m2/t,
yielding a 1.75 10-3 m2/t and 52.5 10-3 m2a/t respectively. A lifetime of 30 years is assumed. Mining
facilities are inventoried as unspecific module according to part IX “Non Ferrous Metal Winning –
Auxiliary Processes”.
Tab. 5.2 Assumptions for infrastructure used in underground mining exploration. As lifetime ten years are estimated,
and 5000 working hours yearly.
Action Share Machine Throughput Unit Infrastructure

Open pit 70 % Conveyor belts (2000 t / h) 100'000'000 10’000 1E-4
Underground 30% Conveyor belts (2000 t / h) 100'000'000 2’000 2E-5
Mix 7.6E-5
Mining
According to Adelhardt & Saiger (1999) 70 % of the manganese ore world production originates from
surface mining operations. The overburden – 2.61 kg per kg ore concentrate extracted – is disposed in
piles near the mine, the standard module for disposal of non-sulphidic overburden was chosen, which
includes the land use and transformation through the disposal. Refilled overburden is assumed to be
burden free. No overburden is refilled according to Adelhardt & Saiger (1999). In the world-wide av-
erage 0.25 m3/t crude ore of the pit water is used for the beneficiation. The remaining pit water is not
inventoried as output but it's emissions are inventoried together with the emissions of the beneficiation
step. The specific use of explosives vary between 0.1 kg (South America) and 0.3 kg (Africa) per
tonne raw ore with a world-wide average of 0.26 kg / t. Diesel use ranges from 0.7 (South America) to
2,3 (Australia) with an average of 1.2 litre per tonne ore exploited. Electricity demand is comprised in
beneficiation.
Beneficiation
Once the ore is mined, it is ground and classified, no flotation occurs. The separated gangue is dis-
posed on heaps, the concentrated ore is fed to the metallurgy, which is assumed to be on-site. The re-
spective standard module “disposal, non-sulphidic tailings, off-site” is chosen. (s. “Non Ferrous Metal
Winning – Auxiliary Processes”, chapter “Disposal of Tailings and Overburden”).
tionally the dust emissions due to mining activities. Hilbrans & Hinrichs (1999) assume for the nickel
mining a share of 60% hydroelectricity due to the operations high and constant energy demand. The
value comprises the electricity demand of the mining, which is expected to be small compared to the
beneficiation.
Steel: The steel input (abrasion) for chromite milling is documented in IPPC (2002) for a single Fin-
nish chromite mine (Kemi site) and amounts to 300 g / t ground ore. Since non information on steel
consumption in the beneficiation of manganese ore is available and the beneficiation process works
analogous to chromite winning, this value is also taken in this study. The mill (rods, balls, etc.) con-
sists of special austenitic manganese steel of Hadfield grade, which is inventoried as chromium steel as
a proxy.

Water use & Effluents: Water use for mining and beneficiation is documented in Adelhardt & Saiger
(1999) and is inventoried in this study as originating from river without pumping activities. As no data
on pit water use are given, the specific water use for chromite exploitation of 1.5 m3 / t raw ore mined
was chosen. The water input for beneficiation is given as 0.8 m3 / t raw ore mined. As no data on
emissions to water could be found, the recommendations of the world bank for base metal sites was
used (Tab. 5.3). These recommended maximum concentrations are multiplied with the sum of effluent
volumes from mining and beneficiation (3.5 m3/t ore). The value of 150 mg / l COD is assumed to
amount to 150% of BOD (Zimmermann et al. (1996)). Values for TOC are calculated using the stan-
dard procedure in Frischknecht et al. (2003) approximating the unknown organic matter with the sim-
plified formula for organic matter C6H12O6. This gives 58.7 mg / l TOC and DOC respectively.
Dust: A case study in Anonymous (1998a) on a projected open cut gold-copper mining and processing
facility in Australia predict annual dust emissions between 0.18 kg/t and 0.46 kg/t of dust produced
per tonne of ore mined. The prediction was made using the Emission Estimation Technique Manual
for Mining (NPI (2001)). The maximum value of 0.46 kg/t is chosen for total dust emissions, the value
for underground mining is assessed with 80% of this amount. The dust composition is approximated
with the manganese ore composition from Tab. 3.1 and the demands on quality for metallurgical grade
ore. For PM10 values, the ratio TSP : PM10 : PM2.5 of manganese ore mining 101.72 : 50 : 5 g / t ore
taken from the CEMIP-database 1 , i. e. one tonne emitted dust corresponds to 492 kg PM10 and 49 kg
PM2.5 respectively. The airborne emissions resulting from mining and beneficiation activities are
summed up in Tab. 5.4. In Ecoinvent, for particulates differential classes are used adding up to the to-
tal dust emission, i. e. 49 kg PM<2.5, 442 kg PM2.5-10 and 508 kg PM>10 per tonne of total dust.
Data quality
Data for major mass fluxes are satisfactory, while the emissions to water and to air are based on as-
sumption. The emissions to water may additionally be overestimated due to different system bounda-
ries of reported effluent volume from Adelhardt & Saiger (1999) and recommendations in Anonymous
(1998b). Furthermore, high uncertainties are associated with the chosen level of impurities of Pb, As,
Cu and Zn in the manganese ore resulting in emissions to water and to air. The emissions from tailings
and overburden are assessed in another module, which itself contains rough assumptions of a global
overburden and tailings composition and a simplified disposal scenario. Emissions tough can vary
considerably from case to case.
Overall, the data quality is not satisfactory for this module. For specific purposes such as the use as
battery grade ore in the battery manufacture further investigations concerning process specific emis-
sions are necessary.
1
http://www.air.sk/tno/cepmeip/em_factors.php, SNAP-code 040616, accessed 01.03.2003

Tab. 5.3 Effluent levels to be achieved during operation and after mine closure according to World Bank Group,
Anonymous (1998b) (Maximum).
Parameter Maximum (mg/l) This Study (mg/l)

Mining Type Base metal
pH 6–9
TSS 50 50
Oil and grease 10 10
Cyanide 1.0 -b
Free 0.1 -b
Weak acid dissociable (WAD) 0.5 -b
COD 150 150
Arsenic 0.1 0.1 a
Cadmium 0.1 -c
Chromium (hexavalent) 0.1 -c
Chromium, ion -c
Chromium (in solid) -c
Copper 0.5 0.1 a
Iron 3.5 3.5
Lead 0.2 0.1 a
Mercury 0.01 -c
Nickel 0.5 -c
Zinc 2 0.1 a
Total metals 10 -
a
Metals, which are frequent impurities in manganese ore. Rough assumption.
b
manganese ore is not sulphidic, therefore no use of this substance and consequently
no emission is assumed.
c
Maximum values relates to the whole range of base metal extraction. This specific
emission is not assumed to be of importance in manganese extraction..

taken from Anonymous (1998a), partition of fractions of particulate matter from the CEMIP-database (mining
of manganese ore), the dust composition correspond to the average manganese ore from “3.1 Manganese
ore” in this study. The impurities As, Pb, Cu and Zn are based on quality demands for metallurgical manga-
nese ore: Mn ≥ 48%, Fe ≤ 6%, As ≤ 0.18%, Cu+Pb+Zn ≤ 0.3% (Adelhardt & Saiger (1999)), assuming single
emission factors for Cu ≤ 0.1%, Pb ≤ 0.1%, Zn ≤ 0.1%.

air mined ore
Iron g / t ore 24.5
Manganese g / t ore 185.2
Phosphorous g / t ore 0.4
Aluminium g / t ore 7.1
Magnesium g / t ore 2.8
Silica g / t ore 13.5
Calcium g / t ore 13.0
Arsenic g / t ore 0.4
Lead g / t ore 0.4
Copper g / t ore 0.4
Zinc g / t ore 0.4
PM<2.5 g / t ore 21.3
PM2.5-10 g / t ore 191
PM>10 g / t ore 220

Source for
ecoinvent
Deviation
Comment
Standard
Category
Location
category
structure
Remarks
Output
name in
General
Process
Modul
Name
value
Input
Infra-
mean
value
Sub-
Type
Mean
Unit
69% yield assumed (31% Mn loss Manganese, 35.7% in
Manganese in in mining / beneficiation), 1.72kg sedimentary deposit,
 resource in ground 6.13E-01 kg (A delhardt 1999) 1 1.13 (2,2,3,1,1,2,4)
ground crude ore and 2.61kg overburden 14.2% in crude ore,
per kg ore concentrate in ground
1.2 l / raw ore. Only machines, reported
diesel f uel construction diesel, burned in
 electricity demand inventoried in machinery No GLO 7.97E-02 MJ (A delhardt 1999) 1 2.08 values w /
(energy) processes building machine
beneficiation. range
reported
construction civil
explosive  0.26 kg explosives / t raw ore No RER blasting 4.52E-04 kg (A delhardt 1999) 1 1.15 values w /
processes engineering
range
exploration
exploration inf rastructure, 2000 m construction conveyor belt, at (5,5,1,1,1,5,9);
infrastructrue,  machinery Y es RER 1.30E-07 m estimation 1 3.32
belts assumed processes plant rough estimate
conveyor belts
infrastructure of
non-ferrous metal, (5,5,1,1,1,5,9);
mining  metals extraction Y es GLO 8.30E-12 unit estimation 1 3.32
mine, underground rough estimate
(underground)
infrastructure of non-ferrous metal, (5,5,1,1,1,5,9);

 metals extraction Y es GLO 1.94E-11 unit estimation 1 3.32
mining (surf ace) mine, surface rough estimate
mining
(3,2,3,1,1,2,8);
Land
Transformation, f rom (Martens 2002), calculation
transf ormation  resource land 3.00E-06 m2 1 2.03
unknow n calculations based on
f rom nature
reported value
(3,2,3,1,1,2,8);
Land Transformation, to
(Martens 2002), calculation
transrormation  resource land mineral extraction 3.00E-06 m2 1 2.03
calculations based on
to mine site
reported value
(3,2,3,1,1,2,7);
Land use f or Occupation, mineral (Martens 2002), calculation
 resource land 9.01E-05 m 2a 1 1.53
mining extraction site calculations based on
reported value
overburden, no overburden assumed to be w aste residual disposal, non-sulfidic

 No GLO 2.61E+00 kg (A delhardt 1999) 1 1.16 (3,2,3,1,1,2,6)
disposed refilled management material landf ill overburden, of f-site
(1.5 - 0.15) m3/t pit w ater per

crude ore (0.15 m3/t used in
beneficiation) w ithout treatment.
The composition is given in Tab. (A delhardt 1998),
 eff luents, mining 2.32E-03 m3 1 1.13 (2,2,3,1,1,2,4)
5.3. The emissions are invnetoried estimation
w ith the emisisons of benef iciation.
The uncertainties of the single
emissions are high.
Fig. 5.2 Flows for “manganese concentrate, at beneficiation – part mining” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg of man-
ganese concentrate.

Ge neral Flow Informa tion Representation in ecoinvent Uncertainty Informations
Source for
ecoinvent
Deviation
Comment
Standard
Category
Location
category
structure
Remarks
Output
name in
General
Process
Modul
Name
value
Input
Infra-
mean
value
Sub-
Type
Mean
Unit
electricity, f rom   A ssumption: 40% electricity f rom electricity, medium
grid, total w aste heat grid. V alue comprising electricity electricity production mix No UCTE voltage, production 3.02E-03 k Wh (A delhardt 1999) 1 1.16 (3,2,3,1,1,2,2)
demand of mining. UCTE, at grid
hydropow er,   A ssumption: 60% hydropow er. electricity,
hydro
total w aste heat V alue comprising electricity pow er plants No RER hydropow er, at run- 4.53E-03 k Wh (A delhardt 1999) 1 1.16 (3,2,3,1,1,2,2)
pow er
demand of mining. of -river pow er plant
f resh and 
0.8 m3/t crude ore (not including
processw ater resource in w ater Water, river 1.37E-03 m3 (A delhardt 1999) 1 1.13 (2,2,3,1,1,2,12)
the recycled pit w ater)
(benef iciation)
steel  chromium steel 18/8,
abrasion in mill (steel balls,etc.) metals extraction No RER 5.15E-04 kg (IPPC, 2002) 1 1.62 (5,5,1,1,1,5,4)
at plant
beneficiation
inf rastrukture  A luminiumhydroxide production as aluminium hydroxide,

metals extraction Yes RER 1.44E-11 unit rough estimate 1 3.32 (5,5,1,1,1,5,9)
benef iciation proxy plant
 tailings, disposed (3,2,3,1,1,2,6);
w aste residual disposal, non-sulf idic
no tailings assumed to be ref illed No GLO 7.10E-01 kg (A delhardt 1999) 1 1.16 reported
management material landf ill tailings, of f -site
values
 ef f luents, composition in Tab. 5.3, emissions
(3,2,3,1,1,2,6);
benef iciation w ith high uncertainties. No 1.20E-03 m3 (A delhardt 1999) 1 1.16
estimated
treatment of w aste w ater.
 Dust Including emissions f rom mining
particulate emissions represented
in f ractions PM<2.5, PM2.5-10,
7.42E-04 kg (NPI, 2001) 1 1.59 (3,2,1,1,3,4,25)
PM>10, including metals. These
emissions are inventoried as
show n in Tab 5.4, they are not
show n in this f igure.
 manganese manganese,
concentrate 42.4% Mn metals extraction No GLO concentrate, at 1.00E+00 kg (A delhardt 1999)
benef iciation
Fig. 5.3 “manganese concentrate, at beneficiation – part beneficiation” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg manganese
concentrate.

5.2 Winning of high carbon ferromanganese

Generally, high-carbon ferromanganese and silicomanganese are produced from a blend of manga-
nese-containing ores, and in the case of silicomanganese, slag and silica are added. High-carbon fer-
romanganese can be converted to medium-carbon manganese by an oxygen blowing process, and sili-
comanganese can be further refined into medium- or low-carbon ferromanganese as well as manga-
nese metal.

Ferromanganese can be produced in blast furnaces in a manner similar to pig iron and spiegeleisen;
however, in the Western world only four producers employ this method. In the former Soviet Union,
the majority of the high-carbon ferromanganese is produced in blast furnaces. The choice of the use of
blast furnaces over electric furnaces is based on the relative price of coke and electricity. Blast fur-
naces are usually used where the cost of power is high in relation to coke. In blast furnaces, coke is
used both as a reductant and as the energy source. The coke rate in blast furnaces is higher than in
submerged arc furnaces, which use electricity as the power source. The product produced from blast
furnaces generally contains 76 % Mn and 16 % Fe; the ferromanganese produced in the former Soviet
Union is generally of a lower grade
According to Adelhardt & Saiger (1999), 15% (related to the manganese contained) of the manganese
production was done by blast furnace and 75% by electric arc furnaces to produce silico- or ferroman-
ganese, the rest being done mainly by electrolysis for other purposes.
The raw materials required for the production of high-carbon ferromanganese are manganese ores,
fluxes such as limestone, dolomite, or silica, and solid fuels and reductants such as coke.
In order to produce ferromanganese of the required grade a single ore is seldom suitable because the
desired Mn/Fe ratio of the charge determines the Mn content of the final product. Ores from various
sources are therefore blended to achieve the ideal ratio and to limit the contents of the deleterious
components silica, alumina (bauxit), and phosphorus in the raw material mix.
Blast furnace
In comparison to iron making, high gas temperatures are required in ferromanganese production be-
cause the reduction of manganese(II) oxide takes place at a higher temperature than is required for the
reduction of wustite. This is achieved by oxygen enrichment of the hot blast. The recovery of manga-
nese in the alloy is usually 75 – 85 %. This is influenced by the MnO content of the slag, the slag-to-
metal ratio, and losses in the flue gases. The MnO content of the slag is highly dependent on the ba-
sicity ratio (CaO + MgO)/SiO2.
The reduction of the higher manganese oxides to manganese(II) oxide takes place in the upper zone of
the shaft (reduction of manganese oxides with carbon monoxide and carbon). These generally occur
below 900 °C and are indirect. The reactions are exothermic, and the heat generated causes high top
temperatures and necessitates water cooling of the furnace top. The reduction of manganese(II) oxide
MnO + C  Mn + CO is highly endothermic, in contrast to the weakly endothermic reduction of
wustite. This requires higher temperatures and, consequently, higher coke rates are required for the
smelting of ferromanganese in blast furnaces.
Electric arc furnaces
The majority of producers of ferromanganese in the Western world use submerged arc electric fur-
naces. Although electric furnaces have lower capacities than blast furnaces they have the advantage
that the heat requirement is provided by electricity, and coke and coal are added to the feed only as re-
ductants. Recently built electric furnaces have capacities of 75 – 90 MVA. Smaller furnaces are still
popular with producers because they offer flexibility in that they can be switched more easily between
different products than their larger counterparts. Electric furnaces used are generally circular and have

three electrodes. The reduction of manganese(II) oxide occurs by the contact of carbon with the mol-
ten oxide in the slag phase. The overall reaction is :
MnO + 10/7 C  1/7 Mn7C3 + CO
Large electric furnaces are usually completely closed at the top, and the CO-rich gas leaves the furnace
at approximately 290 °C and is cleaned in cyclones and venturi scrubbers. The gas is then either
flamed off to the atmosphere or, more recently, is used to generate electricity. Smaller furnaces may be
open, the gas then is usually withdrawn by fans and cleaned in a bag-filter plant. In this case the gases
are completely burnt in the furnace and have no commercial value.
After the metal is tapped from the furnace it is cast into moulds formed from ferromanganese fines or
cast iron and allowed to solidify. The alloy is then removed, crushed, and screened into various size
fractions, depending on the requirements of the user. An alternative to this practice is the use of a cast-
ing machine. In this case the metal is tapped directly onto a moving train consisting of small molds.
The metal then solidifies and is ejected from the mould at the end of the strand.
Adelhardt & Saiger (1999) distinguishes between a electric arc furnace process with calcareous flux
and one without flux, where the energy consumption is smaller due to reduction of the ore by excess
CO. The yield of the process without flux is smaller with 70% compared with 75% of the process with
calcareous flux, due to a slag richer in MnO. However, the manganese rich slag – sometimes over 25%
manganese – of the process without flux is used in the production of silicomanganese, enhancing the
overall yield considerably.
Silicomanganese
The production of silicomanganese is very similar to the ferromanganese production. The main differ-
ence is in the type of flux – silica instead of lime – and of course the composition of the product. Also
the energy demand is some what higher than in the production of ferromanganese. Many plants pro-
duce alternately ferrosilicomanganese and then – processing the developed slag – silicomanganese.
How ever, for purposes of inventorying silicomanganese it is suitable to take ferromanganese as
proxy.
5.2.2 Emission
Air: The main environmental impact by producing ferro-alloys is the emission of dust and fume from
the smelting processes. Dust emissions occur as well from storage, handling and the pre-treatment of
raw materials where fugitive dust emissions play an important role. Depending on the raw material
and the process used other emissions to air are SO2, NOx, CO-gas, CO2, polycyclic aromatic hydrocar-
bon (PAH), VOCs, and volatile metals. The formation of dioxins in the combustion zone and in the
cooling part of the off-gas treatment system (de-novo synthesis) may be possible.
Water: Water emissions are produced from cooling, granulation and other processes and site related
effluents. The wastewater generated by a wet cleaning abatement system plays also an important role.
5.2.3 Waste
Process residues and by-products with a significant amount is slag, filter dust and sludge and spent re-
fractories. These materials are already recycled and reused to a large extent where it is possible. Rich
slag, that means slag with a relatively high proportion of metal oxides, is used as raw material in other
ferro-alloy productions.

5.2.4 Ferromanganese in ecoinvent

This part is mainly based on Adelhardt & Saiger (1999). General process description was taken from
Wellbeloved et al. (1997), process specific emissions from IPPC (2001), IPCC (1996), EPA (1998)
and EC (2002).
Adelhardt & Saiger (1999) differentiates between a blast furnace process (15% of the production re-
spective to manganese contained) and the two processes with electric arc furnaces (with calcareous
flux and without flux, total of 60% of the production) and finally the production of silicomanganese in
electric arc furnaces (13%), totalling to 88% of the total manganese processed to high-carbon ferro-
manganese. In this study an aggregated mix is used, which represent an average production of ferro-
manganese and silicomanganese. 2% is processed to low-carbon ferromanganese, while 10% is used in
the non-ferrous metallurgy, the chemical industry and the battery manufacture (Fig. 5.1). It is suitable
also for the production of silicomanganese, which is produced in a similar way, although using silica
instead of lime and about 10% more energy respective to the manganese contained (Tab. 5.5).
The module describes the world average consumption mix for ferromanganese in Europe in 1994. The
situation for Germany described in Adelhardt & Saiger (1999) is taken as proxy for Europe. It is de-
signed for the use of the metal as alloying element. The layout of the modules with general flow in-
formation, remarks, sources, values and uncertainty information are shown in Fig. 5.4 (inputs) and Fig.
5.5 (output). Ecoinvent meta information for this process are summarised in Tab. 5.1 on page 11. The
following paragraphs describe calculations, sources and assumptions chosen in this study. The func-
tional unit of this process is one kg of ferromanganese with an average manganese content of 74.5%.
Infrastructure and transport: The production infrastructure is represented by the non-ferrous metal
smeltery that is describes in part “Non ferrous metal winning, auxiliary processes” of this report. It is
assumed that the raw materials manganese ore, silica, dolomite, lime, coke and the disposals slag and
cake are transported by lorry over 50 km.
Energy, raw-materials and output: A weighted average of the different inventoried processes – blast
furnace and electric arc furnace- is calculated. An overview over different in- and output values from
different sources is given in Tab. 5.5. Slag is assumed to be sold to silicomanganese production, dust
as reported in Adelhardt & Saiger (1999) is assumed to build process residue, which has to be dis-
posed into a landfill. This is inventoried in this study with the disposal of non-ferrous slag as proxy.
Emissions to air: Air emissions are taken from various sources, they are summarised and discussed in
Tab. 5.6.
Water use and emissions: Water use for European ferrochrome production is reported in IPPC
(2001), the average of 20 m3 (10 m3 – 30 m3) per tonne is chosen as an approximation, resulting in 10
m3 of effluents (volume of process water). The effluent has to be treated in a communal waste-water
treatment plant, a process which is inventoried separately in the chosen module “treatment, sewage,
unpolluted, to wastewater treatment, class 3”. Emissions to water are inventoried separately each as
entry according to reported values of effluents after internal process water treatment and before waste
water treatment plant (Tab. 5.7).
Production and import to Europe. As proxy for the consumption pattern the imports to Germany in
1995 are chosen (Tab. 5.8). Only the ferromanganese imports were considered, assuming, that the im-
ported manganese ore is destined mainly to the chemical industry and the battery manufacture and
only a small amount is used for blending in the domestic ferromanganese production. This results in a
transport intensity of 16 tkm ship and 2 tkm freight train per kg ferromanganese (Tab. 5.9).
Data quality
Process data is satisfactory and valid for the whole world-wide primary ferromanganese production.
Process specific emissions to air and water had to be taken from an other source, which cover other

geography and technology. Nevertheless the overall quality of the module is fair. The import pattern
may have changed in the recent years with a trend to longer import distances.
Tab. 5.5 Production of ferromanganese and silicomanganese. In- and outputs for the production of ferromanganese
with different processes. The last two columns relate to the production of 1 kg manganese alloy using a
weighted mix for ferromanganese according to the share in global production. (Adelhardt & Saiger (1999))
Data for silicomanganese show, that the production of ferromanganese is suitable as a proxy inventory.
Silico- Silico-
Ferroman-
Inventory Unit Ferromanganese manga- manga-
ganese e)
nese nese
Blast Electric arc furnace Electric Total, per 1 kg Total, per
furnace No flux Calc. Flux arc fur- 1 kg
nace
Share in global production 17% 23% 45% 15%
Inputs:
Manganese concentrate t 2.8 3.4 3.1 3.0 2.33E-03 1.98E-03
Electricity kWh 26 3784 4581 4844 2.57 3.25
Process water a) m3 10 10 10 10 0.0075 0.0067
Cooling water a) m3 10 10 10 10 0.0075 0.0067
Fossil energy (coke) kg 2000 400 180 500 0.45 0.34
Electrode graphite kg 23 30 38 1.66E-02 2.55E-02
Silica sand kg 500 0.34
Dolomite kg 100 c) 0.015
Lime kg 100 c) 60 520 0.23
Outputs:
Slag kg 1440 2100 2320 1950 1.55 1.31
Dust kg 130 10 10 10 0.025 6.71E-03
Effluents b) m3 10 10 10 10 7.45E-03 6.71E-03
SO2 d) kg 1.74 6.78E-02 6.78E-02 7.60E-02 2.99E-04 5.10E-05
Ferromanganese kg 1390 1330 1330 1490 1.00 1.00
Manganese concentration % 72% 75% 75% 67% 72.5% 73.4%
a)
values for process water are taken from IPPC (2001) for the similar process of ferrochrome production
b)
volume for effluent equals the process water used.
c)
only sum dolomite plus lime reported, a split 50/50 is assumed.
d)
emission factor for SO2 are taken from IPPC (2001) (see Tab. 5.6)
e)
weighted average of blast furnace and electric arc furnace production..

Tab. 5.6 Emissions to air in the production of 1 tonne ferromanganese. Values are form different sources: BREF
(IPPC (2001)), IPCC (IPCC (1996)), EPA (EPA (1998)) and EEA (EC (2002)). Chosen values for this study are
the average of the given range.
Emission Unit BREF IPPC EPA EEA This study

to air
Low High Low High Low b) High a)
NOx kg / t 0.05 11.7 5.88
NMVOC kg / t 1.7 1.7
CO kg / t 1 e) 1
PM<2.5 kg / t 0.24 17 8.62
PM2.5-10 kg / t 0.16 7 3.58
PM>10 kg / t 0.08 4 2.04
TSP kg / t 0.48 28 14.2 c)
SO2 kg / t 0.05 f) 1.25 f) 0.3 f)
PCDD μg / t 1 d) 1
a)
No abatement
b)
Bag house
c)
Average of EPA values are chosen for world wide mix.
d)
SNAP-Code 040309, „Other - Installations for smelting and casting of non ferrous metals“
e)
Value taken from ferrosilicon production
f)
values for Electric arc furnace (low) and blast furnace (high) resulting in a weighted average value (see Tab. 5.5)
CO2 emission is not inventoried separately as it is included in the emissions of the "hard coal, burned..." inputs
Tab. 5.7 Emissions to water in the production of 1 tonne ferromanganese. The values are reported values for ferro-
chromium production (IPPC (2001)) chosen as approximation. Emissions are before final waste-water treat-
ment and after internal abatement including lamella thickener, vacuum filter an a sand and carbon filter.
Based on the chosen values, final emissions into water are calculated.
Emission to Water Unit IPPC (2001) This Study Resulting values af-
(before treatment) Low High ter treatment
TSS kg / t 5 10 7.5 a) 0.375
PAH mg / t 300 1000 650 b) 65
waste from internal wastewater treatment
dry cake kg / t 200 400 300 c) 300
a)
An elimination of 95% TSS assumed, resulting in an emission after treatment of 375 g TSS / t
b)
Based on Zimmermann et al. (1996) transfer coefficients of 10% to water is assumed, resulting in emis-
sions of 65 mg PAH / t. Resulting emissions to air due to metabolisation as CO2 and NMVOC are ne-
glected.
c)
Dry cake results from internal waste-water treatment. In this study it is assumed to be disposed as inert
waste to an inert material landfill.

Tab. 5.8 Manganese import pattern for Germany in 1995 (Adelhardt & Saiger (1999)).
Manganese consumption pattern for Germany 1995

ore & concentrate 11'355 3%
manganese metal 24'433 7%
Ferromanganaese 198'334 53%
silicomanganese 135'336 36%
other 5'942 2%
Total imports 375'400
Tab. 5.9 Import of manganese into Germany for different products and resulting transport intensity in [tkm]. Mix is
taken as proxy for the European situation. Values are from (Adelhardt & Saiger (1999)) and estimations
2
based on distances provided by “dataloy systems” .
Manganese ore Australia South Africa Morocco other total

Manganese ore t 3'123 2'702 2'157 3'372 11'355
ship sea km 15'500 15'500 1'700 10'900 11'512 tkm / t
train km 1'200 1'400 b) 1'500 1'367 1'354 tkm / t
Ferromanganese France South Africa Norway other total
Ferromanganese t 60'680 40'280 30'440 66'930 198'330
ship sea km 16'100 a) 15'500 16'300 a) 15'967 15'964 tkm / t
train km 2'300 c) 1'400 b) 2'300 c) 2'000 2'016 tkm / t
Manganese metal China Ukraine South Africa other total
Manganese t 11'923 6'792 3'616 2'101 24'433
ship sea km 15'500 3'000 16'300 11'600 11'768 tkm / t
train km 2'500 300 2'300 1'700 1'753 tkm / t
a)
15'500 km sea transport included to include bias for previously transported raw materials
b)
800 km transport by freight train included to include bias for previously transported raw materials
c)
2'000 km transport by freight train included to include bias for previously transported raw materials
2
http://www.dataloy.com/, acessed 01.03.2003

Source for
ecoinvent
Deviation
Comment
Standard
Category
Location
category
structure
Remarks
Output
name in
General
Process
Modul
Name
value
Input
Infra-
mean
value
Sub-
Type
Mean
Unit
manganese,
manganese
 metals extraction No GLO concentrate, at 2.33E+00 kg (A delhardt 1998) 1 1.13 (2,2,3,1,1,2,4)
concentrate
benef iciation
(3,2,3,1,1,2,2);
electricity, medium
(A delhardt 1998), estimate based
electricity, grid   w aste heat 40% electricity f rom grid electricity production mix No UCTE voltage, production 1.03E+00 k Wh 1 1.16
estimations on reported
UCTE, at grid
ferromanganese, high-coal, 74.5% Mn, at regional storage
value
(3,2,3,1,1,2,2);
electricity,
hydro (A delhardt 1998), estimate based
hydropow er   w aste heat 60% electricity f rom hydropow er pow er plants No RER hydropow er, at run- 1.54E+00 k Wh 1 1.16
pow er estimations on reported
of -river pow er plant
value
hard coal, burned in
f ossile energy heating
 hard coal No RER industrial f urnace 1- 4.50E-01 MJ (A delhardt 1998) 1 1.13 (2,2,3,1,1,2,3)
(coke) systems
10MW
construction limestone, milled, (IPPC, 2001), (5,2,3,1,1,2,4);
lime  others No CH 2.32E-01 kg 1 1.52
materials packed, at plant estimations estimate
construction (IPPC, 2001), (5,2,3,1,1,2,4);
dolomite  others No RER dolomite, at plant 1.49E-02 kg 1 1.52
materials estimations estimate
Electrode anode, aluminium
 metals extraction No RER 1.66E-02 kg (A delhardt 1998) 1 1.13 (2,2,3,1,1,2,3)
graphite electrolysis
Water, w ell, in
process w ater  resource in w ater 7.45E-03 m3 (IPPC, 2001) 1 1.11 (2,3,2,3,1,3,12)
ground
Water, cooling,
cooling w ater  resource in w ater unspecif ied natural 7.45E-03 m3 (IPPC, 2001) 1 1.11 (2,3,2,3,1,3,2)
origin
(4,4,1,1,1,5,5);
transport 50 km f or concentrate, silica, lime, transport calculation
 road No RER transport, lorry 32t 2.30E-01 tk m estimate 1 2.11
(ressources) dolomite, coke, slag systems based on
assumptions
(2,2,3,1,1,2,5);
Transport
transport transport, calculation
f erromanganes  ship No OCE 1.60E+01 tk m (A delhardt 1999) 1 2.02
systems transoceanic tanker based on
e over sea
assumptions
(2,2,3,1,1,2,5);
Transport
transport calculation
f erromanganes  train No RER transport, f reight, rail 2.02E+00 tk m (A delhardt 1999) 1 2.02
systems based on
e by train
assumptions
production non-f errous metal,
 metals extraction Y es GLO 6.88E-12 unit rogh estimate 1 3.32 (5,5,1,1,1,5,9)
inf rastructure smelter
Fig. 5.4 Flows for “ferromanganese – inputs” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg ferromanganese.

Source for
ecoinvent
Deviation
Comment
Standard
Category
Location
category
structure
Remarks
Output
name in
General
Process
Modul
Name
value
Input
Infra-
mean
value
Sub-
Type
Mean
Unit
Inventoried as w astew ater treatment, sew age,
treatment w /o pollution. For overall w aste w astew ater unpolluted, to
 Eff luents No CH 7.45E-03 m3 (A delhardt 1998) 1 1.13 (2,2,3,1,1,2,6)
process, w aterborn emission see management treatment w astew ater
Tab. 5.7 treatment, class 3
disposal, inert w aste,
dust in w et from w et scrubber, deposed in w aste inert material
 No CH 5% w ater, to inert 2.53E-02 kg (A delhardt 1999) 1 1.13 (2,2,3,1,1,2,6)
scrubber landfill management landfill
material landfill
ferromanganese, high-coal, 74.5% Mn, at regional storage
low population reported

 NOx air Nitrogen oxides 5.88E-03 kg (IPPC, 2001) 1 1.99
density values
NMV OC, non-
low population methane volatile
 NMV OC air 1.70E-03 kg (IPPC, 2001) 1 1.52 (2,3,2,3,1,3,16)
density organic compounds,
unspecif ied origin
from calcination of dolomite and
carbon dioxide
lime. CO2 f rom burning of f uels is low population
 from dolomite and air Carbon dioxide, fossil 1.09E-01 kg calculation 1 1.3 (4,1,1,1,1,5,14)
inventoried in the "heat at.. " density
lime
module
Dust is the sum of particulate
emissions represented in fractions
PM<2.5, PM2.5-10, PM>10, based on
 Dust (TSP) including metals. These emissions 1.42E-02 kg (IPPC, 2001) 1 1.97 reported
are inventoried as show n in Tab values
5.7, they are not show n in this
figure.
based on
low population
 SO2 air Sulf ur dioxide 3.00E-04 kg (IPPC, 2001) 1 4.17 reported
density
values
low population Carbon monoxide,
 CO air 1.00E-03 kg (IPPC, 2001) 1 5.01 (2,3,2,3,1,3,17)
density f ossil
low population 2,3,7,8-
 PCDD de-novo synthesis air 1.00E-12 kg (IPPC, 2001) 1 3.01 (2,3,2,3,1,3,21)
density tetrachlorodibenzo-p-
dioxin
sold to silicomanganese
 slag 1.55E+00 kg (A delhardt 1998) 1 1.13 (2,2,3,1,1,2,6)
production, not inventoried
f erromanganese,
 ferromanganese metals extraction No RER high-coal, 74.5% Mn, 1.00E+00 kg (A delhardt 1998)
at regional storage
Fig. 5.5 Flows for “ferromanganese – outputs” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg ferromanganese.

5.3 Production of metallic manganese

Manganese metal is mainly used in the iron, steel and aluminium industry. In aluminium production
for instance manganese is used as an alloying element improving the product hardness. For the pro-
duction of pure manganese metal several processes are known (Wellbeloved et al. (1997), IPPC
(2001)):
- Electrolysis of aqueous manganese salts
- Electrothermal decomposition of manganese ores
- Silico-thermic reduction of manganese ores or slag
- Alumino-thermic reduction of manganese ores or slag
- Distillation of ferro-manganese
- Electrolysis of fused salts

From the above listed processes the first two are the most important ones. The other processes are not
economically significant.
Electrolysis of aqueous manganese salts: The production of manganese metal by the electrolysis of
aqueous manganese salts requires at first a milling of the manganese ore. Milling increases the active
surface and ensures sufficient reactivity in both the reduction and the subsequent leaching steps. After
milling the manganese ore is fed to a rotary kiln where the reduction and calcination takes place. This
process is carried out at about 850 - 1000 ºC in a reducing atmosphere. As a reducing agent, several
carbon sources can be used e.g. anthracite, coal, charcoal and hydrocarbon oil or natural gas. The cal-
cined ore needs to be cooled below 100 ºC to avoid a further re-oxidation. The subsequent leaching
process is carried out with recycled electrolyte, mainly sulphuric acid. After leaching and filtration the
iron content is removed from the solution by oxidative precipitation and the nickel and cobalt are re-
moved by sulphide precipitation. The purified electrolyte is then treated with SO2 in order to ensure
plating of γ-Mn during electrolysis. Electrolysis is carried out in diaphragm cells. The cathode is nor-
mally made of stainless steel or titanium. For the anode lead-calcium or lead-silver alloy can be used.
After an appropriate reaction time the cathodes are removed from the electrolysis bath. The manganese
that is deposited on the cathode starter-sheet is stripped off mechanically and then washed and dried.
The metal is crushed to produce metal flakes or powder or granulated, depending on the end use.
Electrothermal decomposition of manganese ores:. The electrothermal process is the second impor-
tant process to produce manganese metal in an industrial scale. The electrothermal process takes place
as a multistage process. In the first stage manganese ore is smelted with only a small amount of reduc-
tant in order to reduce mostly the iron oxide. This produces a low-grade ferro-manganese and a slag
that is rich in Mn-oxide. The slag is then smelted in the second stage with silicon to produce silico-
manganese. The molten silicomanganese can be treated with liquid slag from the fist stage to obtain
relatively pure manganese metal. For the last step a ladle or shaking ladle can be used. The manganese
metal produced by the electrothermal process contains up to 98% of Mn.
5.3.1 Emissions & Wastes

The general manufacturing of refractory metals normally takes place at high temperatures. The envi-
ronmental impact on air, water and land can be classified as follows.
Emissions to air: Dust and fume emissions from smelting, hard metal and carbide production; Other
emissions to air are ammonia (NH3), acid fume (HCl), hydrogen fluoride (HF), VOC and heavy met-
als.
Emissions to water: Overflow water from wet scrubbing systems; Wastewater from slag and metal
granulation; Blow down from cooling water cycles.

Wastes: Dust, fume and sludge; Slag.
5.3.2 “manganese, pure metal, at regional storage” in ecoinvent

The module describes the manganese metal consumed in Europe in 1994. It is designed solely for the
use of the metal in special applications like sputtering or alloying element. This module is explicitly
not to be used as alloying element in bulk iron or steel industry. The overall quality of this module is
poor due to a lack of data.
The layout of the module with general flow information, remarks, sources, values and uncertainty in-
formation is shown in Fig. 5.6. Ecoinvent meta information for this process are summarised in Tab.
5.1 on page 11. The following paragraphs describe calculations, sources and assumptions chosen in
this study. The functional unit of this process is one tonne of manganese consumed in Europe. No in-
formation is available on the share of the two major production processes, a ratio electrother-
mal/electrolysis of 75/25 is assumed.
Electrolytic process
No information on material input and output were available. The energy demand is stated in
Wellbeloved et al. (1997) to be 9 - 12 kWh per kg manganese. Considerable amounts of chemicals
ought to be used for reduction and leaching, but no detailed information is available. Hence none of
these emissions are taken in account. For grinding prior to electrolysis, a value of 20 kWh electrical
energy per tonne processed ore is assumed, reported in IPPC (2002) for base metal sites. A yield of
60 % is roughly estimated.
Electrothermal process
The production of bulk manganese by this method is a well protected secret and little information ap-
pears in the literature. The process is a refinement of the silicothermic reduction of manganese. It re-
sembles that of low-carbon ferromanganese in electric arc furnaces of 1 – 3 MW power consumption.
The raw material used is either low-iron manganese ore or a manganese slag concentrate. The silicon
is in the form of a special silicomanganese which is made from a manganese slag concentrate or from
pure ores. Reduction as complete as possible is ensured by the addition of lime, which forms a basic
slag (Wellbeloved et al. (1997)).
In this study it is assumed, that 75 % of the Mn contained in the raw material originates from ore with
a special grade, which is processed in a way similar to high-coal ferromanganese, the rest being slag,
which is burden free. No information on the further refining process of manganese is available and no
estimations are made in this study. Adelhardt & Saiger (1999) reports efficiencies between 60 % and
65 %, for this study the average of 62.5% was chosen.
Import to Europe: Since some transport intensity is already included in the 1.2 kg ferromanganese
used for the electrothermic process, no further transportation is inventoried within this module. This
approach is obviously not correct, but it will not influence the data quality of the inventoried process,
which is already poor.
Data quality
This process is mainly based on rough estimates. The process specific emissions could be considerable
but due to lack of data none are considered. This module only is to be used, where it plays a minor
role, like in sputtering. If the impact of the modules use is considered to be high, more specific inves-
tigations have to be conduced.

Source for
ecoinvent
Deviation
Comment
Standard
Category
Location
category
structure
Remarks
Output
name in
General
Process
Modul
Name
value
Input
Infra-
mean
value
Sub-
Type
Mean
Unit
f erromanganese,
f erromanganes (5,5,1,1,1,5,4);
 metals extraction No RER high-coal, 74.5% Mn, 1.21E+00 kg Calculation 1 1.62
e rough estimate
regional storage at regional storage
manganese, at
(5,5,1,1,1,5,4);
high silica slag  not inventoried 9.68E-01 kg Calculation 1 1.62
rough estimate
manganese manganese,
(5,5,1,1,1,5,4);
concentrate f or  metals extraction No GLO concentrate, at 9.83E-01 kg Calculation 1 1.62
rough estimate
electrolysis benef iciation
electricity, medium based on
Energy f or
 electricity production mix No UCTE voltage, production 2.63E+00 k Wh (Wellbeloved 197) 1 1.14 theoretical
electrolysis
UCTE, at grid values
Inf rastructure, non-f errous metal,
 metals extraction Y es GLO 1.06E-14 unit rogh estimate 1 3.32 (5,5,1,1,1,5,9)
unspecif ied smelter
low population
 w aste heat air Heat, w aste 9.47E+00 MJ 1 1.16 (3,2,3,1,1,2,13)
density
manganese, at
 manganese metal global average metals extraction No RER 1.00E+00 kg
regional storage
Fig. 5.6 Flows for “Manganese, pure metal, at regional storage” and its representation in the ecoinvent database. Values correspond to the functional unit of 1 kg manganese metal.

6. Literature
6 Literature
Adelhardt & Saiger (1999) Adelhardt W. and Saiger H. (1999) Stoffmengenflüsse und Energiebedarf bei
der Gewinnung ausgewählter mineralischer Rohstoffe; Teilstudie Mangan. In:
www.ecoinvent.ch.
Anonymous (1998a) Anonymous (1998a) Dust Control. In: Best Practice Environmental Manage-
ment in Mining. Environment Australia, Online-Version under: http://
www.ea.gov.au/industry/sustainable/mining/booklets/dust/.
Anonymous (1998b) Anonymous (1998b) Pollution Prevention and Abatement Handbook. The World
Bank Group, Online-Version under: http://lnweb18.worldbank.org/essd/ ess-
dext.nsf/51ByDocName/PollutionPreventionandAbatementHandbook.
sion: 22.08.2002.
IPCC (1996) IPCC (1996) Reference Manual. In: Revised 1996 IPCC Guidelines, Vol. 3,
Online-Version under: http://www.ipcc-nggip.iges.or.jp/public/gl/invs1.htm.
0/733169
Jones (1994) Jones T. S. (1994) Manganese Material Flow Pattern. IC-9399. U.S. Department
of the Interior, Bureau of mines, Online-Version under: http://pubs.usgs.gov/
of/of01-304/.

6. Literature
USGS (1998) USGS (1998) Manganese Recycling in the United States in 1998. U.S. Depart-
ment of the Interior, U.S. Geological Survey. Retrieved at 01.03.2003 from
http://pubs.usgs.gov/of/of01-304/
U.S. Geological Survey. Retrieved at 01.03.2003 from http://minerals.usgs.gov/
minerals/pubs/mcs/. Date of last revision: 26-Feb-2003@09:36.
Wellbeloved et al. (1997) Wellbeloved D. B., Craven P. M. and Waudby J. W. (1997) Manganese and
Manganese Alloys. In: Ullmann's encyclopedia of industrial chemistry (ed.
Anonymous). 5th edition on CD-ROM Edition. Wiley & Sons, London.
Zürich.

Part VIII
Lead, Zinc and the Co-Products

Cadmium and Indium
Data v2.1 (2009)
Authors: Mischa Classen, Empa Dübendorf

Wolfram Scharnhorst, Empa Dübendorf.
Review Roberto Dones, PSI Villigen

Hans-Jörg Althaus, Empa Dübendorf
Summary
Comments on this part

The production of lead, zinc, cadmium and indium is inventoried in the modules described in this report. Data
reported are mainly based on literature and theoretical models.
The Swiss Federal Office for the Environment (FOEN) commissioned Empa in 2005 to identify the datasets that
have to be improved in order to improve the overall quality and applicability of metals inventories in the metal
processing and electronic industry. Based on that preparatory study the significance of the metals extracted from
combined lead and zinc ores was identified. In order to improve and extend datasets the original parts have been
reworked completely for ecoinvent version 2.0. It has been extended by a series of metals that are mainly pro-
duced from intermediate raw materials originating from lead-zinc extraction: cadmium, silver and indium.
This part corresponds to the original report in Althaus et al. (2004) on lead and zinc. The inventory modelling
has been done consistently with multi output modules using economic allocation. All the errors reported in pre-
vious versions 1.x as documented in Frischknecht et al. (2006) have been corrected. Notably the emission fac-
tors to air for primary zinc production have been corrected and updated with the revised source (EEA, 2003).
Likewise the values in Tab. 6.19 have been corrected as reported in the changes to v1.1.

Table of Contents
Table of Contents
COMMENTS ON THIS PART ................................................................................................I
1 INTRODUCTION ....................................................................................................... 1

2.1 Lead....................................................................................................................................... 2
2.2 Lead as secondary resource................................................................................................... 3
2.3 Zinc........................................................................................................................................ 4
2.4 Cadmium ............................................................................................................................... 5
2.5 Indium ................................................................................................................................... 5
2.5.1 Primary sources: Indium ores................................................................................................ 6
2.5.2 Secondary sources................................................................................................................. 6
2.5.3 Indium Supply....................................................................................................................... 7
3 COUPLED PRODUCTION........................................................................................... 9
4 CHARACTERISATION OF THE MATERIALS ................................................................. 10

4.1 Lead..................................................................................................................................... 10
4.2 Zinc...................................................................................................................................... 10
4.3 Cadmium ............................................................................................................................. 11
4.4 Indium ................................................................................................................................. 12

5.1 Lead..................................................................................................................................... 13
5.2 Zinc...................................................................................................................................... 14
5.3 Cadmium ............................................................................................................................. 15
5.4 Indium ................................................................................................................................. 16

6.1 Mining of zinc-lead-deposits............................................................................................... 26
6.1.2 Waste................................................................................................................................... 26
6.1.3 Emission.............................................................................................................................. 26
6.1.4 Mining, zinc-lead deposits in ecoinvent.............................................................................. 26
6.1.5 Data quality ......................................................................................................................... 32
6.2 Primary Lead Production Process and Infrastructure .......................................................... 35
6.2.2 Waste................................................................................................................................... 36
6.2.3 Emission.............................................................................................................................. 36
6.2.4 Smelting, primary lead production in ecoinvent ................................................................. 36
6.2.5 By-products and allocation ................................................................................................. 40
6.2.6 Data quality ......................................................................................................................... 41
6.3 Secondary Lead Production process and infrastructure ...................................................... 44
6.3.1 Lead, secondary, at plant in ecoinvent ................................................................................ 44
6.3.2 Data quality ......................................................................................................................... 45

Table of Contents
6.4 Lead, at regional storage in ecoinvent................................................................................. 48

6.5 Primary Zinc Production Process ........................................................................................ 50
6.5.2 Waste................................................................................................................................... 51
6.5.3 Emission.............................................................................................................................. 52
6.5.4 Smelting, primary zinc production in ecoinvent ................................................................. 52
6.5.5 By-products and allocation ................................................................................................. 55
6.5.6 Data quality ......................................................................................................................... 58
6.6 Cadmium Production process and infrastructure ................................................................ 61
6.6.1 Cadmium, primary, at regional storage in ecoinvent .......................................................... 61
6.6.2 Data Quality ........................................................................................................................ 62
6.7 Indium Production Process and Infrastructure .................................................................... 64
6.7.1 Concentration of Indium in the solution ............................................................................. 64
6.7.2 Refining of Raw Indium...................................................................................................... 66
6.7.3 Indium production in ecoinvent .......................................................................................... 67
6.7.4 Data Quality Considerations ............................................................................................... 70
7 LITERATURE ......................................................................................................... 71

1. Introduction
1 Introduction
Zinc and lead are mined from complex ores that also contain a number of minor metal values, some of
which are almost exclusively produced from such complex ores (Fig. 1.1). Therefore the extraction of
the mentioned metal values is not independent from each other and depends mainly on the carrier
metal at which the extraction of an ore body is aimed at. The coupled occurrence leads to a modelling
approach that relies on multi-output inventory modules. The allocation of environmental burdens to
the reference flows is done by economic criteria. Chapter 6 System Characterisation gives an over-
view to this topic.
Fig. 1.1 The metal wheel displaying combination of jointly occurring metals in ore deposits (Reuter et al., 2004)


2.1 Lead
Lead, like copper and zinc, is a markedly chalcophilic element. Its mean abundance in the Earth's crust
is 16 g/t (0.0016 %). Enrichments occur only in the pneumatolytic and especially the hydrothermal
sulphide phase in the form of vein (lode), impregnated, and particularly metasomatic deposits. Sedi-
mentary enrichments are of little importance.
Galena or lead sulphide [CAS 1314-87-0], PbS, theoretically 86.6 % Pb, is by far the most important
lead mineral and the sole primary one. Other common forms of naturally occurring lead are usually
associated with or derived from galena deposits. Frequent impurities in lead minerals are zinc, copper,
arsenic, tin, anitmony, silver, gold, and bismuth. Of the total world lead and zinc mining output, mixed
lead-zinc ore deposits are the most important and account for ca. 70 % of the total production of both
metals. In second place are deposits that contain predominantly lead ores (ca. 20% of total produc-
tion), and the remainder (ca. 10%) is obtained as a by-product from zinc, copper – zinc, and other de-
posits (Sutherland et al., 1997). For sulphidic deposits of the zinc-lead-copper type Krauss et al.
(1999) reports an average composition in parts of metal by the mass of 2.4 Zn : 1.4 Pb : 1 Cu. Mined
lead ores contain approximately 6 % Pb in average (Ayres et al., 2002).
Tab. 2.1 Mine production and reserves of lead in 2005. “Reserves” is referred to as “the part of the reserve base
and production practices” (USGS, 2006b).
Lead Mine production Reserves Reserve base

2005 [kt Pb / a] 2006 [kt Pb / a] [kt Pb] [kt Pb]
(estimated)
United States 426 430 8100 20000
Australia 776 780 15000 28000
Canada 73 79 2000 9000
China 1000 1050 11000 36000
India 58 60 NA NA
Ireland 64 65 NA NA
Kazakhstan 44 55 5000 7000
Mexico 130 140 1500 2000
Morocco 31 42 500 1000
Peru 319 320 3500 4000
Poland 48 60 NA 5400
South Africa 42 50 400 700
Sweden 61 61 500 1000
Other countries 198 170 19000 30000
World total
(rounded) 3270 3360 67000 140000

2.2 Lead as secondary resource

Recycling of lead is performed at a high level world wide (USGS, 2006b). The production of secon-
dary lead, for example in the US, has risen from about 23’131 tonnes in 1907 1 (USGS, 2006c) to
1’100’000 tonnes in 2005 (Gabby, 2006.) At present it is estimated that about 85.0% lead used can
economical be recovered (Thornton et al., 2001). Because of the permanently increasing demand (de-
mand is higher than lead is available), the price of lead raised comparably (as general tendency for
large price fluctuations can also be observed in Fig. 1).
1400
Lead value
1200
1000
800
Lead [$]
600
400
200
0
1880 1900 1920 1940 1960 1980 2000 2020
Time [years]
Fig. 2.1 Value of lead (adopted from DiFrancesco et al., 2006).
On a global level, the refining of secondary lead has increased from ~ 1’500’000 tons in 1976 to ap-
proximately 2’800’000 tons in 1998 (Dzioubinski & Chipman, 1999). Key source for secondary lead
are lead-acid batteries contributing with 92.0% to the total secondary lead production of the US
(Gabby (2004)). The treatment of such batteries is enforced by adequate governmental directives, for
instance in France (100.0%), Denmark (99.9%) and Portugal (75.0%) (Thornton et al., 2001). Beside
automotive batteries, secondary lead is recovered from used lead-acid rechargeable batteries as for in-
stance applied for back-up power supply in antenna stations of mobile phone networks (Scharnhorst,
2005).
In recent years, the primary production has seen a shift from pure primary production to more flexible
technologies that are able to treat increasingly also secondary materials (QSL furnace at Stolberg fac-
itlity, Kivcet furnace at Trail). Hence, the distinction between processing primary and secondary lead
is more and more blurred. The UK based consultancy CHR Metals estimate that the amount of secon-
dary lead production has raised to 58% of the world lead supply (Roberts, 2003). The ILZSG-statistics
indicate a further decrease in the share of primary (mine) production compared to the total reported
metal production, viz. an increase in secondary (post consumer) production (Tab. 2.2).
1
The earliest documented statistics on secondary lead.

Tab. 2.2 Lead production and consumption worldwide. Statistics taken from ILZSG (2007) and USGS (2006b).
Mine Production Metal Production Metal Consump- Ratio mine pro-

tion duction / metal
production
Year Thousand tonnes Thousand tonnes Thousand tonnes %
2000 3100 *
2001 3100 *
2002 2910 *
2003 3111 (2950 *) 6764 6835 46%
2004 3129 (3150 *) 6957 7279 45%
2005 3415 (3270 *) 7637 7800 45%
2006 3492 (3360, est. *) 8024 8031 44%
* USGS (2006b)
2.3 Zinc
With an average concentration in the earth's crust of 65 g/t (0.0065 %), zinc is the 24th element in or-
der of abundance. It occurs only in the chemically combined state. Like copper and lead, zinc is a
strongly chalcophilic element. It usually occurs as sulphide, mostly deposited from hydrothermal solu-
tions and simic and sialic magmas. Sedimentary deposits are of less importance.
By far the most important zinc mineral today, and probably also the only primary ore, is zinc blende.
The other minerals listed here are of only local importance. Native zinc is practically nonexistent.
Isomorphous components of zinc blende such as cadmium, gallium, germanium, indium, and thallium
do not form deposits of their own. Some of these metals are recovered during zinc ore processing.
Zinc blende (or sphalerite, ZnS) has the theoretical composition 67.09 % Zn, 32.91 % S. It is light yel-
low to black in colour, usually black to brown depending on the iron content, crystallises in the regular
tetrahedral system, and has a Mohs hardness of 3 – 4 and a density of 3.9 – 4.1 g/cm3. The most im-
portant impurity is FeS, which is always present at concentrations between 0.3 and >20 %. Sulphides
of Pb, Cd, Mn, and Cu are often present. Zinc blende also often contains small amounts of As, Sn, Bi,
Co, Hg, In, Tl, Ga, and Ge, and nearly always contains Ag and often Au. The hexagonal modification
of Zinc blende is known as Wurzite (conchoidal or radiating blende). Its density is 3.98 g/cm3. Other
less important minerals are: Smithsonite (zinc spar, calamine) ZnCO3; Hemimorphite , Zn2SiO4 · H2O;
Willemite , Zn2SiO4; Franklinite , (Zn,Fe,Mn)O · (Fe2,Mn2)O3, zinc bloom (hydrozincite), ZnCO3 · n
Zn(OH)2; red zinc ore (zincite), ZnO; and troostite (bainite), (Zn,Mn)2SiO4 (Graf & Schleininger,
1997).
According to Ayres et al. (2002), the average concentration of the material currently mined is 3-11%.
The annual mine production and world-wide reserves are listed in Tab. 2.3.
Approximately 30% of the yearly zinc consumption in Europe is secondary or recycled zinc. Ap-
proximately 50% of this secondary zinc is recycled within the industry. This is particularly true in the
galvanising and brass sector. Scrap arising from the production or processing of products can be recy-
cled almost immediately. Residues and scrap, which are relevant and significant to the secondary zinc
industry, include dust from copper alloy making, residues from the die casting industry, ashes, bottom
and top drosses from the galvanising industry, old roofing and other sheet materials, non-ferrous frac-
tion from the shredding of old cars and of other mainly steel containing products, dust from electric
arc steel making and cast iron making and residues from chemical uses of zinc and burnt tyres (IPPC,
2001).

Tab. 2.3 Mine production and reserves of zinc in 2006. “Reserves” is referred to as “the part of the reserve base
and production practices” (USGS, 2006b).
Zinc Mine production Reserves Reserve base

2005 [kt Zn / a] 2006 [kt Zn / a] [kt Zn] [kt Zn]
(estimated)
Australia 1'330 1'400 33'000 80'000
Canada 755 725 11'000 31'000
China 2'450 2'500 33'000 92'000
Kazakhstan 400 450 30'000 35'000
Mexico 470 450 8'000 25'000
Peru 1'200 1'210 16'000 20'000
Other countries 2'400 2'500 59'000 87'000
World total
(rounded) 9'800 10'000 220'000 460'000
2.4 Cadmium
Cadmium is widely distributed. It occurs in the earth's crust with a content estimated to be between
0.08 and 0.5 ppm. In top soil cadmium content usually lies between 0.1 and 1 ppm. None of known
cadmium minerals such as cadmium sulfide (greenockite, 77.6 % Cd), cadmium carbonate (61.5 %
Cd) and cadmium oxide (87.5 % Cd) is of industrial importance because the deposits are too small.
Only zinc minerals in which cadmium is found as an isomorphic component, with concentrations rang-
ing from 0.05 to 0.8 %, averaging about 0.2 % have economic significance for cadmium recovery. In
addition, lead and copper ores contain small amounts of cadmium, which can be separated during the
roasting and smelting processes (Schulte-Schrepping & Piscator, 2002).
The bulk of the cadmium being recovered is associated with ores of sphalerite (ZnS) that is extracted
for its zinc content. Cadmium therefore is a by-product from the zinc extraction where it accumulates
in flue dust (pyrometallurgical processes) and purification precipitates (hydrometallurgical processes).
Estimated world resources of cadmium were about 6 million tons in 2006, based on identified zinc re-
sources of 1.9 billion tons containing about 0.3% cadmium (USGS, 2007).
The recycling of cadmium is still on a low level. It basically occurs via the recovery of EAF dusts
from zinc manufacturing. Hawkins et al. (2006) estimate that in the U.S. less than 20% of the cad-
mium in the product waste stream is currently collected for recycling. On a world wide average this
number appears to be lower, a level of 10% seems to be a reasonable assumption.
This consideration, together with the assertion that 80% of the primary cadmium produced originates
from zinc extraction (USGS, 2007) and under the assumption of stable stocks, leads to the estimate
that 72% (=80% primary from zinc * 90% primary in supply) of the cadmium supply may be related to
the primary zinc production.
2.5 Indium
Indium does not occur as pure metal. Although widely distributed, it generally occurs in low concen-
trations. The content of indium in the earth crust is estimated to be 0.1 ppm. Indium is found as a trace
element in many minerals and is accumulated during the extraction of the main metal values such as
zinc, lead and copper.

2.5.1 Primary sources: Indium ores

Sphalerite is the most important indium-containing mineral followed by lead and copper sulfides. The
amount of indium in sphalerite may vary widely even within a deposit. The indium concentration is
typically 10 – 20 ppm, but may be as high as 1 % (Felix, 1996). An overview over different deposit
types and ore grade / tonnage relationship is given in Fig. 2.2.
Due to its low concentrations indium is not mined for itself but is recovered as an important by-
product of zinc metal processing operations. It accumulates during extraction in various intermediate
products and waste. Due to the complex technology required to recover indium, only a few operations
are able to produce the metal. The main producers of indium (Umicore, Dowa Mining, Nippon Min-
ing, Sumitomo Metal Mining, Teck Cominco, Falconbridge) are all into zinc processing. However,
Schwarz-Schampera & Herzig (2002) stress that the resource base is underestimated when only look-
ing at zinc bearing minerals, since noteworthy deposits combined with tin (casserite) exist.
Fig. 2.2 Distribution of ore grade and reserve for different indium bearing deposits (Schwarz-Schampera & Herzig,
2002).
2.5.2 Secondary sources

Worldwide only small amounts of Indium (mainly Indium Tin Oxide (ITO) scrap) are recycled. The
reason for the low recycling rate is the lack of infrastructure for collecting indium-containing products,
although recycling is supposed to expand significantly if the current price of indium is sustained. In-
dium tin oxide (ITO) use is highly inefficient, as only about 15% of ITO is utilised in the LCD pro-
duction process; the rest is scrap. The major problem in recycling the ITO scrap is the high cost asso-
ciated with the process. The process takes about 12 weeks from collection of scrap to fabrication of
secondary indium products. A large increase in ITO scrap could be difficult for the recycling industry
to handle because of large capital costs, environmental restrictions, and storage space (USGS, 2006a).
Taking an average concentration of 2 ppm of ITO in collected backlights 2 , 90% recovery rate and 95%
reprocessing efficiency, the required amount of scrap backlights to gain a kg indium would amount to
585 tonnes. Even with prices as high as 1’000 $ per kg ITO a global collection system is not probable
to evolve, and therefore no post consumer scrap recycling may be considered.
2
Content in 17’ backlight (1.8 kg) 1.76 ppm, in 15’ backlight (1.1 kg) 2.05 ppm for total film thickness of 500 nm.

2.5.3 Indium Supply

The supply of indium mainly depends on a few zinc extracting operations in Canada, Russia, China
and Japan (Stevens, 2007). Due to a sharp rise in ITO demand in LCD fabs the primary producer may
not meet the demand. This development led to a dramatic increase of the price starting in 2003 (Fig.
2.3). The supply gap is filled with releases of strategic stocks and the processing of new scrap. In 2004
Teckcominco (2005) estimated that almost 50% of the supply was met by secondary (industry scrap)
sources (Fig. 2.4). Stevens (2007) estimates that existing potentials for future indium supply (“x-
factors”) may be activated by enhancing the recovery efficiency by 5%, treatment of secondary mate-
rials and a slowing down of the demand by LCD fabs. He estimates a share in reclaimed secondary
(prompt scrap) sources of 60% in 2008 (Fig. 2.5).
Fig. 2.3 Indium prices with comments (Teckcominco, 2005; Metal Bulletin, 2006).

Fig. 2.4 Supply of Indium in 2004. Recycling of prompt scrap (ITO targets and the like) plays a vital role in the indium
supply (Teckcominco, 2005).
Fig. 2.5 Estimated sources of Indium. Assumption that all potentials for indium recovery are seized and that the de-
mand will slow down (Stevens, 2007). FPD = Flat panel display, CAGR = compound annual growth rate.

3. Coupled Production
3 Coupled Production
The metals lead, zinc, cadmium, silver (contained in the parkes process crust of primary lead produc-
tion) and indium are mainly found coupled in deposits. A compilation of world production is given in
Tab. 3.1.
Namely the cadmium metal is almost exclusively from Zinc ores. Today approximately 80% of world
cadmium production (=90% of the supply) was derived from mining, smelting, and refining of zinc,
and the remaining 20% came from copper and lead smelting and the recycling of cadmium products
(USGS, 2007). The ratio between the world mine production of Cadmium and Zinc is quite stable over
the last 20 years at about 2 kg Cd produced per tonne extracted Zinc. This corresponds to the average
concentration of Cd found in worked Zn ores of about 0.3%.
Tab. 3.1 World production in metric tonnes per year compiled from Kelly & Matos (2007) through 2004, for 2005 and
2006 the respective mineral commodity summaries from USGS (2006b) were taken.
Lead Silver Zinc Cadmium Indium Cd / Zn In / Zn Ag / Pb

Mine Refinery Refinery
year t/a t/a t/a t/a t/a Ratio Ratio ratio
1986 3'240'000 13'000 6'840'000 19'100 41 0.28% 6.01E-06 4.01E-03
1987 3'430'000 14'000 7'190'000 19'000 54 0.26% 7.44E-06 4.08E-03
1988 3'420'000 15'500 6'770'000 21'900 106 0.32% 1.57E-05 4.53E-03
1989 3'400'000 16'400 6'820'000 21'400 115 0.31% 1.69E-05 4.82E-03
1990 3'370'000 16'600 7'150'000 20'200 118 0.28% 1.65E-05 4.93E-03
1991 3'260'000 15'600 7'270'000 20'900 140 0.29% 1.93E-05 4.79E-03
1992 3'200'000 14'900 7'250'000 19'600 140 0.27% 1.93E-05 4.66E-03
1993 2'900'000 14'100 6'910'000 18'700 140 0.27% 2.03E-05 4.86E-03
1994 2'800'000 14'000 7'050'000 18'200 145 0.26% 2.06E-05 5.00E-03
1995 2'710'000 14'900 7'280'000 20'100 239 0.28% 3.28E-05 5.50E-03
1996 2'920'000 15'100 7'480'000 18'900 200 0.25% 2.67E-05 5.17E-03
1997 3'100'000 16'500 7'540'000 20'300 230 0.27% 3.05E-05 5.32E-03
1998 3'060'000 17'200 7'570'000 20'200 230 0.27% 3.04E-05 5.62E-03
1999 3'080'000 17'600 7'960'000 20'000 215 0.25% 2.70E-05 5.71E-03
2000 3'200'000 18'100 8'770'000 20'300 335 0.23% 3.82E-05 5.66E-03
2001 3'120'000 18'900 8'910'000 20'000 408 0.22% 4.58E-05 6.06E-03
2002 2'870'000 18'500 8'520'000 18'000 366 0.21% 4.30E-05 6.45E-03
2003 3'140'000 18'400 9'600'000 18'500 256 0.19% 2.67E-05 5.86E-03
2004 3'110'000 19'700 9'600'000 18'700 406 0.19% 4.23E-05 6.33E-03
2005 3'270'000 19'300 9'800'000 19'400 500 0.20% 5.10E-05 5.90E-03
2006 3'360'000 19'500 10'000'000 20'900 480 0.21% 4.80E-05 5.80E-03


4.1 Lead
The element lead (Pb, CAS 7439-92-1) is a lustrous metal. When freshly cut, surfaces are bluish white
but oxidise readily to the familiar dull grey colour. Lead has four naturally occurring isotopes: 208Pb
(51 – 53 %), 206Pb (23.5 – 27 %), 207Pb (20.5 – 23 %) and 204Pb (1.35 – 1.5 %). Three of the natural
isotopes are radioactive (208Pb, end product of thorium decay; 206Pb and 207Pb, from uranium series). A
number of unstable radioactive isotopes of Pb also exist. Lead is a preferred shielding material against
gamma rays and X-rays; if sufficiently pure, it does not become radioactive when used for shielding
against neutron sources (Sutherland et al., 1997). Some physical properties are listed in Tab. 4.1.
Tab. 4.1 Some physical properties of lead (Sutherland et al., 1997).
Physical properties of lead

Melting point °C 327.4
Latent heat of fusion, J/g 23.4
Boiling point °C 1741
Latent heat of vaporisation, J/g 862
Vapour pressure, kPa 980 °C 0.133
Density, g/cm3 20 °C 11.336
Mean specific heat, J g–1 K–1 0–100 °C 0.131
Specific electrical resistance, /cm 20 °C 20.65
Brinell hardness (20 °C) 2.5–3.0
4.2 Zinc
The remarkably low boiling point of 906 °C is important in pyrometallurgical zinc production proc-
esses. Zinc can be worked at temperatures as low as 120 °C, and its good deformability is the basis for
some of its uses. Zinc is resistant to air because of the self-protective coating formed. It is also resis-
tant to halogens, but is rapidly corroded by HCl gas. Because of its high surface area, zinc dust is
much more reactive and can even be pyrophoric, e.g., reacting vigorously at elevated temperatures
with the elements oxygen, chlorine, and sulphur (Graf & Schleininger, 1997). Zinc is traded in differ-
ent purity grades, which are specified by different authorities. In Northern America, SHG is the high-
est purity zinc grade, followed by HG and finally PW. Comparable specifications for primary zinc
grades can be found in Europe (European Standard EN 1179, Tab. 4.2), in Japan (Japanese Industrial
Standard JIS H 2107), and in Australia (Australian Standard AS 1242).
3
Tab. 4.2 Composition of different zinc grades in Europe, data provided by IZA .
EN 1179 Grade Pb Fe Cd Al Cu Sn Total Zinc, min

Designation non-zinc
Z1 0.003 0.002 0.003 0.001 0.001 0 0.005 100
Z2 0.005 0.003 0.005 - 0.002 0 0.01 100
Z3 0.03 0.02 0.01 - 0.002 0 0.05 100
Z4 0.45 0.05 0.01 - - - 0.5 99.5
Z5 1.4 0.05 0.01 - - - 1.5 98.5

3
Tab. 4.3 Some physical properties of zinc, data provided by IZA .
Physical properties of zinc

Atomic number 30
Atomic weight 65.37
Density (at 25°C) 7140 kg/m3
Melting point 419.5°C (692.7 K)
Boiling point (760 mm Hg) 907°C (1180 K)
Modulus of elasticity 70’000 MN/m2
Specific heat (20°C) 0.382 kJ/kg*K
Latent heat of fusion (419.5°C) 100.9 kJ/kg
Latent heat of vaporisation (906°C) 1.782 MJ/kg
Heat capacity
- solid Cp = 22.40 + 10.5 x 1/1000 1/TJmol (298 - 692.7 K)
- liquid Cp = 31.40 1/Jmol
- gas Cp = 20.80 1/Jmol
Linear coefficient of thermal expansion
39.7 µm/m*K
(polycrystalline 20-250°C)
Volume coefficient of thermal expansion (20 - 400°C) 0.89 x 1/1000000/K
Thermal conductivity: solid (18°C) 113W/m*K
Electrical resistivity (polycrystalline at 20°C) 59.0 nΩ·m
Standard electrode potential (against H2 electrode) -0.762 V
4.3 Cadmium
Features of Cadmium metal are compiled in Tab. 4.4. Due to its toxicity the dissipative use has been
cut down and regulated.
4
Tab. 4.4 Properties of cadmium .
Natural cadmium levels in the environment

Atmosphere 0.1 to 5 ng/m³
Earth's crust 0.1 to 0.5 µg/g
Marine sediment ~1 µg/g
Sea-water ~0.1 µg/l
Cadmium physical proper-
ties
Form white, soft metal
Characteristics malleable, ductile and flexible
Melting point 321°C
Atomic number 48
Atomic weight 112.41
Density 8.64 g/cm³
3
http://www.iza.com/properties.html, access date may 2007
4
http://www.cadmium.org/, access date april 2007

4.4 Indium
Indium is a very soft, silvery-white metal that has a bright luster. As a pure metal indium emits a high-
pitched "cry" when it is bent. Both gallium and indium are able to wet glass. Indium has a low melting
point (429.75 K) but a high boiling point (2353.15 K). Indium becomes superconducting at 3.37 K.
Metallic indium is not oxidized by air or oxygen at room temperature. It reacts directly with arsenic,
antimony, the halogens, oxygen, phosphorus, sulfur, selenium, and tellurium when heated. Indium dis-
solves only slowly in cold dilute mineral acids, but more readily in hot dilute or concentrated acids.
Alkalis do not attack the massive metal. Indium forms alloys with most other metals. Extensive solid
solutions are formed with lead, thallium, and mercury. Some properties of indium are compiled in Tab.
4.5 (Wikipedia Contributors, 2007; Felix, 1996).
Tab. 4.5 Some properties of the element indium (Wikipedia Contributors, 2007; Felix, 1996).
Properties Unit Value

−3
Density (near r.t.) g·cm 7.31
Liquid density at m.p. g·cm−3 7.02
Melting point °C 156.6
Boiling point °C 2072
−1
Heat of fusion kJ·mol 3.281
Heat of vaporization kJ·mol−1 231.8
Heat capacity (25 °C) J·mol−1·K−1 26.74
Standard atomic weight g·mol−1 114.818(3)
Electron configuration [Kr] 4d10 5s2 5p1
Electrons per shell 2, 8, 18, 18, 3
Electrical resistivity (20 °C) n Ω·m 83.7
Thermal conductivity (300 K) W·m−1·K−1 81.8
Thermal expansion (25 °C) μm·m−1·K−1 32.1
Young's modulus GPa 11
Mohs hardness 1.2
Brinell hardness MPa 8.83
CAS registry number 7440-74-6
Standard electrode potentials
In3+ + 3e B In V –0.338
In3+ + 2e B In+ –0.40
In2+ + e B In+ –0.40
In3+ + e B In2+ –0.49


5.1 Lead
Generally toxicity concerns and substitution by plastics and copper have reduced the market consid-
erably. Main applications of lead are the following (Sutherland et al., 1997):
Batteries: The battery market is by far the largest end use of lead. Batteries for starting, lighting and
ignition (SLI) represent the predominant use of lead. Traction batteries are used to power electric ve-
hicles. Stationary batteries are used for standby and emergency power supply.
Pigments and Lead Compounds: After batteries, the next largest use of refined lead is in glass mak-
ing, ceramic glazes, plastic stabilisers, and paints. The lead pigment market is now very small.
Semi finished Products: Rolled and extruded lead, mainly sheet and pipe, account for about 8 % of
total lead use.
Cable Sheathing: A major area of lead use was in the protection of electrical and telecommunications
cables, but this market has declined drastically due to substitution by aluminium and plastics.
Lead Alloys: Solders, type metal, bearing alloys containing lead still account for about 4 % of the lead
consumption.
Gasoline Additives: In 1972 the use of tetraethyl lead for improving the octane rating of gasoline was
nearly 400’000 t/a. Since then, this sector has declined by 70 % due to restrictions through environ-
mental legislation.
Shot and Ammunition: Shot and ammunition account for about 2 % of lead consumption and repre-
sent a steady market.
Miscellaneous: Uses that do not fall into the above categories account for about 4 % of lead use, and
include products featuring the high density of lead such as wheel weights, yacht keels, ornamental
items, stained glass, and massive radiation shielding.
Fig. 5.1 Applications of lead in 2001 (LME website5)
5
http://www.lme.com/lead_industryusage.asp, accessed in may 2007

5.2 Zinc
Due to its manifold properties, Zinc is used in a broad field of applications (Tab. 5.1). The quantities
of first- and end-use are given in Fig. 5.2. Main uses are as follows (Graf & Schleininger, 1997):
In the industrialised countries, almost 50 % of the zinc consumed is used for corrosion protection.
Commercial zinc (98.5 – 99.5 %) is usually used for hot-dip galvanising in production and jobbing
plants, and high-purity zinc (99.95 %) for the continuous galvanising of steel strip or wire. The metal-
lurgical processes used to produce the less pure grades of zinc means that lead is the main impurity
present. This is an advantage, as the presence of 1 % lead reduces the surface tension of zinc by ca.
40 % compared with that of the high-grade product. The components being galvanised are therefore
more effectively wetted by the molten zinc.
Pressure die casting is another major area of use and is dominated by alloys of the type ZnAl4Cu1 and
modifications thereof. Approximately 50 % of all zinc pressure die castings go to the automobile in-
dustry.
Zinc-based chemicals, including zinc dust, account for ca. 7 % of the worldwide zinc consumption.
Zinc oxide is quantitatively by far the most important zinc-based chemical product, followed by zinc
dust, zinc sulphate, and zinc chloride in order of importance and quantity. All other zinc compounds
are of minor importance.
Fig. 5.2 Zinc consumption in 2000: first-use and end-use.

6
Tab. 5.1 Properties, first- and end-use of zinc, data provided by IZA .
Property First-Use End-Use

- Reactivity with iron Corrosion protection (galva- Building/construction, energy/power, street furni-
- Corrosion resistance nising, zinc thermal spraying, ture, agriculture, automotive/transport
- Electrochemical electroplating, zinc-rich
paints)
- Low melting point Die casting and gravity cast- Automotive equipment, household appliances,
- Fluidity ing fittings, toys, tools, etc.
- Capacity for surface treatment
- Strength
- Alloying characteristics Brass (copper-zinc alloy), Building/construction, fittings, automotive and
aluminium alloys, magnesium electrical components, etc.
alloys
- Formability Rolled zinc Building/construction

- Resistance to corrosion
- Electrochemical Batteries Automotive/transport, computers, medical
equipment, consumer products
- Chemical Zinc oxide, zinc stearate Tyres, all rubber goods, paint pigments, ceramic
glazes, electrostatic copying paper
- Essential nutrient Zinc compounds Food industry, animal feed, fertilisers
- Healing Zinc compounds Pharmaceutical industry, cosmetics industry
5.3 Cadmium
In recent years, the consumption pattern of cadmium in its various applications has shifted away from
the market areas of pigments, stabilisers and coatings to rapidly growing applications in Ni-Cd batter-
ies 7 .
7
Fig. 5.3 Uses of Cadmium in different applications (reference year 2003 )
6
http://www.iza.com/zwo_org/, access date 01.03.03
7
http://www.cadmium.org, accessed in April 2007

Description of the different uses of Cadmium 7:

Nickel-Cadmium Batteries. Cadmium hydroxide is utilised as one of the two principal electrode ma-
terials in Ni-Cd batteries which have extensive applications in the railroad and aircraft industry for
starting and emergency power and in consumer applications such as cordless power tools, cellular
telephones, camcorders, portable computers, portable household appliances and toys.
Cadmium Pigments. Cadmium sulphide and cadmium sulphoselenide are utilised as bright yellow to
deep red pigments in plastics, ceramics, glasses, enamels and artists colours. They are used in applica-
tions where high temperature or high pressure processing is required.
Cadmium Stabilisers. Cadmium-bearing stabilisers retard the degradation processes in polyvinyl-
chloride (PVC) which occur upon exposure to heat and ultraviolet light. These stabilisers contain or-
ganic cadmium salts, usually carboxylates such as cadmium laurate or cadmium stearate, which are in-
corporated into PVC before processing.
Cadmium Coatings. Cadmium coatings are utilised on steel, aluminium, and certain other non-
ferrous metal fasteners and moving parts to provide the best available combination of corrosion resis-
tance, particularly in salt and alkali media, and lubricity or low coefficient of friction. They are also
employed in many electrical or electronic applications where a combination of corrosion resistance
and low electrical resistivity are required. In addition, cadmium coatings exhibit excellent plating
characteristics on a wide variety of substrates, have good galvanic comparability with aluminium, and
are readily solderable.
Alloys and Minor Uses. Cadmium alloys include (a) electrical conductivity alloys, (b) heat conduc-
tivity alloys, and (c) electrical contact alloys. Other minor uses of cadmium include cadmium telluride
and cadmium sulphide in solar cells, and other semiconducting cadmium compounds in a variety of
electronic applications.
5.4 Indium
Today virtually all of the LCD produced use ITO as conducing layer. ITO is the transparent conductor
of choice in terms of high transmissivity of visible light with low electric resistivity compared to other
transparent conductors, such as tin oxide (SnO2) and zinc oxide (ZnO) (Granqvist & Hultaker, 2002).
The sputtering of the substrate is done by the display manufacturers. Due to the very large cost of
building TFT factories, the number of major panel vendors is small. Several of the best known are
Sharp Corporation, Samsung Electronics, LG.Philips LCD and AU Optronics. As of April 2006, the
three biggest manufacturers are LG.Philips and Samsung with 22% of the market, followed by AU
Optronics with a 19% share (Wikipedia Contributors, 2006).
Jansseune (2003) estimates that in 2002 45% of the indium supplied went as ITO into coating,
whereas 2004 an estimated 72% of the indium was used for ITO coatings (Jorgenson & George,
2004). End 2006 Hewett (2006) estimated the share of ITO in the total indium supply to 80%. In 2004
perhaps 65 %of the indium used in coating goes into display devices, and the remainder, into transpar-
ent heat reflection. ITO coatings also are used in architectural glass and solar panel applications (high
infrared reflectance), cathode-ray tubes (CRTs), low-pressure sodium lamps, and windshield glass.
(Jorgenson & George, 2004).
In 2005 the USGS estimates that 70% of the Indium was used in coatings, electrical components and
semiconductors, 12%; solders and alloys, 12%; and research and other, 6% (USGS, 2006a).

Fig. 5.4 In 2002, the semiconductor industry accounted for 15% of indium consumption, equivalent to around 50,000
kg. Credit: USGS (Jansseune, 2003).
Fig. 5.5 In 2004 already almost 80% of the indium went into ITO coatings (Teckcominco, 2005).
In 2004 Teckcominco (2005) estimated about 640 tonnes of indium in the end-use, of which ITO coat-
ings amounted to 79% (Fig. 5.5).
Coatings
Indium's largest use is for coatings. Metallic indium ingots are processed into an oxide and then mixed
with tin-oxide to form indium-tin-oxide (ITO). This material is used to produce sputtering targets.
These targets are placed in a vacuum and the ITO is "sputtered" onto glass for use in display devices
such as laptop computer screens, LCD watches, appliance display panels and portable phones.
ITO is also used in traditional applications such as low-pressure sodium lamps, architectural glass and
oven doors (heat reflector applications).
New applications for indium coatings include high definition televisions and thin screen computer
monitors (Teckcominco, 2006).
Application of ITO coatings is a wasteful process and only around 20-30% of the material actually
ends up on the substrate (Jansseune, 2005). The rest is recycled as prompt scrap.
Alloys
Indium has a relatively low melting point at 157°C. This enables it to be utilized in alloy production
such as AgInSbTe sputtering target manufacturing for the electronics industry (Teckcominco, 2006).

Electronics & Semiconductors

Indium is used as a doping agent for germanium (Ge) and gallium arsenide (GaAs) semiconductors.
The largest use for high purity indium is for the production of indium phosphide (InP) for use in the
semiconductor industry. InP is also used in photonic applications, such as lasers, photo detectors, opti-
cal modulators and amplifiers, and in electronic applications, such as, HEMT’s (high electron mobility
transistors), HBT’s (heterojunction bipolar transistors), and FET’s (field effect transistors), millimeter
wave and microwave circuits, and high-speed digital integrated circuits. (Teckcominco, 2006)

This chapter describes the mining of sulphidic zinc-lead deposits and the related extraction of lead and
zinc but also minor metal values (metals contained in the ore, whose extraction is profitable) such as
silver, cadmium and indium for which these deposits represent an important source for worldwide
primary production (Silver: 20%, Cadmium: 80%, Indium: ~100%). Therefore, the inventory model-
ling involves a series of multi-output processes in which all burdens are attributed by economic alloca-
tion procedures based on Guinée et al. (2004) and Althaus & Classen (2005) using the average prices
from 2004 through 2006 reported by the U.S. Geological Survey (see Tab. 3.1):
6.1. The module “mining, zinc-lead-deposit” delivers the flows
- “lead, concentrate, at beneficiation” (containing silver) and
- “zinc, concentrate, at beneficiation” (containing cadmium and indium)
6.2. The module “smelting, primary lead production” processes the concentrate to the flows
- “lead, primary, at plant” and
- “parkes process crust, from desilverising of lead” as an intermediate step for the silver chain
6.3. “lead, secondary, at plant”
6.4. “lead, at regional storage” (a blend of primary and secondary lead)
6.5. The module “smelting, primary zinc production” processes the concentrate to the flows
- “zinc, primary, at regional storage” (primary production)
- “cadmium sludge, from zinc electrolysis, at plant” as raw material for cadmium production, and
- “leaching residues, indium rich, from zinc circuit, at smelter” as raw material for indium production.
6.6. “cadmium, primary, at plant”
6.7. “indium, at regional storage”
A graphic representation is shown in Fig. 6.1, the boxes with rounded corners represent reference
flows from multi-output processes, bold lines denote datasets to be used by the LCA practitioner.

mining, zinc-lead-
deposit (GLO)
zinc, concentrate, at lead, concentrate, at

beneficiation (GLO) beneficiation (GLO)
smelting, primary zinc smelting, primary

production (GLO) lead production
(GLO)
zinc, primary, at parkes process crust, from lead, secondary, at

regional storage desilverising of lead (GLO) plant (RER)
(GLO)
lead, primary, at plant

cadmium sludge, from leaching residues, indium (GLO)
zinc electrolysis, at plant rich, from zinc circuit, at
(GLO) smelter (GLO)
cadmium, primary, at indium, at regional storage To silver lead, at regional storage

production
plant (GLO) (RER) (RER)
chain
Fig. 6.1 Graphic representation of the modules and flows resulting from the mining of zinc-lead deposits.

Tab. 6.1 Meta information for the mining of zinc and lead bearing ore.
Name mining, zinc-lead-deposit

Location GLO
Unit kg
The module includes mining and beneficiation steps with the mining infrastructure
Included Processes
and disposal of overburden and tailings (see part "Auxiliary Processes").
Amount 1
Local Name Abbau Zink-Blei-Lagerstätte
Synonyms
The module describes the global production mix for jointly produced zinc- and
lead-concentrates from sulphidic deposits. The multi output-process “exploitation,
zinc-lead-deposit” delivers the co-products “lead, concentrate, at beneficiation” and
ence function
“zinc, concentrate, at beneficiation”. It is designed for the use of the concentrates
as raw material for the ultimate production of lead and zinc.
Start Date 1994
End Date 2003
Other Period Text
This module represents a global production mix. Some agents used like lime or
sodium cyanide originate from data sets valid for Switzerland or Europe. Proc-
Geography text
esses like blasting and transport originate from data sets valid for Switzerland or
Europe.
This module is approximated with values from nickel-mining. Mining is assumed to
be underground, followed by froth beneficiation. Overburden and tailings are dis-
Technology text
posed inside the mining panels. For electricity mix a mix of 25% coal and 75% gas
fired plants was chosen.
Production Volume
Uncertainty Adjustments see Geography and Technology

Tab. 6.2 Meta information for the lead production.
Name smelter, primary lead production lead, secondary, at plant lead, at regional storage
Location GLO RER RER
Unit kg kg kg
Data Set Version 2 2.0 2
Included Processes The module includes the pro- Collection, sorting and remelting The module includes the pro-
duction of primary lead with the of the lead contained in lead duction of primary and secon-
sinter/blast furnace (ISP) and acid batteries. dary lead for Europe.
direct smelting process, the dis-
posal of slag and final refining of
lead
Local Name Verhüttung, Primärblei Produk- Blei, sekundär, ab Werk Blei, ab Regionallager
tion
Synonyms Recycling of lead acid (PbA)
batteries // Recycling von Bat-
terien
General Comment The module describes the pri- The module describes the pro- This dataset describes the sup-
to reference func- mary production of lead in duction of secondary lead in ply mix of lead used in Europe.
tion Europe. The multi-output- Europe. The feed of secondary It is modelled with a blend of
process "Smelter, primary lead material consists of scrap lead 25% primary and 75% secon-
production" delivers the co- acid (PbA) batteries form auto- dary production, complemented
products "Lead, primary, at motives. The data refers to one with the transport service re-
plant" and "Parkes process big operation in Europe that op- quired to transport the material
crust, from desilverising of erates with representative tech- to the regional storage. In the
lead".. The flow "Lead, …" is nology. global average the secondary
part of the respective European production amounts only 65%.
supply mix. The by-product
"Parkes process crust, …" re-
ceives part of the burden and
enters the silver production
chain.
Start Date 2000 2000 2000
End Date 2005 2005 2005
Data Valid For En- 1 1 1
tire Period
Geography text This module represents the pro- This module represents the This module represents the
duction of primary lead on a European production based on consumption mix of lead in
global average. a single major site in Belgium. Europe.
Technology text A mix of 56% direct smelting The referred operation uses a A mix of 25% primary and 75%
and 44% sinter/blast furnace shaft furnace with post combus- secondary production.
(ISP) is chosen. For emission tion, which is the usual technol-
control 56% improved and 44% ogy for secondary smelters.
limited control is chosen.
Representativeness 0.8 %
[%]
Production Volume 1'398'000 t in 1994 5'000'000 t in 2005
Sampling Procedure literature literature literature
Uncertainty Adjust- none none none

ments

Tab. 6.3 Meta information for the zinc production process
Name smelting, primary zinc production

Location GLO
Unit kg
The module includes the production of high grade (SHG) primary zinc by
pyro- and hydrometallurgical processes, the disposal of the slag and the
Included Processes
treatment of the waste water. No secondary production - which amounts
to 30% in Europe - is considered.
Amount 1
Local Name Verhüttung, Primärzink Produktion
Synonyms
The module describes the primary production of zinc in Europe. The
multi-output-process "smelting, primary zinc production" delivers the
three co-products "zinc, primary, at regional storage", "cadmium sludge,
General Comment to reference func- from zinc electrolysis, at plant" and "leaching residues, indium rich, from
tion zinc circuit, at smelter". The flow "zinc, …" represents the European
supply mix of primary zinc. The by-products "leaching residues, …" and
"zinc, …" receive part of the burden and enter the primary production
chain.
CAS Number
Start Date 1994
End Date 2003
Other Period Text
This module represents the consumption mix of primary zinc in Europe.
Geography text It is assumed that 75% of the concentrates are imported, while all zinc is
produced within Europe.
A mix of 80% hydrometallurgical and 20% pyrometallurgical production
Technology text is chosen. For emission control 80% improved (hydro) and 20% limited
control (pyro) is chosen.
Production Volume 10'000'000 t in 2006

Tab. 6.4 Meta information for the production of primary cadmium, at regional storage.
Name cadmium, primary production, at regional storage

Location RER
Unit kg
Included Processes The module includes the treatment of cadmium sludge from hydro-
metallurgical zinc operations and subsequent refining by electrolysis.
Transports and infrastructure are considered.
Amount 1
Local Name Cadmium, Primärproduktion, ab Regionallager
Synonyms
General Comment to reference func- The module describes the primary production of cadmium metal
tion (99.9%) on a global average. No secondary sources from post con-
sumer scrap are considered. Only the most important source, the cad-
mium sludges from hydrometallurgical zinc operations, is reflected with
the dataset. For the use in applications where cadmium is a major con-
stituent – e.g. in NiCd battery systems – the dataset should be com-
plemented with additional specific emission factors and the secondary
production process..
Start Date 2000
End Date 2005
Other Period Text
Geography text This module represents the production of primary cadmium on a global
average. The inventory however is compiled with European average
data, e.g. for electricity production.
Technology text Cadmium sludge formed in hydrometallurgical zinc operations by purifi-
cation of zinc solution. Subsequent purification and oxidation of the
sponge. Cadmium cathodes formed by electrowinning. Other sources
from pyrometallurgical zinc wastes and post consumer scrap (batteries)
disregarded.
Production Volume

Tab. 6.5 Meta-information for the dataset “indium, at regional storage”
Name indium, at regional storage

Location RER
Unit kg
Included Processes The module includes the production of indium from zinc extraction operations. Infra-
structure and transports to Europe are considered. Most process specific emissions
are missing.
Amount 1
Local Name Indium, ab Regionallager
Synonyms
General Comment to ref- The module describes the primary production of indium metal (99.999%, 5N) on a
erence function European average. No secondary sources from post consumer scrap are consid-
ered. Only the most important source of indium raw material, the hydrometallurgical
zinc operations, is reflected with the dataset. This dataset is based on the respec-
tive flow from multi-output module "smelting, primary zinc production". This dataset
is not to be used when indium is a substantial part of the system modelled, viz. for
comparisons of special solder pastes.
CAS Number
Start Date 2000
End Date 2005
Data Valid For Entire Pe- 1
riod
Other Period Text
Geography text This module represents the production of primary indium for the consumption in
Europe. The inventory however is compiled with global average data, e.g. for zinc-
production.
Technology text Indium containing residues from hydrometallurgical zinc operations are leached.
Subsequent solvent extraction and precipitation of sponge indium. Indium cathodes
are cast and electrorefined, followed by vacuum refining to obtain 5N quality. Sol-
vent-extraction and electrowinning has been extrapolated from copper refining.
Production Volume 480 t in 2006

6.1 Mining of zinc-lead-deposits

About 70 % of zinc and the same amount of lead lead is won from a sulphidic ore, where the elements
occur jointly. They are also jointly extracted.

Zinc-lead ores are assumed to be mined mostly underground. The ore is transferred to the benefici-
ation. Widening of the tunnels is mainly done by blasting. The overburden – material, which does not
contain metal-bearing ore – is partly deposed off-site and partly refilled into the tunnels.
After mining, the ore is firstly ground. In a next step it goes through gravity concentration to separate
the metallic particles from the metal bearing minerals. After this first concentration step, flotation is
carried out to remove the gangue from the sulphidic minerals. For neutralisation lime is added. In the
flotation several organic chemicals are used as collector, frother, activator, depressor and flocculant.
Sometimes cyanide is used as depressant. Tailings are usually transferred to tailing heaps or ponds.
Concentrates are produced containing around 50% zinc and 50% lead.
6.1.2 Waste
The overburden is disposed of close to the mine. Tailings are deposed as piles and in ponds. In both
the overburden and the tailings acid rock drainage occurs over a long period of time. These effects are
handled in the part IX (Non Ferrous Metal Winning – Auxiliary Processes, chapter Disposal of Tail-
ings and Overburden) of this report.
6.1.3 Emission
The major emissions are mineral born pollutants in effluents and dust. The prevalently underground
mining operations generate roughly 80 % of the dust emissions from surface operations, since the ma-
jor dust sources do not take place underground. Rain percolates through the overburden and generates
metal leachates to groundwater.
the ore itself. Flotation produces effluents containing several organic agents 8 used. Some of these
chemicals evaporate and account for VOC emissions to air. In particular, xanthates decompose hydro-
lytically to release carbon disulphide. Tailings effluent contains additional sulphuric acid from acid
rock drainage.
6.1.4 Mining, zinc-lead deposits in ecoinvent

Since the extraction of zinc and lead is very similar to the extraction of other sulphidic deposits like
nickel and copper bearing ores, and no information on the specific mining processes for zinc-lead de-
posits is available, the data for the exploitation of nickel provided in Hilbrans & Hinrichs (1999) was
used. Mass flows were adjusted to values specific to zinc-lead extraction, according to information
given in Ayres et al. (2002).
8

The module describes the global production mix for jointly produced zinc- and lead-concentrates from
sulphidic deposits. The multioutput-process “exploitation, zinc-lead-deposit” delivers the co-products
“lead, concentrate, at beneficiation” and “zinc, concentrate, at beneficiation”. It is designed for the use
of the concentrates as raw material for the production of lead and zinc. The layout of the modules with
general flow information, remarks, sources, values and uncertainty information is shown in Fig. 6.3
and Fig. 6.4. The following paragraphs describe calculations, sources and assumptions chosen in this
study. The functional unit of this process is one kg of added zinc- and lead-concentrates. The meta-
information for this dataset is compiled in Tab. 6.1
Fig. 6.2 gives an overview of the general assumptions made in the mass flows of this module. It is as-
sumed, that two virtually distinctive concentrate products are produced. This of course is not true in
real production, but it is a formal necessity for the further use in ecoinvent in the modules for zinc and
lead production.
The range of concentrate composition described varies considerably. Depending on the processing
technology, a high degree (60-75% of a single metal, lead or zinc) concentrate may be required. Most
smelters accept lower grade concentrates that range 50-60% Pb (e.g. Trail, Mt. Isa). In this study a
content of 55% Pb in the concentrate is chosen. The range of zinc content reported in zinc concentrates
is much narrower than for lead. A compilation of concentrates of 80 mines in (Rodier, 1990) showed a
range between 45 and 60% Zn, with a mean value of 53%. Reuter (2005) states an average zinc con-
centrate to have 50% Zn and uses 54% in his calculations.
In order to comply with the global production figures the assumption is made that not all of the minor
metals cadmium and indium are not entirely recovered. Only a share of the extracted volume is
worked for these metals, the other operations discard these metal values with the gangue. This is rep-
resented by the percentages in white colour in the input arrow in Fig. 6.2.
mining /
ore: 8'028 kg total concentrate: 1'852 kg
beneficiation
yield
Pb 401 kg @ 5% 100% Pb: 95% 693 kg Pb 381 kg @ 55%
Zn 723 kg @ 9% 100% Zn: 85% Ag 0.556 kg @ 802 ppm
Ag 0.59 kg @ 73 ppm 100% Ag: 95%
Cd 24.1 kg @ 0.3% 17% Cd: 95% 1159 kg Zn 614 kg @ 53%
In 0.40 kg @ 50 ppm 39% In: 80% Cd 3.84 kg @ 0.3%
In 0.125 kg @ 108 ppm
overburden unwanted minerals: 852 kg

and gangue:
6'176 kg
Fig. 6.2 Basic assumptions for the mining process of zinc- and lead ore from sulphidic deposits. Values relate to the
production of the sum of 1 tonne metals contained in the concentrate, they are mainly taken from Ayres et
al. (2002) and IPPC (2001).

The infrastructure used for ore exploitation is inventoried within this module, whereas the infrastruc-
ture like buildings and land-use for the operation is inventoried in another appropriate module, com-
prising both mining and beneficiation facilities. The mobile equipment used for exploitation like drill-
ing machines, scoop trams and underground trucks is inventoried through their diesel consumption
The module used for this purpose “diesel, burned in building machine” includes the infrastructure of
the machines. Additionally, 2000 m conveyor belts are assumed with a capacity of 2000 t/h which
yields 2*10-5 m conveyor belt required per tonne mined raw ore.
Since mining is underground and no heap leaching is done, no land use for exploitation is assumed.

Tab. 6.6 Infrastructure in underground mining, assumptions used. A lifetime of ten years and 5000 working hours
yearly are estimated.
action machine throughput lifetime unit Infrastructure

Mass (t) Length (m) m/t ore mined
Transport Conveyor belts 100'000'000 2000 2*10-5
Mining
mine, for which the standard module for disposal of sulphidic overburden has been chosen, which in-
cludes the land use and land transformation by the disposal. Overburden refilled into the mine is as-
sumed to be burden free. The degree of the refilling is documented for extraction of nickel-ores in
Hilbrans & Hinrichs (1999). This source is also used for water use and effluent volume from mining.
The use of freshwater was inventoried as river water, whereas the pit water was assumed to be
groundwater, which has to be pumped out. The composition of a generic effluent from mines was
taken from IPPC (2002), its values are shown in Tab. 6.8. No waste-water treatment is considered. No
information on the energy production in mining is available. In many cases the electricity used in the
often remote mining operations is made on site by gas or coal fired power plants. In this study a split
of 75% gas fired and 25% coal fired plants was assumed based on own estimations. Specific values
and corresponding inventory flows are shown in Tab. 6.7. Cement, explosives and sand are assumed to
be transported by lorry over 100 km.
For refilling the pit, apart from overburden additional stowing material – mostly sand – and cement are
used.
Beneficiation
Once the ore is mined, it is ground and conditioned for subsequent concentration. Considerable
amounts of lime and flotation agents are added. The separated gangue is disposed in tailings ponds, the
concentrated ore is fed to the metallurgy, which is assumed to be on-site.
tionally the dust emissions due to mining activities. Like in the mining step, a share of 60% hydroelec-
tricity was chosen according to Hilbrans & Hinrichs (1999).
Lime: Lime input is documented in Hilbrans & Hinrichs (1999) for the beneficiation of lead-/zinc-
sulphide, the input of 554 g / t ore is taken from the specific input in nickel beneficiation.
Reagents: No information on the nature and mass of the floating agents used is available. For copper
mining a range of 25 – 300 g collector and 25 – 250 g frothing reagent per ton sulphidic ore is recom-
mended (Krauss et al., 1999). This is confirmed by the values in Tab. 6.9. Collector and frother are
balanced in ecoinvent as unspecified organic chemical. According to NPI (2001) xanthates hydrolyti-
cally decompose to 0.53 mg carbon disulphide per mg xanthate. With 189 mg xanthates from Tab. 6.9
this is about 100 mg carbon disulphide per t mined ore. This again is modelled as unspecific organic
chemical.
cals. Their emission to water is covered by reported emissions from mining sites shown in Tab. 6.8.
For pH adjustment and leaching, sulphuric acid is considered with 3.74 kg / t ore mined (Tab. 6.10).
As sulphuric acid is a by-product from the metallurgy, its production is regarded in ecoinvent as bur-
den free. Other acids are neglected.
Since composition and environmental fate of activator, depressant and sulphuric acid are uncertain,
reported data for total effluents shown in Tab. 6.8 are taken.

is used as leachate and as depressant for certain sulphidic minerals. A cyanide recovery of 80 – 90% is
reported due to abatement and recovery in the Golden Cross mine in New Zealand (Anonymous,
1998). Here a value of 80% is chosen for cyanide-abatement technology. With a reported use of 35 g
NaCN per tonne ore mined (Tab. 6.9) this means a direct emission of 3.75 g CN- per tonne ore mined.
An overview of the reagents accounted for is presented in Tab. 6.9.
Effluents: Additional to direct emissions due to the use of chemicals, emissions to water are chosen
according to Tab. 6.8. These values are from base metal sites, thus only an approximation for the real
effluent concentrations in the zinc / lead extraction. It is assumed here that the effluents also include
acid rock drainage (ARD). ARD after mine closure is modelled in the module “disposal, sulphidic tail-
ings, off-site”.
Dust: The dust emissions are approximated in both amount and particle distribution approximated
with specific values from copper mining from the CEPMEIP-Database 9 . According to IAEA 10 the
dust composition is approximated with concentrations of the zinc-lead-ore (Tab. 6.11). The dust emis-
sions are summed up in Tab. 6.11. In ecoinvent, for particulates three different classes are used, whose
total corresponds to the total dust emission.
Tab. 6.7 specific input and effluent values used and their corresponding value per t concentrate established in this
inventory. The specific values are approximated by values from sulphidic nickel deposits (Hilbrans &
Hinrichs, 1999).
per t ground per t ore per t con-

(ore plus o- centrate
verburden)
Amount ground treated t 7.30
Amount ore retrieved t 4.33
stowing material, sand t 0.284 2.07
cement t 0.0225 0.165
explosive kg 1.56 11.4
diesel fuel (energy) MJ 69.8 510
el. Energie, total kWh 39.6 172
fresh and processwater (mining) m3 0.347 1.51
effluents, mining m3 0.423 3
effluents, benefication m3 1.38 6.00
9
http://www.air.sk/tno/cepmeip/em_factors.php, accessed in 2003
10
The IAEA Database of Natural Matrix Reference Materials, http://www-naweb.iaea.org/nahu/external/e4/nmrm/, accessed
01.03.2003

Tab. 6.8 Emissions to water from mining and beneficiation. Composition of effluents from mining activities, values
are approximations from total emissions to water from base metal sites in (IPPC, 2002).
Compound Unit Range This study Number of

(geometric Samples
mean)
Ca g / m3 686 686 1
SO4 g / m3 1750 - 3175 2357 2
COD g / m3 2.24 - 48.5 10.4 2
Solids g / m3 2.38 - 11.3 5.17 4
Al mg / m3 69.3 - 108 86.3 2
As mg / m3 0.264 - 14.1 2.95 3
Cd mg / m3 0.0901 - 1.02 0.316 4
Co mg / m3 0.742 - 0.823 0.781 2
Cr mg / m3 0.0311 - 9.62 0.547 2
Cu mg / m3 1.24 - 44.9 7.91 5
Fe mg / m3 61.7 - 1326 291 3
Mn mg / m3 24.7 24.7 1
Hg mg / m3 0.0155 - 0.12 0.0377 3
3
Ni mg / m 0.792 - 1338 24.3 3
Pb mg / m3 0.0155 - 20 2.80 4
Zn mg / m3 5.37 – 225 75.9 5
N g / m3 2.50 - 1786 22.8 3
Tab. 6.9 Consumption of reagents in base metal mineral processing plants. Data taken from (IPPC, 2002).

(geometric samples
mean)
MIBC, Dowfroth
Dithiophosphate
phate

chosen in this study. For further details read in text and Tab. 6.9.

Carbon dioxide air emission g/t 244
Tab. 6.11 Emissions to air resulting from mining activities per tonne mined ore. Approximated values derived from
reported emission factors from copper extraction and a single ore composition. Values for total dust emis-
sion and, partition of fractions of particulate matter are taken from the CEMIP-database, the dust composi-
tion correspond to a specific nickel sulphide with values taken from the IAEA-database (see text).
Emission to air Unit Emission / t ore

Cadmium g / t ore 0.102
Lead g / t ore 5.1
Zinc g / t ore 9.2
Sulphur g / t ore 10.2
PM<2.5 g / t ore 5.0
PM2.5-10 g / t ore 45
PM>10 g / t ore 52
total dust g / t ore 101.72
Allocation
According to the allocation principle pursued throughout the inventory modelling in this report, an
economic allocation for the joint by-products lead and zinc concentrate is envisioned. Since beside the
main constituents lead and zinc, also silver, cadmium and indium are present in the products, also
these values have to be considered. Cadmium is considered contaminant and therefore does not affect
the allocation procedure, since it does not receive a value within the zinc production. The cadmium
containing wastes become valuable only after refining.

Tab. 6.12 Overview over the relevant parameters to determine the allocation split: composition of the by-products and
value of the metals contained.
Zinc concentrate Lead Concentrate Value in $ / kg

feed in kg 0.699 0.301
Pb in kg - 0.166 1.26
Zn in kg 0.370 - 3.09
Ag in kg - 3.29E-04 360
Cd in kg 0.0023 - contaminant
In in kg 6.03E-05 - 855
Value contained in $ 1.19 0.33
Allocation by value 79% 21%
Allocation by mass 70% 30%
6.1.5 Data quality

Data uncertainty is derived using the pedigree matrix as described in Frischknecht et al. (2004). The
inventory is fully based on values from nickel-mining reported in Hilbrans & Hinrichs (1999), which
may eventually bias the assertions. For specific conclusions further investigation especially on the ef-
fluents composition and tailings emissions has to be done. The uncertainty of emissions that are
adopted from nickel mining is arbitrarily reflected by a geometric standard deviation of 10. The choice
of electricity mix bases on assumptions that have to be improved with further investigations.

in ecoinvent
Modul name
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
Zinc, in sulphidic Zinc, 9.0% in sulfide, Zn 5.3%, Pb,
 85% yield resource in ground 3.90E-01 kg own assumptions 1 1.58 (4,2,2,1,4,4,12)
ore Ag, Cd, In, in ground
Lead, in sulphidic Lead, 5.0% in sulfide, Pb 3.0%, Zn,
ore Ag, Cd, In, in ground
Silver, in Silver, 0.007% in sulfide, Ag

sulphidic ore 0.004%, Pb, Zn, Cd, In, in ground
Cadmium, in 95% yield, in only 17% of the oredeposits Cadmium, 0.30% in sulfide, Cd
 resource in ground 1.30E-02 kg own assumptions 1 1.58 (4,2,2,1,4,4,12)
sulphidic ore Cd is extracted 0.18%, Pb, Zn, Ag, In, in ground
Indium, in 80% yield, in only 39% of the oredeposits Indium, 0.005% in sulfide, In
 resource in ground 2.17E-04 kg own assumptions 1 1.58 (4,2,2,1,4,4,12)
sulphidic ore In is extracted 0.003%, Pb, Zn, Ag, Cd, in ground
fresh and
processwater  surface water is chosen. resource in water Water, river 2.54E-03 m3 (Hilbrans, 1999) 1 1.53 (2,2,2,1,4,4,12)
(mining)
pit water  resource in water Water, well, in ground 3.09E-03 m3 (Hilbrans, 1999) 1 1.53 (2,2,2,1,4,4,12)
Electricity from
 Assumption: 25% electricity from coal hard coal power plants No US electricity, hard coal, at power plant 4.29E-02 kWh (Hilbrans, 1999) 1 1.58 (4,2,2,1,4,4,2)
coal
Electricity from electricity, natural gas, at power
 Assumption: 75% electricity from gas natural gas power plants No US 1.29E-01 kWh (Hilbrans, 1999) 1 1.58 (4,2,2,1,4,4,2)
gas plant
mining
diesel fuel construction

 machinery No GLO diesel, burned in building machine 5.10E-01 MJ (Hilbrans, 1999) 1 1.53 (2,2,2,1,4,4,4)
(energy) processes
stowing material, additional material, no excavation construction
 additives No CH sand, at mine 2.07E+00 kg (Hilbrans, 1999) 1 1.53 (2,2,2,1,4,4,4)
sand included materials
construction portland calcareous cement, at
cement  binder No CH 1.65E-01 kg (Hilbrans, 1999) 1 1.53 (2,2,2,1,4,4,4)
materials plant
construction
explosive  civil engineering No RER blasting 1.14E-02 kg (Hilbrans, 1999) 1 1.53 (2,2,2,1,4,4,4)
processes
transports, lorry, transportation of cement, explosives and transport calculations based

 road No RER transport, lorry >16t, fleet average 2.25E-01 tkm 1 2.11 (4,2,2,1,3,4,5)
mining sand 100km assumed. systems on (Hilbrans, 1999)
exploitation
exploitation infrastructure, 2000 m belts construction
infrastructrue,  machinery Yes RER conveyor belt, at plant 1.46E-07 m own assumptions 1 1.33 (4,2,2,1,3,4,4)
assumed processes
conveyor belts
infrastructure of
non-ferrous metal mine,
underground  including beneficiation facilities metals extraction Yes GLO 2.17E-10 unit own assumptions 1 3.32 (5,5,1,1,1,5,9)
underground
mining operation
overburden, waste
 residual material landfill No GLO disposal, sulfidic tailings, off-site 2.04E+00 kg (Hilbrans, 1999) 1 1.53 (2,2,2,1,4,4,6)
disposed management
 overburden, refilled not inventoried, no ecological burden 9.26E-01 kg (Hilbrans, 1999) 1 1.53 (2,2,2,1,4,4,6)
Detailed composition see Tab. 5.10, p.

 effluents, mining 30 (variability of the single emissions has 3.09E-03 m3 (Hilbrans, 1999) 1 1.53 (2,2,2,1,4,4,6)
with stdev95 of 10)
Fig. 6.3 Flows for "exploitation, zinc-lead deposit" (GLO) (part mining) and its representation in the ecoinvent database

in ecoinvent
Modul name
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
electricity,
 included in mining
benefication
fresh and
processwater  resource in water Water, well, in ground 6.00E-03 m3 (Hilbrans, 1999) 1 1.53 (2,2,2,1,4,4,12)
(benefication)
construction
lime  others No CH limestone, milled, packed, at plant 2.40E-03 kg (Krauss, 1999) 1 2 reported values
materials
(4,2,2,1,3,4,5);
transports, lorry, transportation of lime and chemicals, transport calculation based on
 road No RER transport, lorry >16t, fleet average 1.88E-03 tkm calculation 1 2.11
benefication 100km assumed. systems values for lime and
chemicals
infrastrukture
 included in infrastructure for mining
beneficiation
floating agents,
 Collector and frother chemicals organics No GLO chemicals organic, at plant 9.79E-04 kg (IPPC, 2002) 1 2 reported values
organic
floating agents,
 Depressant and activator chemicals inorganics No GLO chemicals inorganic, at plant 3.34E-03 kg (IPPC, 2002) 1 2 reported values
inorganic
beneficiation
Sodium cyanide  chemicals organics No RER hydrogen cyanide, at plant 1.52E-04 kg (IPPC, 2002) 1 2 reported values
burden-free, since byproduct from
Sulphuric Acid  1.62E-02 kg (IPPC, 2002) 1 2 reported values
metallurgy
not clear, wether net disposal or partly re-
waste (2,2,2,1,4,4,6); reported
 tailings, disposed filled in mine. Here assumed as net residual material landfill No GLO disposal, sulfidic tailings, off-site 2.21E+00 kg (Hilbrans, 1999) 1 1.53
management values
disposal
 tailings, refilled not inventoried, no ecological burden 1.13E+00 kg (Hilbrans, 1999) 1 1.53 (2,2,2,1,4,4,6)
Detailed composition see Tab. 6.9, p. 30
effluents, (2,2,2,1,4,4,6); reported
 (variability of the single emissions has 6.00E-03 m3 (Hilbrans, 1999) 1 1.53
benefication values
with stdev95 of 10)
(IPPC, 2002),
 Emission Cyanide From sodium cyanide, 80% control water river Cyanide 1.62E-05 kg 1 1.79 (2,2,2,3,4,4,33)
(Anonymous 1998)
0.53 kg carbondisulfide out of 1 kg

emission
 xanthate through hydrolytical air low population density Carbon disulfide 4.33E-04 kg (NPI, 2001) 1 2.26 (2,2,2,5,4,4,23)
carbondisulfide
decomposition
including emissions from mining
 Dust activities. Inventoried are the emissions 4.41E-04 kg (NPI, 2001) 1 1.59 (3,2,1,1,3,4,25)
given in Tab. 6.12, p 31
Concentrate contains 55% lead (0.21 kg (NPI, 2001),

 Lead concentrate Pb, yield = 95%) and 0.08% silver (0.30 metals extraction No GLO lead concentrate, at beneficiation 3.74E-01 kg (naweb, 2003),
g Ag, yield = 95%) cemip
Concentrate contains 53% zinc (0.33 kg

(NPI, 2001),
Zn, yield = 85%), 0.33% cadmium (2.07
 Zinc concentrate metals extraction No GLO zinc concentrate, at beneficiation 6.26E-01 kg (naweb, 2003),
g Cd, yield = 95%) and 0.01% indium
cemip
(0.07 g In, yield = 80%)
Fig. 6.4 Flows for "exploitation, zinc-lead deposit" (part beneficiation) and its representation in the ecoinvent database

6.2 Primary Lead Production Process and Infrastructure

There are two basic pyrometallurgical processes available for the production of lead from lead or
mixed lead-zinc-sulphide concentrates: sinter oxidation / blast furnace reduction route or Direct Smelt-
ing Reduction Processes.

Both processes may also be used for concentrates mixed with secondary raw materials. Both processes
are followed by a refining step to produce the final product with the required purity (Sutherland et al.,
1997; IPPC, 2001).
The sinter oxidation / blast furnace reduction involves two steps:
1) a sintering oxidative roast to remove sulphur with production of PbO; and,
2) blast furnace reduction of the sinter product.
The objective of sintering lead concentrates is to remove as much sulphur as possible from the galena
and the accompanying iron, zinc, and copper sulphides, while producing lump agglomerate with ap-
propriate properties for subsequent reduction in the blast furnace (a type of a shaft furnace).
As raw material feed, lead concentrates are blended with recycled sinter fines, secondary material and
other process materials and pelletised in rotating drums. Pellets are fed onto sinter machine and ig-
nited. The burning pellets are conveyed over a series of wind-boxes through which air is blown. Sul-
phur is oxidised to sulphur dioxide and the reaction generates enough heat to fuse and agglomerate the
pellets.
Sinter is charged to the blast furnace with metallurgical coke. Air and/or oxygen enriched air is in-
jected and reacts with the coke to produce carbon monoxide. This generates sufficient heat to melt the
charge. The gangue content of the furnace charge combines with the added fluxes or reagents to form
a slag.
For smelting bulk lead-zinc-concentrates and secondary material, frequently the Imperial Smelting
Furnace is used. Here, hot sinter and pre-heated coke as well as hot briquettes are charged. Hot air is
injected. The reduction of the metal oxides not only produces lead and slag but also zinc, which is
volatile at the furnace operating temperature and passes out of the ISF with the furnace off-gases. The
gases also contain some cadmium and lead. The furnace gases pass through a splash condenser in
which a shower of molten lead quenches them and the metals are absorbed into the liquid lead, the
zinc is refined by distillation.
The Direct Smelting Reduction Process does not carry out the sintering stage separately. Lead sul-
phide concentrates and secondary materials are charged directly to a furnace and are then melted and
oxidised. Sulphur dioxide is formed and is collected, cleaned and converted to sulphuric acid. Carbon
(coke or gas) and fluxing agents are added to the molten charge and lead oxide is reduced to lead, a
slag is formed. Some zinc and cadmium are “fumed” off in the furnace, their oxides are captured in the
abatement plant and recovered.
Several processes are used for direct smelting of lead concentrates and some secondary material to
produce crude lead and slag. Bath smelting processes are used: the ISA Smelt/Ausmelt furnaces
(sometimes in combination with blast furnaces), Kaldo (TBRC) and QSL integrated processes are used
in EU and Worldwide. The Kivcet integrated process is also used and is a flash smelting process. The
ISA Smelt/Ausmelt furnaces and the QSL take moist, pelletised feed and the Kaldo and Kivcet use
dried feed.
Refining: Lead bullion may contain varying amounts of copper, silver, bismuth, antimony, arsenic
and tin. Lead recovered from secondary sources may contain similar impurities, but generally anti-
mony and calcium dominate.

There are two methods of refining crude lead: electrolytic refining and pyrometallurgical refining.
Electrolytic refining uses anodes of de-copperised lead bullion and starter cathodes of pure lead. This
is a high-cost process and is used infrequently. A pyrometallurgical refinery consists of a series of ket-
tles, which are indirectly heated by oil or gas. Over a series of separation processes impurities and
metal values are separated from the lead bouillon.
6.2.2 Waste
The production of metals is related to the generation of several by-products, residues and wastes,
which are also listed in the European Waste Catalogue (Council Decision 94/3/EEC). The ISF or di-
rect smelting furnaces also are significant sources of solid slag. This slag has been subjected to high
temperatures and generally contains low levels of leachable metals, consequently it may be used in
construction. Solid residues also arise as the result of the treatment of liquid effluents. The main waste
stream is gypsum waste (CaSO4) and metal hydroxides that are produced at the wastewater neutralisa-
tion plant. These wastes are considered to be a cross-media effect of these treatment techniques but
many are recycled to pyrometallurgical process to recover the metals. Dust or sludge from the treat-
ment of gases are used as raw materials for the production of other metals such as Ge, Ga, In and As,
etc or can be returned to the smelter or into the leach circuit for the recovery of lead and zinc. Hg/Se
residues arise at the pre-treatment of mercury or selenium streams from the gas cleaning stage. This
solid waste stream amounts to approximately 40 - 120 t/y in a typical plant. Hg and Se can be recov-
ered from these residues depending on the market for these metals.
6.2.3 Emission
The main emissions to air from zinc and lead production are sulphur dioxide, other sulphur com-
pounds and acid mists; nitrogen oxides and other nitrogen compounds, metals and their compounds;
dust; VOC and dioxins. Other pollutants are considered to be of negligible importance for the industry,
partly because they are not present in the production process and partly because they are immediately
neutralised (e.g. chlorine) or occur in very low concentrations. Emissions are to a large extent bound
to dust (except cadmium, arsenic and mercury that can be present in the vapour phase).
Metals and their compounds and materials in suspension are the main pollutants emitted to water. The
metals concerned are Zn, Cd, Pb, Hg, Se, Cu, Ni, As, Co and Cr. Other significant substances are fluo-
rides, chlorides and sulphates. Wastewater from the gas cleaning of the smelter and fluid-bed roasting
stages are the most important sources.
6.2.4 Smelting, primary lead production in ecoinvent

No contiguous study on the lead production is yet available, therefore this module is based partly on
values reported in Ayres et al. (2002), IPPC (2001), Xiao et al. (2003), and EEA (1999a).
The module describes the primary production of lead in Europe. The multi-output-process "smelting,
primary lead production" delivers the co-products "lead, primary, at plant" and "parkes process crust,
from desilverising of lead". The flow "lead, …" is part of the respective European supply mix. The by-
product "parkes process crust, …" receives part of the burden and enters the silver production chain.
The modules including general flow information, remarks, sources, values, and uncertainty informa-
tion are shown in Fig. 6.5 (inputs) and Fig. 6.6 (outputs). The following sections describe calculations,
sources and assumptions chosen in this study. The meta-information for this dataset is compiled in
Tab. 6.2.
The smelting process is modelled as a combination of the two predominant processes
 Sinter oxidation / blast furnace reduction (44% of world production, Tab. 6.13), and the newer

 Direct smelting process (QSL, Kivcet, etc., 56% of world production, Tab. 6.14)
The inputted raw materials and the emission factors are combined according to the production share in
order to for the inventory in ecoinvent (Fig. 6.5 and Fig. 6.6). No quantitative compilation of electric-
ity production is available, therefore the UCTE production mix was chosen as approximation (year
2000: nuclear 15%; hydro 4%; fossil 81% 11 ).
The two predominant processes differ also in emission factors per tonne lead produced. Where the
older sinter oxidation / blast furnace processes exhibit limited abatement possibilities, higher emis-
sions occur from these (Tab. 6.15). The more recent direct smelting processes in contrast have im-
proved emission control (Tab. 6.16). The combination of the two process technologies combined ac-
cording to their production shares gives the emission factors used in this inventory (Tab. 6.17). Some
of the emissions generally reported in literature and stated in the previous section – Ge, Ga, In, Se,
acid mists, VOC (air); fluorides, chlorides and sulphates (water) – are not covered by this inventory
because no emission factors could be found. The dioxin emissions are approximated with the reported
emission factor of 50 μg-TEQ / t secondary zinc production in EC (2002).
Tab. 6.13 Lead production: Sinter oxidation / blast furnace reduction. Input / output and energy demand reported from
various sources. ISF = Imperial smelting furnace. 44 % of the lead production, emission factors in this table
are combined with 56 % direct smelting in Tab. 6.14
Source Ayres et al. IPPC IPPC (2001) IPPC (2001) This study
(2002) (2001)
Type Unknown ISF 4) Shaft Shaft furn.
furn.(min) (max)
Inputs
Sec. materials kg 926
Concentrate kg 3312 926
Total raw material kg 3312 1852 1855 1)
Fluxes kg 2691 n. a. n. a. n. a. 2691 1)
Lime kg 92 n. a. n. a. n. a. 92 1)
Coking Coal kg 667 741 150 225 446 2)
Natural Gas MJ 1815 2541 1089
Electricity kWh 180 300 120
Iron kg n. a. n. a. 120 170 145
Sulphur kg 3 n. a. n. a. n. a. 3
Air (Oxygen source) kg 12334 n. a. n. a. n. a. 12334
Outputs
Slag kg 3661 519 519 3)
Sulfuric acid kg 2021 1296 2021
Lead, refined kg 1000 259 1000 1000 1000
Lead+Zinc kg 1000
1)
Corresponds to the raw material composition given by preceding inventory “mining, zinc-lead deposits”.
2)
Arithmetic mean, processes assumed to be evenly partitioned.
3)
value of IPPC (2001) taken, assuming that this figure represents internal recycling in contrary to the total
value calculated in Ayres et al. (2002)
4)
Values normalised to the sum of 1 t metal (lead and zinc) produced.
n. a. = no values were reported, but not zero.
11
http://www.ucte.org/

Tab. 6.14 Lead production: Direct smelting. Input / output and energy demand per tonne lead reported in IPPC (2001).
56 % of the lead production, emission factors are combined with 44 % blast furnace / ISF in Tab. 6.13
Type QSL-1 QSL-2 Kivcet TBRC ISA This study

Inputs
Raw material kg n. a 1530 n. a n. a 1356 1855 1)
Fluxes kg n. a 235 n. a n. a 39 137 2)
Coal kg 100 141 105 4 79 77 2)
Natural Gas MJ 726 695 0 0 0 178 2)
Fuel Oil MJ 0 0 1030 1236 0 567 2)
Electricity kWh 0 0 250 500 0 188 2)
Oxygen kg 471 541 642 200 148 374 2)
Nitrogen kg n.a. 147 0 0 0 37 2)
Outputs
Slag kg n. a 589 n. a n. a 111 350 2)
Sulphuric acid kg n. a 706 n. a n. a 278 492 2)
Lead, bullion kg 1059 3)
Lead, refined kg 1000 1000 1000 1000 1000 1000
1)
Corresponds to the raw material composition given by proceeding inventory “mining, zinc-lead deposits”.
2)
Arithmetic mean of the two QSL-processes (one mean value) and the other three processes
3)
Only the output of lead bullion was reported, which was normalised to the amount of refined lead, assum-
ing a lead concentration in the bullion of 94.5% and a refining yield of 100%.
Tab. 6.15 Lead: Emissions to air per tonne lead produced – limited control. Values are collected from various
sources; they correspond to the conventional sinter-oxidation / blast furnace reduction process.
Source Reported emis- Proposed emis- Reported emis- This study

sions in IPPC sion factor in sion factor in
(2001), tab. 5.17 EEA (1999a) EEA (1999a) 1)
Sulphur dioxide kg / t 73.2 73.2
Lead g/t 256.31 800 573 573
Cadmium g/t 5.91 10 6.67 6.67
Antimony g/t 4.76 4.76
Arsenic g/t 2.26 3 8.87 8.87
Mercury g/t 0.177 3 2) 0.177
1)
Geometric mean of a range of values reported in Poland and the value of a plant in Sweden.
2)
Value based on one single value of unknown abatement efficiency, therefore value of IPPC (2001) chosen.

Tab. 6.16 Lead: Emissions to air per tonne lead produced – improved control. Values are collected from various
sources. They correspond to the direct smelting process.
Source im- Reported emissions in Proposed Reported This study

proved abate- IPPC (2001), tab. 5.17 emission fac- emission fac-
ment tor in EEA tor in EEA
(1999a) (1999a) 1)
Sulphur dioxide kg / t 1.61 1.61
Lead g/t 16.7 200 200 1081)
Cadmium g/t 0.0466 1 0.6 0.6
Antimony g/t 0.317 0.317
Arsenic g/t 0.0642 0.3 0.3
Mercury g/t 0.0100 3 2) 0.01
1)
Mean of considered processes in IPPC and EAA.
2)
Value based on one single value of unknown abatement efficiency, therefore value of IPPC (2001) chosen.
Tab. 6.17 Emission factors per tonne lead produced used in this inventory. Combination of limited and improved
emission control according to the technologies and their production share (see Tab. 6.15 and Tab. 6.16).
Limited control Improved control This Study

Sinter oxidation / Direct smelting Mix
blast furnace
Share world production 44% 56% mix
Sulphur dioxide kg / t 73.2 1.61 33
Lead g/t 573 60 312
Cadmium g/t 6.67 0.60 3.26
Antimony g/t 4.76 0.317 2.26
Arsenic g/t 8.87 0.300 4.06
Mercury g/t 0.177 0.010 0.083
Dioxin μg / t 50 50 50
Tab. 6.18 Emission to air: solid particles. Values taken from CEPMEIP-Database (SNAP 030304). Values might be
overestimated considering that TSP from fuels burned in the smelting process are included in the shown
figures. Additionally, TSP is accounted for in the respective modules for burning of fuels whose require-
ments are in Tab. 6.13 and Tab. 6.14.
Solid Particle Emission Unit Medium / high emission Low emission

(sinter / blast furnace) (direct smelting)
TSP g/t 600 120
PM10 g/t 500 114
PM2.5 g/t 300 60
Differential classes used in ecoinvent
PM<2.5 g/t 300 60
PM2.5-10 g/t 200 54
PM>10 g/t 100 6

Tab. 6.19 Emissions to water. Due to lack of data the values for a Chinese zinc smelter reported in Xiao et al. (2003)
are used.
Emission to water Unit Value (unspecific)

Hg g/t 0.043
Cd g/t 2.6
As g/t 1
Pb g/t 18
Cu g/t 5
COD g/t 583
6.2.5 By-products and allocation

Argentiferous lead and lead-zinc ores, which often occur in association, are distributed widely. In al-
most all silver-producing countries, they constitute an important fraction of the silver produced by
mining and account for ca. 50 % of world silver production (Renner et al., 2002).
Mine production remains by far the largest contribution to silver supply, normally accounting for
around two-thirds of the total. Beside mine production, post consumer scrap, disinvestment, govern-
ment sales and producer hedging are sources for supply. Old scrap normally makes up a little over a
fifth of the supply (Klapwijk et al., 2006).
The ratio between the world supply of silver and the world mine production of lead amounted in the
last twenty years between four to six kilograms silver per tonne lead mined (s. Tab. 3.1). This leads to
the assumption that in a world average per tonne lead mined 2.3 kg silver is jointly extracted and re-
fined.
By the so-called Parkes process a silver containing crust arises where silver is accumulated and sube-
quently refined further (Renner et al., 2002). A metallurgical process Ag yield of 97.4% is assumed
(Reuter, 2005). For more details on the process, see the respective chapter on Silver in this report.
However, the allocation of the subprocesses lead- and silver production is not evident to solve, since
only the final commodity prices are known but no in-firm prices for the curst and for the raw lead bul-
lion, on which the allocation would rely. Also the share of the products in total costs – gross sales
method as proposed in Guinée et al. (2004) and Guinée et al. (2001) – is not easy to estimate. The by-
product allocation problem – the joint production of silver and lead – is tackled ere by a subdivision of
the sub-processes (according to Ekvall & Finnveden, 2001). The silver production from the parkes
process crust is treated in a distinct inventory. Hence the burdens have to be allocated between the lead
and the silver containing crust.
As starting point for an estimation it is assumed that the profit margin for the production of silver cor-
responds to the company’s performance. Taking an arbitrary performance value of 10% - which re-
flects the general profit expectations of major enterprises – the value of the crust corresponds conse-
quently to 10% of the proceeds from the silver sales. Therefore 10 % of the silver price is taken to es-
timate the value of the silver containing Parkes process crust.
The composition and the corresponding values contained in the by-products result in a split of 4% to
the silver and 96% to the lead (Tab. 6.20).
However, the resources (elementary flows) have to be allocated by mass to the respective by-products,
in order to keep up with the resource balance in the final commodity. Since the concentrate required
by this multi-output-process is allocated also by revenue, the therein contained resources are not cor-
rectly attributed and have to be corrected with a corrective exchange flow (Tab. 6.21).

Tab. 6.20 By-products of primary lead production: Composition, flows and value to derive the revenue-based alloca-
tion factors (Values are the year averages from 2004 through 2006, reported by the U.S. Geological Survey)
Lead / Silver Process lead con- parkes lead Value in

bearing concen- Yield centrate process $ / kg
trate crust
composition input output output
Pb 98.0% 55% 39% 100% 1.15
Ag 97.4% 0.08% 18% 0% 27
Amount kg 1.86 0.0081 1
Value contained in $ $ 0.04 1.15
Allocation by Value 3% 97%
Tab. 6.21 Individual correction factors to of the respective by-products in the lead production.
correction from economic to per mass according to economic allocation

Resources demand in the feed Lead Parkes process crust
Concentrate kg 1.86 97% 3%
Pb kg 1.074 1.039 0.0354
Ag kg 0.00157 0.00152 0.000052
Resources attributed in by-product according to per mass allocation
Pb 1.074
Ag 0.00157
Resource exchange flow difference to be corrected
Pb 0.03536 -0.03536
Ag -0.001515 0.001515
6.2.6 Data quality

Data uncertainty is derived using the Pedigree matrix as described in Frischknecht et al. (2004). Some
of the emissions that are described in the literature as typical for lead, could not be covered in this in-
ventory due to missing emission factors, namely Ge, Ga In and Se into air and , Zn, Se, Cu, Ni, Co and
Cr to water. Also salts – fluorides, chlorides and sulphates – are not included due to missing data. Fur-
ther investigations on these issues have been done to improve the quality of this dataset. However, for
the purpose of a background database the quality is sufficient. The choice of electricity mix bases on
assumptions that have to be improved with further investigations.

in ecoinvent
Modul name
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
Concentrate contains 55% lead
(1.02 kg Pb, yield = 98%) and lead concentrate, at
Concentrate  metals extraction No GLO 1.86E+00 kg Calculation 1 1.33 (3,3,2,1,3,5,4)
0.08% silver (1.49 g Ag, yield = beneficiation
97%)
smelting, primary lead production (inputs)
Pb resource in concentrate fully
Correction Pb in resource correction, PbZn, lead,
 allocated to the Pb bearing by- metals extraction No GLO 3.54E-02 kg calculation 1 1.33 (3,3,2,1,3,5,4)
Conc -> Pb positive
product. Correction Lead
Pb resource in concentrate fully
Correction Pb in resource correction, PbZn, lead,
 allocated to the Pb bearing by- metals extraction No GLO 3.54E-02 kg calculation 1 1.33 (3,3,2,1,3,5,4)
Conc -> Ag negative
product. Correction Silver.
Ag resource in concentrate fully
Correction Ag in resource correction, PbZn,
 allocated to the Ag bearing by- metals extraction No GLO 1.52E-03 kg calculation 1 1.33 (3,3,2,1,3,5,4)
Conc -> Pb silver, negative
product. Correction Lead
Ag resource in concentrate fully
Correction Ag in resource correction, PbZn,
 allocated to the Ag bearing by- metals extraction No GLO 1.52E-03 kg calculation 1 1.33 (3,3,2,1,3,5,4)
Conc -> Ag silver, positive
product. Correction Silver
Lime (total)  lime + 50% of "Fluxes" others No CH 6.69E-01 kg (Ayres, 2002) 1 1.25 (3,3,2,1,3,3,4)
materials plant
construction
silica sand  50% of "Fluxes" additives No DE silica sand, at plant 6.28E-01 kg (Ayres, 2002) 1 1.25 (3,3,2,1,3,3,4)
materials
Electricity from UCTE mix: nuclear 15%; hydro 4%; electricity, medium voltage,
 electricity production mix No UCTE 1.58E-01 kWh (IPPC, 2002) 1 1.33 (3,3,2,1,3,5,2)
Grid fossil 81% production UCTE, at grid
natural gas, burned in industrial
Natural Gas  natural gas heating systems No RER 5.77E-01 MJ (IPPC, 2002) 1 1.25 (3,3,2,1,3,3,1)
furnace >100kW
Sulfur  not inventoried, negligible 1.32E-03 kg (Ayres, 2002) 1 1.25 (3,3,2,1,3,3,4)
Oxygen  chemicals inorganics No RER oxygen, liquid, at plant 2.10E-01 kg (IPPC, 2002) 1 1.25 (3,3,2,1,3,3,4)
Nitrogen  chemicals inorganics No RER nitrogen, liquid, at plant 2.07E-02 kg (IPPC, 2002) 1 1.25 (3,3,2,1,3,3,2)
Iron oxide minerals added, iron ore, 65% Fe, at

iron  metals extraction No GLO 6.36E-02 kg (IPPC, 2002) 1 1.25 (3,3,2,1,3,3,2)
inventoried as iron ore concentrate beneficiation

Fuel Oil  oil heating systems No RER industrial furnace 1MW, non- 2.84E-01 MJ (IPPC, 2002) 1 1.25 (3,3,2,1,3,3,1)
modulating
28.6 MJ energy contained per kg hard coal, burned in industrial
Coal  hard coal heating systems No RER 6.97E+00 MJ (IPPC, 2002) 1 1.25 (3,3,2,1,3,3,2)
coal furnace 1-10MW
Assumption - 600km : concentrate,

transport
Transport by train  coal; 100km : lime, silica, sulphur, train No RER transport, freight, rail 5.46E+00 tkm standard distances 1 1.31 (4,3,1,1,3,3,2)
systems
oxygen, nitrogen, iron

Transport by lorry transport transport, lorry >16t, fleet
 coal; 50km : lime, silica, sulphur, road No RER 9.62E-01 tkm standard distances 1 1.31 (4,3,1,1,3,3,2)
(32t) systems average
oxygen, nitrogen, iron
Fig. 6.5 Flows for "Smelting, primary lead production" (inputs) and its representation in the ecoinvent database

Modul name
in ecoinvent
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
Proxy for disposal of lead smelter waste

smelting, primary lead production (outputs)
Slag residual material landfill No CH 0% water, to residual material 4.24E-01 kg (IPPC, 2002) 1 1.25 (3,3,2,1,3,3,6)
slag. management
landfill
used, e.g. in beneficiation --> not
 Sulphuric Acid 1.16E+00 kg (IPPC, 2002) 1 1.25 (3,3,2,1,3,3,4)
inventoried
(1,3,1,3,3,4,35);
 Cu water river Copper, ion 5.00E-06 kg (Xiao, 2003) 1 5.07 values from Zn-
production
(1,3,1,3,3,4,35);
 Pb water river Lead 1.80E-05 kg (Xiao, 2003) 1 5.07 values from Zn-
production
(1,3,1,3,3,4,35);
 As water river Arsenic, ion 1.00E-06 kg (Xiao, 2003) 1 5.07 values from Zn-
production
(1,3,1,3,3,4,35);
 Hg water river Mercury 4.30E-08 kg (Xiao, 2003) 1 5.07 values from Zn-
production
(1,3,1,3,3,4,35);
 Cd water river Cadmium, ion 2.60E-06 kg (Xiao, 2003) 1 5.07 values from Zn-
production
(1,3,1,3,3,4,32);
 COD water river 5.83E-04 kg (Xiao, 2003) 1 1.58 values from Zn-
Demand
production
(1,3,1,3,3,4,32);
BOD5, Biological Oxygen calculated, based
 BSB5 water river 3.89E-04 kg 1 1.58 values from Zn-
Demand on (Xiao, 2003)
production
(1,3,1,3,3,4,32);
calculated, based
 TOC water river TOC, Total Organic Carbon 2.28E-04 kg 1 1.58 values from Zn-
on (Xiao, 2003)
production
(1,3,1,3,3,4,32);
calculated, based
 DOC water river DOC, Dissolved Organic Carbon 2.28E-04 kg 1 1.58 values from Zn-
on (Xiao, 2003)
production
 Lead air low population density Lead 3.12E-04 kg (IPPC, 2002) 1 5.04 (3,4,2,1,1,4,22)
 Antimony air low population density Antimony 2.26E-06 kg (IPPC, 2002) 1 5.04 (3,4,2,1,1,4,22)
 Mercury air low population density Mercury 8.31E-08 kg (IPPC, 2002) 1 5.04 (3,4,2,1,1,4,22)
 Arsenic air low population density Arsenic 4.06E-06 kg (IPPC, 2002) 1 5.04 (3,4,2,1,1,4,22)
 PM<2.5 air low population density Particulates, < 2.5 um 1.65E-04 kg (IPPC, 2002) 1 3.04 (3,4,2,1,1,4,27)
Particulates, > 2.5 um, and <
 PM2.5-10 air low population density 1.18E-04 kg (IPPC, 2002) 1 1.28 (3,4,2,1,1,4,28)
10um
 PM>10 air low population density Particulates, > 10 um 4.72E-05 kg (IPPC, 2002) 1 1.46 (3,4,2,1,1,4,29)
 Cadmium air low population density Cadmium 3.26E-06 kg (IPPC, 2002) 1 1.55 (3,4,2,1,1,4,16)
Dioxins, measured as 2,3,7,8- European Dioxin
 TCDD air low population density 5.00E-11 kg 1 10 reported
tetrachlorodibenzo-p-dioxin Inventory
Carbon dioxide
 0.44 kg CO2 / kg lime air unspecified Carbon dioxide, fossil 2.94E-01 kg (IPPC, 2002) 1 1.58 (4,2,2,3,4,4,14)
from lime addition
 Sulphur Dioxide air low population density Sulfur dioxide 3.30E-02 kg (IPPC, 2002) 1 1.19 (3,4,2,1,1,4,15)
Parkes Process parkes process crust, from

 Contains 18% Ag metals extraction No GLO 8.05E-03 kg
crust desilverising of lead
 Refined lead metals extraction No GLO lead, primary, at plant 1.00E+00 kg
Fig. 6.6 Flows for "Smelting, primary lead production " (outputs) and its representation in the ecoinvent database

6.3 Secondary Lead Production process and infrastructure

Secondary lead originates to a predominant share from spent lead acid (PbA) batteries. Gabby (2004)
estimates 92% for this source in the U.S. in 2004. Lead acid batteries’ recycling is very similar to the
primary lead production process. The main differences are in material preparation before reduction,
which affects plant size, since there is no need for sintering (Espinosa et al., 2004).
The processing of this sort of scrap can be done by different technologies. The classical way makes
use of short or long rotary furnace (Various Authors, 1996). More recently the convergence of primary
and secondary producer has led to hybrid systems that process simultaneously both lead ore and metal-
lic post consumer scrap lead. One example is the QSL process at the Bleihütte Binsfelderhammer in
Stolberg, Germany (Berzelius, 1993).
6.3.1 Lead, secondary, at plant in ecoinvent

This dataset describes the recovery of secondary lead from spent lead acid batteries as done at
Campine, Belgium (Campine NV, 2006). The company uses a shaft furnace with post combustion
(Quirijnen, 1999). The compilation of the unit process LCI as applied in ecoinvent is displayed in Tab.
6.22. The meta-information is reported in Tab. 6.2, at the beginning of this chapter.
Typically this technology produces 5000 t / a sulphuric acid (15% concentration), 25’000 t lead bul-
lion (98% Pb), 1200 t / a slags (1% Pb) and 3000 t / a raw lead matte (10% Pb) to be shipped to pri-
mary smelters. Overall Pb yield is typically 98.8% at the plant level and 99.8% after reworking the
matte (Quirijnen, 1999).
The operation treats junk batteries and plates but also lead cable sheathing, drosses and sludges, leaded
glass and balancing weights. From this feed it manufactures mainly antimonial lead up to 10% Sb, cal-
cium-aluminium lead alloys with or without tin and soft lead with low and high copper content. All
these products are the result of a refining and alloying step to meet the compliance with the designa-
tions desired. For the purpose of this LCI-module all these alloys are subsumed under the reference
flow “lead, secondary, at plant”. The company deliveries amounted in 2004 and 2005 around 40’000 t
lead per year, which represents ca. 0.8% of the world secondary lead production (estimated 5’000’000
t in 2005 (USGS, 2006b)).
The inventory data for the process has been adopted from the Battery LCA commissioned by the UK
Department for Environment Food and Rural Affairs (Defra), prepared by Fisher et al. (2006). It
represents data for Campine Ltd from 2004 (Tab. 6.22).
The values for the energy carriers have been converted in effective heat delivered with the respective
energy content and energy efficiency as given in ecoinvent Centre (2005) and Frischknecht et al.
(2004). The datasets consider the system specific combustion emissions, therefore the originally re-
ported CO2 emissions – mainly due to the combustion – are omitted for LCI calculation. It is not clear
whether the emissions from SO2 emissions are due to the combustion. Assuming a heavy fuel burner
in Jungbluth (2003) with quite high SO2 emissions (0.005 kg SO2 per kg CO2 emitted in off-gas,), only
35 % of the reported SO2 emissions may be explained. Therefore it is assumed, that the SO2 emissions
originate from a process other than combustion and is counted for as indicated in Fisher et al. (2006).
The infrastructure has been approximated with the generic module for non-ferrous metal operations
with 2'650'000 t/a output and 50 a life time. Transportation of the raw material is considered in the
dataset “iron scrap, at plant” that is used as approximation for collection and preparation of the battery
scrap. The rest of the auxiliary material, energy carrier, as well as the disposal of the slag is assumed
to occur by lorry over 100 km distance.

Tab. 6.22 Unit process data for “lead, secondary, at plant” as given in Fisher et al. (2006).
Flow Quantity Unit Comments / Inventory used

INPUTS
Lead acid batteries (incl acid) 1000 kg Approximated with the dataset for collection
and preparation of iron scrap.
Limestone 5.8 kg Limestone, milled.
Iron scrap 4.0 kg Iron scrap.
NaOH 350 kg 50% NaOH.
Sodium nitrate 0.4 kg Generic inorganic chemicals.
Sulphur 0.9 kg Secondary Sulphur.
Iron chloride 0.9 kg Iron (III) chloride (30%).
Slag 150 kg Reused from process
Electricity 35.2 kWh Grid Electricity, Medium Voltage.
Natural gas 16.2 kg 45.4 MJ per kg energy content, energy effi-
ciency 95%.
Coke 20.0 kg Hard coal coke with 28.6 MJ per kg energy
content, energy efficiency 70%.
Process water 770 kg Treated rainwater, reused through process.
OUTPUTS
Flue dust for internal reuse 13.6 kg Reused in process
Return slag for internal reuse 150 kg Reused in process
Sulphuric acid for internal reuse 71.0 kg Sulphuric acid. ecoinvent, Europe,
SO2 7.1 kg -
CO2 (fuel combustion) 500 kg Contained in dataset for gas combustion.
Pb 0.00127 kg -
Sb 0.0000056 kg -
Excess slag to landfill 44.0 kg Disposal to landfill.
Lead to processor 650 kg Reference Flow
6.3.2 Data quality

Data uncertainty is derived using the Pedigree matrix as described in Frischknecht et al. (2004). As
there is one facility that has to stand for a variety of technologies applied, the reliability and correct-
ness of the data set is limited and may not be applicable to life cycle assessment studies of products
where the production of secondary lead becomes important.

in ecoinvent
Modul name
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
Approximated with the dataset for
Lead acid
 collection and preparation of iron metals extraction No RER iron scrap, at plant 1.54E+00 kg Fisher et al (2006) 2 0.08 normal dist +/- 5%
batteries
scrap.

Limestone  others No CH 8.92E-03 kg Fisher et al (2006) 1 1.13 (1,4,1,3,1,3,3)
materials plant
Iron scrap  metals extraction No RER iron scrap, at plant 6.15E-03 kg Fisher et al (2006) 1 1.13 (1,4,1,3,1,3,4)
lead, secondary, at plant (input)
sodium hydroxide, 50% in H2O,

NaOH  Corresponds to 0.538 kg NaOH chemicals inorganics No RER 2.69E-01 kg Fisher et al (2006) 1 1.13 (1,4,1,3,1,3,4)
production mix, at plant
sodium sulphate, powder,

Sodium nitrate  chemicals inorganics No RER 6.15E-04 kg Fisher et al (2006) 1 1.13 (1,4,1,3,1,3,4)
Sulphur  chemicals inorganics No RER secondary sulphur, at refinery 1.38E-03 kg Fisher et al (2006) 1 1.13 (1,4,1,3,1,3,4)
Corresponds to 1.4E-3 kg iron iron (III) chloride, 40% in H2O,

Iron chloride  chemicals inorganics No CH 3.20E-04 kg Fisher et al (2006) 1 1.13 (1,4,1,3,1,3,4)
chloride at plant
Reused from process, not

Slag  2.31E-01 kg Fisher et al (2006) 1
inventoried.

Electricity  electricity production mix No UCTE 5.42E-02 kWh Fisher et al (2006) 1 1.13 (1,4,1,3,1,3,2)
45.4 MJ per kg energy content, heat, natural gas, at industrial

Natural gas  natural gas heating systems No RER 1.08E+00 MJ Fisher et al (2006) 1 1.13 (1,4,1,3,1,3,1)
energy efficiency 95% furnace >100kW
Hard coal coke with 28.6 MJ per kg

heat, hard coal coke, at stove 5-
Coke  energy content, energy efficiency hard coal heating systems No RER 6.15E-01 MJ Fisher et al (2006) 1 1.13 (1,4,1,3,1,3,1)
15kW
70%
Treated rainwater, reused through

Process water  1.18E+00 kg Fisher et al (2006)
process, not inventoried.
Transport of the auxiliary materials

Transport transport transport, lorry >16t, fleet
 and disposed waste (total 0.61 kg) road No RER 6.13E-01 tkm Calculations 1 2.16 (4,4,1,3,3,5,5)
auxiliaries systems average
over a distance of 1 km.
Approximated with generic module

Infrastructure of for non-ferrous metal operations
 chemicals inorganics No RER sulphuric acid, liquid, at plant 7.55E-12 kg Assumptions 1 3.15 (4,4,1,3,3,5,9)
facilities. with 2'650'000 t output and 50 a life
time.
Excess slag to Proxy for disposal of lead smelter waste residual material
 No CH 0% water, to residual material 6.77E-02 kg Fisher et al (2006) 1 1.4 (4,4,1,3,3,5,6)
landfill slag. management landfill
landfill
Fig. 6.7 Unit process inventories of the data set “Lead, secondary, at plant (inputs)”.

Modul name
in ecoinvent
mean value
Mean value
Source for
Comment
Deviation
Category
structure
Standard
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
 waste heat Calculated from electricity use. air high population density No RER Heat, waste 1.95E-01 MJ Calculations 1 1.13 (1,4,1,3,1,3,13)
lead, secondary, at plant
Flue dust for

 not inventoried 2.09E-02 kg Fisher et al (2006)
internal reuse
Return slag for

internal reuse
(outputs)
Sulphuric acid for

internal reuse
 SO2 air high population density No RER Sulfur dioxide 1.09E-02 kg Fisher et al (2006) 1 1.52 (1,4,1,3,1,3,31)
Already ontained in datasets for

CO2 (fuel
 gas and coke combustion, 7.69E-01 kg Fisher et al (2006)
combustion)
hence not inventoried.
 Pb air high population density Lead 1.95E-06 kg Fisher et al (2006) 1 1.52 (1,4,1,3,1,3,31)
 Sb air high population density Antimony 8.62E-09 kg Fisher et al (2006) 1 1.52 (1,4,1,3,1,3,31)
 Lead to processor metals extraction No RER lead, secondary, at plant 1.00E+00 kg Fisher et al (2006) 1
Fig. 6.7 (Continued) Unit process inventories of the data set “Lead, secondary, at plant (outputs)”.

6.4 Lead, at regional storage in ecoinvent

This dataset describes the supply-mix of lead used in Europe. Following the recycled content approach
it consists of a blend of the datasets for primary and secondary production, complemented with the
transport service required to transport the material to the regional storage. The meta-information is
compiled on Tab. 6.2 at the beginning of this chapter.
As indicated in Section 2.2 Lead as secondary resource the primary production of lead has seen a de-
cline in recent years due to more stringent regulations in the use of lead. Since no actual (2007) figures
are available, the value of 58% secondary lead in 2001 is taken as starting point. CHR Metals assumed
in 2001 the decline in global lead mine production with a small increase thereafter and a respectively
higher secondary production that satisfies the gap in supply (Roberts, 2003). The mine production data
shows such a small decline in 2002 and a slow increase after that year. On the other hand the reported
metal production increased faster, indicating that – under the assumption that no considerable stocks
are being consumed, viz. steady state situation – also the secondary production has risen moderately.
This reasoning leads to an estimated 65% secondary lead in the global supply mix, compared to 70 –
80% as reported for the U.S. in the years from 2003 – 2006 (USGS, 2006b). A value of 75% is
adopted for the European dataset (country-code RER).
Average transport distances have been assumed to depict the European situation (Tab. 6.23).
Tab. 6.23 Transport of primary and secondary lead to Europe. Assumptions.
Primary lead Secondary lead

Lorry km 100 100
Train km 200 200
Ship km 5000 0

Modul name
in ecoinvent
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
A global share of 25%
Primary Lead  primary production metals extraction No GLO lead, primary, at plant 2.50E-01 kg Roberts (2003) 4 Uniform dist
lead, at regional
assumed.
A global share of 75%
Secondary Lead  secondary production metals extraction No RER lead, secondary, at plant 7.50E-01 kg Roberts (2003) 4 Uniform dist
assumed.
storage
Assumed distance: 100

transport
Transport lorry  km for primary, 100 km road No RER transport, lorry >16t, fleet average 1.00E-01 tkm Assumptions 1 2.32 (5,5,1,1,1,5,5)
systems
for secondary
Assumed distance: 200
transport
Transport Train  km for primary, 200 km train No RER transport, freight, rail 2.00E-01 tkm Assumptions 1 2.32 (5,5,1,1,1,5,5)
systems
for secondary
Assumed distance:
transport
Transport Ship  5000 km for primary, 0 ship No OCE transport, transoceanic freight ship 1.25E+00 tkm Assumptions 1 2.32 (5,5,1,1,1,5,5)
systems
km for secondary
lead, at regional
 metals extraction No RER lead, at regional storage 1.00E+00 kg Assumptions 1
storage
Fig. 6.8 Unit process inventories of the data set Lead, at regional storage.

6.5 Primary Zinc Production Process

Zinc can be produced from primary raw materials by pyrometallurgical or hydrometallurgical methods
(Fig. 6.9). Pyrometallurgical methods are used in other parts of the world but have gradually lost their
importance and are not used in EU for simple zinc concentrates. Determining factors for this tendency
are the need for an extra distillation stage to obtain high-grade zinc and the relatively low zinc extrac-
tion efficiency. The pyrometallurgical Imperial Smelting Furnace process (ISF) is however still of im-
portance in EU because it enables complex lead-zinc concentrates and secondary material to be treated
simultaneously, yielding saleable lead and zinc. It can also consume residues from other processes
(IPPC, 2001).
Fig. 6.9 General scheme of zinc production (Dove & Boustead, 1998). The pyrometallurgical and hydrometallurgical
production processes account for 20% and 80% of the world production, respectively (IPPC, 2001).

The pyrometallurgical route is used for feed of mixed zinc/lead concentrates and secondary material
and uses the Imperial Smelting Furnace (ISF), a special form of a blast furnace. The electrothermic
furnace is used in other parts of the world. Zinc produced in the Imperial Smelting Furnace may con-
tain varying amounts of cadmium, lead, copper, arsenic, antimony and iron. The process uses a refin-
ing stage. Zinc from the ISF is refined by reflux distillation in columns containing a large number of
refractory trays (New Jersey Distillation). The lower ends of the columns are heated externally by
natural gas. The upper ends are not heated and run cool enough to reflux the higher boiling point met-
als before vapours pass to a condenser. The New Jersey distillation column is also used for secondary
zinc materials. Distillation proceeds in two stages; first the separation of zinc and cadmium from lead
and then separation of cadmium from zinc. In the first stage, molten zinc is fed into a column where all
the cadmium and a high proportion of the zinc are distilled. The mixture is condensed and fed directly
to a second column. This column is operated at a slightly lower temperature to distil mainly cadmium,
which is condensed as a zinc-cadmium alloy. The alloy is transferred to a cadmium refinery. The

metal run-off from the bottom of the second column is high-grade zinc (SHG) of 99.995% purity. The
run-off metal from the first stage is zinc with lead, tin, arsenic, iron, antimony and copper impurities.
This alloy is cooled to separate lead, which is recycled to the ISF splash condenser and an inter-
metallic compound of iron, zinc and arsenic, which is recycled to the ISF itself. The zinc is then
treated with sodium to remove residual arsenic and antimony as sodium arsenides and antimonides,
respectively, which are also recycled to the ISF. The zinc produced in this way is of a lower grade
(GOB), but free of cadmium, and is used mainly for galvanising (IPPC, 2001).
Fig. 6.10 Process scheme, part of the pyrometallurgical process route (rectification). The residues of the first stage
distillation are processed to GOB Zinc (lowest quality, “Good Ordinary Brand”), which is used for galvanis-
ing, where as the SHG Zinc (highest quality, “Special High Grade”) is won by a second distillation (IPPC,
2001)
The hydrometallurgical route is used for zinc sulphide, oxide, carbonate or silicate concentrates, mak-
ing about 80% of the total world output. The majority of the EU production facilities use the hydro-
metallurgical process, with a total production capacity of 1’665’000 t/a in 1997. Sulphide concentrates
are roasted first in fluidised bed roasters to produce zinc oxide and sulphur dioxide. Roasting is an
exothermic process and no additional fuel is used to sustain the reaction, the heat generated is recov-
ered. The zinc oxide (calcine) passes from the furnace and is collected and cooled. Roaster gases are
treated in hot electric precipitators to remove dust (which is passed to the calcine). Other dust and
volatile metals such as Hg and Se are removed in a gas cleaning train that incorporates scrubbing sys-
tems and wet EPs. The sulphur dioxide is then converted to sulphuric acid in a conventional recovery
system. Leaching of the calcine is carried out in a number of successive stages using a gradually in-
creasing strength of hot sulphuric acid. The initial stages do not dissolve significant amounts of iron
but the later ones do. The leaching process is carried out in a variety of reactors using open tanks.
Leaching may be stopped after the neutral leach. The leach residue is sent to an ISF and added to the
sinter feed. Zinc, lead and silver are recovered as metals, sulphur as H2SO4. Instead of an ISF, a Waelz
Kiln may be used but SO2 absorption is necessary in such a case. Other metals are also dissolved dur-
ing the process and are removed after leaching. Iron is the major impurity, which is precipitated in
three basic forms; Jarosite, Goethite or Haemetite (IPPC, 2001). Zinc produced hydrometallurgically
does not need to be refined (Graf & Schleininger, 1997).
6.5.2 Waste
The production of metals is related to the generation of several by-products, residues and wastes,
which are also listed in the European Waste Catalogue (Council Decision 94/3/EEC). The electrolytic
production of zinc is one of the main sources of solid waste in the non-ferrous industry. Relatively

large quantities of iron based solids are generated by the leaching process. Jarosite and goethite are
classified as hazardous waste because of the content of leachable elements such as Cd, Pb and As. The
leaching and purification processes and electrolysis of zinc and the refining stages of lead also gener-
ate other metal rich solids. These are usually rich in a specific metal and are recycled to the appropri-
ate production process. The ISF or direct smelting furnaces also are significant sources of solid slag.
This slag has been subjected to high temperatures and generally contains low levels of leachable met-
als, consequently then may be used in construction. Solid residues also arise as the result of the treat-
ment of liquid effluents. The main waste stream is gypsum waste (CaSO4) and metal hydroxides that
are produced at the wastewater neutralisation plant. Some of these wastes are recycled to pyrometal-
lurgical process to recover the metals. Dust or sludge from the treatment of gases are used as raw ma-
terials for the production of other metals such as Ge, Ga, In and As etc or can be returned to a smelter
or into the leach circuit for the recovery of lead and zinc. Hg/Se residues arise at the pre-treatment of
mercury or selenium streams from the gas cleaning stage. This solid waste stream amounts to ap-
proximately 40 - 120 t/y in a typical plant. Hg and Se can be recovered from these residues depending
on the market demand for these metals (IPPC, 2001).
6.5.3 Emission
The emissions to air can escape the process either as stack emissions or as fugitive emissions depend-
ing on the age of the plant and the technology used. Stack emissions are normally monitored continu-
ously or periodically and reported. The main emissions to air from zinc production are sulphur dioxide
(SO2), other sulphur compounds and acid mists; oxides of nitrogen (NOx) and other nitrogen com-
pounds; metals and their compounds; dust; VOC and dioxins (IPPC, 2001).
Main emissions to water are metals and their compounds and materials in suspension are the main pol-
lutants emitted to water. The metals concerned are Zn, Cd, Pb, Hg, Se, Cu, Ni, As, Co and Cr. Other
significant substances are fluorides, chlorides and sulphates (IPPC, 2001).
6.5.4 Smelting, primary zinc production in ecoinvent

No contiguous study on the zinc production is yet available, therefore this module is based in parts on
values reported in Ayres et al. (2002), Xiao et al. (2003), IPPC (2001) and EEA (1999b).
The multi-output-process "smelting, primary zinc production" delivers the three co-products "zinc,
primary, at regional storage", "cadmium sludge, from zinc electrolysis, at plant" and "leaching resi-
dues, indium rich, from zinc circuit, at smelter". The flow "zinc, …" represents the European supply
mix of primary zinc. The by-products "leaching residues, …" and "zinc, …" receive part of the burden
and enter the primary production chain.
The module describes the production of primary SHG zinc used for galvanising in Europe, it does not
consider secondary production, which accounts in Europe for 30% of the metal consumed (IPPC,
2001). The dataset is designed for the use of zinc in coating processes, and is not suitable for compari-
sons of products containing zinc in a high degree.
The meta-information to the multi-output process "smelting, primary zinc production " is given in Tab.
6.3. The layout of the modules with general flow information, remarks, sources, values and uncer-
tainty information are shown in Fig. 6.12 (inputs) and Fig. 6.13 (outputs). The following paragraphs
describe calculations, sources and assumptions chosen in this study. The functional unit of this process
is one kg of zinc. The meta-information is compiled above in Tab. 6.3 in the beginning of this chapter.
Consumption mix: The consumption mix – i.e. the pattern of the production processes producing the
zinc actually used for galvanising – is approximated with the production mix world-wide of 80% hy-
drometallurgical and 20% pyrometallurgical route. Low quality zinc (GOB) is mainly produced via the
pyrometallurgical route without a final refining step (Graf & Schleininger, 1997). According to IPPC
(2001) 75% of the concentrates used in European production are imported, for which a distance of

5000 km transport by ship is assumed. Since Europe is the world leader in zinc production, no imports
of refined zinc is assumed.
Demand in materials and energy: The input of zinc concentrate was taken from Ayres et al. (2002)
to be in line with the material flow. Most other values are taken from Xiao et al. (2003), which inven-
toried one hydro- and one pyrometallurgical smelter in China. Electricity values had to be taken from
IPPC (2001) since the detailed values presented in Xiao et al. (2003) could not be reproduced. Never-
theless they are shown together with the actually chosen values and the resulting mix in Tab. 6.24.
Also the coal gas consumption reported by Xiao et al. (2003) seems to be too high compared to the
values for natural gas reported by IPPC (2001). Since no information on the composition and heating
value of the coal gas was available, the numbers for natural gas given by IPPC were taken in this
study. There are indications that the total energy consumption indicated by IPPC (2001) is 27% and
31% of the values for the Hydro and the Pyro process reported in Xiao et al. (2003). However, since
Xiao et al. (2003) does not characterise energy carrier and end energy uses sufficiently his results re-
main ambiguous an are not taken. No quantitative compilation of electricity production is available,
therefore for pyrometallurgical processes the UCTE production mix was chosen as approximation
(year 2000: nuclear 15%; hydro 4%; fossil 81% 12 ). Since the electrolysis plants depend on cheap and
readily available electricity, they often are located near hydropower dams often owned by the metal-
lurgical enterprise. A compilation of electricity sources for the aluminium electrolysis is given in the
part “Aluminium” of this report, where a specific dataset for the electricity production for the alumin-
ium electrolysis is established. This dataset is chosen to approximate the zinc electrolysis.
Emissions and disposal. Wastewater is reflected by the direct emissions and the volume reported in
Xiao et al. (2003) and IPPC (2001). An additional process covers solely the treatment process itself –
infrastructure, mixing and pumping, without accounting for specific emissions from wastewater– as if
the treated wastewater was unpolluted (Doka (2003)). Emissions are summed up in Tab. 6.25 (particu-
late emission to air) and Tab. 6.26 (other emissions to air and to water). The dioxin emissions are ap-
proximated with the reported emission factor of 50 μg-TEQ / t secondary zinc production in EC
(2002).
12
http://www.ucte.org/

Tab. 6.24 Zinc production: input and output according to different sources. The values chosen in this study reflect the
production split of 80% hydro- and 20% pyrometallurgical production.
Type Hydro Zinc Elec- Pyro (ISP) ISP This study, split
trolysis 80 hydro : 20 py-
(Hydro) ro
Source Xiao et al. IPPC (2001) Xiao et al. IPPC (2001)
(2003) (2003)
unit per t Zn per t Zn per t Zn per t Zn per t Zn
concentrate kg 1855 1)
Water t 16.2 130 39.0
Diesel oil t 6.0E-4 2.5E-3 6.4E-4
steam t 1.24 0.53 1.09
Oxygen t 0.54 0.108
Jarosite t 0.350 0.280
Solid wastes t 0.323 0.582 0.3753)
Total coal t 0.801 2.82 1.1 0.22 2)
Coal gas m3 769 3247 Not defined
Natural gas m3 160 32.0 2)
total fuel MJ 40274 148309
(coal/gas)
Electricity kWh 3835 4100 918 1050 3490 2)
total energy MJ 54080 151614
1)
Corresponds to the chosen content (55%) and the process yield of 98%.
2)
Energy values given in Xiao et al. (2003) could not be reproduced on the basis of the publication; the
values reported from IPPC (2001) were taken instead.
3)
Including indium and cadmium containing residues and sludges.
Fehler! Textmarke nicht definiert.

Tab. 6.25 Zinc: solid particulates released to air. Values taken from CEPMEIP-Database . Values
might be overestimated – TSP through fuels burned in the smelting process are included in this figures, but
are also accounted for in the respective modules for fuel burning.
Solid Particle Emission Unit Medium / high emission Low emission

(pyrometallurgy, 80%) (hydrometallurgy. 20%)
TSP g / t Zn 500 200
PM10 g / t Zn 400 180
PM2.5 g / t Zn 300 160
Differential classes used in ecoinvent
PM<2.5 g / t Zn 300 160
PM2.5-10 g / t Zn 100 20
PM>10 g / t Zn 100 20

Tab. 6.26 Zinc: Emissions to water and air, except particulates & emissions from fuel combustion. Values correspond
to the production of 1 t zinc with hydro- and pyrometallurgical processes. The values chosen in this study
reflect the production split of 80% hydro- and 20% pyrometallurgical production. Some values are not re-
ported by Xiao et al. (2003); for calculations they are approximated with zero.
Unit Hydro Electroly- Pyro (ISP) ISP This study, split 80

sis hydro : 20 pyro
Source Xiao et al. EEA (2003), Xiao et al. EEA
(2003) EEA (1999b) (2003) (2003)
Emission to air
Zn kg 0.1 10.02.080
SO2 kg 19.1 11.0 17.5
Pb kg 0.097 0.005 0.5 0.104
As kg 1.58E-2 1.26E-2
Hg kg 3.47E-2 2.00E-2 4.0E-3
TCDD kg 5E-11
Emission to water
COD kg 0.583 Not reported 0.466
BSB5 kg 0.389 Not reported 0.311
TOC kg 0.228 Not reported 0.182
DOC kg 0.228 Not reported 0.182
Hg kg 4.30E-5 6.90E-4 1.72E-4
Cd kg 2.60E-3 8.00E-3 3.68E-3
As kg 1.00E-3 1.90E-3 1.18E-3
Pb kg 1.80E-2 1.38E-1 4.20E-2
Cu kg 5.00E-3 4.00E-3
Zn kg Not reported 2.29E-1 4.58E-2
Fluoride kg 1.49E-1 2.98E-2
6.5.5 By-products and allocation

Of the many by-products that are jointly extracted with the zinc, indium and cadmium are modelled
within this inventory. For these metals the zinc extraction is the dominant source for raw-material. In
order to attribute a share of the environmental burden from the joint extraction, an allocation factor for
the three products zinc, indium and cadmium has to be determined. In accordance with Althaus &
Classen (2005) an allocation procedure by revenue is applied.
However, the allocation of the subprocesses indium-, cadmium and zinc-production is not evident to
solve, since only the final commodity prices are known but no in-firm prices for the metal containing
residues and for the raw zinc, on which the allocation would rely. Also no share of the products in to-
tal costs – gross sales method as proposed in Guinée et al. (2004) and Guinée et al. (2001) – is easy to
estimate. The by-product allocation problem – the joint production of cadmium, indium and zinc – is
reduced here by a subdivision of the sub-processes (according to Ekvall & Finnveden (2001)). The in-
dium production from the leaching residues and the cadmium production from sludge respectively is
treated in a distinct inventory. Hence it is the task to allocate the burdens between the zinc, the sludge
and and the leaching residues.
As a starting point for an estimation it is assumed, that the profit margin for the production of silver
corresponds to the company’s performance. Taking a performance value of 10%, the value of the in-
dium containing residues corresponds consequently to 10% of the proceeds from the indium sales.
Therefore 10 % of the indium price is taken to estimate the value of the residues. Since Cadmium is to

be regarded as contaminant that has to be isolated from the main product anyway, the cadmium con-
taining sludge is regarded as burden free (cut-off).
By-product Cadmium in flue dust and purification precipitate

There are mainly two different mechanisms how cadmium accumulates during the zinc production
process. During roasting of the concentrate in the pyro-route the cadmium vaporises and accumulates
in the flue dust that is collected in scrubbers or ESP. The second source is the cadmium sponge that is
precipitated in the course of zinc solution purification prior to electrowinning. In both cases cadmium
arises as unwanted element that has to be separated from the main product (hydro) and has to be re-
tained for controlled handling (pyro) due to its toxicity. Due to the marginal position of the pyrometal-
lurgical route, here only the cadmium sludge from zinc solution purification is modelled in this multi-
output module.
With worldwide roughly 20’000 tons refined cadmium per year in 2006 (Tab. 3.1), the assumption
that 72% of the cadmium refinery output originates from zinc extraction (see Reserves and Resources
of Materials) would mean 13’500 tonnes that are direct by-products from the 10’000’000 tonnes pri-
mary zinc produced. In terms of mass this means that for every tonne of zinc extracted 1.5 kg cad-
mium is refined. This results in a back-calculated value of 0.29% Cd in a concentrate of 53% Zn con-
tent.
According to Fthenakis (2004) the cadmium content of the zinc concentrate is around 0.3% to 0.5%.
This value is in the same order as the concentrate in the raw ore and implies that only in 20% of the
co-extracted zinc is processed to ingots. The remaining 80% is supposed to be discarded, e.g. due to
low concentration or missing markets.
The resulting cadmium sludge from zinc solution purification is assumed to consist of 50% water,
45% cemented cadmium and the rest of zinc dust. It is further processed in a cadmium plant, where the
sludge is dewatered, oxidised and refined by means of electrowinning (see 6.6 Cadmium Production
process and infrastructure).
Despite of Cadmium being a commodity traded at international exchanges, it has to be regarded as
contaminant at the stage where it arises in the zinc metallurgy. According to the common allocation
practice therefore no upstream burdens are to be attributed to the by-product cadmium (Guinée et al.,
2004).
By-product Indium in purification solutions
Indium is produced by a few companies only, all of which are in the zinc production. In those deposit
where indium is present the concentration ranges between 10 and 100 ppm with 50 ppm being a com-
mon value of those operations recovering indium (e.g. Brunswick).
It is very uncertain to which amount secondary indium is recovered, because until 2002 it was not a
technological issue due to favourable indium prices (Jansseune, 2003). The sky-rocketing prices and
the connected ramping up of production facilities will have changed this picture by 2007 but no as-
sured estimate is available in the open literature. Therefore, an amount of 30% secondary production is
arbitrarily chosen to assess the coupling of indium production to the zinc extraction.
The trend for increasing indium recovery within the zinc industry is displayed in Fig. 6.11. Based on
the last years, a ratio of 45 g refined Indium per tonne extracted zinc may be established assuming
steady state stocks and optimal recovery measures. Under the assumption of correspondingly 70% in-
dium from primary production, 3.4E-05 kg indium per kg zinc was chosen as joint production factor.
The indium containing intermediate product is a solution from wet-chemical purification processes. In
the hydrometallurgical zinc recovery process at Akita mine (Dowa Mining Co, Ltd.), gallium and in-
dium are recovered by three stages of ion exchange columns. The feed solution to the third ion ex-
change column contains the relative concentrations Ga: 0.26, In: 0.11, Zn: 1.0 g /l (Teramoto et al.,
1993). A value of 0.1 g / l is chosen for further calculations.

Given an extraction yield of 60% and under the assumption, the indium contained in the residue
amounts to 10% of the sales proceeds, a share of 0.09% of the burden is allocated to the indium con-
taining by-product.
6.0E-05
In production in tpa / Zn
production in tpa
5.0E-05
4.0E-05
3.0E-05
2.0E-05
1.0E-05
0.0E+00
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
Fig. 6.11 Quotient of worldwide indium refinery production and zinc mine production as base for the assessment of
the mulit-output relation in ecoinvent USGS, 2006b). For explanations see in text above.
Tab. 6.27 By-products of primary zinc production: Composition, flows and value to derive the value-based allocation
factors
Metal Process zinc concen- cadmium leaching re- Zinc, primary Value
Yield trate sludge sidues, In (output)
(input) (output) (output)
% Content Content Content Content $ / kg
Zn 97 % 53% 100% 1.88
Cd 99 % 45% -
In 60 % 0.01% 77.5
Amount in kg 1.91 0.00338 0.34 1.00
Value contained (in $) - 0.003 1.88
Allocation by Value - 0.14% 99.86%
Correction of resource flow allocation

Even though overall allocation is performed according to economic revenue, the resources are allo-
cated by mass to the respective by-product to balance the mass flow. Since the concentrate required by
this multi-output-process is allocated also by revenue, the therein contained resources are not correctly
attributed and have to be corrected with a corrective exchange flow (Tab. 6.28).

Tab. 6.28 Individual correction factors to allocate the resource depletion to the respective by-products in order to bal-
ance the mass flow.
correction from economic to according to economic allocation

per mass
Resources attributed for in the feed 0% 0.14% 99.86%
Concentrate kg 1.91 Cd-sludge In-leachate Zinc
Zn kg 1.19 - 0.00166 1.19
Cd kg 0.0396 - 5.55E-05 0.0395
In kg 6.60E-04 - 9.24E-07 6.59E-04
Resources attributed in by-product according to per mass allocation
Zn 1.19
Cd 0.0396
In 6.60E-04
Resource exchange flow difference to be corrected
Zn - -0.00166 1.66E-03
Cd 0.0396 -5.55E-05 -3.95E-02
In - 6.59E-04 -6.59E-04
6.5.6 Data quality

Data uncertainty is derived using the Pedigree matrix as described in Frischknecht et al. (2004). Some
of the less important emissions that are described in the literature as typical for lead, could not be cov-
ered in this inventory due to missing emission factors, viz. Ge, Ga In and Se into air and Se, Ni, Co
and Cr to water. Also some salts –chlorides and sulphates – are not included due to missing data. Fur-
ther investigations on these issues have been done to improve the quality of this dataset. However, for
the purpose of a background database the quality is sufficient.
The inventory does not reflect the portion of secondary material in the zinc used. For specific conclu-
sions further investigation has to be done. Further more, the choice of electricity mix bases on ap-
proximations that have to be improved with further investigations.

in ecoinvent
Modul name
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
Concentrate contains 53% zinc
(1.01 kg Zn, yield = 99%), 0.33%
zinc concentrate, at
Concentrate  cadmium (1.74 g Cd, yield = 99%) metals extraction No GLO 1.91E+00 kg Calculation 1 1.33 (3,3,2,1,3,5,4)
beneficiation
and 0.01% indium (0.06 g In, yield
smelting, primary zinc production (inputs)
= 60%)
Cd resource in concentrate fully
Correction Cd in Cadmium, 0.30% in sulfide, Cd Calculation from
 allocated to the Cd bearing by- resource in ground -5.55E-05 kg 1 1.33
Conc -> In 0.18%, Pb, Zn, Ag, In, in ground concentrate input
product. Correction Indium.
Cd resource in concentrate fully
Correction Cd in Cadmium, 0.30% in sulfide, Cd Calculation from
 allocated to the Cd bearing by- resource in ground -3.95E-02 kg 1 1.33
Conc -> Zn 0.18%, Pb, Zn, Ag, In, in ground concentrate input
product. Correction Zinc
Zn resource in concentrate fully
Correction Zn in Zinc, 9.0% in sulfide, Zn 5.3%, Calculation from
 allocated to Zinc. Correction resource in ground - kg 1 1.33
Conc -> Cd Pb, Ag, Cd, In, in ground concentrate input
Cadmium.
Correction Zn in Zn resource in concentrate fully Zinc, 9.0% in sulfide, Zn 5.3%, Calculation from
 resource in ground -1.66E-03 kg 1 1.33
Conc -> In allocated to Zinc. Correction Indium Pb, Ag, Cd, In, in ground concentrate input
Correction Zn in Zn resource in concentrate fully Zinc, 9.0% in sulfide, Zn 5.3%, Calculation from
 resource in ground 1.66E-03 kg 1 1.33
Conc -> Zn allocated to Zinc. Correction Zinc. Pb, Ag, Cd, In, in ground concentrate input
In resource in concentrate fully Indium, 0.005% in sulfide, In

Correction In in Calculation from
 allocated to Indium. Correction resource in ground 0.003%, Pb, Zn, Ag, Cd, in - kg 1 1.33
Conc -> Cd concentrate input
Cadmium. ground
inventoried as diesel in building construction diesel, burned in building

Diesel and Gasoil  machinery No GLO 4.19E-02 MJ (Xiao, 2003) 1 1.25 (3,3,2,1,3,3,2)
machine processes machine
washing steam, for chemical processes,

steam  auxiliary agents No RER 1.09E+00 kg (Xiao, 2003) 1 1.25 (3,3,2,1,3,3,1)
agents at plant
Total coal  hard coal heating systems No RER 6.93E+00 MJ (IPPC, 2002) 1 1.25 (3,3,2,1,3,3,1)
furnace 1-10MW
Electricity UCTE mix: nuclear 15%; hydro 4%; electricity, medium voltage,
 electricity production mix No UCTE 2.10E-01 kWh (IPPC, 2002) 1 2.07 (3,3,2,1,5,5,2)
Pyrometallurgy fossil 81% production UCTE, at grid
Electricity Approximated with aluminium electricity, medium voltage,
 electricity supply mix No GLO 3.28E+00 kWh (IPPC, 2002) 1 2.07 (3,3,2,1,5,5,2)
Electrolysis electrolysis aluminium industry, at grid
natural gas, burned in industrial
Natural Gas  natural gas heating systems No RER 1.60E+00 MJ (IPPC, 2002) 1 1.25 (3,3,2,1,3,3,1)
furnace >100kW
Water  resource in water Water, river 3.90E-02 m3 (Xiao, 2003) 1 1.25 (3,3,2,1,3,3,2)
Oxygen  chemicals inorganics No RER oxygen, liquid, at plant 1.08E-01 kg (Xiao, 2003) 1 1.25 (3,3,2,1,3,3,4)
Assumption - 600km : concentrate, transport

Transport by train  train No RER transport, freight, rail 1.88E+00 tkm standard distances 1 1.31 (4,3,1,1,3,3,2)
coal, zinc; 100km : jarosite, oxygen systems

Transport by lorry transport transport, lorry >16t, fleet
 coal, zinc; 50km : jarosite, oxygen, road No RER 3.18E-01 tkm standard distances 1 1.31 (4,3,1,1,3,3,2)
(32t) systems average
solid waste
Fig. 6.12 Flows for "Zinc for coating, at regional storage" (inputs) and its representation in the ecoinvent

Modul name
in ecoinvent
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
Portion of the waste fraction that is
waste disposal, inert waste, 5% water,
 Solid wastes not processed for byproduct metals inert material landfill No CH 3.20E-02 kg (Xiao, 2003) 1 1.25 (3,3,2,1,3,3,6)
management to inert material landfill
but disposed of.
waste wastewater treatment, sewage, unpolluted,

smelting, primary zinc production (outputs)  Effluents No CH 3.90E-02 m3 1 1.25 (3,3,2,1,3,3,6)
management treatment to wastewater treatment, class 3
 Cu (water) water river Copper, ion 4.00E-06 kg (Xiao, 2003) 1 5.02 (1,3,1,3,1,4,35)
 Zn (water) water river Zinc, ion 4.58E-05 kg (Xiao, 2003) 1 5.02 (1,3,1,3,1,4,35)
 Pb (water) water river Lead 4.20E-05 kg (Xiao, 2003) 1 5.02 (1,3,1,3,1,4,35)
 As (water) water river Arsenic, ion 1.18E-06 kg (Xiao, 2003) 1 5.02 (1,3,1,3,1,4,35)
 Hg (water) water river Mercury 1.72E-07 kg (Xiao, 2003) 1 5.02 (1,3,1,3,1,4,35)
 Cd (water) water river Cadmium, ion 3.68E-06 kg (Xiao, 2003) 1 5.02 (1,3,1,3,1,4,35)
 Fluoride (water) water river Fluoride 2.98E-05 kg (Xiao, 2003) 1 1.58 (1,3,1,3,3,4,32)
 COD (water) water river 4.66E-04 kg (Xiao, 2003) 1 1.6 (3,3,1,3,3,4,32)
Demand
BOD5, Biological Oxygen
 BSB water river 3.11E-04 kg (Xiao, 2003) 1 1.65 (3,5,1,3,3,4,32)
Demand
 TOC water river TOC, Total Organic Carbon 1.82E-04 kg (Xiao, 2003) 1 1.65 (3,5,1,3,3,4,32)
 DOC water river DOC, Dissolved Organic Carbon 1.82E-04 kg (Xiao, 2003) 1 1.65 (3,5,1,3,3,4,32)
low population
 Pb (air) air Lead 1.04E-04 kg (Xiao, 2003) 1 5.04 (3,4,2,1,1,4,22)
density
low population
 Zn (air) air Zinc 2.08E-03 kg (Xiao, 2003) 1 5.04 (3,4,2,1,1,4,22)
density
low population
 Hg (air) air Mercury 2.29E-05 kg (Xiao, 2003) 1 5.04 (3,4,2,1,1,4,22)
density
low population
 As (air) air Arsenic 1.26E-05 kg (Xiao, 2003) 1 5.04 (3,4,2,1,1,4,22)
density
low population
 PM<2.5 air Particulates, < 2.5 um 1.88E-04 kg CEPMEIP 1 3.04 (3,4,2,1,1,4,27)
density
 PM2.5-10 air 3.60E-05 kg CEPMEIP 1 1.28 (3,4,2,1,1,4,28)
density 10um
low population
 PM>10 air Particulates, > 10 um 3.60E-05 kg CEPMEIP 1 1.46 (3,4,2,1,1,4,29)
density
low population Dioxins, measured as 2,3,7,8- European Dioxin
 TCDD air 5.00E-11 kg 1 10 reported
density tetrachlorodibenzo-p-dioxin Inventory
low population
 SO2 (air) air Sulfur dioxide 1.75E-02 kg (Xiao, 2003) 1 1.19 (3,4,2,1,1,4,15)
density
high population
 waste heat Calculated from electricity use. air No RER Heat, waste 1.26E+01 MJ Calculations 1 1.13 (1,4,1,3,1,3,13)
density
Residues, indium leaching residues, indium rich,

 Contains 0.010% In metals extraction No GLO 3.38E-03 kg Calculations
containing from zinc circuit, at smelter
cadmium sludge, from zinc

 Cadmium sludge Contains 45% Cd metals extraction No GLO 3.39E-01 kg Calculations
electrolysis, at plant
zinc, primary, at regional
 Refined zinc metals extraction No RER 1.00E+00 kg
storage
Fig. 6.13 Flows for "zinc for coating, at regional storage" (outputs) and its representation in the ecoinvent database

6.6 Cadmium Production process and infrastructure

In hydrometallurgical zinc refining, cadmium is removed from the feed together with copper by reduc-
tion with zinc dust to give a metallic sludge. Similar cadmium sludge forms in the production of zinc
sulfide pigments or zinc carbonate. These are the most important sources for cadmium refining today
(Schulte-Schrepping & Piscator, 2002).
Fig. 6.14 Sources and process routes for cadmium production. The most common production path is the electrolysis
from processed cadmium sludge from hydrometallurgical Zinc operations (Schulte-Schrepping & Piscator,
2002), modelled in this study.
Further secondary starting materials mentioned above may be added to potential primary raw materials
without causing large alterations in the refining process. The different raw materials are dissolved in a
solution.
Cadmium is extracted by oxidizing the enriched starting materials solutions. In order to oxidize the
cadmium, either air is blown in or manganese dioxide is added. Cadmium is precipitated from the so-
lution as a metal sponge by reduction with zinc dust. The precipitated sponge contains 90 % cadmium
and only 2.5 % zinc.
Today, the largest amounts of cadmium are recovered electrolytically. The cadmium sponge is precipi-
tated directly in the tank with the electrolytic acid. The anode consists of lead, which does not dis-
solve, whereas the cathode is aluminum sheet. The current yield is ca. 93 %; the energy consumption
is 1250 kWh per tonne of cadmium cathode sheet deposited. Typical impurity levels in electrolytic
cadmium sheet are 5 – 10 ppm Cu, 15 – 20 ppm Pb, 5 – 10 ppm Tl, and 5 ppm Zn.
6.6.1 Cadmium, primary, at regional storage in ecoinvent

This dataset gives only a rough estimate on the production of primary cadmium from leaching residues
from hydrometallurgical zinc operations by means of electrolytic refining. It is mainly based on litera-
ture data (Schulte-Schrepping & Piscator, 2002) and is designed for the use of cadmium in electric de-
vices, namely semiconductor applications such as CdTe-films.

The layout of the modules, general flow information, remarks, sources, values and uncertainty infor-
mation are shown in
Fig. 6.15. The following paragraphs describe calculations, sources and assumptions chosen in this
study. The functional unit of this process is one kg of refined cadmium sheet. Tab. 6.4 in the begin-
ning of this chapter summarises the meta-information for this dataset.
Precipitation. The zinc dust used for cementation of the cadmium sponge is recovered later in the
sludge of the electrolysis and will be returned into the zinc cycle. Therefore no account for zinc dust is
done.
Oxidation. The oxidation of the sponge is often done with air, therefore no chemicals consumption is
considered in this step. Also no additional heat for drying is assumed.
Electrolysis. The consumption of 1.25 kWh electricity per kg sheet is considered in this inventory as
reported in Schulte-Schrepping & Piscator (2002). Taking copper electrowinning as approximation,
0.2 m3 electrolyte per tonne is bled (drained) and purified for recycling into electrolytic refining
(Davenport et al., 2002). It is assumed that all of this volume is replenished with fresh electrolyte (180
kg H2SO4 per m3), since no process data on the purification of spent electrolyte are available.
Because of the lack of a complete and contiguous LCI study that treats the production of cadmium and
the absence of problematic processes and auxiliaries employed, no emissions to air and water could be
inventoried.
6.6.2 Data Quality

Data uncertainty is derived using the Pedigree matrix as described in Frischknecht et al. (2004). This
dataset is comparably coarse since it disregards all but the most important environmental impacts at-
tributable to this process. However, since cadmium is not used in applications as pure metal, but di-
luted in alloys or chemicals, this dataset is of sufficient quality. For the use of this inventory in appli-
cations where cadmium is a major constituent – e.g. in NiCd battery systems – the dataset must be
complemented with additional specific emission factors and the secondary production process.

Modul name
in ecoinvent
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
Cadmium starting cadmium sludge, from zinc
 metals extraction No GLO 2.24E+00 kg Calculations 4 uniform dist +/- 5%
material electrolysis, at plant
cadmium, primary,
Electricity for electricity, medium voltage, Schulte-Schrepping
 electricity production mix No UCTE 1.25E+00 kWh 4 1.13 (1,2,3,3,1,3,2)
electrolysis production UCTE, at grid (2002)
Transport over 250 km transport

Transport lorry  road No RER transport, lorry >16t, fleet average Assumptions 1 2.16 (4,4,1,3,3,5,5)
at plant
2.50E-01 tkm
to regional storage. systems
0.2 m3 bled per tonne,

Electrolyte  180 kg H2SO4 per m3 chemicals inorganics No RER sulphuric acid, liquid, at plant 3.60E-02 kg Assumptions 1 1.33 (4,4,1,3,3,3,4)
electrolyte
Approximated with
"chemical plant,
Facilities  organic", which as an chemicals organics Yes RER chemical plant, organics 8.00E-10 unit Assumptions 1 3.15 (4,4,1,3,3,5,9)
output of 50'000 t per
year (25 year lifetime)
Calculated from
 waste heat air high population density No RER Heat, waste 4.50E+00 MJ Calculations 1 1.13 (1,2,3,3,1,3,13)
electricity use.
 Cadmium Primary metals extraction No GLO cadmium, primary, at plant 1.00E+00 kg 1
Fig. 6.15 Input and output flows for "cadmium, primary production, at regional storage"

6.7 Indium Production Process and Infrastructure

Industrially, indium is produced mainly by recovering from by-products of zinc smelting such as
fumes, flue dust, jarosite or leach solution (Kammer et al., 2006). These by-products contain small
amounts of indium together with other accompanying elements. Therefore a complex hydrometallur-
gical process is required for isolating indium from impurities such as arsenic, zinc and cadmium.
These processes highly depend on the composition given by the specific deposit.
Since the recovery processes are optimised to the main intended products – zinc, lead – the indium re-
covery is subject to quite a poor efficiency. Indium concentrates generally at two points in the hydro-
metallurgical zinc extraction process, while roasting the concentrate and in the neutral leach residue.
Jorgenson & George (2004) report that a traditionally low 30 % yield of available indium in treatable
residues has been improved to nowadays 80%. Within the jarosite residues a maximum of 60% can be
dissolved from the neutral leach residue, which leads to an estimated overall extraction efficiency of
around 50% of the indium, i.e. 50% of the indium present in the ore is transferred to the pregnant solu-
tion.
If the indium content in the ore is back calculated based on process efficiencies and the indium refin-
ery production and compared to the indium contained in the extracted lead-zinc ore with 50ppm In, it
can be estimated that only 40% of the indium extracted is treated for recovery. Thus an overall yield of
19% may be estimated which corresponds to an overall yield of 20% indicated in Stevens (2007).
Generally, indium is recovered by removing impurities through various chemical purification methods
such as acid and alkali leaching, solvent extraction, neutralization, sulphidising and cementation are
combined to gain a high indium content solution. From this solution indium is either directly electro-
won (e.g. Okamoto & Takebayashi, 1996) or precipitated with zinc as a refined sponge indium that is
cast into anodes and electro-refined (Felix, 1996; Power Porto, 1990). To overcome undesirable chlo-
rine gas generation while electrowinning, a two-compartment cell is required (Okamoto & Takebaya-
shi, 1996).
6.7.1 Concentration of Indium in the solution

In this first step indium is concentrated from around 500 ppm in byproducts and 50 mg / l in a first
leach stage to 30 g / l by hydrometallurgical means.
Generally indium containing byproducts arise in the extraction of zinc, lead and tin. The lead and tin
metallurgy are quantitatively of minor importance in terms of mine production, although some big
single operations are based on lead-zinc and tin deposits, e.g. Cominco at Trail, Canada. Among the
indium producers in tin circuits, Capper Pass in Hull, U. K. is mentioned (Alfantazi & Moskalyk,
2003; Felix, 1996; Power Porto, 1990). However, this operation has been – last owned by Rio Tinto
Zinc Ltd. – closed down in 1991 and decommissioned. Since then the tin circuits are of minor impor-
tance as source for indium bearing raw materials.
In technical literature quite a few different purification processes are described that follow a row of
leaching, neutralisation and precipitation stages. However, the more recent patent literature gives
strong indications that solvent extraction (SX) techniques are favoured as e.g. proposed by Tomii &
Tsuchida (1981) and studied by Power Porto (1990) and Fortes et al. (2003). This development is
linked to a shift in the zinc extraction technology from pyrolytical processes to hydrometallurgical
processes, where indium accumulates in the leach solution. In a review article Paiva (2001) states, that
commercial grade indium is industrially produced using leaching, cementation and electrorefining
technologies. SX is often used to recover indium from leach residues. More recent work confirms this
tendency and indicates the introduction of more specific ion exchange and extraction resins (Liu et al..
2006; Zhou & Ruan, 2006).

Based on the above mentioned sources it can be concluded that most of the indium produced today is
won in hydrometallurgical zinc operations like the Akita refinery in Akita, JP or the Zhouzhou smelter
in Hunan Province, China, Cominco’s Trail operation (30 tonnes per year 13 ) or Falconbridge’s Kidd
Creek Refinery, Canada. These operations process ores from close-by mines and indium bearing in-
termediate products, prompt scrap and post consumer scrap.
A typical process scheme for this type of operations is given in Fig. 6.16 on the example of the Kidd
Creek refinery. The process is similar to the descriptions given in Kudo & Maruyama (2001) and
Power Porto (1990).
14
Fig. 6.16 Process scheme of Falconbridge’s Kidd Creek refinery
The leach solution which carries indium has usually a concentration of 120 ppm in Zinc residue
(Akita, Alfantazi & Moskalyk, 2003), 50-70 mg / L after first leaching (Akita, Felix, 1996) 690 mg / L
(typical zinc plant after concentration, Fortes et al., 2003), 200 mg / L (Laboratory setting, Liu et al.,
2006, 0.4% (in secondary oxide, 0.68% after preleaching Felix, 1996), 2.7% chloride slag (capper
pass, Felix, 1996). Stevens (2007) indicates a range of 300-500 ppm for high indium concentrates.
This solution is concentrated either by SX or a row of neutralisation-precipitation steps to ca. 1 g / L
(Power Porto, 1990), or 2 g / L (Akita; Felix, 1996). From this solution sponge indium is precipitated
either on aluminium sheets or with zinc powder. The resulting sponge is cast to raw indium anodes
with a purity of 95-99.5 %.
As Tomii & Tsuchida (1981) put forward, a main challenge of the neutralisation process is the re-
moval of accompanying contaminants, therefore this quite unspecific procedure is highly complicated,
costly and does not yield the purity required for subsequent refining to 5N (five nines = 99.999%)
quality. Within the different concentration techniques SX today is regarded as one of the most promis-
ing separation operations (Paiva, 2001).
13
In 2005, http://www.teckcominco.com/operations/trail/production.htm, accessed in april 2007
14
http://www.xstrata.com/, accessed in april 2007
However, because of the presence of ions with similar extraction tendency as indium, it is difficult to
prepare high purity indium (>4N) only by electrowinning unless several combinations of purification
methods are applied (Zhou & Ruan, 2006).
Therefore a sequence of (i) solvent extraction (ii) cementation with zinc powder (iii) anode casting is
considered to be the most appropriate process for gaining raw indium that meets the requirements for
further refining to over 5N quality. This process corresponds to the study set-up in Power Porto (1990)
that is shown in Fig. 6.17.
Fig. 6.17 Study set-up of a semi industrial extraction facility by Power Porto (1990) that reflects an up-to-date solvent
extraction process scheme. Process covers solvent extraction and cementation; the resulting cementate
(“In-Zementat”) is cast further into anodes, which is not depicted in this figure.
6.7.2 Refining of Raw Indium

There are basically three possibilities described for refining raw sponge indium (99.5%) to the re-
quired 5N purity: Vacuum refining, electrorefining zone refining and electrowinning or a combination
of all of these. Kammer et al. (2006) additionally mention distillation of InCl3 and vacuum distillation.
The main challenge is to remove the accompanying contaminants that have a negative effect on the
properties of the semiconductor made from indium. Namely thallium and cadmium have detrimental
effects on the semiconducting properties.
Purification of metals by vacuum refining is widely practiced. Generally vacuum refining is employed
after the electrolytic step to further reduce the level of volatile impurities in the indium metal. Indium
metal has low melting point and high boiling point. Therefore, it is possible to remove the impurity
metals with high vapour pressure by distilling indium metal at relatively low temperature under vac-

uum. The impurity metals, which are not well removed by chemical methods, are easily removed by
vacuum refining at relatively low temperatures (Lee & Sohn, 2003).
6.7.3 Indium production in ecoinvent

The module describes the primary production of indium metal (99.999%, 5N purity) on a European
average. It is based on the respective flow from the multi-output module "smelting, primary zinc pro-
duction". Meta-information for this dataset are given in Tab. 6.5, an overview over the inventory is
given in Fig. 6.18
Due to the strategic importance of the material and because of the heterogeneous raw materials, no
contiguous information on the most common recovery processes are available for LCI modelling.
However, there exist many procedures common with the solvent extraction / electrowinning process
used in copper production. There also pregnant leach solution is subjected to a solvent-solvent extrac-
tion. Based on the process descriptions above, following general steps are covered in this inventory:
- Leaching and solvent extraction with D2EHPA
- Cementation with zinc powder and anode casting
- Electrorefining
- Vacuum distillation
Leaching and solvent extraction (SX)

The inventory bases on the leaching residues from hydrometallurgical zinc extraction as raw materials.
Davenport et al. (2002) states that in the case of SX in copper production 35 g Cu / L are required for
electrowinning. SX has to yield this concentration starting from leach solutions not richer than 1 - 6 g /
L. This concentration factor of around ten times is comparable with the SX of indium, where a solu-
tion of 50 – 200 mg / L is concentrated to 1 to 2 g / l.
In this study it is assumed that the lower concentration level in the case of solvent extraction of indium
leads to a proportional increase of throughput compared to SX of copper. The leach solution of indium
is 20 to 30 times less concentrated than in the copper extraction. It is assumed in this study, that the
required accumulation of indium requires consequently 20 to 30 times the volume put through. Hence
the specific emissions as well as energy and material consumption documented for the solvent extrac-
tion of copper are multiplied by 25 and are taken as approximation for the SX of indium. The values
for electricity consumption for pumping, mixing, etc. as well as the “crud” generation (solid particles
associated with oil/water dispersions that are disposed of) are taken from Krauss et al. (1999). Direct
emissions were estimated based on reported yearly air emissions of an U.S. SX/EW copper operation
in Arizona (ADEQ, 2006) and the reported copper output of the facility 15 .
Cementation and anode casting

Crude indium is formed by the precipitation of a sponge. Here, it is assumed that the sponge contains
20%wt moisture and is dried prior to anode casting. Drying is assumed to be effected in a spray dryer
used in ceramic industry (38kW at 250 kg per hour Timokhova (2002), Melting is assessed using the
heat of fusion and assuming a system efficiency of 70%.
15
2006 BHP Billiton Annual Report, Information on the Company: Pinto Valley Operation produced 8’500 t in 2006
Electrorefining
No information on electrorefining of indium was available. However, this process – where the anode
dissolves anodically – usually ranges between 0.2 to 0.5 kWh per kg. In this inventory a value as re-
quired in copper refining is chosen. In Davenport et al. (2002, Tab 16.4) for electrorefining of copper a
range of 0.25 to 0.35 kWh / kg is given. Taking the upper value, an electricity consumption of 0.35
kWh per kg indium cathode sheet is assumed.
Belsky et al. (1980) state a current density of 0.015 to 0.03 A / cm2 (150 to 300 A / m2) for the elec-
trorefining of indium, which is comparable to copper (200 to 300 A / m2). Zhou et al. (2005) deter-
mine optimal conditions of 0.2-0.3 V bath voltage but at lower current densities of 50-80 A / m2.
In Davenport et al. (2002) it is stated that 0.1 to 0.2 m3 electrolyte is bled per tonne of copper, i.e. re-
circulated for purification. In Zhou et al. (2005) the electrolyte is a low concentration solution of sul-
phuric acid. We assume a value of 0.15 kg 25% H2SO4 per kg indium refined.
Vacuum refining
A continuous process is described in McNamara et al. (1989) on behalf of Indium Corporation of
America Inc. In the proposed installation liquid indium is heated under vacuum (10-7 to 10-8 torr) to
1’400 °C in order to volatilise remaining impurities. The original patent by Belsky et al. (1980) re-
mains less specific but claims the process duration of two heating phases of 1-5 h at 850-940°C fol-
lowed by 0.5-2 h at 950-1100°C. The long duration of this refining step is also confirmed by Lee &
Sohn (2003) who found slow evaporation rates. In this study we assume that a RF induction heated
Vacuum furnace is employed that uses 0.5 kW per kg load. This assumption corresponds to a batch of
30 kg indium that is heated for 4 hours in a 15kW furnace as often used in lab-scale. Commercial fur-
naces 16 exhibit energy supply from 36 kW (0.01 m3 volume) to 90 kW (0.127 m3 volume). Assuming
that one third of the heated volume can catch the indium load (7.3 g / cm3) a somewhat higher span of
0.5 – 0.3 kW per kg batch size results. Inductotherm Corp. reports 0.15 kW per kg for its Acutrac (R)
furnace 17 . However, bearing in mind that the metal is very costly and no large batch sizes are sup-
posed to be processed, the upper value of this span is justified.
Additional power consumption for vacuum preparation is disregarded due to a lack of reliable sources
and because this particular process is expected to be of minor importance compared to the overall im-
pacts, Also the inert gas atmosphere is supposed to be of minor importance for the environmental im-
pacts and is therefore not inventoried.
Transport and Infrastructure

The indium metal is assumed to be transported by lorry to and from the airport (250 km each), by
plane over 5’000 km to Europe. Additional weight due to packaging is accounted for with additional
50%.
The infrastructure is estimated based on the module "chemical plant, organics", that has a built area of
about 4.2 ha and an average output of 50'000 t/a (Althaus et al., 2003).
16
http://www.vacuumfurnace.com/, accessed in may 2007
ecoinvent v2.1 report No. 10
17
http://www.inductothermgroup.com/, accessed in may 2007 - 68 -
Modul name
in ecoinvent
mean value
Mean value
Source for
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
General
Output
Name
Input
Infra-
Type
Sub-
Unit
leaching residues, indium rich, from
Leach solution  metals extraction No GLO 1.00E+04 kg 1 1.32 (3,2,1,2,3,5,3)
zinc circuit, at smelter
Electricity for electricity, medium voltage,

 Based on copper SX electricity production mix No UCTE 8.75E+00 kWh Krauss et al 1999 1 1.38 (4,2,1,2,3,5,2)
mixing (SX) production UCTE, at grid
Organic Crud formation, based

 chemicals organics No GLO chemicals organic, at plant 4.25E-02 kg Krauss et al 1999 1 1.3 (4,2,1,2,1,5,3)
Chemicals on copper SX
NMVOC, non-methane volatile

based on a copper SX
 VOCs air high population density organic compounds, unspecified 9.17E-02 kg 1 1.62 (4,2,1,2,1,5,16)
operation
origin
indium, at regional storage
based on a copper SX
 H2SO4 mist operation, approx with air low population density Sulfuric acid 4.34E-02 kg 1 1.62 (4,2,1,2,1,5,18)
HCl (Electrolyte)
Spray dryer 38 kW @ electricity, medium voltage,

drying energy  electricity production mix No UCTE 1.57E-01 kWh Timokhova 2002 1 1.38 (4,2,1,2,3,5,2)
250 kg / h production UCTE, at grid
heat of fusion 3.281 kJ / electricity, medium voltage,

melting energy  electricity production mix No UCTE 9.92E-03 kWh 1 1.3 (4,2,1,2,1,5,2)
mol, 70% efficiency production UCTE, at grid
Replenishment of
sulphuric acid  electrolyte, H2SO4, 25% chemicals inorganics No RER sulphuric acid, liquid, at plant 3.75E-02 kg Davenport 2002 1 1.38 (4,2,1,2,3,5,3)
conc.
Electrowinning Based on Copper electricity, medium voltage,

 electricity production mix No UCTE 3.50E-01 kWh Davenport 2002 1 1.38 (4,2,1,2,3,5,2)
electricity Electrorefining production UCTE, at grid
Vacuum Based on different

distillation  furnace types from 0.7 electricity production mix No UCTE 2.00E+00 kWh Estimate 1 1.3 (4,2,1,2,1,5,2)
electricity to 0.15 kW per kg.
2 x 250 km distance,
Transport by transport
 with 50%wt additional road No RER transport, lorry >16t, fleet average 7.50E-01 tkm Assumption 1 2.33 (5,3,1,1,3,5,5)
truck to dealer systems
packaging
Transport by 5'000 km, with 50%wt transport

 airplane No RER transport, aircraft, freight 7.50E+00 tkm Assumption 1 2.33 (5,3,1,1,3,5,5)
plane to region additional packaging systems
Approximated with
"chemical plant,
Facilities  organic", which as an chemicals organics Yes RER chemical plant, organics 8.00E-10 unit Assumption 1 3.32 (5,1,1,1,3,5,9)
output of 50'000 t per
year (25 year lifetime)
Calculated from
 waste heat air high population density No RER Heat, waste 4.06E+01 MJ 1 1.05 (1,1,1,1,1,1,3)
electricity use.
 Indium metal 5N metals extraction No RER indium, at regional storage 1.00E+00 kg 1
Fig. 6.18 Flows for "Indium, at regional storage" and their representation in the ecoinvent database

6.7.4 Data Quality Considerations

The production process is rather new and not much is known from public sources about quantitative
process measures. Therefore the data has been established based on theoretical reasoning and rough
assumptions, extrapolating the data from copper refining. Except for the sub-process of solvent extrac-
tion which employs organic extraction solution and emits VOC and H2SO4 mist, no direct emissions
from other sub-processes have been considered. Infrastructure has been approximated with the related
technology of chemical plants. The uncertainty of the used environmental flows has been calculated
with the pedigree approach. The data does comply with the demands for the use in electronic devices.
This dataset is not to be used when indium is a substantial part of the system modelled (e.g. for com-
parisons of special solder paste).

7. Literature
7 Literature
ADEQ (2006) ADEQ (2006) TECHNICAL REVIEW AND EVALUATION AIR QUALITY
PERMIT NO. 38400 BHP Copper Inc. – Pinto Valley Operations. Arizona De-
partment of Environmental Quality, Online-Version under: http://www.az-
deq.gov/download/38400tsd.pdf.
Alfantazi & Moskalyk (2003) Alfantazi A. M. and Moskalyk R. R. (2003) Processing of indium: a review. In:
Minerals Engineering, 16(8), pp. 687-694.
Althaus & Classen (2005) Althaus H. J. and Classen M. (2005) Life Cycle Inventories of Metals and Meth-
odological Aspects of Inventorying Material Resources in ecoinvent. In: Int J
LCA, 10(1), pp. 43-49.
Anonymous (1998) Anonymous (1998) Pollution Prevention and Abatement Handbook; Base Metal
and Iron Mining. World Bank Group. Retrieved from
http://wbln0018.worldbank.org/essd.nsf/Globalview/PPAH/$File/48.basem.pdf
Products and Byproducts. MMSD - Mining, Minerals and Sustainable Develop-
ment, Online-Version under: http://www.iied.org/mmsd/mmsd_pdfs/ayres_lca_
main.pdf.
Belsky et al. (1980) Belsky A. A., Eljutin A. V., Zubkov V. N., Konjukov A. I., Krasinskaya L. I.,
Nekrasova E. A., Nizharadze K. S., Markova L. F., Mirskaya M. G., Frolov A.
V. and Judin E. A. (1980) Process of Ultrapurification of Indium. Patent Number
U.S. 4,287,030 held by
Berzelius (1993) Berzelius (1993) QSL - Die umweltfreundliche Bleigewinnung. Berzelius
Stolberg GmbH, Stolberg.
Campine NV (2006) Campine NV (2006) Corporate website from http://www.campine.be
Davenport et al. (2002) Davenport W. G., King M., Schlesinger M. and Biswas A. K. (2002) Extractive
metallurgy of copper. 4th Edition. Pergamon, Oxford. ISBN 0-08-044029-0.
Online-Version under: http://www.engineeringvillage2.org/controller/serv-
let/OpenURL?genre=book&isbn=0080440290.
DiFrancesco et al. (2006) DiFrancesco C. A., Smith G. R. and Gabby P. N. (2006) U.S. Geological Sur-
vey: Lead Statistics. U.S. Geological Survey, Washington, D.C., Online-Version
under: http://minerals.usgs.gov/ds/2005/140/lead.pdf.
Doka (2003) Doka G. (2003) Life Cycle Inventories of Waste Treatment Services. Final re-
port ecoinvent 2000 No. 13. EMPA St. Gallen, Swiss Centre for Life Cycle In-
Dove & Boustead (1998) Dove W. T. and Boustead I. (1998) The Effect of Sulphur on Primary Zinc Eco-
profile Calculations. In proceedings from: The Third International Conference
on EcoBalance, Tsukuba, Japan, 25rd-27th November. Online-Version under:
http://www.boustead-consulting.co.uk/publicat.htm.
Dzioubinski & Chipman (1999) Dzioubinski O. and Chipman R. (1999) Trends in Consumption and Produc-
tion: Selected Minerals. In: DESA Discussion Paper No. 5. United Nations - De-
partment of Economic and Social Affairs, Online-Version under:
http://www.un.org/ esa/sustdev/publications/esa99dp5.pdf.

7. Literature

sion: 22.08.2002.
ecoinvent Centre (2005) ecoinvent Centre (2005) ecoinvent data v1.2. In: Final reports ecoinvent 2000
No. 1-15. Swiss Centre for Life Cycle Inventories, Dübendorf, CH, Online-
Version under: www.ecoinvent.ch.
EEA (1999a) EEA (ed.) (1999a) Primary Lead Production (B334-1). In: EMEP / CORINAIR
EEA (1999b) EEA (ed.) (1999b) Primary Zinc Production (B335-1). In: EMEP / CORINAIR
EEA (2003) EEA (ed.) (2003) Primary Zinc Production (B335-1). In: EMEP / CORINAIR
Emission Inventory Guidebook. February 2003 Edition Edition. European En-
viornmental Agency, Online-Version under: http://reports.eea.eu.int/techni-
cal_report_2001_3/en/en/tab_content_RLR.
Ekvall & Finnveden (2001) Ekvall T. and Finnveden G. (2001) Allocation in ISO 14041- A Critical Review.
In: Journal of Cleaner Production, 9(3), pp. 197-208.
Espinosa et al. (2004) Espinosa D. C. R., Bernardes A. M. and Tenorio J. A. S. (2004) An overview on
the current processes for the recycling of batteries. In: Journal of Power
Sources, 135(1-2), pp. 311-319.
Felix (1996) Felix N. (1996) Indium and Indium Compounds. In: Ullmann's Encyclopedia of
Industrial Chemistry. Fifth Edition Edition. Wiley-Interscience.
Fisher et al. (2006) Fisher K., Wallén E., Laenen P. P. and Collins M. (2006) Battery Waste Man-
agement Life Cycle Assessment. Environmental Resources Management ERM,
Ltd.
Fortes et al. (2003) Fortes M. C. B., Martins A. H. and Benedetto J. S. (2003) Indium recovery from
acidic aqueous solutions by solvent extraction with D2EHPA: a statistical ap-
proach to the experimental design. In: Brazilian Journal of Chemical Engineer-
ing, 20, pp. 121-128, Online-Version under: http://www.scielo.br/scielo.php?
script=sci_arttext&pid=S0104-66322003000200005&nrm=iso
Frischknecht et al. (2004) Frischknecht R., Jungbluth N., Althaus H.-J., Doka G., Dones R., Hischier R.,
Hellweg S., Nemecek T., Rebitzer G. and Spielmann M. (2004) Overview and
Fthenakis (2004) Fthenakis V. M. (2004) Life cycle impact analysis of cadmium in CdTePV pro-
duction. In: Renewable & Sustainable Energy Reviews, 8(4), pp. 303-334,
Online-Version under: http://dx.doi.org/10.1016/j.rser.2003.12.001.
Gabby (2004) Gabby P. N. (2004) Minerals Year Book 2004: Lead. U.S. Geological Survey,
Washington, D.C., Online-Version under: http://www.minerals.usgs.gov/miner-
als/pubs/commodity/lead/leadmyb04.pdf.

7. Literature
Gabby (2006) Gabby P. N. (2006) Mineral Commodity Summary: Lead. U.S. Geological Sur-
vey, Washington, D.C., Online-Version under:
http://minerals.usgs.gov/minerals/ pubs/commodity/lead/index.html#mcs.
Graf & Schleininger (1997) Graf P. and Schleininger R. (1997) Betonrecycling; Stoffliche Auswirkungen
von Alternativstoffen in der Zement- und Betonproduktion auf die Zusammen-
setzung eines mehrfach rückgebauten Betons. ETH Zürich, Zürich.
Granqvist & Hultaker (2002) Granqvist C. G. and Hultaker A. (2002) Transparent and conducting ITO films:
new developments and applications. In: Thin Solid Films, 411(1), pp. 1-5,
Online-Version under: http://dx.doi.org/10.1016/S0040-6090(02)00163-3.
Guinée et al. (2001) Guinée J. B., Gorrée M., Heijungs R., Huppes G., Kleijn R., de Koning A., van
Oers L., Wegener Sleesewijk A., Suh S., Udo de Haes H. A., de Brujin H., van
Duin R., Huijbregts M. A. J., Lindeijer E. W., Roorda A. A. H., van der Ven B.
L. and Weidema B. P. (2001) Life Cycle Assessment; An Operational Guide to
the ISO Standards; Final Report, May 2001. Ministry of Housing, Spatial Plan-
ning and Environment (VROM) and Centrum voor Milieukunde (CML),
Rijksuniversiteit, Den Haag and Leiden, Online-Version under: http://www.lei-
denuniv.nl/cml/lca2/index.html.
Guinée et al. (2004) Guinée J. B., Heijungs R. and Huppes G. (2004) Economic Allocation: Exam-
ples and Derived Decision Tree. In: Int J LCA, 9(1), pp. 23-33.
Hawkins et al. (2006) Hawkins T. R., Matthews H. S. and Hendrickson C. (2006) Closing the loop on
cadmium - An assessment of the material cycle of cadmium in the US. In: Inter-
national Journal of Life Cycle Assessment, 11(1), pp. 38-48, Online-Version un-
der: http://dx.doi.org/10.1065/lca2006.01.234.
Hewett (2006) Hewett C. (2006) A volatile year. In: Metal Bulletin, Online-Version under:
http://www1.metalbulletin.com/story_2003.asp?storycode=1332427.
ILZSG (2007) ILZSG (2007) Lead and Zinc Statistics. International Lead and Zinc Study
Group, Online-Version under: http://www.ilzsg.org/.
0/733169
Jansseune (2003) Jansseune T. (2003) The indium market and compound semiconductors. In:
Compound Semiconductor Week. Number September.
Jansseune (2005) Jansseune T. (2005) Indium price soars as demand for displays continues to
grow. In: Compound Semiconductor Week. Number September.
Jorgenson & George (2004) Jorgenson J. D. and George M. W. (2004) Indium. In: USGS Open File Reports
2004-1300. U.S. Geological Survey, Online-Version under: http://minerals.
usgs.gov/minerals/pubs/commodity/indium.
Jungbluth (2003) Jungbluth N. (2003) Erdöl. In: Sachbilanzen von Energiesystemen: Grundlagen
für den ökologischen Vergleich von Energiesystemen und den Einbezug von E-
nergiesystemen in Ökobilanzen für die Schweiz (ed. Dones R.). Final report

7. Literature
ecoinvent 2000 No. 6, Paul Scherrer Institut Villigen, Swiss Centre for Life Cy-
cle Inventories, Dübendorf, CH. Online-Version under: www.ecoinvent.ch.
Kammer et al. (2006) Kammer U., John R., Monse G. and Wiese U. (2006) Nebenmetalle. In: Winna-
cker/Küchler. Chemische Technik: Prozesse und Produkte. Band 6b Metalle,
Vol. 6b. 2006 Edition. Wiley-VCH, Weinheim.
Kelly & Matos (2007) Kelly T. D. and Matos G. R. (2007) Historical Statistics for Mineral and Mate-
rial Commodities in the United States. In: Data Series 140. U. S. Geological
Survey, Online-Version under: http://minerals.usgs.gov/ds/2005/140/.
Kudo & Maruyama (2001) Kudo Y. and Maruyama Y. (2001) Gallium And/Or Indium Separation and Con-
centration Method. Patent Number U.S. 6,319,483 B1, held by Dowa Mining
Co., Ltd.
Lee & Sohn (2003) Lee M. S. and Sohn K. Y. (2003) Comparison of indium purification between
vacuum refining and electrowinning. In: Journal of Materials Science, 38(24),
pp. 4843-4848.
Liu et al. (2006) Liu J. S., Chen H., Chen X. Y., Guo Z. L., Hu Y. C., Liu C. P. and Sun Y. Z.
(2006) Extraction and separation of In(III), Ga(III) and Zn(II) from sulfate solu-
tion using extraction resin. In: Hydrometallurgy, 82(3-4), pp. 137-143, Online-
Version under: http://dx.doi.org/10.1016/j.hydromet.2006.03.008.
McNamara et al. (1989) McNamara M. F., Slattery J. A. and Witt A. F. (1989) Process of Ultrapurifica-
tion of Indium. Patent Number U.S. 4,828,608, held by Indium Corporation of
America
Metal Bulletin (2006) Metal Bulletin (2006) MB Prices & Archive from http://www1.metalbul-
letin.com/prices_mbprices_results.asp?mode=category
Okamoto & Takebayashi (1996) Okamoto H. and Takebayashi K. (1996) Method for Recovering Indium by
Electrowinning and Aparatus Therefor. Patent Number U.S. 5,543,031, held by
Nippon Mining & Metals Co., Ltd.
Paiva (2001) Paiva A. P. (2001) Recovery Of Indium From Aqueous Solutions By Solvent
Extraction. In: Separation Science and Technology, 36(7), pp. 1395 - 1419, On-
line-Version under: http://dx.doi.org/10.1081/SS-100103878.
Power Porto (1990) Power Porto G. F. (1990) Abtrennung von Gallium, Germanium und Indium im
Rahmen der hydrometallurgischen Zinkgewinnungsprozesse. Aachen.
Quirijnen (1999) Quirijnen L. (1999) How to implement efficient local lead-acid battery recy-
cling. In: Journal of Power Sources, 78(1-2), pp. 267-269.
Renner et al. (2002) Renner H., Schlamp G., Hollmann D., Lüschow H. M., Rothaut J., Knödler A.,
Hecht C., Schlott M., Drieselmann R., Peter C. and Schiele R. (2002) Silver, Sil-
ver Compounds, and Silver Alloys. In: Ullmann's Encyclopedia of Industrial

7. Literature
Interscience, Online-Version under: http://dx.doi.org/10.1002/14356007.

a24_107.
Reuter (2005) Reuter M. A. (2005) The metrics of material and metal ecology harmonizing the
resource, technology and environmental cycles. Elsevier, Amsterdam. ISBN 0-
444-51137-7.
Reuter et al. (2004) Reuter M. A., Boin U., Rem P., Yang Y., Fraunholcz N. and Schaik A. v. (2004)
The Optimization of Recycling: Integrating the Resource, Technological, and
Life Cycles. In: JOM, 56, pp. 33-37, Online-Version under:
http://www.ingentaconnect.com/content/tms/jom/2004/00000056/00000008/art0
0007
Roberts (2003) Roberts H. (2003) Changing patterns in global lead supply and demand. In:
Journal of Power Sources, 116(1-2), pp. 23-31.
Rodier (1990) Rodier D. D. (1990) An Overview of Silver and Trace Metal Recovery Strate-
gies in the Zinc Industry. In proceedings from: Lead-Zinc '90, Dallas, Texas.
Scharnhorst (2005) Scharnhorst W. (2005) Life-Cycle Assessment of Mobile Telephone Networks,
with Focus on the Environmental Relevance of the End-of-Life Phase. PhD the-
sis. Swiss Federal Institute of Technology, Lausanne.
Schulte-Schrepping & Piscator (2002) Schulte-Schrepping K.-H. and Piscator M. (2002) Cadmium and
Cadmium Compounds. In: Ullmann's Encyclopedia of Industrial Chemistry.
Online version, posting date: September 15, 2000 Edition. Wiley-Interscience,
Online-Version under: http://dx.doi.org/10.1002/14356007.a04_499.
Schwarz-Schampera & Herzig (2002) Schwarz-Schampera U. and Herzig P. M. (2002) Indium geology,
mineralogy, and economics. Springer, Berlin. ISBN 3-540-43135-7.
Stevens (2007) Stevens L. (2007) Indium as a Critical Metal. Indium Corporation. Retrieved
from http://dels.nas.edu/besr/docs/Stevens.pdf
Sutherland et al. (1997) Sutherland C. A., Milner E. F., Kerby R. C., Teindl H. and Melin A. (1997)
Lead. In: Ullmann's encyclopedia of industrial chemistry (ed. Anonymous). 5th
edition on CD-ROM Edition. Wiley & Sons, London.
Teckcominco (2005) Teckcominco (2005) Indium, Germanium and Cadmium Market Overview. In
proceedings from: Minor Metals, Lisbon.
Teckcominco (2006) Teckcominco (2006) Indium Applications at December from http://products.
teckcominco.com/Products/IndiumApplications.html
Teramoto et al. (1993) Teramoto M., Kariya A., Yonehara T., Tokunaga M., Ohnishi N., Matsuyama H.
and Miyake Y. (1993) Separation of Gallium and Indium by Supported Liquid
Membranes Containing 2-Bromodecanoic Acid as Carrier - Design of Supported
Liquid Membrane Module Based on Batch Permeation Experiments. In: Hydro-
metallurgy, 33(1-2), pp. 1-15.
Thornton et al. (2001) Thornton I., Rautiu R. and Brush S. (2001) Lead - the facts. IC consultants Ltd.,
London, Online-Version under: http://www.ldaint.org/factbook/index.htm.
Timokhova (2002) Timokhova M. I. (2002) Certain specifics of quasi-isostatic molding (a review).
In: Glass and Ceramics, 59(1-2), pp. 22-26, Online-Version under: <Go to
ISI>://000177012300007.
Tomii & Tsuchida (1981) Tomii K. and Tsuchida H. (1981) Solvent Extraction Recovery Process for In-
dium. Patent Number U.S. 4,292,284, held by Toho Aen Kabushiki Kaisha
USGS (2006a) USGS (2006a) Indium. In: Mineral Commodity Summaries. U.S. Geological
Survey.

7. Literature
USGS (2006b) USGS (2006b) Mineral Commodity Summaries. US. Geological Survey. Re-
USGS (2006c) USGS (2006c) U.S. salient lead statistics. U.S. Geological Survey. Retrieved
from http://minerals.usgs.gov/minerals/pubs/commodity/lead/stat/tbl1.txt
USGS (2007) USGS (2007) Cadmium. In: Mineral Commodity Summaries:. U. S. Geological
Survey, Online-Version under: http://minerals.usgs.gov/minerals/pubs/commod-
ity/cadmium/cadmimcs07.pdf.
Various Authors (1996) Various Authors (1996) Lead. In: Ullmann's Encyclopedia of Industrial Chem-
istry. Fifth Edition Edition. Wiley-Interscience, Online-Version under:
http://dx.doi.org/10.1002/14356007.a15_193.pub2.
Wikipedia Contributors (2007) Wikipedia Contributors (2007) Indium from http://en.wikipedia.org/w/in-
dex.php?title=Indium&oldid=125116630. Date of last revision: 23 April 2007
09:03 UTC.
Wikipedia Contributors (2006) Wikipedia Contributors (2006) TFT LCD. Wikipedia, The Free Encyclopedia.
Retrieved at 13 December 2006 from http://en.wikipedia.org/w/index.php?ti-
tle=TFT_LCD&oldid=93781736. Date of last revision: 12 December 2006.
Xiao et al. (2003) Xiao X., Songwen X., Xueyi G., Kelong H. and Ryoichi Y. (2003) LCA Case
Study of Zinc Hydro and Pyro-Metallurgical Process in China. In: International
Journal for Life Cycle Assessment, Online first, pp. 1 - 5.
Zhou et al. (2005) Zhou Z.-H., Ruan J.-M. and Mo H.-B. (2005) Preparation of 6N high-purity in-
dium by method of physical-chemical purification and electrorefining. In: Jour-
nal of Materials Science, 40(24), pp. 6529-6533 as pdf-File under:
10.1007/s10853-005-1817-y, Online-Version under: http://dx.doi.org/10.1007/
s10853-005-1817-y
Zhou & Ruan (2006) Zhou Z. H. and Ruan J. M. (2006) Purification of InCl3 electrolyte by ion ex-
change. In: Mineral Processing and Extractive Metallurgy : IMM Transactions
section C, 115, pp. 123-126, Online-Version under: http://dx.doi.org/10.1179/
174328506X99934

Part IX
Gold and Silver

Data v2.1 (2009)

Matthias Tuchschmid, ESU-services Ltd.
Mireille Faist Emmenegger, ESU-services Ltd.
Wolfram Scharnhorst, EMPA Dübendorf
Review: Roberto Dones, PSI, Villigen
Summary

This part corresponds to version the part in version 2.0 without changes.

The Swiss Federal Office for the Environment (FOEN) commissioned Empa in 2005 to identify the
datasets that have to be improved in order to improve the overall quality and applicability of metals
inventories in the metal processing and electronic industry. Based on that preparatory study Gold and
Silver were identified as widely applied metals with potentially high impact and therefore of high
relevance.
Summary
This part documents the inventories for primary and secondary silver and gold, respectively, as well as
the production of copper telluride cement, which is a by-product of the treatment of copper anode
slimes in order to recover silver. The inventories for primary gold are compiled country specifically
according to the environmental reports of the world leading gold producers. These inventories also
cover the production of silver from auriferous ores in a multi-output module. The rest of the primary
silver production is reflected by the processing of silver bearing by-products of lead / zinc (Parkes
crust) and copper (anode slimes) production. Secondary gold and silver production is reflected by the
recovery processes in complex integrated smelters (Boliden, SE).

Table of Contents
Table of Contents
SUMMARY ......................................................................................................................I
1 INTRODUCTION ....................................................................................................... 5

2.1 Gold....................................................................................................................................... 6
2.1.1 Gold from primary resources ................................................................................................ 6
2.1.2 Gold from secondary resources............................................................................................. 7
2.2 Silver ..................................................................................................................................... 8
2.2.1 Silver as by-product of lead/zinc, copper and gold production............................................. 8
2.2.2 Silver from old scrap............................................................................................................. 9
2.3 Allocation for Metals............................................................................................................. 9
3 CHARACTERISATION OF THE MATERIALS ................................................................. 11

3.1 Gold..................................................................................................................................... 11
3.2 Silver ................................................................................................................................... 11

5.1 Production of gold and silver from auriferous ores............................................................. 15
5.1.1 Gold and silver mining........................................................................................................ 15
5.1.2 Gold and silver production, Sweden (SE)........................................................................... 16
5.1.3 Gold and silver production, Papua New Guinea (PG) ........................................................ 27
5.1.4 Gold and silver production, Chile (CL) .............................................................................. 33
5.1.5 Gold and silver production, Peru (PE) ................................................................................ 39
5.1.6 Gold production, Canada (CA) ........................................................................................... 46
5.1.7 Gold production, United States of America (US) ............................................................... 52
5.1.8 Gold production, South Africa (ZA)................................................................................... 61
5.1.9 Gold production, Australia (AU) ........................................................................................ 65
5.1.10 Gold production, Tanzania (TZ) ......................................................................................... 71
5.1.11 Data quality for primary gold / silver production................................................................ 76
5.1.12 Primary gold, production mix (GLO) ................................................................................. 77
5.1.13 Production of primary silver from auriferous ores, production mix ................................... 77
5.2 Secondary gold production process and infrastructure ....................................................... 79
5.2.1 Waste................................................................................................................................... 82
5.2.2 Emissions ............................................................................................................................ 82
5.2.2.1 Gold / Silver, secondary, from electronic scrap in ecoinvent.............................................. 82
5.2.3 Data quality ....................................................................................................................... 102
5.3 Production of Silver from Lead-Zinc ores......................................................................... 102
5.3.1 Silver from lead production in ecoinvent.......................................................................... 102
5.3.2 Data quality ....................................................................................................................... 104
5.4 Production of silver and raw copper telluride from copper ores ....................................... 106
5.4.1 Processing of anode slime, primary copper production, in ecoinvent............................... 106
5.4.2 Data quality ....................................................................................................................... 109
5.5 Gold, at regional storage ................................................................................................... 111
5.6 Silver, at regional storage.................................................................................................. 111

Table of Contents
6 LITERATURE ........................................................................................................114

1. Introduction
1 Introduction
This life cycle inventory (LCI) of gold and silver is meant to be used as background data sets for the
preparation of life cycle assessments (LCA) of technical products such as electronic components. This
LCI includes data sets of gold mining and refining in Australia, Canada, Chile, Peru, Papua New
Guinea, South Africa, Sweden, Tanzania and the USA, as well as of the gold supply to Europe. It also
includes data sets of silver as a by-product of gold production in Chile, Peru, Papua New Guinea and
Sweden.
Data in this report are based mostly on environmental reports of large multinational companies. How-
ever, it must be assumed that these sources represent rather the best practices for gold mining. Various
NGOs like Berne Declaration, Greenpeace 1 etc. report on severe impacts of gold mining on environ-
ment, local communities or human rights concerning projects or mines of important companies like
Barrick Gold and Newmont Mining, the two largest gold companies of the world today. Controversial
issues are for example a mine project in Chile involving relocating/destroying of three glaciers in the
Andes which are of crucial importance to the Pascua Lama Highlands, or the development of a mine in
Nevada, USA in the Wester Shoshone Lands, which is expected to have a negative effect on the re-
gion’s scarce water resources and to destroy Shoshone sacred places. The nodirtygold campaign 2 de-
nounces among others the problem of tailings dam failures which release large amounts of cyanide-
laden tailings in rivers. Such impacts of the gold mining are very difficult to quantify and include in
LCI due lacking of data and also because of the punctual character of the events. Therefore it must be
assumed that the environmental impacts of gold mining are rather underestimated in the following data
sets, which consider only regular operation.
1
See also http://www.evb.ch/researchcs, http://www.evb.ch/researchubs, http:/www.greenpeace.ch, etc.
2
See http://www.nodirtygold.org

2.1 Gold
2.1.1 Gold from primary resources
“Gold abundance in the Earth’s crust is 0.002 per 106 atoms of silicon. Gold average economic abun-
dance in the ore is today in the range of 0.1 to 0.2 oz/ton or 3.42-6.84 gram per ton of ore, but also de-
posits with a tenth of those values, are exploited.” (Gasparrini, 1993)
In the last years gold was produced in following countries:
3
Tab. 2.1: Annual share of gold mine production .
2000 2001 2002 2003 2004 Average

South Africa 16.8% 15.4% 15.5% 14.7% 14.0% 15.3%
Russia 11.5% 12.1% 12.5% 12.9% 13.9% 12.5%
USA 13.7% 13.1% 11.7% 10.9% 10.6% 12.0%
Australia 11.5% 11.1% 10.7% 11.1% 10.7% 11.0%
Indonesia&PG 7.8% 9.1% 8.1% 8.0% 6.8% 8.0%
China 7.0% 7.2% 7.5% 8.0% 8.8% 7.7%
Africa without South Africa 6.1% 6.5% 7.0% 6.6% 6.7% 6.6%
Peru 5.2% 5.4% 6.2% 6.7% 7.1% 6.1%
Canada 6.1% 6.2% 6.0% 5.5% 5.3% 5.8%
South and Central America
without Peru/Brazil/Chile 5.5% 5.3% 5.1% 5.8% 5.7% 5.5%
Asia 2.3% 2.2% 2.6% 2.5% 2.6% 2.5%
Brazil 2.0% 1.7% 1.6% 1.6% 1.7% 1.7%
Chile 2.1% 1.7% 1.5% 1.5% 1.6% 1.7%
Tanzania 0.6% 1.2% 1.7% 1.9% 2.1% 1.5%
Europe (incl. Turkey) 0.9% 1.1% 1.2% 1.2% 1.4% 1.2%
New Zealand 0.4% 0.4% 0.4% 0.4% 0.3% 0.4%
Middle East 0.2% 0.2% 0.2% 0.4% 0.4% 0.3%
India 0.2% 0.1% 0.1% 0.1% 0.2% 0.2%
Pacific 0.2% 0.2% 0.2% 0.1% 0.2% 0.2%
World Production (kg) 2'567'348 2'563'867 2'548'471 2'548'693 2'431'367 2'532'671
The total volume of gold production is rather stable. However, substantial changes occurred by the
country. South Africa’s production diminished between 2000 and 2004, whereas production in Russia,
China and Peru increased considerably.
This life cycle inventory is based on data from Australia, Canada, Chile, Peru, Papua New Guinea,
South Africa, Sweden, Tanzania and the United States of America. The data set „gold primary, at re-
finery“ takes into account the average production shares 2000-2004 of these countries (Tab. 2.2). Pro-
duction in African countries is approximated by South African and Tanzanian production, production
in South and Central America by Chilean and Peruvian production, production in Asia (incl. Middle
East) and Pacific by production in Papua New Guinea, and production in Europe by Swedish produc-
tion. The production from Russia / former USSR is approximated with Canadian production, due to
similar geographical and meteorological conditions.
3
http://www.indexmundi.com/en/commodities/minerals/gold/gold_table08.html.

Tab. 2.2: Shares of the countries in mine production and the respective share in data set „gold primary, at refinery“
Average
2000-2004 This project Remark
South Africa
15.3% 18.6% incl. half of African countries
USA
12.0% 12.0%
Australia 11.0% 11.4% incl. New Zealand
Canada 5.8% 18.4% incl. Russia
Papua New Guinea 2.7% 18.8% Pacific, Asia, Middle East, excluding Russia
Chile 1.7% 5.3% incl. half of South- and Central American countries
Peru 6.1% 9.7% incl. half of South- and Central American countries
Tanzania 1.5% 4.8% incl. half of African countries
Sweden 0.2% 1.2% Europe (incl. Turkey)
Rest of the World 46%
Total 100% 100%
2.1.2 Gold from secondary resources

Among other precious metals, such as silver, palladium, platinum, etc., gold is contained in varying
amounts in metallic scrap of any kind. Depending on the application, for instance printed wiring board
assemblies may contain between 0.03 wt% (Goosey & Kellner, 2002) and 0.089 wt% (Motorola,
2005). Other authors report gold contents ranging from 0.035 wt% up to 0.18 wt% (Zhang et al.,
2000). Other sources of secondary gold are: scrap jewellery, dental applications, plated rejected com-
ponents, porcelain scrap, and mirrors (George, 2006; Syed, 2006).
In the U.S. the production of secondary gold has increased from 16.1 tonnes in 1964 to 45.0 tonnes in
2004 and thus accounts for about 14.8 % of the total gold production (Geoffroy & Cardarelli, 2005;
George, 2006). Due to the scarcity of the metal and because of the drastically increased demand, the
price of gold (and secondary gold) has risen dramatically since the late 1970ies.
25000000
Gold price
20000000
15000000
Gold [$/t]
10000000
5000000
0
1880 1900 1920 1940 1960 1980 2000 2020
Time [years]
Fig. 1 Value of gold (adapted from George (2006)).
Key sources for the refinement of secondary gold, however, are new and old metallic scrap. In 2000 it
was estimated that about 280 tonnes of gold were present in electronic goods (Holliday & Goodman,
2002). Gold can be extracted from the scrap in a number of subsequent steps (depending on the re-
quired degree of purity) with a purity between 90.0 % (Doré) and 99.99 % (Boliden Contech A.B.,
1995). Again, depending on the intended application of the final product, the recovered gold is either
added as bullion or higher grade secondary gold to primary gold. Along with the extraction of gold

from new and old electronic scrap, typically silver, palladium and (in case) selenium are co-extracted
(Boliden, 2000; Boliden, 2002a; Boliden Contech A.B., 1995).
2.2 Silver
The extraction of silver from true silver ores – in which silver is practically the only valuable metal –
is by far not the biggest source for supply. Depending on the price of silver, it can be cost-effective to
process materials containing 0.01 % (100 ppm) silver, or even less. The most important primary
sources of silver are traditionally argentiferous lead and lead-zinc ores that are widely distributed.
They constitute the most important fraction of the mine production. However, since the mine produc-
tion of lead declined in the last years, these deposits became less important for the worldwide supply.
The separation and recovery of silver during lead production is profitable even for a silver content of a
few ppm. Argentiferous copper ores, which sometimes also contain nickel and cobalt, satisfy another
important share of the world silver demand. Even very low contents of silver can be recovered profita-
bly from copper ores, since almost all the copper is electrorefined, and the silver remains in the anode
slimes. The largest argentiferous copper reserves in the world are in Canada (Sudbury, Ontario, No-
randa-Rouyn, Quebec). Extraction of silver from gold ores is carried out mainly in the Republic of
South Africa, and also in Canada, Alaska, and Sweden. The quantity obtained from these deposits is
around 10-20% of the world mine production. Silver-containing tin deposits are concentrated in Bo-
livia (Cerro de Pasco, Potosi) but also occur in Burma and Argentina. Silver produced from tin ores
amounts to only 2 % of world silver production. (Renner et al., 2002b).
2.2.1 Silver as by-product of lead/zinc, copper and gold production

Most of the world silver is produced as by-product from lead/zinc, copper and gold bearing deposits.
The share of the different ore deposits that constitute the world mine production changes over the
years (Tab. 2.3). Recently, the Silver Institute (2002, p. 26) estimates the breakdown by ore type as
follows: 25 % silver ores, 15% gold ores, 24 % copper ores, 35% lead, zinc, and lead-zinc ores and 1%
other ores (Butterman & Hilliard, 2004, JMB, 2002) Nigel H Maund estimated the supply in 2004 as
follows 4 : Lead - Zinc deposits (mostly Mexico and Cannington, and smaller Chinese contributions)
31%; Copper deposits (Kupferschiefer of Poland and porphyry copper mines) 25%; Gold deposits (in
particular Carlin type) 14% Primary silver deposits (in particular "Idaho type") 28%.
This LCI corresponds to the available data from Klapwijk ((2006, p.87)) as shown in Tab. 2.4, that
specifies the share of the origin countries for the data set „ silver mix, from combined gold / silver
production, at refinery“.
4
“Silver – An Economic Geologist Perspective”, http://www.gold-eagle.com/editorials_04/nmaund072004.html

Tab. 2.3 Silver supply differentiated into sources.
Source Renner et JMB (2002) Nigel H Klapwijk et Primary produc- Mix primary and
5
al. (2002b) (Silver insti- Maund al. (2006) tion assumed in secondary
tute) this study
year 1996 2002 2004 2006
Lead / Zinc 50% 35% 31% 31% 24%
Copper 30% 24% 25% 25% 20%
Silver 10% 25% 28% 28% 22%
Gold 10% 15% 14% 14% 11%
other 1% 2% 2% 2%
Secondary 21% 21%
Mine 70%
Hedging 2%
Gov sales 7%
Tab. 2.4 Shares of the countries in the data set „silver mix, from combined gold / silver production, at regional stor-
age“, Source: Klapwijk et al. (2006)
This
% project: Remark
incl. Central & South america
Central & South America Chile 28.8% 32.0% Chile excluding Peru
Peru 13.1% 31.3% Peru incl. half of USA/Canada

others 3.2%
Asia 14.8% 14.8% Papua New Guinea incl. Asia
North America USA 14.4%
Canada 22.0%
others 3.7% 21.9% Sweden incl. half of USA/Canada
World production (kg) 2.36E+06 100.0%
2.2.2 Silver from old scrap

According to the Silver Institute 21% of the silver demand was satisfied from old scrap (Klapwijk et
al., 2006) (as Secondary in Tab. 2.3).
2.3 Allocation for Metals

Metal prices are used as the allocation parameter for polymetallic mines. Prices in USD of copper,
lead, zinc, gold and silver stem from Boliden for the year of 1996 and 2001, the prices for 2005 are
from USGS (2006). The prices per tonne in the yearly average 2004-2006 are compiled from Kelly &
Matos (2007) through 2004 and the respective 2007 mineral commodity summaries from USGS
(2006) for values after 2004. These prices have been used in this project for the allocation of different
metals.
5
“Silver – An Economic Geologist Perspective”, http://www.gold-eagle.com/editorials_04/nmaund072004.html

Prices were relatively stable between 1996 and 2001, whereas a strong demand in the years 2005 to
2006 lead to a sharp increase in 2006 (see Tab. 2.5). Relative shares remained about constant.
Tab. 2.5: Price of gold, silver, copper, lead and zinc 1996, 2001, 2005 and the average prices 2004-2006, used in this
project (sources see text).
Prices of metals Average 2004-2006, used

1996 2001 2005 in this project
US$ / kg US$ / kg US$ / kg US$ / kg
Prices of metals 10'714 8'682 14'295 15'682
Silver 157 140 239 269
Gold 2.1 0.8 1.4 1.9
Silver 0.7 1.5 3.7 4.5
Zinc 1.1 0.5 1.0 1.2


For gold and silver the term precious metals is used because of their rarity and corrosion resistance.
Gold and silver are often mined together in polymetallic mines.
3.1 Gold
Apart from copper, Gold is the only metallic element with a colour different from a grey tone. It is a
soft yellow metal with a moderately high melting point (1064°C) and a high electrical and thermal
conductivity exceeded only among the elements by silver and copper. It is the most malleable and duc-
tile of the metals. One ounce of gold can be beaten in to foil some 300 square feet in area. Tab. 3.1
displays some physical properties of the element gold.
Tab. 3.1 Some physical properties of gold (Renner et al., 2002a).
Property Value
Melting Point 1064.43°C
Boiling Point 2808°C
Density at 20°C 19.32 g/cm3
Density at 1065°C 17.32 g/cm3
Vapor pressure at 1064°C 0.002 Pa
Atomic volume at 20°C 10.21 cm3/mol
Electrical resistivity at 0°C 2.06×10–6 Ω cm
Thermal conductivity at 0°C 3.14 Wcm–1K–1
Specific heat 0.138 Jg–1K–1
Enthalpy of fusion 12.77 kJ/mol
Enthalpy of vaporization 324.4 kJ/mol
Tensile strength 127.5 N/mm2
3.2 Silver
“Silver has a moderately high melting point (961.9°C) and the highest electrical and thermal conduc-
tivity of all metals. When in the metallic form, silver is unaffected by water or oxygen, but is altered
by ozone with the formation of silver oxide, and by sulphur and other sulphur compounds with the
formation of silver sulphide” (Gasparrini, 1993). Tab. 3.2 summarises some physical properties of ele-
mental silver.
Tab. 3.2 Some physical properties of silver (Renner et al., 2002b).
Physical Property Value

Melting point 961.9 ◦C
Boiling point 2212 ◦C
Specific heat capacity at 25 ◦C 0.23 J kg−1 K−1
Thermal conductivity 418 W m−1 K−1
Vapor pressure at 1030 ◦C 1.33 Pa
Density at 20 ◦C 10.49 g/cm3
Density of liquid at mp 9.35 g/cm3
Brinell hardness 26
Modulus of elasticity 82 000 MPa
Tensile strength 140 MPa
Resistivity at 0 ◦C 1.465 Ωcm


Gold is used mainly for jewellery, electronics and other industrial and decorative appliance (Tab. 4.1).
In electric/electronic devices, it is mostly used as connector or as solder paste (Goosey & Kellner,
2002). The latter one is typically a gold alloy incorporating Si (98%Au, 2%Si), Ge (87.5%Au,
12.5%Ge), In (82%Au, 18%In) or Sn (80%Au, 20%Sn). The principal users of silver are industrial ap-
plications, photography and jewellery (Fig. 4.1).
6
Tab. 4.1 World Gold demand in different application fields (World Gold Council ).
7
Fig. 4.1 World silver demand in different application fields (Silver Institute ).
6
http://www.gold.org/ accessed in july 2007
7
http://www.silverinstitute.org/supply/, accessed in july 2007

The production of silver is heavily interlaced with the production of other metals, namely lead and
zinc, copper and gold. This is taken into account by establishing a complex LCI model (Fig. 5.1, for
supply mix see chapter 5.6 Silver, at regional storage). However, not every source contributing to the
worldwide production of gold and silver could be evaluated in ecoinvent version 2.0.
Due to a considerable lack of systematic information on processes in the artisanal mining of gold, no
inventory on these processes could be established. The production of silver as main product – which
accounts for 22% of the world supply – has been disregarded in this study for two reasons: first, a
coverage of 78% seems fair enough to represent silver. Second, the production of silver as main-
product occurs with the same technologies as the production of silver as by-product and thus can be
approximated by these datasets. An overview over the system boundaries chosen for the inventories in
this chapter is given in Fig. 5.2.
Silver Gold
Mining of gold Artisanal mining
Primary silver Primary silver bearing ores of gold
from copper from lead and
production lead-zinc cycle
Primary silver
Primary silver from from gold bearing
argentiferous ores 20% ores
24% 38%
13%
11%
22%
Recycling of
Silver metal scrap containing Gold metal
21% precious metals 50%
Silver Gold powder, from

Gold compounds melt atomisation
Compounds
galvanic Gold powder, from
applications precipitation
Silver powder for solder pastes, sintered films, Gold powder for
powdermetallurgy p/m alloys powdermetallurgy
Fig. 5.1 General overview of Silver and Gold production.

mining, gold-silver- primary mining and gold, at

zinc-lead-copper refining, gold-silver beneficiation
deposit deposit
SE
PG, CL, PE CA, US, ZA, AU, TZ
gold, at beneficiation
silver, at silver, from combined gold-silver gold, from combined gold-silver

beneficiation production, at refinery production, at refinery
refining, copper-
gold, at precious metal gold, at refinery
zinc-lead-gold-
silver, in smelter plant
SE CA, US, ZA, PG,

CL, TZ, AU, PE
silver, from
combined gold-
silver production,
silver, at precious at refinery
metal plant PG, CL
silver, primary, from

copper production, at
refinery
secondary
secondary copper gold, primary, at
silver, from conversion lead, secondary, from refinery
combined gold- electronic and electric scrap
silver production, recycling, at plant CA, US, ZA, PG,
at refinery (GLO) CL, TZ, AU, PE
silver, primary, from metal values from electric waste,

lead/zinc production, in blister-copper, at converter
at refinery
secondary copper nickel, secondary, from electronic and
refining gold, primary, at
electric scrap recycling, at refinery
refinery
silver, at regional copper, secondary, from electronic and
storage electric scrap recycling, at refinery GLO
disposal, building, fibre board, to

final disposal
precious metal
silver,
refining, secondary
secondary, at gold, secondary, at gold, at regional
copper
precious precious metal refinery storage
metal refinery
palladium, secondary, at precious

metal refinery
Fig. 5.2 System lay-out in ecoinvent data v2.0. Boxes with rounded edges indicate reference flows from multi-output
processes.

5.1 Production of gold and silver from auriferous ores

5.1.1 Gold and silver mining
The description of gold production is based on Newmont (2004).
Exploration
The gold mining process begins with geologists discovering a deposit. After finding indications of
gold, drill rigs are used to obtain samples from below the surface. These samples are analyzed for their
gold content. If there is enough gold in the deposit to be produced economically, a mine is permitted
and developed, if environmental regulations are fulfilled.
Open Pit Mining
The ore is mined in four steps: drilling, blasting, loading and hauling. In the case of a surface
mine, a pattern of holes is drilled in the pit and filled with explosives. The explosives are deto-
nated in order to break up the ground so large shovels or front-end loaders can load it into haul
trucks.
Underground Mining
Some ore bodies are more economically mined underground. In this case, a tunnel called an adit or
a shaft is dug into the earth. Sort tunnels leading from the adit or shaft, called stopes, are dug to
access the ore. The surface containing the ore, called a face, is drilled and loaded with explosives.
Following blasting, the broken ore is loaded onto electric trucks and taken to the surface. Once
mining is completed in a particular stope, it is backfilled with a cement compound.
Ore and Waste Haulage
The haul trucks transport the ore to various areas for processing. The grade and type of ore deter-
mine the processing method used. Higher-grade ores are taken to a mill. Lower grade ores are
taken to leach pads. Some ores may be stockpiled for later processing.
Heap leaching
Heap leaching
The ore is crushed or placed directly on lined leach pads where a dilute cyanide solution is applied
to the surface of the heap. The solution percolates down through the ore, where it leaches the gold
and flows to a central collection location. The solution is recovered in this closed system. The
pregnant leach solution is fed to electrowinning cells and undergoes the same steps as described
below from Electro-winning.
Ore Processing
Milling
The ore is fed into a series of grinding mills where steel balls grind the ore to a fine slurry or pow-
der.
Oxidization and leaching
Some types of ore require further processing before gold is recovered. In this case, the slurry is
pressure-oxidized in an autoclave before going to the leaching tanks or a dry powder is fed
through a roaster in which it is oxidized using heat before being sent to the leaching tanks as a
slurry. The slurry is thickened and runs through a series of leaching tanks. The gold in the slurry
adheres to carbon in the tanks.
Stripping
The carbon is then moved into a stripping vessel where the gold is removed from the carbon by
pumping a hot caustic solution through the carbon. The carbon is later recycled.

Electro-winning
The gold-bearing solution is pumped through electro-winning cells or through a zinc precipitation
circuit where the gold is recovered from the solution.
Smelting
The gold is then melted in a furnace at about 1’064°C and poured into moulds, creating doré bars.
Doré bars are unrefined gold bullion bars containing between 60% and 95% gold.
Refining
The bars are sent to a refinery for further processing into pure gold.
5.1.2 Gold and silver production, Sweden (SE)

The following data on gold and silver origin from the reports of the Swedish company Boliden
(2006a), that operates mines and smelters in Sweden, Finland, Norway, Ireland and the Netherlands
(copper, zinc, lead, gold and silver production). This LCI includes the data of the Swedish mines
Aitik, Boliden and Garpenberg, which produce gold, silver and other metals.
Mining infrastructure
This data set includes land use as well as rough estimates on buildings and machines necessary in gold
mining. The open pit mine “Aitik” produces the large part the ore. Therefore data for buildings and
machines are based on the (generic) data set “non-ferrous metal, mine, surface” (and not “non-ferrous
metal, mine, surface underground”). For details about infrastructure in mining, see the corresponding
section of this report. It is assumed, that the quantities of machinery as well as surface and volume of
buildings are equal to the generic mine. This approach is rough, and the uncertainty is rather high. As
the mined ore from the generic mine is some orders of magnitude lower than the mine in focus, a pro-
portional scaling by milled ore will lead to a higher mistake (Tab. 5.1). The lifetime of mines is as-
sumed to be 50 years.
Tab. 5.1: Quantity of machinery as well as surface and volume of buildings for a “non-ferrous metal mine, surface”,
Source: see corresponding section in this report
Dataset Unit Quantitiy per mine

(lifetime 50y)
infrastructure, mine, surface

3
building, multi-storey m 2.75E+5
2
building, hall m 1.25E+4
industrial machine, heavy, unspecified, at plant kg 8.50E+5
conveyor belt, at plant m 6.00E+2
Land use is based on the aggregated figures from the environmental report (Tab. 5.2). Not the entire
area of the mine is in active use at one time; the value in Tab. 5.2 is assumed to be the average share of
the mine area in use (see line “Disturbed area in active use”). The total area is occupied during an av-
erage of about 11 years according to the following formula:
Disturbed area in active use 4.65 * 10 7

* lifetime of mine  * 50 y  11.6 y , the occupation of the mine is
Area disturbed during life of mine 1.98 * 108
calculated as Disturbed area in active use *50 y  Occupation, mineral extraction site

Tab. 5.2 Land use of the Aitik, Boliden Area and Garpenberg Mines in 2005 (Boliden, 2006d)
Land use Aitik Boliden Area Garpenberg Sweden Remark:
Type of mine: open pit underground, underground

open pit
Area disturbed during m2 n.a. n.a. n.a. 1.98E+08 Transformation, from unknown and
life of mine Transformation, to mineral
extraction site
Disturbed area in m2 n.a. n.a. n.a. 4.65E+07 Occupation, mineral extraction site,
active use during 50 years
Area available for m2 n.a. n.a. n.a. 4.53E+07 added to "Transformation, to
revegetation unknown"
Area revegetated m2 n.a. n.a. n.a. 1.19E+06 Transformation, to heterogeneous,
agricultural
Tab. 5.3 Unit process raw data of the polymetallic mines in Sweden
UncertaintyType
StandardDeviati
InfrastructurePr
Location
mine, gold-
on95%
ocess
Unit
Name silver-zinc- GeneralComment
lead-copper
Location SE
Unit unit
product mine, gold-silver-zinc-lead-copper SE 1 unit 1.00E+0
technosphere building, multi-storey RER 1 m3 2.75E+5 1 2.39 (4,5,5,5,5,5); Rough estimates
building, hall CH 1 m2 1.25E+4 1 2.39 (4,5,5,5,5,5); Rough estimates
industrial machine, heavy, unspecified, at plant RER 1 kg 8.50E+5 1 2.39 (4,5,5,5,5,5); Rough estimates
conveyor belt, at plant RER 1 m 6.00E+2 1 2.39 (4,5,5,5,5,5); Rough estimates
(1,3,2,1,1,3); Environmental
resource, land Transformation, from unknown - - m2 1.98E+8 1 2.01
report, Boliden 2005
Transformation, to mineral extraction site - - m2 1.98E+8 1 2.01
Occupation, mineral extraction site - - m2a 2.33E+9 1 1.51
Transformation, from mineral extraction site - - m2 1.98E+8 1 2.01
Transformation, to heterogeneous, agricultural - - m2 1.19E+6 1 2.01
Transformation, to unknown - - m2 1.97E+8 1 2.01
Mining
The energy and water requirements as well as the emissions are allocated to the different metals on the
basis of their economic value calculated with the prices shown in Tab. 2.5. Tab 5.4 summarizes the
concentrate production (expressed in metal quantity) of the considered plants for the year 2005.

Tab 5.4: Production (quantity of metals) in year 2005 of the mines Aitik, Bolden and Garpenberg mines (Boliden,
2006a)
Production Aitik Boliden Area Garpenberg Sweden

Ore: kg 2.04E+10 1.69E+09 1.10E+09 2.32E+10
Production Gold kg 1.84E+03 2.43E+03 2.03E+02 4.47E+03 Price per kg: 15682.1 US $
Production Silver kg 4.13E+04 8.72E+04 9.76E+04 2.26E+05 Price per kg: 268.7 US $
Production Zinc kg 8.73E+07 5.84E+07 1.46E+08 Price per kg: 1.9 US $
Production Copper kg 6.56E+07 2.07E+07 5.65E+05 8.69E+07 Price per kg: 4.5 US $
Production Lead kg 3.04E+06 2.07E+05 3.25E+06 Price per kg: 1.2 US $
Head grade
Gold kg/kg ore 2.02E-06 5.17E-05 8.87E-05 1.93E-07
Silver kg/kg ore 5.44E-04 1.39E-02 2.38E-02 9.75E-06
Zinc kg/kg ore n.a. 5.17E-02 5.31E-02 6.28E-03
Copper kg/kg ore 3.22E-03 1.23E-02 5.14E-04 3.75E-03
Lead kg/kg ore n.a. 1.80E-03 1.88E-04 1.40E-04
Assumption, based on global
Recovery rate of metals from ore 90% data
The recovery rate (proportion of valuable material physically recovered in the processing of ore 8 ) is
taken into account in order to define the resource extraction. It is estimated 90% on the basis of global
data. The average concentration (head grade, expressed in kg metal per kg ore) of metals in ore for
Sweden is calculated with the quantity of ore milled and the corresponding production rate.
The turnover of the production sites (Table 5.5) is calculated with the metal quantity multiplied with
the prices (2005) in Tab. 2.5. The share in turnover is used as allocation factor for the respective metal.
Table 5.5: Calculated turnover (prices 2005) and derived allocation factors for the Swedish mines
Aitik Boliden Area Garpenberg Sweden Allocation factor SE

Production Gold US $ 2.89E+07 3.81E+07 3.18E+06 7.01E+07 8.8 %
Production Silver US $ 1.11E+07 2.34E+07 2.62E+07 6.07E+07 7.6 %
Production Zinc US $ 1.64E+08 1.10E+08 2.73E+08 34.3 %
Production Copper US $ 2.94E+08 9.28E+07 2.53E+06 3.89E+08 48.8 %
Production Lead US $ 3.50E+06 2.38E+05 3.74E+06 0.5 %
Total US $ 3.34E+08 3.22E+08 1.42E+08 7.97E+08 100 %
Energy and water requirements as well as emissions of the considered mines are summed up in Table
5.6. Water emissions are assumed to be emitted to rivers.
8
The recovery rate is generally stated as a percentage of the material recovered compared to the total material originally
present ([Barrick, 2007 #2767]).

Table 5.6: Total energy, material requirements, and emissions of the Swedish multi-metal mines in 2005 (Boliden,
2006a); n.a. = not available
Energy Aitik Boliden Area Garpenberg Sweden Remark:

Energy consumption kWh 4.32E+08 2.41E+08 1.23E+08 7.97E+08 Share is calculated with
global percentage of Boliden-
Group
Electricity kWh n.a. n.a. n.a. 5.74E+08 SE, electricity mix
Coal/Coke MJ n.a. n.a. n.a. 3.15E+08 in LCI: hard coal, burned in
industrial furnace
Diesel (mobile MJ n.a. n.a. n.a. 1.72E+08 Diesel, burned in building
equipment) machine
Natural Gas MJ n.a. n.a. n.a. 8.60E+07 in LCI: natural gas, burned in
industrial furnace, >100kW
Oil MJ n.a. n.a. n.a. 1.15E+08 heavy fuel oil, burned in

industrial furnace 1MW, non-
modulating
Materials
Cyanide kg n.a. n.a. n.a. 2.32E+05 estimated with US-Data,
based on processed Gold, in
LCI as sodium cyanide
Lime kg n.a. n.a. n.a. 5.18E+05 estimated from Steward
(1999) by processed gold and
silver, accounted for as
limestone, milled
Caustic kg n.a. n.a. n.a. 9.54E+04 estimated from Steward
silver, accounted for sodium
hydroxide
Carbon kg n.a. n.a. n.a. 1.39E+05 estimated from Steward
silver, accounted for as
Charcoal, at plant
Zinc kg n.a. n.a. n.a. 2.37E+04 Scaled from Steward (1999)
by processed gold, accounted
for as zinc, primary, at
regional storage
Sulphuric Acid kg n.a. n.a. n.a. 5.45E+03 estimated from Steward
silver
Hydrochloric Acid kg n.a. n.a. n.a. 1.12E+05 estimated from Steward
silver
Explosives kg n.a. n.a. n.a. 2.07E+07 Share of Tara-mines is
excluded, calculated by milled
ore, partly ANFO explosives,
in LCI: "blasting"
Water resources
Water use m3 1.00E+06 5.25E+06 6.27E+05 6.87E+06 Share is calculated with global
percentage of Boliden-Group
Seawater m3 n.a. n.a. n.a. 3.51E+06 In LCI as water, salt, ocean
Surface Water Use m3 n.a. n.a. n.a. 1.58E+06 In LCI as Water, river
Groundwater Use m3 n.a. n.a. n.a. 2.75E+05 In LCI as Water, well, in
ground
Municipal Water m3 n.a. n.a. n.a. 6.87E+04 in LCI as tap, water
Recirculated Water m3 n.a. n.a. n.a. 1.44E+06 recycled Water, not
accounted for in inventory
Emissions to Water
Zinc kg 8.80E+01 1.70E+03 3.76E+02 2.16E+03
Copper kg 8.20E+01 1.55E+02 1.60E+01 2.53E+02
Lead kg 8.00E-01 9.00E+00 1.26E+02 1.36E+02
Emissions to Air
Total based on Aitik-Mine,
Zinc kg 4.50E+00 n.a. 0.00E+00 5.11E+00 scaled by total milled ore
Copper kg 1.63E+01 n.a. 0.00E+00 1.85E+01 scaled by total milled ore
Lead kg 5.30E+00 n.a. 0.00E+00 6.02E+00 scaled by total milled ore
Carbon Dioxide Calculation, based on data
(from explosives) kg n.a. n.a. n.a. 3.30E+06 from Newmont (2005)

The unit process raw data of metal mining in Sweden are summarized in Tab. 5.7 and Tab. 5.8. Cya-
nide consumption is estimated with the average of US-Data from Barrick (2006b) and Newmont
(2005b) (scaled by the produced gold), other material requirements are estimated with Australian data
(scaled by the produced gold and silver). Material requirements are estimated only for the gold and
silver production. Since the module “blasting” does not include the CO2 emissions from explosives,
they are calculated with the factor of the mines of Newmont (2005b) (0.16 kg CO2 per kg explosive).
Transport of materials is calculated on the base of standard distances (Frischknecht et al., 2004). Data
about tailings disposal is approximated with data from Papua New Guinea.
Tab. 5.7: Unit process raw data of the metal mining in Sweden
InfrastructureProcess
StandardDeviation95
UncertaintyType
Location
mining, gold-
Unit
Name silver-zinc-lead- GeneralComment
%
copper deposit
Location SE
Unit a
allocated gold, from combined metal production, at beneficiation SE 0 kg 4.47E+3
products silver, from combined metal production, at beneficiation SE 0 kg 2.26E+5
copper, from combined metal production, at beneficiation SE 0 kg 8.69E+7
lead, from combined metal production, at beneficiation SE 0 kg 3.25E+6
zinc, from combined metal production, at beneficiation SE 0 kg 1.46E+8
technosphere electricity, medium voltage, at grid SE 0 kWh 5.74E+8 1 1.14 (2,4,1,3,1,3); Environmental report
hard coal, burned in industrial furnace 1-10MW RER 0 MJ 3.15E+8 1 1.14 (2,4,1,3,1,3); Environmental report
diesel, burned in building machine GLO 0 MJ 1.72E+8 1 1.14 (2,4,1,3,1,3); Environmental report
natural gas, burned in industrial furnace >100kW RER 0 MJ 8.60E+7 1 1.14 (2,4,1,3,1,3); Environmental report
heavy fuel oil, burned in industrial furnace 1MW, non-modulating RER 0 MJ 1.15E+8 1 1.14 (2,4,1,3,1,3); Environmental report
(1,3,3,5,3,3); Calculation, based about
sodium cyanide, at plant RER 0 kg 2.32E+5 1 1.27
average US-data
limestone, milled, packed, at plant CH 0 kg 5.18E+5 1 1.27 thesis on Australian precious metal
production, data for 1999
sodium hydroxide, 50% in H2O, production mix, at plant RER 0 kg 9.54E+4 1 1.27 thesis on Australian precious metal
zinc, primary, at regional storage RER 0 kg 2.37E+4 1 1.27 thesis on Australian precious metal
charcoal, at plant GLO 0 kg 1.39E+5 1 1.27 thesis on Australian precious metal
sulphuric acid, liquid, at plant RER 0 kg 5.45E+3 1 1.27 thesis on Australian precious metal
blasting RER 0 kg 2.07E+7 1 1.14 (2,4,1,3,1,3); Annual report
hydrochloric acid, 30% in H2O, at plant RER 0 kg 1.12E+5 1 1.27 thesis on Australian precious metal
(4,5,na,na,na,na); Calculation, based
transport, lorry >16t, fleet average RER 0 tkm 2.35E+6 1 2.09
about standard distances
(4,5,na,na,na,na); Calculation, based
transport, freight, rail RER 0 tkm 1.92E+7 1 2.09
about standard distances
mine, gold-silver-zinc-lead-copper SE 1 unit 2.00E-2 1 3.02 (2,4,1,3,1,3); Environmental report
tap water, at user RER 0 kg 6.87E+4 1 1.14 (2,4,1,3,1,3); Environmental report
disposal, sulfidic tailings, off-site GLO 0 kg 7.28E+8 1 1.14 (2,4,1,3,1,3); Environmental report
emission air, low
Copper - - kg 1.85E+1 1 5.02 (2,4,1,3,1,3); Environmental report
population density
Carbon dioxide, fossil - - kg 3.30E+6 1 1.14 (2,4,1,3,1,3); Environmental report
Lead - - kg 6.02E+0 1 5.02 (2,4,1,3,1,3); Environmental report
Zinc - - kg 5.11E+0 1 5.02 (2,4,1,3,1,3); Environmental report
(2,4,1,3,1,3); Environmental report,
Heat, waste - - MJ 2.06E+9 1 1.14
Calculation
emission water, river Copper, ion - - kg 2.53E+2 1 3.02 (2,4,1,3,1,3); Environmental report
Zinc, ion - - kg 2.16E+3 1 5.02 (2,4,1,3,1,3); Environmental report
resource, in water Water, salt, ocean - - m3 3.51E+6 1 1.14 (2,4,1,3,1,3); Environmental report
Water, well, in ground - - m3 2.75E+5 1 1.14 (2,4,1,3,1,3); Environmental report
Water, river - - m3 1.58E+6 1 1.14 (2,4,1,3,1,3); Environmental report
Gold, Au 9.7E-4%, Ag 9.7E-4%, Zn 0.63%, Cu 0.38%, Pb 0.014%, in
resource, in ground - - kg 4.97E+3 1 1.14 (2,4,1,3,1,3); Environmental report
ore, in ground
Silver, Ag 9.7E-4%, Au 9.7E-4%, Zn 0.63%, Cu 0.38%, Pb 0.014%, in
- - kg 2.51E+5 1 1.14 (2,4,1,3,1,3); Environmental report
ore, in ground
Zinc, Zn 0.63%, Au 9.7E-4%, Ag 9.7E-4%, Cu 0.38%, Pb 0.014%, in
ore, in ground
Copper, Cu 0.38%, Au 9.7E-4%, Ag 9.7E-4%, Zn 0.63%, Pb 0.014%,
in ore, in ground
Lead, Pb 0.014%, Au 9.7E-4%, Ag 9.7E-4%, Zn 0.63%, Cu 0.38%, in
ore, in ground

Tab. 5.8: Allocation factors applied on mining data of the different metals. Material requirements were calculated on
the basis of average data for gold and silver processing and extrapolated with the mined quantity of gold
and silver. They are therefore only allocated to these two metals
gold, from silver, from copper, from lead, from zinc, from
Location
mining, gold-
combined metal combined metal combined metal combined metal combined metal
Unit
Name silver-zinc-lead-
production, at production, at production, at production, at production, at
copper deposit
beneficiation beneficiation beneficiation beneficiation beneficiation
Location SE SE SE SE SE SE
InfrastructureProcess 0 0 0 0 0 0
Unit a kg kg kg kg kg
allocated gold, from combined metal production, at beneficiation SE 0 kg 4.47E+3 100 - - - -
products silver, from combined metal production, at beneficiation SE 0 kg 2.26E+5 - 100 - - -
copper, from combined metal production, at beneficiation SE 0 kg 8.69E+7 - - 100 - -
lead, from combined metal production, at beneficiation SE 0 kg 3.25E+6 - - - 100 -
zinc, from combined metal production, at beneficiation SE 0 kg 1.46E+8 - - - - 100
technosphere electricity, medium voltage, at grid SE 0 kWh 5.74E+8 8.8 7.6 48.8 0.5 34.3
hard coal, burned in industrial furnace 1-10MW RER 0 MJ 3.15E+8 8.8 7.6 48.8 0.5 34.3
diesel, burned in building machine GLO 0 MJ 1.72E+8 8.8 7.6 48.8 0.5 34.3
natural gas, burned in industrial furnace >100kW RER 0 MJ 8.60E+7 8.8 7.6 48.8 0.5 34.3
heavy fuel oil, burned in industrial furnace 1MW, non-modulating RER 0 MJ 1.15E+8 8.8 7.6 48.8 0.5 34.3
sodium cyanide, at plant RER 0 kg 2.32E+5 54.0 46.0 - - -
limestone, milled, packed, at plant CH 0 kg 5.18E+5 54.0 46.0 - - -
sodium hydroxide, 50% in H2O, production mix, at plant RER 0 kg 9.54E+4 54.0 46.0 - - -
zinc, primary, at regional storage RER 0 kg 2.37E+4 54.0 46.0 - - -
charcoal, at plant GLO 0 kg 1.39E+5 54.0 46.0 - - -
sulphuric acid, liquid, at plant RER 0 kg 5.45E+3 54.0 46.0 - - -
blasting RER 0 kg 2.07E+7 8.8 7.6 48.8 0.5 34.3
hydrochloric acid, 30% in H2O, at plant RER 0 kg 1.12E+5 54.0 46.0 - - -
transport, lorry >16t, fleet average RER 0 tkm 2.35E+6 8.8 7.6 48.8 0.5 34.3
transport, freight, rail RER 0 tkm 1.92E+7 8.8 7.6 48.8 0.5 34.3
mine, gold-silver-zinc-lead-copper SE 1 unit 2.00E-2 8.8 7.6 48.8 0.5 34.3
tap water, at user RER 0 kg 6.87E+4 8.8 7.6 48.8 0.5 34.3
disposal, sulfidic tailings, off-site GLO 0 kg 7.28E+8 8.8 7.6 48.8 0.5 34.3
emission air, low
Copper - - kg 1.85E+1 8.8 7.6 48.8 0.5 34.3
population density
Carbon dioxide, fossil - - kg 3.30E+6 8.8 7.6 48.8 0.5 34.3
Lead - - kg 6.02E+0 8.8 7.6 48.8 0.5 34.3
Zinc - - kg 5.11E+0 8.8 7.6 48.8 0.5 34.3
Heat, waste - - MJ 2.06E+9 8.8 7.6 48.8 0.5 34.3
emission water, river Copper, ion - - kg 2.53E+2 8.8 7.6 48.8 0.5 34.3
Lead - - kg 1.36E+2 8.8 7.6 48.8 0.5 34.3
Zinc, ion - - kg 2.16E+3 8.8 7.6 48.8 0.5 34.3
resource, in water Water, salt, ocean - - m3 3.51E+6 8.8 7.6 48.8 0.5 34.3
Water, well, in ground - - m3 2.75E+5 8.8 7.6 48.8 0.5 34.3
Water, river - - m3 1.58E+6 8.8 7.6 48.8 0.5 34.3
Gold, Au 9.7E-4%, Ag 9.7E-4%, Zn 0.63%, Cu 0.38%, Pb 0.014%, in
resource, in ground - - kg 4.97E+3 100.0 - - - -
ore, in ground
Silver, Ag 9.7E-4%, Au 9.7E-4%, Zn 0.63%, Cu 0.38%, Pb 0.014%, in
- - kg 2.51E+5 - 100.0 - - -
ore, in ground
Zinc, Zn 0.63%, Au 9.7E-4%, Ag 9.7E-4%, Cu 0.38%, Pb 0.014%, in
- - kg 9.66E+7 - - 100.0 - -
ore, in ground
Copper, Cu 0.38%, Au 9.7E-4%, Ag 9.7E-4%, Zn 0.63%, Pb 0.014%,
- - kg 3.61E+6 - - - 100.0 -
in ore, in ground
Lead, Pb 0.014%, Au 9.7E-4%, Ag 9.7E-4%, Zn 0.63%, Cu 0.38%, in
- - kg 1.62E+8 - - - - 100.0
ore, in ground
Refining
The total production in 2005 of the smelter in Rönnskär amounted to 223’400 tonnes of copper,
27’000 tonnes of lead, 35’000 tonnes of zinc, 430 tonnes of silver and 17 tonnes of gold. A high share
of the production is recycled metal, e.g. 74%for gold. Since the production of secondary gold and sil-
ver from scrap is investigated in a dedicated chapter of this report, only material requirement and
emissions from the treatment of primary material is considered (see Tab. 5.10). Gold and Silver are
treated in the PGM plant in Rönnskär, the other metals undergo other treatments.

Tab. 5.9 Metal production volume 2005 from primary material (Boliden, 2006b)
Production from Scrap from Mines Remark:
concentrate from kg 6.47E+08 82% from Boliden (2005)

Mines
Secondary Raw kg 1.41E+08 18% from Boliden (2005)
Materials
contained lead kg n.a. 2.22E+07 calculated from output with share of concentrate input, in
LCI: lead, at beneficiation
n.a. 1.83E+08 calculated from output with share of concentrate input, in
contained copper kg LCI: copper, at beneficiation
n.a. 2.87E+07 calculated from output with share of concentrate input, in
contained Zinc kg LCI: zinc, at beneficiation
containted gold kg 1.25E+04 4.47E+03 from Boliden(2005), in LCI: gold, at beneficiation
contained silver kg 2.04E+05 2.26E+05 from Boliden (2005), in LCI: silver, at beneficiation
The allocation on the basis of the revenue (estimated with the prices of Tab. 2.5) results in 54% of the
requirements and emissions attributed to gold and 46% attributed to silver for the PM-plant (Precious
metal plant). The allocation for the refining of the other metals results in 6% attributed to zinc, 91% to
copper and 3% to lead (see Tab. 5.10).
Tab. 5.10: Metal production volume in 2005, calculated turnover and shares in turnover of the different products in the
Rönnskär Smelter (Boliden, 2006b). The revenue is calculated on the basis of the metal prices in Tab. 2.5
(average from the years 2004-2006). The PM-Plant is a sub process of the refinery, dedicated to the refine-
ment of gold and silver.
Allocation of metals refining metals Calc. Turnover

Production Gold kg 4.47E+03 7.01E+07 6.8% refining gold-silver 53.6% Price per kg: 15682.1 US$
(PM-plant)
Production Silver kg 2.26E+05 6.07E+07 5.9% 46.4% Price per kg: 268.7 US$
Production Zinc kg 2.87E+07 5.39E+07 5.2% refining, copper- 6.0% Price per kg: 1.9 US$
Production Copper kg 1.83E+08 8.21E+08 79.6% zinc-lead 91.2% Price per kg: 4.5 US$
Production Lead kg 2.22E+07 2.55E+07 2.5% 2.8% Price per kg: 1.2 US$
Total 1.03E+09 100%
The total annual energy and material requirements as well as emissions of the refinery are summarized
in Tab. 5.11. LPG use is approximated with the data set “natural gas, burned in industrial furnace”.
The SO2 emissions of the combustion of heavy fuel oil and coal are already included in the data set
“heavy fuel oil, burned in industrial furnace 1MW, non-modulating” and “hard coal, burned in indus-
trial furnace 1-10MW”, respectively. Particulates fractions are determined with standard data for non-
ferrous metal industry 9 . Transport requirements are estimated to be 250 tkm lorry and 250 tkm train
(average distance from the mines to the smelter, according to environmental report), for fossil fuels an
average energy density of 40 MJ/kg is used. Where available, data from the PGM-plant is used to de-
termine the requirement of gold- and silver treatment (Boliden, 2004). In this case, the energy re-
quirements of the PGM-Plant are indicated in Tab. 5.11 as share of the whole refining process.
9
PM2.5 82%, PM10 92% (http://www.iiasa.ac.at/~rains/cgi-bin/rains_pm, retrieved 6.4.2004).

Tab. 5.11 Annual energy consumption, material requirements and emissions of the Rönnskär smelter (Boliden, 2004;
Boliden, 2006b)
Energy refining copper-zinc- refining gold-silver, Remark:

lead, Boliden (2006) Boliden (2004)
Electricity kWh 4.95E+08 99.67% 1.62E+06 0.33% SE, electricity mix, Source: Boliden (2004), (2006)
Oil MJ 3.76E+08 99.76% 9.06E+05 0.24% In LCI as: heavy fuel oil, burned in industrial
furnace, Source: Boliden (2004), (2006)
Coke MJ 0.00% 7.55E+04 100% in LCI as: hard coal, burned in industrial furnace 1-
10MW, Source: Boliden (2004), (2006)
Natural Gas MJ 1.48E+07 99.96% 5.59E+03 0.04% in LCI: natural gas, burned in industrial furnace,
>100kW, Source: Boliden (2004), (2006)
Materials
Lime kg 1.89E+06 no lime is used for gold/silver purification, in LCI
as "limestone, milled"
Caustic kg 1.89E+07 no caustic soda is used for gold/silver purification,
in LCI as "sodium hydroxid"
Carbon kg 4.48E+07 no carbon is used for gold/silver purification, in LCI
as "charcoal"
Ammonia kg 1.40E+06 no ammonia is used for gold/silver purification, in
LCI as "ammonia, liquid"
Oxygen Nm
3 6.90E+07 Source: Boliden (2006)
Oxygen kg 9.31E+07 Recalculation with an average density of 1.35
kg/m3
Seawater m
3 8.17E+02 100% data for PGM-Plant, allocated to gold and silver, in
LCI as water, salt, ocean (Boliden, 2004)
Emission to Water
Copper kg 3.86E+02 Source: Boliden (2006)
Lead kg 4.19E+02 Source: Boliden (2006)
Zinc kg 1.97E+03 Source: Boliden (2006)
Cadmium kg 4.93E+01 Source: Boliden (2006)
Arsenic kg 9.86E+02 Source: Boliden (2006)
Mercury kg 1.15E+02 Source: Boliden (2006)
Nickel kg 1.56E+02 Source: Boliden (2006)
Emissions to air
Dust kg 2.63E+04 99.75% 6.45E+01 0.25% Source: Boliden (2004), (2006)
Copper kg 1.15E+03 Source: Boliden (2006)
Lead kg 2.79E+04 Source: Boliden (2006)
Zinc kg 4.19E+03 Source: Boliden (2006)
Cadmium kg 5.75E+01 Source: Boliden (2006)
Arsenic kg 1.89E+02 Source: Boliden (2006)
Mercury kg 1.07E+02 Source: Boliden (2006)
Sulphur Dioxide kg 3.16E+06 not accounted in LCI, as the emissions are
included in the datasets of diesel and naturalgas
Selenium kg 1.61E+01 data for PGM-Plant, allocated to gold and silver,

Source: Boliden (2004)
Dioxin kg 1.81E-03 Source: Boliden (2006)
Waste
combustible Waste kg 1.72E+05 disposal, municipal solid waste, 22.9% water, to
municipal incineration
Wood waste kg 3.90E+05 disposal, wood untreated, 20% water, to municipal
incineration
Metals kg 2.79E+05 not accounted in LCI, assumed to be recycled
Paper kg 5.75E+03 disposal, paper, 11.2% water, to municipal
incineration
Corrugated kg 3.61E+04 disposal, packaging cardboard, 19.6% water, to
cardboard municipal incineration
Landfill waste kg 2.09E+05 disposal, sulfidic tailings, off-site
Oil/water kg 1.85E+05 disposal, refinery sludge, 89.5% water, to
hazardous waste incineration

Tab. 5.12: Unit Raw data for “refining, copper-zinc-lead-gold-silver in smelter”
StandardDeviation95
UncertaintyType
Location
refining, copper-
Unit
Name zinc-lead-gold- GeneralComment
%
silver, in smelter
Location SE
Unit a
allocated zinc, from combined metal production, at refinery SE 0 kg 2.87E+7
products copper, from combined metal production, at refinery SE 0 kg 1.83E+8
lead, from combined metal production, at refinery SE 0 kg 2.22E+7
gold, from combined metal production, at refinery SE 0 kg 4.47E+3
silver, from combined metal production, at refinery SE 0 kg 2.26E+5
technosphere zinc, from combined metal production, at beneficiation SE 0 kg 2.87E+7 1 1.13 (1,4,1,3,1,3); Environmental report
copper, from combined metal production, at beneficiation SE 0 kg 1.83E+8 1 1.13 (1,4,1,3,1,3); Environmental report
lead, from combined metal production, at beneficiation SE 0 kg 2.22E+7 1 1.13 (1,4,1,3,1,3); Environmental report
gold, from combined metal production, at beneficiation SE 0 kg 4.47E+3 1 1.13 (1,4,1,3,1,3); Environmental report
silver, from combined metal production, at beneficiation SE 0 kg 2.26E+5 1 1.13 (1,4,1,3,1,3); Environmental report
electricity, medium voltage, at grid SE 0 kWh 4.95E+8 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2004 / 2006
heavy fuel oil, burned in industrial furnace 1MW, non-
RER 0 MJ 3.76E+8 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2004 / 2006
modulating
hard coal, burned in industrial furnace 1-10MW RER 0 MJ 3.76E+8 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2004 / 2006
natural gas, burned in industrial furnace >100kW RER 0 MJ 1.48E+7 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2004 / 2006
limestone, milled, packed, at plant CH 0 kg 1.89E+6 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2006
sodium hydroxide, 50% in H2O, production mix, at plant RER 0 kg 1.89E+7 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2006
charcoal, at plant GLO 0 kg 4.48E+7 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2006
ammonia, liquid, at regional storehouse RER 0 kg 1.40E+6 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2006
oxygen, liquid, at plant RER 0 kg 9.31E+7 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2004 / 2006
facilities anode refinery, secondary copper SE 1 unit 2.67E-2 1 3.55 (5,5,2,2,4,5); Rough Estimation
facilities precious metal refinery SE 1 unit 7.52E-2 1 3.55 (5,5,2,2,4,5); Rough Estimation
transport, lorry >16t, fleet average RER 0 tkm 2.61E+7 1 2.09 (4,5,na,na,na,na); Estimation
transport, freight, rail RER 0 tkm 2.61E+7 1 2.09 (4,5,na,na,na,na); Estimation
resource, in water Water, salt, ocean - - m3 8.17E+2 1 1.30 (4,5,na,na,na,na); Estimation
emission air, low population (1,4,1,3,1,3); Calculation based on environmental report,
Particulates, < 2.5 um - - kg 2.16E+4 1 3.02
density Boliden 2004
(1,4,1,3,1,3); Calculation based on environmental report,
Particulates, > 2.5 um, and < 10um - - kg 2.63E+3 1 2.02
Boliden 2004
(1,4,1,3,1,3); Calculation based on environmental report,
Particulates, > 10 um - - kg 2.10E+3 1 1.52
Boliden 2004
Copper - - kg 1.15E+3 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Lead - - kg 2.79E+4 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Zinc - - kg 4.19E+3 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Cadmium - - kg 5.75E+1 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Arsenic - - kg 1.89E+2 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Mercury - - kg 1.07E+2 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Dioxins, measured as 2,3,7,8-tetrachlorodibenzo-p-dioxin - - kg 1.81E-3 1 3.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Selenium - - kg 1.61E+1 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2004 / 2006

Heat, waste - - MJ 1.78E+9 1 1.13 (1,4,1,3,1,3); Calculation
emission water, ocean Copper, ion - - kg 3.86E+2 1 3.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Lead - - kg 4.19E+2 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Zinc, ion - - kg 1.97E+3 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Cadmium, ion - - kg 4.93E+1 1 3.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Arsenic, ion - - kg 9.86E+2 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Mercury - - kg 1.15E+2 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
Nickel, ion - - kg 1.56E+2 1 5.02 (1,4,1,3,1,3); Environmental report, Boliden 2006
disposal, municipal solid waste, 22.9% water, to municipal
technosphere CH 0 kg 1.72E+5 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2006
incineration
disposal, wood untreated, 20% water, to municipal
CH 0 kg 3.90E+5 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2006
incineration
disposal, paper, 11.2% water, to municipal incineration CH 0 kg 5.75E+3 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2006
disposal, packaging cardboard, 19.6% water, to municipal
incineration
disposal, sulfidic tailings, off-site GLO 0 kg 2.09E+5 1 1.13 (1,4,1,3,1,3); Environmental report, Boliden 2006
disposal, refinery sludge, 89.5% water, to hazardous waste
incineration

Tab. 5.13 Allocation factors of the data set “refining, copper-zinc-lead-gold-silver in smelter” (Boliden, 2004; Boliden,
2006b)
zinc, from copper, from lead, from gold, from silver, from
Location
refining, copper- combined combined combined combined combined
Unit
Name zinc-lead-gold- metal metal metal metal metal
silver, in smelter production, at production, at production, at production, at production, at
refinery refinery refinery refinery refinery
Location SE SE SE SE SE SE
InfrastructureProcess 0 0 0 0 0 0
Unit a kg kg kg kg kg
allocated zinc, from combined metal production, at refinery SE 0 kg 2.87E+7 100 - - - -
products copper, from combined metal production, at refinery SE 0 kg 1.83E+8 - 100 - - -
lead, from combined metal production, at refinery SE 0 kg 2.22E+7 - - 100 - -
gold, from combined metal production, at refinery SE 0 kg 4.47E+3 - - - 100 -
silver, from combined metal production, at refinery SE 0 kg 2.26E+5 - - - - 100
technosphere zinc, from combined metal production, at beneficiation SE 0 kg 2.87E+7 100.0 - - - -
copper, from combined metal production, at beneficiation SE 0 kg 1.83E+8 - 100.0 - - -
lead, from combined metal production, at beneficiation SE 0 kg 2.22E+7 - - 100.0 - -
gold, from combined metal production, at beneficiation SE 0 kg 4.47E+3 - - - 100.0 -
silver, from combined metal production, at beneficiation SE 0 kg 2.26E+5 - - - - 100.0
electricity, medium voltage, at grid SE 0 kWh 4.95E+8 6.0 90.9 2.8 0.18 0.15
RER 0 MJ 3.76E+8 6.0 91.0 2.8 0.1 0.1
modulating
hard coal, burned in industrial furnace 1-10MW RER 0 MJ 3.76E+8 - - - 53.6 46.4
natural gas, burned in industrial furnace >100kW RER 0 MJ 1.48E+7 6.0 91.1 2.8 0.0 0.0
limestone, milled, packed, at plant CH 0 kg 1.89E+6 6.0 91.2 2.8 - -
sodium hydroxide, 50% in H2O, production mix, at plant RER 0 kg 1.89E+7 6.0 91.2 2.8 - -
charcoal, at plant GLO 0 kg 4.48E+7 6.0 91.2 2.8 - -
ammonia, liquid, at regional storehouse RER 0 kg 1.40E+6 6.0 91.2 2.8 - -
oxygen, liquid, at plant RER 0 kg 9.31E+7 5.2 79.6 2.5 6.8 5.9
facilities anode refinery, secondary copper SE 1 unit 2.67E-2 5.2 79.6 2.5 6.8 5.9
facilities precious metal refinery SE 1 unit 7.52E-2 5.2 79.6 2.5 6.8 5.9
transport, lorry >16t, fleet average RER 0 tkm 2.61E+7 5.2 79.6 2.5 6.8 5.9
transport, freight, rail RER 0 tkm 2.61E+7 5.2 79.6 2.5 6.8 5.9
resource, in water Water, salt, ocean - - m3 8.17E+2 - - - 53.6 46.4
emission air, low population
Particulates, < 2.5 um - - kg 2.16E+4 6.0 91.0 2.8 0.1 0.1
density
Particulates, > 2.5 um, and < 10um - - kg 2.63E+3 6.0 91.0 2.8 0.1 0.1
Particulates, > 10 um - - kg 2.10E+3 6.0 91.0 2.8 0.1 0.1

Copper - - kg 1.15E+3 5.2 79.6 2.5 6.8 5.9
Lead - - kg 2.79E+4 5.2 79.6 2.5 6.8 5.9
Zinc - - kg 4.19E+3 5.2 79.6 2.5 6.8 5.9
Cadmium - - kg 5.75E+1 5.2 79.6 2.5 6.8 5.9
Arsenic - - kg 1.89E+2 5.2 79.6 2.5 6.8 5.9
Mercury - - kg 1.07E+2 5.2 79.6 2.5 6.8 5.9
Dioxins, measured as 2,3,7,8-tetrachlorodibenzo-p-dioxin - - kg 1.81E-3 5.2 79.6 2.5 6.8 5.9
Selenium - - kg 1.61E+1 - - - 53.6 46.4

Heat, waste - - MJ 1.78E+9 5.2 79.6 2.5 6.8 5.9
emission water, ocean Copper, ion - - kg 3.86E+2 5.2 79.6 2.5 6.8 5.9
Lead - - kg 4.19E+2 5.2 79.6 2.5 6.8 5.9
Zinc, ion - - kg 1.97E+3 5.2 79.6 2.5 6.8 5.9
Cadmium, ion - - kg 4.93E+1 5.2 79.6 2.5 6.8 5.9
Arsenic, ion - - kg 9.86E+2 5.2 79.6 2.5 6.8 5.9
Mercury - - kg 1.15E+2 5.2 79.6 2.5 6.8 5.9
Nickel, ion - - kg 1.56E+2 5.2 79.6 2.5 6.8 5.9
disposal, municipal solid waste, 22.9% water, to municipal
technosphere CH 0 kg 1.72E+5 5.2 79.6 2.5 6.8 5.9
incineration
disposal, wood untreated, 20% water, to municipal
CH 0 kg 3.90E+5 5.2 79.6 2.5 6.8 5.9
incineration
disposal, paper, 11.2% water, to municipal incineration CH 0 kg 5.75E+3 5.2 79.6 2.5 6.8 5.9
disposal, packaging cardboard, 19.6% water, to municipal
CH 0 kg 3.61E+4 5.2 79.6 2.5 6.8 5.9
incineration
disposal, sulfidic tailings, off-site GLO 0 kg 2.09E+5 5.2 79.6 2.5 6.8 5.9
disposal, refinery sludge, 89.5% water, to hazardous waste
CH 0 kg 1.85E+5 5.2 79.6 2.5 6.8 5.9
incineration

Tab. 5.14 Ecospold Meta information of metal production in Sweden
Name
mine, gold-silver-zinc-lead-copper mining, gold-silver-zinc-lead-copper deposit refining, copper-zinc-lead-gold-silver, in smelter
Location SE SE SE
Unit unit a a
Type 1 5 5
Version 2.0 2.0 2.0
energyValues 0 0 0
LanguageCode en en en
LocalLanguageCode de de de
Person 44 44 44
QualityNetwork 1 1 1
DataSetRelatesToProduct 1 1 1
IncludedProcesses
This data set includes the extraction of metals
This data set includes combined mining of and by products from concentrates and recycled
Land use, buildings and machinery of a gold, silver, copper, zinc and lead in Sweden raw materials. It includes energy and water
combined metal mine incl. energy and material use, water and air consumption, concentrate inputs, transports,
emissions of metals and land use. waste disposal, infrastructure use as well as
emissions into air and water.
Amount 1 1 1
LocalName Raffination, Kupfer-Zink-Blei-Gold-Silber in
Mine, Gold-Silber-Zink-Blei-Kupfer Abbau Gold-Silber-Zink-Blei-Kupfer-Lagerstätte
Metallhütte
Synonyms
GeneralComment
This multi-output process "mining, gold-silver-
zinc-lead-copper deposit" delivers the five co-
The multi-output process "refining, copper-zinc-
products "gold/silver/copper/zinc/lead, from
This data set includes land use of the Swedish lead-gold-silver in smelter" delivers the five co-
combined metal production, at beneficiation".
considered mines incl. revegetated areas as products "copper/zinc/lead/gold/silver from
The metals consisting of 1kg metal in the
well as generic estimates of buildings and combined metal production, at refinery". This
product, further refining is required for pure
machinery of the mine based on the data set data set describes the raffination of gold and
metal.
non-ferrous metal, mine, surface/underground. silver in a Swedish smelter. Allocation is based
This data set includes the mining in Sweden in
on economic value.
open pit and underground mines incl. land use.
Allocation is based on economic value.
InfrastructureIncluded 1 1 1
Category metals metals metals
SubCategory extraction extraction extraction
LocalCategory Metalle Metalle Metalle
LocalSubCategory Gewinnung Gewinnung Gewinnung
CASNumber
StartDate 2004 2004 2004
EndDate 2006 2006 2006
DataValidForEntirePeriod 1 1 1
OtherPeriodText Period of public environmental report Period of public environmental report Period of public environmental report
Text Material requirements for gold and silver are
Land use for Sweden, other inputs based on Energy use and emissions are from Swedish
estimated with Australian data, cyanide
rough assumptions. mines.
consumption is estimated with US-Data
Text Assumptions for buildings and machinery
based on the data set "non-ferrous metal,
open pit mines and underground mines modern swedish refinement plant
mine, surface/underground with 50 years life
time.
Percent 20
ProductionVolume >2.32E7 t ore >2.32E7 t ore >2.36E+5 t of metals
SamplingProcedure environmental report for landuse only environmental report environmental report
Extrapolations Buildings and machinery are based on global
assumptions. The figures were extrapolated on See Geography, air emissions are estimated
none
the basis of the ore production (2.32E10 kg per with data of one mine
year)
Person 43 43 43
DataPublishedIn 2 2 2
ReferenceToPublishedSource 10 10 10
Copyright 1 1 1
AccessRestrictedTo 0 0 0
CompanyCode
CountryCode
PageNumbers
Validator 51 51 51
Details automatic validation in Excel automatic validation in Excel automatic validation in Excel
OtherDetails none none none

5.1.3 Gold and silver production, Papua New Guinea (PG)

Data stem from the environmental report of Placer Dome Inc. (Placer Dome Inc., 2002b; Placer Dome
Inc., 2003b) and Barrick (2006b) for the Misima Mine and the Porgera Mine. Both are open pit mines.
The Misima Mine was in operation until 2004 and produced gold and silver. In January 2006, the Por-
gera Mine was sold to the Canadian company of Barrick. Actual data (2006) from Barrick is used
where available, no environmental report has been published in the years 2004 and 2005. Since no ac-
tual production takes place in the Misima mine, the data of the last year of full activity (year 2000) is
used as one production site of gold in Papua New Guinea (see Tab. 5.15 for production values).
The Porgera Mine is located in the Enga Province of the Papua New Guinea highlands about 550 kilo-
metres north west of the capital Port Moresby and produces only gold.
The Misima mine is situated on an island 190 km east from the main island of Papua New Guinea. The
distance to the capital Port Moresby is about 650 km. The tailings of Misima are disposed of with a
pipeline in deep-sea (Placer Dome Inc., 2000).
Tab. 5.15 shows the production volume in 2000/2006, calculated turnover and shares in turnover of
the different products of the Misima and the Porgera Mine.
The recovery rate (proportion of valuable material physically recovered in the processing of ore) is
taken into account in order to define the resource extraction. The average concentration (head grade,
expressed in kg metal per kg ore) of metals in ore for Papua New Guinea is calculated with the quan-
tity of ore milled and the corresponding production rate.
Tab. 5.15: Production volume 2000 / 2006, calculated turnover and shares in turnover of the different products of the
Misima and the Porgera Mine (Placer Dome Inc., 2001, Barrick, 2006a), Basis: prices 2004-2006 according to
Tab. 2.5
Production Misima Porgera Papua New

2000 2006 Guinea
Ore milled kg 6.10E+09 5.23E+09 1.13E+10
Production Gold kg 6.77E+03 1.68E+04 2.36E+04
Production Silver kg 2.40E+04 2.40E+04
Value Gold US$ 1.06E+08 2.63E+08 2.63E+08 Value of 1kg Gold: 15680 US$
Value Silver US$ 6.45E+06 6.45E+06 Value of 1kg Silver: 269 US$
Allocation Gold 97.6%
Allocation Silver 2.4%
Head grade
Gold kg/kg ore 1.11E-06 3.21E-06 2.08E-06
Silver kg/kg ore 3.93E-06 2.12E-06
Recovery rate 90% based on reports of Placer-Dome Inc.
(2002)
mining. It is assumed, that the quantities of machinery as well as surface and volume of buildings are
equal to the generic mine. This approach is rough, and the uncertainity is rather high. As the mined ore
from the generic mine is some orders of magnitude lower than the mine in focus, a proportional scal-
ing by milled ore will lead to a higher mistake (see Tab. 5.1: Gold and Silver production in Sweden).
The lifetime of mines is assumed to be 50 years.
Land use is based on the figures from the environmental report (Tab. 5.16). Not all the area of the
mine is in active use (disturbed area) at one time; the value in Tab. 5.16 is assumed to be the average
share of the mine area in use. Therefore, the total area is occupied during an average of about 25 years
according to the following formula:

Disturbed area in active use 1.17 *10 7

* lifetime of mine  * 50 y  25 y , the occupation of the mine is
Area disturbed during life of mine 2.32 *10 8
Tab. 5.16: Land use of the Misima and Porgera Mines in 2002 (Placer Dome Inc., 2002a)
Land use Misima Porgera Papua New Remark

2000 2006 Guinea
Area disturbed during m2 6.11E+06 1.71E+07 2.32E+07 Transformation, from unknown and
life of mine Transformation, to mineral extraction site,
Disturbed area in m2 2.53E+06 9.21E+06 1.17E+07 Occupation, mineral extraction site during 50
active use years
Area available for m2 1.80E+05 1.74E+06 1.92E+06 added in "Transformation, to unknown"
revegetation
Area revegetated m2 3.40E+06 6.16E+06 9.56E+06 Transformation, to heterogeneous, agricultural
Tab. 5.17: Unit process raw data of the gold and silver mine in Papua New Guinea
Uncertainty
Infrastructu
StandardD
eviation95
reProcess
Location
Type
Unit mine, gold
Name GeneralComment
%
and silver
Location PG
Unit unit
product mine, gold and silver PG 1 unit 1.00E+0
report, Placer Dome Inc. 2003
Mining and refining

The recovery processes of the Misima Mine are cyanide leach and carbon in pulp (CIP), in the Porgera
Mine pressure oxidation and cyanide leach. Both mines produce doré.
The following table (Tab. 5.18) sums up the total requirements and emissions of the Misima and Por-
gera Mines. For the Porgera Mine data from 2006 is used where available, the consumption of some
material requirements is scaled with the weight of processed gold and data from the year 2002. For the
Misima Mine data from 2000 is used, when the Misima Mine was still producing (Placer Dome Inc.,
2001). Since the module “blasting” does not include the CO2 emissions from explosives, they are cal-
culated with the factor of the mines of Newmont (2005b) (0.16 kg CO2 per kg explosive). The emis-
sions to air and in water from the Misima and Porgera mines are as declared in the environmental re-
ports, only liquid effluents have been modelled. Emissions to water head from the Misima-Mine to the
ocean, whereas from Porgera they are into the river.

Tab. 5.18: Material and energy requirements of the Misima Mine in 2000 (Placer Dome Inc., 2001) and the Porgera Mine
in 2002 / 2006 (Placer Dome Inc., 2002b, Barrick, 2006b)
Energy Misima Porgera Papua New

2000 2006 Guinea Remark
Electricity for mining 1.14E+08 1.14E+08 Electricity is produced by the mine on site or with own
power stations, environmental impacts are included in
1)
kWh diesel-electric moduls
Electricity for refining kWh n.a. n.a. 5.28E+04 assumed with Möbius / Wohlwill-electrolysis, electricity
is produced by the mine on site, added to "Diesel,
burned in diesel-electric set" with thermal-electrical
efficiency of 35%
Diesel (generators) MJ 2.77E+07 2.77E+07 Diesel, burned in diesel-electric generating set
2)
Diesel (mobile MJ 7.15E+08 2.96E+09 3.68E+09 Diesel, burned in building machine

equipment)
Petrol l 5.28E+04 5.28E+04 converted in transport, van <3.5t (0.38l/tkm)
Oil MJ 9.87E+08 9.87E+08 heavy fuel oil, burned in industrial furnace 1MW, non-
modulating
LPG MJ 3.43E+05 5.33E+09 5.33E+09 natural gas, burned in industrial furnace
Materials
Acetylen m3 2.69E+03 2.69E+03 Density at 15°C: 1.095 kg/m3
Cyanide kg 2.37E+06 6.05E+05 2.98E+06 In LCI as sodium cyanide
kg 1.60E+07 1.34E+07 2.94E+07 Value of Porgera scaled from consumption in 2003 by
Lime processed gold.
kg 7.88E+04 1.96E+05 2.74E+05 Scaled from Steward (1999) by processed gold,
Caustic accounted for sodium hydroxide
Flocculent / Flotation 1.76E+05 5.25E+05 7.01E+05
chemicals kg approximated as organic chemicals
kg 2.36E+05 5.53E+05 7.89E+05 Value of Porgera scaled from consumption in 2003 by
Carbon processed gold.
kg 1.96E+04 4.86E+04 6.82E+04 Scaled from Steward (1999) by processed gold,
Zinc accounted for as zinc, primary, at regional storage
kg 8.74E+04 8.74E+04 Value of Porgera scaled from consumption in 2003 by
Sulphuric Acid processed gold.
2.20E+05 8.74E+04 3.07E+05 Value of Porgera scaled from consumption in 2003 by
Hydrochloric Acid kg processed gold.
2.83E+06 3.96E+06 6.79E+06 Value of Porgera scaled from consumption in 2003 by
Steel grinding media kg processed gold.
Explosives kg 1.56E+06 3.43E+06 4.99E+06 Partly ANFO explosives, in LCI: "blasting"
Water resources
m3 3.30E+06 3)
3.30E+06 in LCI: Water, unspecified natural origin
Water from bore fields
Groundwater m3 2.85E+07 2.85E+07 in LCI: Water, well, in ground
4)
Emissions to Water
5)
Arsenic kg 2.40E+01 4.87E+02 5.11E+02 Corrected value for Misima
Value of Porgera scaled from emission in 2003 by
Cadmium kg 9.50E+01 2.30E+03 2.39E+03 milled ore
Copper kg 9.76E+04 3.52E+04 1.33E+05 milled ore
Cyanide kg 1.70E+05 4.57E+04 2.16E+05 milled ore
Lead kg 2.83E+02 6.77E+01 3.51E+02 milled ore
Mercury kg 0.00E+00 5.43E+00 5.43E+00 milled ore
Nickel kg 6.67E+02 3.11E+04 3.18E+04 milled ore
Zinc kg 6.36E+03 8.41E+05 8.47E+05 milled ore
Emissions to Air
Carbon Dioxide (from Calculation, based on data from Newmont (2005)
explosives) kg 2.50E+05 5.49E+05 7.99E+05
Zinc kg n.a. n.a. 2.50E+00 based on Swedish data, scaled by milled ore
Copper kg n.a. n.a. 9.05E+00 based on Swedish data, scaled by milled ore
Lead kg n.a. n.a. 2.94E+00 based on Swedish data, scaled by milled ore
Placer Dome Inc. (2002), in LCI: Disposal, sulphidic
Disposal kg 6.09E+09 1.96E+09 8.05E+09 tailings
1
Electricity in Misima is produced from diesel and heavy fuel oil. Porgera's principal source of power is
supplied by a 73-kilometre transmission line from the gas fired and PJV-owned Hides Power Station.
The station has a total output of 62 megawatts (“MW”). A back up diesel power station is located at

the mine and has an output of 13MW. The average power requirement of the mine is about 60 MW.
The environmental impacts of electricity use in Papua New Guinea are considered in the modules
“diesel, burned in diesel-electric generating set” and “heavy fuel oil, burned in industrial furnace”.
2
An 18 MW diesel fired power station supplies electrical power. Diesel was used in the station due to
the unavailability of previously supplied heavy fuel oil.
3
Process water is supplied from pit dewatering bores and in-pit water. Potable water is sourced from
boreholes in the coastal limestone. The figure in the environmental report is 3.33E+12 l. However, as
the discharge in 2002 is about 7.0E+9 l, the value in the report is assumed to be 1000 times too high
and corrected to 3.33E+09 l.
4
The main water supply of the mine is the Waile Creek Dam, located approximately 7 kilometres from
the mine. The reservoir has a capacity of approximately 717, 000 m3 of water. Water for the grinding
circuit is also extracted from Kogai Creek, which is located adjacent to the grinding circuit. The mine
operates four water treatment plants for potable water and five sewage treatment plants.
5
The figure in the environmental report from 2000 was 0.24 kg. However, 2001-2003, the values are
23-47kg, so the figure for 2000 is assumed to be 100 times too small and is corrected to 24 kg.
The figures are allocated on the basis of the allocation factors shown in Tab. 5.15. Air emissions (zinc,
copper and lead) are approximated with Swedish data. As Swedish data stem from a combined proc-
ess, they were allocated to gold and silver for Sweden and scaled by the milled ore for Papua New
Guinea. Transport distances are estimated to be 600 km for the materials on land (corresponds to the
average distance from capital Port Moresby to the mines).
Since the mines produce doré gold (unrefined gold bullion bars containing between 60% and 95%
gold), the refinery process is less complex than the refinery process of Rönnskär in Sweden. It is as-
sumed, that the gold doré-bars undergo the treatment of Wohlwill electrolysis (for details see 5.2.2.1,
“Gold / Silver, secondary, from electronic scrap in ecoinvent”). Due to the lack of specific data, the
figure of refinement from electronic scrap is used: 1.63 kWh per kg gold (Renner et al., 2002a). The
refinery of silver requires 0.6 kWh per kg of silver (Möbius electrolysis). No emissions were assumed,
since on the one hand the high value of the processed material requires an efficient emission control
and on the other hand, refinement is a hydrometallurgical process with low impact. The infrastructure
is approximated with the facility of precious metal refinery. The unit process raw data are summarized
in Tab. 5.19.

Tab. 5.19: Unit process raw data of gold and silver mining and refining in Papua New Guinea
UncertaintyType
StandardDeviati
InfrastructurePr
Location
mining and
on95%
ocess
Unit
Name refining, gold- GeneralComment
silver deposit
Location PG
Unit a
allocated gold, from combined gold-silver production, at refinery PG 0 kg 2.36E+4
products silver, from combined gold-silver production, at refinery PG 0 kg 2.40E+4
technosphere diesel, burned in diesel-electric generating set GLO 0 MJ 2.82E+7 1 1.14 (2,4,2,3,1,3); Environmental report
transport, van <3.5t RER 0 tkm 1.51E+5 1 2.02 (2,4,2,3,1,3); Environmental report
RER 0 MJ 9.87E+8 1 1.14 (2,4,2,3,1,3); Environmental report
modulating
natural gas, burned in industrial furnace >100kW RER 0 MJ 5.33E+9 1 1.14 (2,4,2,3,1,3); Environmental report
acetylene, at regional storehouse CH 0 kg 2.78E+3 1 1.14 (2,4,2,3,1,3); Environmental report
sodium cyanide, at plant RER 0 kg 2.98E+6 1 1.14 (2,4,2,3,1,3); Environmental report
limestone, milled, packed, at plant CH 0 kg 2.94E+7 1 1.14 (2,4,2,3,1,3); Environmental report
(2,3,2,5,1,3); Calculation, based on
sodium hydroxide, 50% in H2O, production mix, at plant RER 0 kg 2.74E+5 1 1.15 thesis about Australian precious metal
chemicals organic, at plant GLO 0 kg 7.01E+5 1 1.15
Placer Dome Inc. 2003
charcoal, at plant GLO 0 kg 7.89E+5 1 1.15 thesis about Australian precious metal
zinc, primary, at regional storage RER 0 kg 6.82E+4 1 1.15 thesis about Australian precious metal
sulphuric acid, liquid, at plant RER 0 kg 8.74E+4 1 1.15 thesis about Australian precious metal
hydrochloric acid, 30% in H2O, at plant RER 0 kg 3.07E+5 1 1.11
steel, converter, chromium steel 18/8, at plant RER 0 kg 6.79E+6 1 1.11
blasting RER 0 kg 4.99E+6 1 1.11
Placer Dome Inc. 2003, Barrick 2006
(2,4,na,na,na,na); Calculation, based on
estimated distance of 600km
disposal, sulfidic tailings, off-site GLO 0 kg 8.05E+9 1 1.30 (4,5,na,na,na,na); Environmental report
infrastructure mine, gold and silver PG 1 unit 2.00E-2 1 3.01 (2,3,2,3,1,3); Environmental report
facilities precious metal refinery SE 1 unit 7.83E-3 1 3.09 (4,5,na,na,na,na); Environmental report
emission air, low (2,4,2,3,1,3); Calculation, based on
Copper - - kg 9.05E+0 1 5.02
population density Swedish production
Lead - - kg 2.94E+0 1 5.02
Swedish production
Zinc - - kg 2.50E+0 1 5.02
Swedish production
emission water,
Arsenic, ion - - kg 2.40E+1 1 5.02 (2,4,2,3,1,3); Environmental report
ocean
Cadmium, ion - - kg 9.50E+1 1 3.02 (2,4,2,3,1,3); Environmental report
Copper, ion - - kg 9.76E+4 1 3.02 (2,4,2,3,1,3); Environmental report
Cyanide - - kg 1.70E+5 1 3.02 (2,4,2,3,1,3); Environmental report
Mercury - - kg 0 1 5.02 (2,4,2,3,1,3); Environmental report
Nickel, ion - - kg 6.67E+2 1 5.02 (2,4,2,3,1,3); Environmental report
emission water, river Arsenic, ion - - kg 4.87E+2 1 5.02 (2,4,2,3,1,3); Environmental report
Cadmium, ion - - kg 2.30E+3 1 3.02 (2,4,2,3,1,3); Environmental report
Copper, ion - - kg 3.52E+4 1 3.02 (2,4,2,3,1,3); Environmental report
Cyanide - - kg 4.57E+4 1 3.02 (2,4,2,3,1,3); Environmental report
Mercury - - kg 5.43E+0 1 5.02 (2,4,2,3,1,3); Environmental report
Nickel, ion - - kg 3.11E+4 1 5.02 (2,4,2,3,1,3); Environmental report
resource, in water Water, well, in ground - - m3 2.85E+7 1 1.14 (2,4,2,3,1,3); Environmental report
Water, unspecified natural origin - - m3 3.30E+6 1 1.14 (2,4,2,3,1,3); Environmental report
resource, in ground Gold, Au 2.1E-4%, Ag 2.1E-4%, in ore, in ground - - kg 2.62E+4 1 1.14 (2,4,2,3,1,3); Estimation
Silver, Ag 2.1E-4%, Au 2.1E-4%, in ore, in ground - - kg 2.67E+4 1 1.14 (2,4,2,3,1,3); Estimation

Tab. 5.20 Allocation factors of the data set “mining and refining, gold-silver deposit”
InfrastructurePr
gold, from silver, from
Location
mining and combined gold- combined gold-
ocess
Unit
Name refining, gold- silver silver
silver deposit production, at production, at
refinery refinery
Location PG PG PG
Unit a kg kg
allocated gold, from combined gold-silver production, at refinery PG 0 kg 2.36E+4 100 -
products silver, from combined gold-silver production, at refinery PG 0 kg 2.40E+4 - 100
technosphere diesel, burned in diesel-electric generating set GLO 0 MJ 2.82E+7 97.6 2.4
diesel, burned in building machine GLO 0 MJ 3.68E+9 97.6 2.4
transport, van <3.5t RER 0 tkm 1.51E+5 97.6 2.4
RER 0 MJ 9.87E+8 97.6 2.4
modulating
natural gas, burned in industrial furnace >100kW RER 0 MJ 5.33E+9 97.6 2.4
acetylene, at regional storehouse CH 0 kg 2.78E+3 97.6 2.4
sodium cyanide, at plant RER 0 kg 2.98E+6 97.6 2.4
limestone, milled, packed, at plant CH 0 kg 2.94E+7 97.6 2.4
sodium hydroxide, 50% in H2O, production mix, at plant RER 0 kg 2.74E+5 97.6 2.4
chemicals organic, at plant GLO 0 kg 7.01E+5 97.6 2.4
charcoal, at plant GLO 0 kg 7.89E+5 97.6 2.4
zinc, primary, at regional storage RER 0 kg 6.82E+4 97.6 2.4
sulphuric acid, liquid, at plant RER 0 kg 8.74E+4 97.6 2.4
hydrochloric acid, 30% in H2O, at plant RER 0 kg 3.07E+5 97.6 2.4
steel, converter, chromium steel 18/8, at plant RER 0 kg 6.79E+6 97.6 2.4
blasting RER 0 kg 4.99E+6 97.6 2.4
transport, lorry >16t, fleet average RER 0 tkm 2.79E+7 97.6 2.4
disposal, sulfidic tailings, off-site GLO 0 kg 8.05E+9 97.6 2.4
infrastructure mine, gold and silver PG 1 unit 2.00E-2 97.6 2.4
facilities precious metal refinery SE 1 unit 7.83E-3 97.6 2.4
emission air, low
Copper - - kg 9.05E+0 97.6 2.4
population density
Lead - - kg 2.94E+0 97.6 2.4
Zinc - - kg 2.50E+0 97.6 2.4

Carbon dioxide, fossil - - kg 7.99E+5 97.6 2.4
emission water,
Arsenic, ion - - kg 2.40E+1 97.6 2.4
ocean
Cadmium, ion - - kg 9.50E+1 97.6 2.4
Copper, ion - - kg 9.76E+4 97.6 2.4
Cyanide - - kg 1.70E+5 97.6 2.4
Lead - - kg 2.83E+2 97.6 2.4
Mercury - - kg 0 97.6 2.4
Nickel, ion - - kg 6.67E+2 97.6 2.4
Zinc, ion - - kg 6.36E+3 97.6 2.4
emission water, river Arsenic, ion - - kg 4.87E+2 97.6 2.4
Cadmium, ion - - kg 2.30E+3 97.6 2.4
Copper, ion - - kg 3.52E+4 97.6 2.4
Cyanide - - kg 4.57E+4 97.6 2.4
Lead - - kg 6.77E+1 97.6 2.4
Mercury - - kg 5.43E+0 97.6 2.4
Nickel, ion - - kg 3.11E+4 97.6 2.4
Zinc, ion - - kg 8.41E+5 97.6 2.4
resource, in water Water, well, in ground - - m3 2.85E+7 97.6 2.4
Water, unspecified natural origin - - m3 3.30E+6 97.6 2.4
resource, in ground Gold, Au 2.1E-4%, Ag 2.1E-4%, in ore, in ground - - kg 2.62E+4 100.0 -
Silver, Ag 2.1E-4%, Au 2.1E-4%, in ore, in ground - - kg 2.67E+4 - 100.0

Tab. 5.21 Ecospold Meta information of gold and silver production in Papua New Guinea
Name mine, gold and silver mining and refining, gold-silver deposit
Location PG PG
InfrastructureProcess 1 0
Unit unit a
Type 1 5
Version 2.0 2.0
energyValues 0 0
LanguageCode en en
LocalLanguageCode de de
Person 44 44
QualityNetwork 1 1
DataSetRelatesToProduct 1 1
IncludedProcesses This data set includes the combined mining and
refining of gold and silver in open pit mines in
Land use, buildings and machinery of a gold
Papua New Guinea incl. energy and material
and silver mine
use, water and air emissions of metals and land
use.
Amount 1 1
LocalName Mine, Gold und Silber Abbau und Raffination, Gold-Silber-Lagerstätte
Synonyms
GeneralComment This multi-output process 'mining and refining,
gold-silver deposit' delivers the two co-products
This data set includes land use of the
'gold, from combined gold-silver-production, at
considered mines incl. revegetated areas as
refinery' and 'silver, from combined gold-silver-
well as estimates of buildings and machinery of
production, at refinery'. This data set includes
the mine based on the data set non-ferrous
the mining in Papua New Guinea in open pit
metal, mine, surface/underground.
mines incl. land use. Air emissions are
estimated.
InfrastructureIncluded 1 1
Category metals metals
SubCategory extraction extraction
LocalCategory Metalle Metalle
LocalSubCategory Gewinnung Gewinnung
Formula
StatisticalClassification
CASNumber
StartDate 2000 2000
EndDate 2006 2006
DataValidForEntirePeriod 1 1
OtherPeriodText Period of public environmental report Period of two public environmental report
Text Material requirements are estimated with
Land use of Papua New Guinea-plant, other
Australian data, Air emissions are estimated
inputs based on rough assumptions.
with Swedish data.
Text
Assumptions for buildings and machinery
based on the data set "non-ferrous metal, mine, Refinement is estimated with electrolysis-data.
surface/underground with 50 years life time.
Percent
ProductionVolume 11.3 Mill. t ore per year 23.57 t gold and 24 t silver per year
SamplingProcedure environmental report for landuse in the year environmental report of one mine in the year
2000 and one mine in 2006 2000 and one mine in 2006
Emissions to water are calculated with older
assumptions. The figures were extrapolated on
data of one mine
the basis of the ore production.
UncertaintyAdjustments none none
Person 44 44
DataPublishedIn 2 2
ReferenceToPublishedSour
10 10
ce
Copyright 1 1
AccessRestrictedTo 0 0
CompanyCode
CountryCode
PageNumbers
Validator 51 51
Details automatic validation in Excel automatic validation in Excel
OtherDetails none none
5.1.4 Gold and silver production, Chile (CL)

Data for the Mine “La Coipa” stem from the environmental report of Placer Dome Inc. (2002b). Since
the mine has been sold in 2006 to Kinross Gold Corporation and Goldcorp Inc., no more recent data is
available. Both enterprises do not publish their environmental data. Therefore, the older data from
2002 is used to model the gold- and silver production in Chile.

The La Coipa Mine produces gold and silver and is an open pit mine. Tab. 5.22 shows the production
volume 2002 and calculated turnover of the La Coipa Mine. The recovery rate (proportion of valuable
material physically recovered in the processing of ore) is taken into account in order to define the re-
source extraction. The average concentration (head grade, expressed in kg metal per kg ore) of metals
in ore for Chile is calculated with the quantity of ore milled and the corresponding production rate.
Tab. 5.22: Production volume 2002, calculated turnover and shares in turnover of the different products of the La
Coipa Mine (Placer Dome Inc., 2003a). Basis: prices 2005 according to Tab. 2.5.
Production La Coipa
Ore milled kg 3.17E+09
Production Gold kg 2.99E+03
Production Silver kg 1.12E+05
Value Gold US$ 4.69E+07 Value of 1kg Gold: 15680 US$
Value Silver US$ 3.01E+07 Value of 1kg Silver: 269 US$
Head grade
Gold kg/kg 1.11E-06
Silver kg/kg 4.16E-05
Recovery rate 85%
equal to the generic mine. This approach is rough, and the uncertainty is rather high. As the mined ore
mine is in active use at one time; the value in Tab. 5.23 is assumed to be the average share of the mine
area in use. Therefore, the total area is occupied during an average of about 43 years, according to
formula:

Tab. 5.23: Land use of the La Coipa Mine in 2002 (Placer Dome Inc., 2004).
Land use La Coipa Remark

Area disturbed during life of mine 2 1.34E+07 Transformation, from unknown and Transformation, to
m
mineral extraction site,
1) 2 1.16E+07 Occupation, mineral extraction site, during 50 years
Disturbed area in active use m
3
Area available for revegetation m 1.72E+06 Transformation to unknown
Area revegetated 2 3.00E+04 added in "Transformation, to heterogeneous, agricultural"
m
1
This figure is 0 in the original data, which is assumed to be a mistake. The area in active use is there-
fore calculated as the area disturbed during life of mine minus area available for revegetation and
area revegetated.

Tab. 5.24: Unit process raw data for the gold and silver mine in Chile.
Infrastructur
StandardDe
Uncertainty
viation95%
eProcess
Location
Type
mine, gold
Unit
Name GeneralComment
and silver
Location CL
Unit unit
product mine, gold and silver CL 1 unit 1.00E+0
Mining and refining

The processing plant consists of primary crushing, a pre-crushing circuit, (semi autogenous ball mill
crushing) grinding, leaching, filtering and washing, Merrill-Crowe plant and doré refinery. The
Merrill-Crowe metal recovery circuit is better than a carbon-in-pulp system for the high-grade silver
material. Tailings are filtered to recover excess water as well as residual cyanide and metals. A dry
tailings disposal system was preferred to a conventional wet tailings impoundment because of site-
specific environmental considerations. The following table (Table 5.25) sums up the total energy and
material requirements of the La Coipa Mine in 2002.

Table 5.25: Energy and material requirements of the La Coipa Mine in 2002 (Placer Dome Inc., 2002b).
Energy La Coipa Remark

Electricity for mining kWh 1.60E+08 Chilean mix 1) approximated with the UCTE mix
Electricity for refining kWh 7.21E+04 assumed with Möbius / Wohlwill-electrolysis, Chilean mix
1) approximated with the UCTE mix
Diesel (generators) MJ 4.18E+08 Diesel, burned in diesel-electric generating set
Diesel (mobile equipment) l 9.44E+07 Diesel, burned in building machine
Materials
Cyanide kg 2.14E+06 In LCI as sodium cyanide
Lime kg 1.89E+05 Scaled from Steward (1999) by processed gold
Caustic kg 3.48E+04 Scaled from Steward (1999) by processed gold, accounted
for sodium hydroxide
Carbon kg 5.07E+04 Scaled from Steward (1999) by processed gold, accounted
for as charcoal
Zinc kg 8.65E+03 Scaled from Steward (1999) by processed gold, accounted
for as zinc, primary, at regional storage
Sulphuric Acid kg 1.99E+03 Scaled from Steward (1999) by processed gold
Hydrochloric Acid kg 4.08E+04 Scaled from Steward (1999) by processed gold
Explosives kg 1.72E+06 Scaled from Peruvian Mine by milled ore, in LCI: "blasting"
Water resources
Water resources m
3 2.87E+04 Scaled from Peruvian Mine by milled ore, in LCI: Water,
unspecified natural origin
Emissions to Water
Arsenic, ion kg 1.24E+02 estimated with Data from PG, scaled by milled ore
Cadmium, ion kg 5.80E+02 estimated with Data from PG, scaled by milled ore
Copper, ion kg 3.21E+04 estimated with Data from PG, scaled by milled ore
Cyanide kg 5.22E+04 estimated with Data from PG, scaled by milled ore
Lead kg 8.49E+01 estimated with Data from PG, scaled by milled ore
Mercury kg 1.32E+00 estimated with Data from PG, scaled by milled ore
Nickel, ion kg 7.70E+03 estimated with Data from PG, scaled by milled ore
Zinc, ion kg 2.05E+05 estimated with Data from PG, scaled by milled ore
Emissions to Air
Carbon Dioxide (from explosives) kg 2.75E+05 Calculation, based on data from Newmont (2005)
Zinc kg 6.99E-01 estimated with Data from SE, scaled by milled ore
Copper kg 2.53E+00 estimated with Data from SE, scaled by milled ore
Lead kg 8.24E-01 estimated with Data from SE, scaled by milled ore
Disposal kg 2.25E+09 estimated with data from PG, in LCI as: Disposal, sulphidic
tailings
1
Chilean electricity mix in 2005 (Nationmaster, 2004): fossil fuel 47%, hydro 51.5%, other 1.5%.
There are existing facilities for water extraction in the Salar de Maricunga, which is located 50 kilome-
tres southeast of La Coipa. There is a water pump system that collects the water from five wells. The
water is accounted for as ground water (water, well, in ground).
Water emissions are approximated with figures of the mines in Papua New Guinea, material use (with
exception of cyanide) with figures of Australian data for gold and silver production (Stewart, 1999),
see also chapter 5.1.9), air emissions (zinc, copper and lead) with Swedish data. As Swedish data stand
for a combined process, they are allocated to gold and silver for Sweden and used in this form for
Chile. Cyanide is accounted for as sodium cyanide, which is commonly used in the gold mining. Since
the module “blasting” does not include the CO2 emissions from explosives, they are calculated with
the factor of the mines of Newmont (2005b) (0.16 kg CO2 per kg explosive). Water emissions are as-
sumed to be emitted to rivers. Transports of materials are estimated to be 600 km by truck. Data on the
amount of tailings disposal is approximated with data of mines in Papua New Guinea.
“Gold / Silver, secondary, from electronic scrap in ecoinvent”). Due to the lack of specific data about
primary refinement, the figure of refinement from electronic scrap is used: 1.63 kWh per kg gold

(Renner et al., 2002a). The refinery of silver requires 0.6 kWh per kg of silver (Möbius electrolysis).
No emissions were assumed, since on the one hand the high value of the processed material requires
an efficient emission control and on the other hand, refinement is a hydrometallurgical process with
low impact. The infrastructure is approximated with the facility of precious metal refinery. The unit
process raw data are summarized in Tab. 5.26.
Tab. 5.26: Unit process raw data of gold and silver mining and refining in Chile
StandardDeviation95
UncertaintyType
Location
mining and
Unit
%
silver deposit
Location CL
Unit a
gold, from combined gold-silver production, at
allocated CL 0 kg 2.99E+3
refinery
silver, from combined gold-silver production, at
products CL 0 kg 1.12E+5
refinery
electricity, medium voltage, production UCTE,
technosphere UCTE 0 kWh 1.60E+8 1 1.14 (2,4,2,3,1,3); Environmental report (Placer-Dome, 2002)
at grid
diesel, burned in diesel-electric generating set GLO 0 MJ 4.18E+8 1 1.14 (2,4,2,3,1,3); Environmental report (Placer-Dome, 2002)
diesel, burned in building machine GLO 0 MJ 9.44E+7 1 1.14 (2,4,2,3,1,3); Environmental report (Placer-Dome, 2002)
(2,4,2,3,1,3); Calculation, based on thesis on Australian
precious metal production, data for 1999
limestone, milled, packed, at plant CH 0 kg 1.89E+5 1 1.14
sodium hydroxide, 50% in H2O, production mix, (2,3,2,5,1,3); Calculation, based on thesis on Australian
RER 0 kg 3.48E+4 1 1.15
at plant precious metal production, data for 1999
charcoal, at plant GLO 0 kg 5.07E+4 1 1.15
zinc, primary, at regional storage RER 0 kg 8.65E+3 1 1.15
sulphuric acid, liquid, at plant RER 0 kg 1.99E+3 1 1.15
blasting RER 0 kg 1.72E+6 1 1.11 (2,3,2,3,1,3); Estimation, based on peruvian Data
(2,4,na,na,na,na); Calculation, based on estimated distance
of 600km
disposal, sulfidic tailings, off-site GLO 0 kg 2.25E+9 1 1.11 (2,3,2,3,1,3); Estimation
infrastructure facilities precious metal refinery SE 1 unit 1.89E-2 1 3.09 (4,5,na,na,na,na); Environmental report
mine, gold and silver CL 1 unit 2.00E-2 1 3.01 (2,3,2,3,1,3); Environmental report
emission air, low
Copper - - kg 2.53E+0 1 5.02 (2,4,2,3,1,3); Estimation, based on Swedish Data
population density
Lead - - kg 8.24E-1 1 5.02 (2,4,2,3,1,3); Estimation, based on Swedish Data
Zinc - - kg 6.99E-1 1 5.02 (2,4,2,3,1,3); Estimation, based on Swedish Data
Carbon dioxide, fossil - - kg 2.75E+5 1 1.14 (2,4,2,3,1,3); Estimation, based on Swedish Data
emission water, ocean Arsenic, ion - - kg 1.24E+2 1 5.02 (2,4,2,3,1,3); Estimation, based on PG-Data
Cadmium, ion - - kg 5.80E+2 1 3.02 (2,4,2,3,1,3); Estimation, based on PG-Data
Copper, ion - - kg 3.21E+4 1 3.02 (2,4,2,3,1,3); Estimation, based on PG-Data
Cyanide - - kg 5.22E+4 1 3.02 (2,4,2,3,1,3); Estimation, based on PG-Data
Lead - - kg 8.49E+1 1 5.02 (2,4,2,3,1,3); Estimation, based on PG-Data
Mercury - - kg 1.32E+0 1 5.02 (2,4,2,3,1,3); Estimation, based on PG-Data
Nickel, ion - - kg 7.70E+3 1 5.02 (2,4,2,3,1,3); Estimation, based on PG-Data
Zinc, ion - - kg 2.05E+5 1 5.02 (2,4,2,3,1,3); Estimation, based on PG-Data
resource, in water Water, unspecified natural origin - - m3 2.87E+4 1 1.14 (2,4,2,3,1,3); Estimation, based on PG-Data
Gold, Au 1.1E-4%, Ag 4.2E-3%, in ore, in
resource, in ground - - kg 3.52E+3 1 1.14 (2,4,2,3,1,3); Estimation
ground
Silver, Ag 4.2E-3%, Au 1.1E-4%, in ore, in
- - kg 1.32E+05 1 1.14 (2,4,2,3,1,3); Estimation
ground

Tab. 5.27 Allocation factors of the data set “mining and refining, gold-silver deposit”
gold, from silver, from
Location
mining and combined gold- combined gold-
Unit
Name refining, gold- silver silver
silver deposit production, at production, at
refinery refinery
Location CL CL CL
Unit a kg kg
allocated CL 0 kg 2.99E+3 100 -
refinery
products CL 0 kg 1.12E+5 - 100
refinery
electricity, medium voltage, production UCTE,
technosphere UCTE 0 kWh 1.60E+8 60.9 39.1
at grid
diesel, burned in diesel-electric generating set GLO 0 MJ 4.18E+8 60.9 39.1

sodium hydroxide, 50% in H2O, production mix,
RER 0 kg 3.48E+4 60.9 39.1
at plant

blasting RER 0 kg 1.72E+6 60.9 39.1
infrastructure facilities precious metal refinery SE 1 unit 1.89E-2 97.6 2.4
mine, gold and silver CL 1 unit 2.00E-2 60.9 39.1
emission air, low
Copper - - kg 2.53E+0 60.9 39.1
population density
Lead - - kg 8.24E-1 60.9 39.1
Zinc - - kg 6.99E-1 60.9 39.1
Heat, waste - - MJ 5.76E+8 60.9 39.1
emission water, ocean Arsenic, ion - - kg 1.24E+2 60.9 39.1
Cadmium, ion - - kg 5.80E+2 60.9 39.1
Copper, ion - - kg 3.21E+4 60.9 39.1
Cyanide - - kg 5.22E+4 60.9 39.1
Lead - - kg 8.49E+1 60.9 39.1
Mercury - - kg 1.32E+0 60.9 39.1
Nickel, ion - - kg 7.70E+3 60.9 39.1
Zinc, ion - - kg 2.05E+5 60.9 39.1
resource, in water Water, unspecified natural origin - - m3 2.87E+4 60.9 39.1
resource, in ground - - kg 3.52E+3 100.0 -
ground
- - kg 1.32E+05 - 100.0
ground

Tab. 5.28 Ecospold Meta information of metal production in Chile
Location CL CL
Unit unit a
Type 1 5
Version 2.0 2.0
energyValues 0 0
LanguageCode en en
Person 44 44
QualityNetwork 1 1
IncludedProcesses This data set includes the combined mining and
refining of gold and silver in an open pit mine in Chile
Land use, buildings and machinery for mining
incl. energy and material use, water and air emissions
of metals and land use.
Amount 1 1
LocalName Mine, Gold und Silber Abbau und Raffination, Gold-Silber-Lagerstätte
Synonyms
GeneralComment This multi-output process 'mining and refining, gold-
This data set includes land use of the considered silver deposit' delivers the two co-products 'gold, from
mines incl. revegetated areas as well as estimates of combined gold-silver-production, at refinery' and
buildings and machinery of the mine based on the 'silver, from combined gold-silver-production, at
data set non-ferrous metal, mine, refinery'. This data set includes the mining in Chile in
surface/underground. open pit mine incl. land use. Air emissions are
estimated.
Formula
CASNumber
StartDate 2002 2002
EndDate 2006 2006
OtherPeriodText Period of public environmental report Period of public environmental report
Text Land use of Chilean-plant, other inputs based on Material requirements are estimated with Australian
rough assumptions. data
Text Assumptions for buildings and machinery based on
the data set "non-ferrous metal, mine, Refinement is estimated with electrolysis-data.
Percent
ProductionVolume 3.2 Mill. t ore per year 2.99 t gold and 112 t silver per year
SamplingProcedure environmental report environmental report
assumptions. The figures were extrapolated on the See Geography
basis of the ore production.
Person 44 44
DataPublishedIn 2 2
ReferenceToPublishedSource 10 10
Copyright 1 1
CompanyCode
CountryCode
PageNumbers
Validator 51 51
5.1.5 Gold and silver production, Peru (PE)

Yanacocha is a gold mine in northern Peru, considered one of the biggest and most profitable in the
world. The Yanacocha-Mine is located in the Cajamarca Province of the Peruvian highlands about 450
kilometres north of the capital Lima and produces gold and silver. Data for the Mine “Yanacocha”
stem from the environmental reports of Newmont (2005a) and Arana (2002). Newmont is the major
shareholder together with Buenaventura, a Peruvian company. In 2005, Yanacocha produced 103’000

kg of gold. Minera Yanacocha consists of five open pit mines, four leach pads, two gold recovery
plants, and a crushing and agglomeration facility.
The recovery rate (proportion of valuable material physically recovered in the processing of ore) is
taken into account in order to define the resource extraction. The average concentration (head grade,
expressed in kg metal per kg ore) of metals in ore for Peru is calculated with the quantity of ore
milled, the corresponding production rate and the recovery rate.
Tab. 5.29 Production, Allocation and Recovery rate of the various metals produced in 2005 by the Yanacocha-Mines
(Newmont, 2005a; Arana, 2002), the prices are from Tab. 2.5.
Yanagocha
Production Gold kg 1.03E+05 Value of 1kg Gold: 15680 US$
Production Silver kg 3.67E+04 Value of 1kg Silver: 269 US$
Head grade
Gold kg/kg ore 1.54E-06
Silver kg/kg ore 5.49E-07
Recovery rate 85%
area in use. Therefore, the total area is occupied during an average of about 42 years, according to the
following formula:

Tab. 5.30: Land use of the Yanacocha-Mine in 2005 (Newmont, 2005a)
Land use Remark

Area disturbed during life of mine 2 3.19E+07 Transformation, from unknown and Transformation, to
m
mineral extraction site,
Disturbed area in active use 2 2.68E+07 Occupation, mineral extraction site, during 50 years
m
Area available for revegetation 2 1.80E+06 Transformation, to unknown
m
Area revegetated 2 3.37E+06 Transformation, to heterogeneous, agricultural
m

Tab. 5.31: Unit process raw data of the gold and silver mine in Peru.
Infrastructur
StandardDe
Uncertainty
viation95%
eProcess
Location
Type
mine, gold
Unit
Name GeneralComment
and silver
Location PE
Unit unit
product mine, gold and silver PE 1 unit 1.00E+0
resource, land Transformation, from unknown - - m2 3.19E+7 1 2.01 (1,3,1,1,1,3); Environmental report, Newmont 2005
Transformation, to mineral extraction site - - m2 3.19E+7 1 2.01 (1,3,1,1,1,3); Environmental report, Newmont 2005
Occupation, mineral extraction site - - m2a 2.68E+7 1 1.51 (1,3,1,1,1,3); Environmental report, Newmont 2005
Transformation, from mineral extraction site - - m2 3.19E+7 1 2.01 (1,3,1,1,1,3); Environmental report, Newmont 2005
Transformation, to heterogeneous, agricultural - - m2 3.37E+6 1 2.01 (1,3,1,1,1,3); Environmental report, Newmont 2005
Transformation, to unknown - - m2 2.86E+7 1 2.01 (1,3,1,1,1,3); Environmental report, Newmont 2005
Mining and refining

The ore mined at Minera Yanacocha is broken up into small pieces and placed in layers, forming a hill
of ore known as a leach pad. A diluted sodium-cyanide solution is dripped onto the leach pad to ex-
tract the gold. The gold-laden solution is collected at the bottom of the pad for processing.
Table 5.25 displays the total energy and material requirements of the Yanagocha-Mines in 2005.

Tab. 5.32: Energy and material requirements of the Yanacocha-Mines in 2005 (Newmont, 2005a)
Yanagocha Remark
Energy
Electricity for mining kWh 3.37E+08 Environmental report, Peruvian mix 1) approximated with
the Brasilian mix
Electricity for refining kWh 1.89E+05 assumed with Möbius / Wohlwill-electrolysis, Peruvian mix
1) approximated with the Brasilian mix
Diesel (mobile equipment) MJ 4.86E+09 Diesel, burned in building machine
Petrol l 4.38E+04 converted in transport, van <3.5t (0.38l/tkm)
Oil MJ 1.41E+08 heavy fuel oil, burned in industrial furnace 1MW, non-
modulating
Materials .
Lime kg 1.19E+08 Environmental report
kg 1.20E+06 Scaled from australian value by processed gold and silver,
Caustic accounted for as Sodium Hydroxid
kg 1.75E+06 Scaled from australian value by processed gold & silver,
Carbon accounted for as Charcoal, at plant
kg 2.99E+05 Scaled from Steward (1999) by processed gold and silver,
accounted for as zinc, primary, at regional storage
Zinc
kg 5.43E+05 Acids (hydrochloric, sulphamic and sulphuric) consumption
is divided in half sulphuric, half hydrochloric (Newmont,
Sulphuric Acid 2006)
Hydrochloric Acid kg 5.43E+05 Environmental report
Explosives kg 4.26E+07 Partly ANFO explosives, approximates with Tovex
Water resources
3 4.40E+05 in LCI: Water, river
Surface Water Use m
3 2.72E+05 in LCI: Water, well, in ground
Groundwater Use m
3 2.77E+06
Mine dewatering Use m recycled Water, not accounted for in inventory
Emissions to Water
Arsenic, ion kg 3.07E+03 Estimation with PG-Data, scaled by milled ore
Cadmium, ion kg 1.44E+04 Estimation with PG-Data, scaled by milled ore
Copper, ion kg 7.96E+05 Estimation with PG-Data, scaled by milled ore
Cyanide kg 1.29E+06 Estimation with PG-Data, scaled by milled ore
Lead kg 2.10E+03 Estimation with PG-Data, scaled by milled ore
Mercury kg 3.26E+01 Estimation with PG-Data, scaled by milled ore
Nickel, ion kg 1.91E+05 Estimation with PG-Data, scaled by milled ore
Zinc, ion kg 5.07E+06 Estimation with PG-Data, scaled by milled ore
Emissions to Air
Arsenic kg 9.07E+00 Environmental Report
Carbon Monoxide kg 9.60E+04 Environmental Report, not accounted for
Carbon Dioxide (from explosives) kg 7.00E+06 Environmental Report
Lead kg 4.54E+01 Environmental Report
Mercury kg 3.45E+02 Environmental Report
Nitrogen Oxide (NOx) kg 1.50E+05 Environmental Report, not accounted for
Particulates kg 1.31E+04 Environmental Report, not accounted for
Sulfur Dioxide (SO2) kg 6.04E+04 Environmental Report, not accounted for
1 Peruvian electricity mix in 2005 (Nationmaster, 2004): fossil fuel 14.5%, hydro 84.7% others: 0.8%,
Brazilian electricity mix: fossil fuel 9.9%, hydro: 83.7, others: 6.4%
Since no reliable measures or values are available, water emissions are approximated with the figures
of the mines in Papua New Guinea. The use of charcoal and zinc is estimated with figures of Austra-
lian data for gold and silver production (Stewart, 1999, see also chapter 5.1.9). Airborne emissions as
carbon monoxide, nitrogen oxide, particulates and sulfur dioxide due to burning of fossil fuels are al-
ready inventoried and therefore not accounted in this LCI. The disposal of charcoal has been ne-
glected. Since the module “blasting” does not include the CO2 emissions from explosives, they are
calculated with the factor of the mines of Newmont (2005b) (0.16 kg CO2 per kg explosive). Cyanide
is accounted for as sodium cyanide, which is commonly used in the gold mining. Water emissions are
assumed to be emitted to rivers. Transports of materials are estimated to be 600 km per truck. Data

about amounts of tailings disposal is approximated with data of gold/silver mines in Papua New
Guinea.
figure of refinement from electronic scrap is used: 1.63 kWh per kg gold (Renner et al., 2002a). The
refinery of silver requires 0.6 kWh per kg of silver (Möbius electrolysis). No emissions were assumed,
since on the one hand the high value of the processed material requires an efficient emission control
and on the other hand, refinement is a hydrometallurgical process with low impact. The infrastructure
is approximated with the facility of precious metal refinery. The unit process raw data are summarized
in Tab. 5.33 .

Tab. 5.33: Unit process raw data of gold and silver mining and refining in Peru
StandardDeviation
UncertaintyType
Infrastructure
Location
mining and
Process
95%
Unit
silver deposit
Location PE
Unit a
allocated PE 0 kg 1.03E+5
refinery
products PE 0 kg 3.67E+4
refinery
technosphere electricity, medium voltage, at grid BR 0 kWh 3.37E+08 1 1.20 (3,4,1,5,1,3); Environmental report
diesel, burned in diesel-electric generating set GLO 0 MJ 7.88E+06 1 1.16 (3,4,1,2,1,3); Environmental report
transport, van <3.5t RER 0 tkm 1.15E+05 1 2.24 (5,3,1,1,1,3); Estimations
(3,3,3,5,1,3); Calculation, based on thesis
RER 0 kg 1.20E+06 1 1.20 on Australian precious metal production,
at plant
data for 1999
hydrochloric acid, 30% in H2O, at plant RER 0 kg 5.43E+05 1 1.14 (3,3,1,1,1,3); Environmental report
zinc, primary, at regional storage RER 0 kg 2.99E+05 1 1.20 on Australian precious metal production,
data for 1999
charcoal, at plant GLO 0 kg 1.75E+06 1 1.20 on Australian precious metal production,
data for 1999
sulphuric acid, liquid, at plant RER 0 kg 5.43E+05 1 1.14 (3,3,1,1,1,3); Environmental report
blasting RER 0 kg 4.26E+07 1 1.09 (1,3,1,1,1,3); Environmental report
transport, lorry >16t, fleet average RER 0 tkm 1.76E+08 1 2.24 (5,3,1,1,1,3); Estimations
infrastructure facilities precious metal refinery SE 1 unit 2.29E-02 1 3.09 (4,5,na,na,na,na); Environmental report
mine, gold and silver PE 1 unit 2.00E-02 1 3.05 (1,3,1,2,3,3); Annual report
emission air, low
Arsenic - - kg 9.07E+00 1 5.04 (3,3,1,5,1,3); Environmental report
population density
Mercury - - kg 3.45E+02 1 5.02 (3,3,1,1,1,3); Environmental report
(3,3,2,5,1,3); Estimated with data from
emission water, river Arsenic, ion - - kg 3.07E+03 1 5.04
combined production, PG
Cadmium, ion - - kg 1.44E+04 1 3.03
Copper, ion - - kg 7.96E+05 1 3.03
Cyanide - - kg 1.29E+06 1 3.03
Lead - - kg 2.10E+03 1 5.04
Mercury - - kg 3.26E+01 1 5.04
Nickel, ion - - kg 1.91E+05 1 5.04
Zinc, ion - - kg 5.07E+06 1 5.04
resource, in ground - - kg 1.21E+05 1 1.17 (3,3,1,5,1,3); Environmental report
ground
ground

Tab. 5.34 Allocation factors of the data set “mining, gold-silver deposit”
gold, from
Infrastructure
silver, from
Location
mining and combined gold-
Process
combined gold-
Unit
Name refining, gold- silver
silver production,
silver deposit production, at
at refinery
refinery
Location PE PE PE
Unit a kg kg
allocated PE 0 kg 1.03E+5 100 0
refinery
products PE 0 kg 3.67E+4 0 100
refinery
technosphere electricity, medium voltage, at grid BR 0 kWh 3.37E+08 99.42 0.6
diesel, burned in diesel-electric generating set GLO 0 MJ 7.88E+06 99.42 0.6
transport, van <3.5t RER 0 tkm 1.15E+05 97.6 2.4
RER 0 kg 1.20E+06 99.42 0.6
at plant

blasting RER 0 kg 4.26E+07 99.42 0.6
infrastructure facilities precious metal refinery SE 1 unit 2.29E-02 97.6 2.4
mine, gold and silver PE 1 unit 2.00E-02 99.42 0.6
emission air, low
Arsenic - - kg 9.07E+00 99.42 0.6
population density
Lead - - kg 4.54E+01 99.42 0.6
Mercury - - kg 3.45E+02 99.42 0.6
Heat, waste - - MJ 1.21E+09 99.42 0.6
emission water, river Arsenic, ion - - kg 3.07E+03 99.42 0.6
Cadmium, ion - - kg 1.44E+04 99.42 0.6
Copper, ion - - kg 7.96E+05 99.42 0.6
Cyanide - - kg 1.29E+06 99.42 0.6
Lead - - kg 2.10E+03 99.42 0.6
Mercury - - kg 3.26E+01 99.42 0.6
Nickel, ion - - kg 1.91E+05 99.42 0.6
Zinc, ion - - kg 5.07E+06 99.42 0.6

resource, in water Water, well, in ground - - m3 4.40E+05 99.42 0.6
Water, river - - m3 2.72E+05 99.42 0.6
resource, in ground - - kg 1.21E+05 100.00 -
ground
- - kg 4.31E+04 - 100.0
ground

5.35 EcoSpold Meta information of gold / silver mining and refining in Peru
Location PE PE
Unit unit a
Type 1 5
Version 2.0 2.0
energyValues 0 0
LanguageCode en en
Person 44 44
QualityNetwork 1 1
IncludedProcesses
This data set includes the combined mining and
refining of gold and silver in open pit mines in Peru
incl. energy and material use, water and air
emissions of metals and land use.
Amount 1 1
LocalName
Mine, Gold und Silber Abbau und Raffination, Gold-Silber-Lagerstätte
GeneralComment
This multi-output process 'mining and refining, gold-
This data set includes land use of the considered silver deposit' delivers the two co-products 'gold,
mines incl. revegetated areas as well as estimates from combined gold-silver-production, at refinery'
of buildings and machinery of the mine based on and 'silver, from combined gold-silver-production,
the data set non-ferrous metal, mine, at refinery'. This data set includes the mining in
surface/underground. Peru in open pit mines incl. land use. Air
emissions are estimated.
StartDate 2004 2004
EndDate 2006 2006
Text Material requirements are estimated with Australian
Land use of Peruvian-plant, other inputs based on
data, water emissions are estimated with Data from
rough assumptions.
Papua New Guinea
Text
Assumptions for buildings and machinery based on
The mining of ore from open pit mines is
the data set "non-ferrous metal, mine,
considered.
Percent
ProductionVolume 78.5 Mill. t ore per year 102.64 t gold and 36.66 t silver per year
See Geography, estimations have been made for
assumptions. The figures were extrapolated on the
refinement of Gold
Person 44 44
DataPublishedIn 2 2
Copyright 1 1
CompanyCode
CountryCode
PageNumbers
Validator 51 51
5.1.6 Gold production, Canada (CA)

Data stem from the environmental report of Placer Dome Inc. (2002b) for the Campbell, Musselwhite
and Porcupine Mines. Since the mines have been sold in 2006 to Kinross Gold Corporation and Gold-
corp Inc., no actual data is available. Both enterprises do not publish their environmental data. There-
fore, data from 2002 is used to model the gold production in Canada.

The mines produce only gold. Campbell and Musselwhite are underground, Porcupine has under-
ground and open pit mines. Tab. 5.36 shows the production volume in year 2002 of the Campbell,
Musselwhite and Porcupine Mines. The recovery rate (proportion of valuable material physically re-
covered in the processing of ore) is taken into account in order to define the resource extraction. The
average concentration (head grade, expressed in kg metal per kg ore) of metals in ore for Canada is
calculated with the quantity of ore milled, the corresponding production rate and the recovery rate.
Tab. 5.36: Production volume in 2002 of the Campbell, Musselwhite and Porcupine Mines (Placer Dome Inc., 2003a).
Production Campbell Musselwhite Porcupine Canada
Ore milled kg 3.57E+08 7.87E+08 1.10E+09 2.24E+09

Production Gold kg 6.01E+03 4.43E+03 3.17E+03 1.36E+04
Head grade
Gold kg/kg 1.98E-05 6.62E-06 3.39E-06 7.14E-06
Recovery rate 85%
Tab. 5.37: Land use of the Campbell, Musselwhite and Porcupine Mines in 2002 (Placer Dome Inc., 2002b).
Land use Remark

Campbell Musselwhite Porcupine Canada
Area disturbed during m 2 1.60E+06 4.44E+06 1.08E+08 1.14E+08 Transformation, from unknown and
life of mine Transformation, to mineral
extraction site,
Disturbed area in m
2 6.80E+05 2.91E+06 4.70E+07 5.06E+07 Occupation, mineral extraction site,
active use during 50 years
Area available for m
3 1.50E+05 1.31E+06 4.80E+07 4.95E+07
revegetation Transformation to unknown
Area revegetated m
2 4.70E+05 8.00E+04 1.08E+07 1.14E+07 added in "Transformation, to
heterogeneous, agricultural"
Area that meets final m
3 3.00E+05 0.00E+00 2.20E+06 2.50E+06 added in "Transformation, to
objectives heterogeneous, agricultural"
formula


Tab. 5.38: Unit process raw data for the gold and silver mine in Canada.
Infrastructur
StandardDe
Uncertainty
viation95%
eProcess
Location
Type
Unit
Name mine, gold GeneralComment
Location CA
Unit unit
product mine, gold CA 1 unit 1.00E+0
Mining and refining

In the Porcupine Mines, gold is recovered using a combination of gravity concentration, milling and
cyanidation techniques. The milling process consists of primary crushing, secondary crushing, rod/ball
mill grinding, gravity concentration, cyanide leaching, carbon-in-pulp gold recovery, stripping, elec-
trowinning and refining.
In the Campbell Mine, the ore from the mine, after crushing and grinding, is processed by gravity
separation, flotation, pressure oxidation, cyanidation and carbon-in-pulp process followed by electro-
winning and gold refining to doré on site.
The Musselwhite Mine uses gravity separation, carbon in pulp, electrowinning and gold refining to
doré on site. The following table (Tab. 5.39) shows the total requirements and emissions of the Camp-
bell, Musselwhite and Porcupine Mines in 2002 as well as the total for Canada.

Tab. 5.39: Annual requirements and emissions of the Campbell, Musselwhite and Porcupine Mines in 2002 (Placer
Dome Inc., 2002b).
Energy Campbell Musselwhite Porcupine Canada Remark

Electricity for kWh 8.08E+07 6.17E+07 1.64E+08 3.07E+08 Canadian mix approximated with the NORDEL
mining mix
1)
Electricity for kWh n.a. n.a. n.a. 2.22E+04 assumed with Möbius / Wohlwill-electrolysis
refining
Diesel (generators) MJ 7.42E+05 2.56E+06 6.70E+06 1.00E+07 Diesel, burned in diesel-electric generating set
2)
Diesel (mobile l 1.02E+07 3.53E+07 9.24E+07 1.38E+08 Diesel, burned in building machine
2)
equipment)
Propane (LPG) MJ 7.10E+07 2.45E+08 6.41E+08 9.57E+08 2)
Transport, van tkm 1.20E+05 3.32E+05 8.69E+05 1.32E+06 Approximation for gasoline, calculated with
<3.5t specific fuel consumption (0.35l / tkm)
2)
Materials
Cyanide kg 2.49E+05 4.15E+05 1.19E+06 1.85E+06 In LCI as sodium cyanide
Lime kg n.a. n.a. n.a. 8.60E+05 Scaled from Steward (1999) by processed gold
Caustic kg n.a. n.a. n.a. 1.58E+05 Scaled from Steward (1999) by processed gold,
accounted for sodium hydroxide
Carbon kg n.a. n.a. n.a. 2.31E+05 Scaled from Steward (1999) by processed gold,
accounted for as charcoal
Zinc kg n.a. n.a. n.a. 3.94E+04 Scaled from Steward (1999) by processed gold,
accounted for as zinc, primary, at regional
storage
Sulphuric Acid kg n.a. n.a. n.a. 9.06E+03 Scaled from Steward (1999) by processed gold
Hydrochloric Acid kg n.a. n.a. n.a. 1.86E+05 Scaled from Steward (1999) by processed gold
Explosives kg n.a. n.a. n.a. 2.16E+05 Scaled from US-Mines by milled ore, in LCI:
"blasting"
Water resources
Groundwater m
3 9.55E+05 2.68E+06 1.10E+06 4.74E+06 in LCI: Water, well, in ground
Water from onsite 5.80E+05 0 0 5.80E+05 Recycled Water: not accounted for in inventory
impoundments
Emissions to
Water
Arsenic, ion kg 8.00E+00 4.10E+00 5.10E+01 6.31E+01 Environmental Report (Placer Dome, 2002)
Cadmium, ion kg 0.00E+00 5.00E-01 1.40E+01 1.45E+01 Environmental Report (Placer Dome, 2002)
Copper, ion kg 1.30E+01 7.60E+00 1.52E+02 1.73E+02 Environmental Report (Placer Dome, 2002)
Cyanide kg 1.80E+01 4.20E+01 3.21E+02 3.81E+02 Environmental Report (Placer Dome, 2002)
Lead kg 7.00E-01 4.90E+00 2.60E+01 3.16E+01 Environmental Report (Placer Dome, 2002)
Mercury kg 1.00E-02 2.00E-02 1.50E-01 1.80E-01 Environmental Report (Placer Dome, 2002)
Nickel, ion kg 4.20E+01 2.10E+00 1.69E+02 2.13E+02 Environmental Report (Placer Dome, 2002)
Zinc, ion kg 2.30E+01 9.70E+00 5.20E+01 8.47E+01 Environmental Report (Placer Dome, 2002)
Emissions to Air
Carbon Dioxide kg n.a. n.a. n.a. 3.45E+04 Calculation, based on data from Newmont
(from explosives) (2005)
Zinc kg n.a. n.a. n.a. 4.95E-01 estimated with Data from SE, scaled by milled
ore
Copper kg n.a. n.a. n.a. 1.79E+00 estimated with Data from SE, scaled by milled
ore
Lead kg n.a. n.a. n.a. 5.83E-01 estimated with Data from SE, scaled by milled
ore
Disposal kg n.a. n.a. n.a. 1.59E+09 estimated with data from PG, in LCI as:
Disposal, sulphidic tailings
1
Electricity mix Canada in 2000 (GENI (2004)): fossil 26%, hydro 61%, nuclear 12%, other 2%. Camp-
bell mine uses only hydro power.
2
The consumption of the different energy sources in Musselwhite and Porcupine is calculated on the
basis of their share in the figures of the Campbell mine.
Material use is approximated with figures from Australian data for gold and silver production
(Stewart, 1999), whereas air emissions (zinc, copper and lead) are modelled with Swedish data. As
Swedish data stem for a combined process, they are allocated to gold and silver for Sweden and used

in this form for Canada. Electricity supply mix of Canada is approximated with the NORDEL electric-
ity mix (Nordic countries power association) that has a related share of hydroelectric power production
(The higher part of fossil electricity generation in NORDEL is somewhat compensated by the use of
exclusively hydro by the Campbell-Mine). Explosive use is estimated with Data from US. Since the
module “blasting” not includes the CO2 emissions from explosives, they are calculated with the factor
of the mines of Newmont (2005b) (0.16 kg CO2 per kg explosive). Water emissions are assumed to be
emitted to rivers. Estimation of transport is based on standard distances. Data about amounts of tail-
ings disposal is approximated with data of mines in Papua New Guinea.
figure of refinement from electronic scrap is used: 1.63 kWh per kg gold (Renner et al., 2002a). No
emissions were assumed, since on the one hand the high value of the processed material requires an
efficient emission control and on the other hand, refinement is a hydrometallurgical process with low
impact. The infrastructure is approximated with the facility of precious metal refinery. The unit proc-
ess raw data are summarized in Tab. 5.40.

Tab. 5.40: Unit process raw data of gold mining and refining in Canada.
Infrastructure Process
StandardDeviation95
UncertaintyType
Location
Unit
Name gold, at refinery GeneralComment
%
Location CA
Unit kg
product gold, at refinery CA 0 kg 1.00E+0
electricity, medium voltage, production
technosphere NORDEL 0 kWh 2.25E+4 1 1.14 (2,4,2,3,1,3); Environmental report (Placer-Dome, 2002)
NORDEL, at grid
diesel, burned in diesel-electric
GLO 0 MJ 7.35E+2 1 1.14 (2,4,2,3,1,3); Environmental report (Placer-Dome, 2002)
generating set
diesel, burned in building machine GLO 0 MJ 1.01E+4 1 1.14 (2,4,2,3,1,3); Environmental report (Placer-Dome, 2002)
natural gas, burned in industrial furnace (2,4,2,3,1,3); Calculation, based on thesis on Australian
RER 0 MJ 7.03E+4 1 1.14
>100kW precious metal production, data for 1999
transport, van <3.5t RER 0 tkm 9.70E+1 1 2.02
limestone, milled, packed, at plant CH 0 kg 6.32E+1 1 1.14
sodium hydroxide, 50% in H2O, (2,3,2,5,1,3); Calculation, based on thesis on Australian
RER 0 kg 1.16E+1 1 1.15
production mix, at plant precious metal production, data for 1999
charcoal, at plant GLO 0 kg 1.70E+1 1 1.15
zinc, primary, at regional storage RER 0 kg 2.89E+0 1 1.15
sulphuric acid, liquid, at plant RER 0 kg 6.65E-1 1 1.15
blasting RER 0 kg 1.59E+1 1 1.11 (2,3,2,3,1,3); Estimation, based on US-Data
transport, lorry >16t, fleet average RER 0 tkm 2.29E+2 1 2.02 (2,4,na,na,na,na); Calculation, based on standard distances
transport, freight, rail RER 0 tkm 4.58E+2 1 2.02 (2,4,na,na,na,na); Calculation, based on standard distances
infrastructure facilities precious metal refinery SE 1 unit 1.65E-7 1 3.02 (2,4,2,3,1,3); Estimation
mine, gold CA 1 unit 1.47E-6 1 3.01 (2,3,2,3,1,3); Environmental report
emission air, low
Copper - - kg 1.32E-4 1 5.02 (2,4,2,3,1,3); Estimation, based on Swedish Data
population density
Lead - - kg 4.28E-5 1 5.02 (2,4,2,3,1,3); Estimation, based on Swedish Data
Zinc - - kg 3.64E-5 1 5.02 (2,4,2,3,1,3); Estimation, based on Swedish Data
Carbon dioxide, fossil - - kg 2.54E+0 1 1.14 (2,4,2,3,1,3); Estimation, based on Swedish Data
emission water, ocean Arsenic, ion - - kg 4.64E-3 1 5.01 (1,3,2,3,1,3); Environmental report (Placer-Dome, 2002)
Cadmium, ion - - kg 1.07E-3 1 3.01 (1,3,2,3,1,3); Environmental report (Placer-Dome, 2002)
Copper, ion - - kg 1.27E-2 1 3.01 (1,3,2,3,1,3); Environmental report (Placer-Dome, 2002)
Cyanide - - kg 2.80E-2 1 3.01 (1,3,2,3,1,3); Environmental report (Placer-Dome, 2002)
Lead - - kg 2.32E-3 1 5.01 (1,3,2,3,1,3); Environmental report (Placer-Dome, 2002)
Mercury - - kg 1.32E-5 1 5.01 (1,3,2,3,1,3); Environmental report (Placer-Dome, 2002)
Nickel, ion - - kg 1.57E-2 1 5.01 (1,3,2,3,1,3); Environmental report (Placer-Dome, 2002)
Zinc, ion - - kg 6.22E-3 1 5.01 (1,3,2,3,1,3); Environmental report (Placer-Dome, 2002)
resource, in water Water, well, in ground - - m3 3.48E+2 1 1.10 (1,3,2,3,1,3); Environmental report (Placer-Dome, 2002)
resource, in ground Gold, Au 7.1E-4%, in ore, in ground - - kg 1.18E+0 1 1.10 (1,3,2,3,1,3); Estimation

Tab. 5.41 Ecospold Meta information of gold production in Canada
Name mine, gold gold, at refinery

Location CA CA
Unit unit kg
Type 1 1
Version 2.0 2.0
energyValues 0 0
LanguageCode en en
Person 44 44
QualityNetwork 1 1
IncludedProcesses This data set describes the mining of gold concentrate in
Canada. It includes energy and water consumption, land
use, transports, waste disposal as well as emissions to
air and water.
Amount 1 1
LocalName Mine, Gold Gold, ab Raffinerie
Synonyms
GeneralComment This data set includes land use of the considered mines
(open pit and underground) incl. revegetated areas as
This data set describes the beneficiation, concentration
well as estimates of buildings and machinery of the mine
and refinement of gold.
based on the data set non-ferrous metal, mine,
surface/underground.
Formula
CASNumber 007440-57-5
StartDate 2003 2003
EndDate 2006 2006
Text Material requirements for gold concentration are
Land use of Canadian mine, other inputs based on rough
estimated with Australian data, air emission are
assumptions.
estimated with Swedish data.
Text Assumptions for buildings and machinery based on the
The mining of ore from open pit and underground mines
data set "non-ferrous metal, mine, surface/underground
is considered.
with 50 years life time.
Percent
ProductionVolume 2.2 Mill. t ore 13.61 t gold
See Geography, estimation of conversion of transport
assumptions. The figures are extrapolated on the basis of
fuel to tkm
the ore production.
Person 44 44
DataPublishedIn 2 2
Copyright 1 1
CompanyCode
CountryCode
PageNumbers
Validator 51 51
5.1.7 Gold production, United States of America (US)

Data stem from the environmental report of Barrick (2006b) for the Bald Mountain, Cortez, Golden
Sunlight, Goldstrike and Round Mountain Mines. Since the mines have been sold in 2006 from the
company of Placer Dome to Barrick, previous environmental reports of Placer Dome Inc. (2002b) are
used as well.
The data about the mines of Nevada and Golden Giant are from Newmont (2005b). The mines are ei-
ther underground or open pits. All mines produce only gold. Tab. 5.42 shows the production volume
2005 of the considered US Mines. The recovery rate (proportion of valuable material physically re-
covered in the processing of ore) is taken into account in order to define the resource extraction. The
average concentration (head grade, expressed in kg metal per kg ore) of metals in ore for United States
is calculated with the quantity of ore milled, the corresponding production rate and the recovery rate.

Land use is based on the figures from the environmental report of Newmont (2005b) and the previous
environmental report of Placer Dome Inc. (2002b). Not all the area of the mine is in active use at one
time; the value in Tab. 5.56 is assumed to be the average share of the mine area in use. Therefore, the
total area is occupied during an average of about 43 years, according to the formula:


Tab. 5.42 Production volume 2005 of the Golden Giant, Nevada, Bald Mountain, Cortez, Golden Sunlight, Goldstrike, and Round Mountain mines (Barrick, 2006b; Newmont,
2005b)
GoldenGiant Nevada Bald Mountain Cortez Golden Sunlight Goldstrike Round Mountain United States Remark
Production Gold kg 5.04E+03 7.46E+04 8.65E+03 1.31E+04 2.89E+03 5.79E+04 2.12E+04 1.83E+05
Recovery rate n.a. n.a. 91% 89% 78% 88% 88% 88% Recovery rate, average value for the United States, based
on data of Placer-Dome, (2002)
Ore milled kg 8.35E+08 2.32E+11 6.04E+09 3.08E+10 3.08E+09 1.28E+10 6.60E+10 3.52E+11
Head grade
Gold kg/kg ore 6.85E-06 3.66E-07 1.79E-06 5.43E-07 1.37E-06 5.84E-06 4.14E-07 5.93E-07
Tab. 5.43 Land use of the Golden Giant, Nevada, Bald Mountain, Cortez, Golden Sunlight, Goldstrike and Round Mountain mines (Barrick, 2006b; Newmont, 2005b)
Land use GoldenGiant Nevada Bald Mountain Cortez Golden Sunlight Goldstrike Round Mountain United States Remark
Area disturbed during m
2 2.72E+06 1.33E+08 1.50E+07 2.89E+07 9.49E+06 3.80E+08 2.50E+08 8.19E+08 Transformation, from unknown and Transformation, to
life of mine mineral extraction site,
Disturbed area in active m
2 2.12E+06 9.87E+07 8.63E+06 2.25E+07 4.14E+06 3.40E+08 2.28E+08 7.04E+08 Occupation, mineral extraction site, during 50 years
use
2 2.00E+04 7.69E+06 1.30E+06 1.05E+06 1.28E+06 3.15E+07 2.80E+07 7.08E+07 Transformation, to unknown
revegetation
Area revegetated m
2 5.80E+05 2.43E+07 5.04E+06 5.30E+06 2.30E+05 0.00E+00 0.00E+00 3.55E+07 Transformation, to heterogeneous, agricultural

Tab. 5.44: Unit process raw data for the gold and silver mine in the USA.
Infrastructur
StandardDe
Uncertainty
viation95%
eProcess
Location
Type
Unit
Location US
Unit unit
product mine, gold US 1 unit 1.00E+0
report, Newmont 2005
Mining and refining

In the Bald Mountain Mines, the ore treatment method is based on conventional heap leaching tech-
nology followed by carbon absorption. The loaded carbon is stripped and refined in the newly com-
missioned refinery on site. Water is supplied by wells located on the mine property. Grid power was
brought to Bald Mountain Mine in 1996. For this purpose, one 27-kilometre 69 KVA power line was
constructed from the Alligator Ridge Mine substation to the grid.
At the Golden Sunlight Mines, the ore treatment plant is based on conventional carbon-in-pulp tech-
nology, with the addition of a Sand Tailings Retreatment (STR) gold recovery plant to recover gold
that would otherwise be lost to tailings. The STR circuit removes the heavier gold bearing pyrite from
the sand portion of the tailings by gravity separation. The gold is refined into doré at the mine. Tailing
from the mill is discharged to an impoundment area where the solids are allowed to settle so the water
can be reused. A cyanide recovery/destruction process was commissioned in 1998. It eliminates the
hazard posed to wildlife at the tailings impoundment by lowering cyanide concentrations below
20 mg/l. Fresh water for ore processing, dust suppression, and fire control is supplied from the Jeffer-
son Slough, which is an old natural channel of the Jefferson River. Ore processing also uses water
pumped from the tailings impoundment. Pit water is treated in a facility located in the mill complex
prior to disposal or for use in dust control. Drinking water is made available by filtering fresh water
through an on-site treatment plant. Electric power is provided from a substation at the south property
boundary. North-Western Energy supplies electricity the substation. Small diesel generators are used
for emergency lighting. A natural gas pipeline supplies gas for heating buildings, a crusher, air scrub-
ber, boiler, carbon reactivation kiln, and refining furnaces.
At the Cortez Mine, three different metallurgical processes are employed for the recovery of gold. The
process used for a particular ore is determined based on grade and metallurgical character of that ore.
Lower grade oxide ore is heap leached, while higher-grade non-refractory ore is treated in a conven-
tional mill using cyanidation and a carbon-in-leach (“CIL”) process. When carbonaceous ore is proc-
essed by Barrick, it is first dry ground, and then oxidized in a circulating fluid bed roaster, followed by
CIL recovery. In 2002 a new leach pad and process plant was commissioned. This facility will be ca-
pable of processing 164 million tonnes of heap leach ore over the life of the asset. Heap leach ore pro-
duction is hauled directly to heap leach pads for gold recovery. Water for process use is supplied from
the open pit dewatering system. Approximately 90 litres per second of the pit dewatering volume is
diverted for plant use. Electric power is supplied by Sierra Pacific Power Company (“SPPC”) through
a 73 kilometre, 120 kV transmission line. A long-term agreement is in place with SPPC to provide

power through the regulated power system. The average power requirement of the mine is about 160
GWh/year.
Further information about the mines and their infrastructure can be found in the corresponding chap-
ters of Barrick (2006b); Newmont (2005b) and Barrick (2006b). The following table (Tab. 5.45)
shows the requirements and emissions of the considered mines as well as the US total.
Material use of zinc and carbon is approximated with figures from Australian data for gold and silver
production (Stewart, 1999). The consumption of Lime and caustic (Sulphuric Acid / Hydrochloric
Acid) was taken from the environmental reports (where available), other ways estimated with the Aus-
tralian values (Stewart, 1999).
The water emissions are estimated with data of Australia, it is assumed that they are emitted to rivers.
Since the module “blasting” does not include the CO2 emissions from explosives, they are calculated
with the factor of the mines of Newmont (2005b) (0.16 kg CO2 per kg explosive). The transport of ma-
terials is calculated with ecoinvent standard distances for Europe. Data of tailing disposal are ap-
proximated with data from Papua New Guinea.

Tab. 5.45 Annual requirements of or attributed to the Golden Giant, Nevada, Bald Mountain, Cortez, Golden Sunlight, Goldstrike and Round Mountain mines and total for US
Energy GoldenGiant Nevada Bald Mountain Cortez Golden Sunlight Goldstrike Round Mountain United States Remark
Electricity, medium voltage, US, share for Barrick-Mines is
Electricity for mining kWh 1.14E+08 1.37E+09 1.44E+06 3.45E+07 2.77E+07 2.04E+08 3.28E+07 1.79E+09 based on actual total energy consumption for older share
of electricity (Placer Dome, 2002)
Electricity for refining kWh n.a. n.a. n.a. n.a. n.a. n.a. n.a. 2.99E+05 assumed with Wohlwill-electrolysis
Diesel, burned in diesel-electric generating set, share for
Diesel (generators) MJ 0.00E+00 3.20E+04 4.66E+07 6.43E+08 1.67E+08 3.04E+09 4.90E+08 4.39E+09 Barrick-Mines is based on actual total energy consumption
for older Diesel in generators (Placer Dome, 2002)
Diesel, burned in building machine, share for Barrick-
Diesel (mobile Mines is based on actual total energy consumption for
MJ 2.01E+07 5.27E+09 4.74E+08 1.71E+09 3.26E+08 4.14E+09 1.66E+09 1.36E+10
equipment) older share of diesel in mobile equipment (Placer Dome,
2002)
converted in transport, van <3.5t (0.38l/tkm), consumption
Petrol l 2.57E+03 1.20E+05 n.a. n.a. n.a. n.a. n.a. 1.23E+05
in Barrick-Mines is included in Diesel (mobile equipment)
Oil MJ 2.66E+06 1.33E+08 n.a. n.a. n.a. n.a. n.a. 1.36E+08 modulating, consumption in Barrick-Mines is included in
Diesel (mobile equipment)
Materials . .
Cyanide kg 1.10E+05 9.38E+06 1.35E+06 4.52E+06 1.97E+06 4.91E+06 1.15E+07 9.49E+06 In LCI as sodium cyanide
Data for Golden Giant and Nevada from environmental
Lime kg 5.09E+05 2.01E+08 5.47E+05 8.29E+05 1.83E+05 3.66E+06 1.34E+06 2.08E+08 report, other data estimatet from Steward (1999) by
processed gold
Caustic kg 0.00E+00 3.29E+04 1.01E+05 1.53E+05 3.37E+04 6.74E+05 2.46E+05 1.24E+06 report, other data estimatet from Steward (1999) by
processed gold
Scaled from Steward (1999) by processed gold, accounted
Carbon kg 8.55E+04 1.27E+06 1.47E+05 2.23E+05 4.91E+04 9.82E+05 3.59E+05 3.11E+06
for as Charcoal, at plant
Scaled from Steward (1999) by processed gold, accounted
Zinc kg 1.46E+04 2.16E+05 2.50E+04 3.80E+04 8.37E+03 1.67E+05 6.12E+04 5.31E+05
for as zinc, primary, at regional storage
Sulphuric Acid kg 2.08E+04 3.59E+07 5.75E+03 8.73E+03 1.92E+03 3.85E+04 1.41E+04 3.60E+07 report, other data estimatet from Steward (1999) by
processed gold
Hydrochloric Acid kg 2.08E+04 3.59E+07 1.18E+05 1.79E+05 3.95E+04 7.89E+05 2.89E+05 3.73E+07 report, other data estimatet from Steward (1999) by
processed gold
Explosives kg 2.39E+05 3.35E+07 5.38E+03 1.52E+04 4.27E+03 1.94E+04 7.21E+03 3.38E+07 Partly ANFO explosives, in LCI: "blasting"
Water resources
3
Surface Water Use m 1.01E+05 0.00E+00 n.a. n.a. n.a. n.a. n.a. 1.01E+05 in LCI:Water, river
3
Groundwater Use m 0.00E+00 8.45E+06 2.09E+05 3.68E+06 5.65E+05 1.23E+07 6.74E+06 3.19E+07 in LCI:Water, well, in ground
3
Mine dewatering Use m 1.90E+04 2.48E+07 n.a. n.a. n.a. n.a. n.a. 2.48E+07 recycled Water, not accounted for in inventory

Tab. 5.46 Annual emissions of or attributed to the Golden Giant, Nevada, Bald Mountain, Cortez, Golden Sunlight, Goldstrike and Round Mountain mines and total for US
Emissions to Water GoldenGiant Nevada Bald Mountain Cortez Golden Sunlight Goldstrike Round Mountain United States
Arsenic kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 9.80E+03 Total based on australian value, scaled on milled ore
Cadmium kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 4.42E+03 Total based on australian value, scaled on milled ore
Copper kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 8.01E+04 Total based on australian value, scaled on milled ore
Cyanide kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 1.31E+05 Total based on australian value, scaled on milled ore
Lead kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 2.70E+05 Total based on australian value, scaled on milled ore
Mercury kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 4.91E+03 Total based on australian value, scaled on milled ore
Nickel kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 4.42E+04 Total based on australian value, scaled on milled ore
Zinc kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 4.91E+05 Total based on australian value, scaled on milled ore
Emissions to Air GoldenGiant Nevada 0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 United States
Total based on Golden Giant, environmental report of
Antimony kg 2.63E+02 n.a. n.a. n.a. n.a. n.a. n.a. 1.11E+05
Newmont (2005)
Arsenic kg 9.07E+00 n.a. n.a. n.a. n.a. n.a. n.a. 3.82E+03
Newmont (2005)
Beryllium kg 0.00E+00 n.a. n.a. n.a. n.a. n.a. n.a. 1.02E+03 Total based on australian value, scaled by milled ore
Cadmium kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 5.42E+01 Total based on australian value, scaled by milled ore
environmental report of Newmont (2005), not accounted
Carbon Monoxide kg 1.40E+03 2.41E+05 n.a. n.a. n.a. n.a. n.a. -
for in LCI as emissions are already included in fossil fuels
Carbon Dioxide (from Total based on Nevada, environmental report of Newmont
kg 7.00E+06 7.00E+06 2.16E+05 7.23E+05 3.14E+05 7.86E+05 1.84E+06 1.40E+07
explosives) (2005)
Chromium kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 8.86E+03 Total based on australian value, scaled by milled ore
Cobalt kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 1.53E+03 Total based on australian value, scaled by milled ore
Lead kg 9.07E+00 n.a. n.a. n.a. n.a. n.a. n.a. 3.82E+03
Newmont (2005)
Manganese kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 4.80E+04 Total based on australian value, scaled by milled ore
Total based on Nevada, environmental report of Newmont
Mercury kg 9.07E+00 3.90E+02 n.a. n.a. n.a. n.a. n.a. 5.91E+02
(2005)
Nitrogen Oxide (NOx) kg 4.96E+03 1.18E+05 n.a. n.a. n.a. n.a. n.a. -

Particulates (PM10) kg 5.21E+04 2.80E+05 n.a. n.a. n.a. n.a. n.a. -
Selenium kg n.a. n.a. n.a. n.a. n.a. n.a. n.a. 2.93E+04 Total based on australian value, scaled by milled ore
Sulfur Dioxide (SO2) kg 9.98E+01 9.79E+04 n.a. n.a. n.a. n.a. n.a. -

Tab. 5.47: Unit process raw data of gold mining and refining in USA.
UncertaintyType
StandardDeviati
Infrastructure
Location
Process
on95%
Unit
Location US
Unit kg
product gold, at refinery US 0 kg 1.00E+00
technosphere electricity, medium voltage, at grid US 0 kWh 9.76E+03 1 1.29 (3,4,1,5,3,3); Environmental report
heavy fuel oil, burned in industrial furnace 1MW,
non-modulating
RER 0 kg 6.76E+00 1 1.17 on Australian precious metal production,
at plant
data for 1999
data for 1999
data for 1999
infrastructure facilities precious metal refinery SE 1 unit 1.65E-07 1 3.05 (1,3,1,1,3,3); Annual report
mine, gold US 1 unit 1.09E-07 1 3.05 (1,3,1,1,3,3); Annual report
emission air, low
Arsenic - - kg 2.08E-02 1 5.03 (3,4,1,1,1,3); Environmental report
population density
emission air, low
Antimony - - kg 6.04E-01 1 5.03 (3,4,1,1,1,3); Environmental report
population density
Beryllium - - kg 5.58E-03 1 5.04
Australian data
Cadmium - - kg 2.96E-04 1 5.04
Australian data
Lead - - kg 2.08E-02 1 5.03 (3,4,1,1,1,3); Environmental report
Chromium - - kg 4.83E-02 1 5.04
Australian data
Cobalt - - kg 8.37E-03 1 5.04
Australian data
Mercury - - kg 3.22E-03 1 5.03 (3,4,1,1,1,3); Environmental report
Manganese - - kg 2.62E-01 1 5.04
Australian data
Selenium - - kg 1.54E-01 1 5.04
Australian data
emission water, river Arsenic, ion - - kg 5.35E-02 1 5.04
Australian data
Cadmium, ion - - kg 2.41E-02 1 3.03
Australian data
Copper, ion - - kg 4.37E-01 1 3.03
Australian data
Cyanide - - kg 7.14E-01 1 3.03
Australian data
Lead - - kg 1.47E+00 1 5.04
Australian data
Mercury - - kg 2.68E-02 1 5.04
Australian data
Nickel, ion - - kg 2.41E-01 1 5.04
Australian data
Zinc, ion - - kg 2.68E+00 1 5.04
Australian data
resource, in water Water, well, in ground - - m3 5.51E-01 1 1.14 (3,3,1,1,1,3); Environmental report
resource, in ground Gold, Au 1.4E-4%, in ore, in ground - - kg 1.14E+00 1 1.14 (3,3,1,1,1,3); Environmental report

Tab. 5.48 Ecospold Meta information of gold production in United States

Location US US
Unit unit kg
Type 1 1
Version 2.0 2.0
energyValues 0 0
LanguageCode en en
Person 44 44
QualityNetwork 1 1
IncludedProcesses This data set describes the mining of gold
concentrate in United States. It includes
Land use, buildings and machinery of mining energy and water consumption, land use,
transports, waste disposal as well as
emissions to air and water.
Amount 1 1
Synonyms
GeneralComment
This data set includes land use of the
considered mines (open pit and underground)
incl. revegetated areas as well as estimates of This data set describes the beneficiation,
buildings and machinery of the mine based on concentration and refinement of gold.
the data set non-ferrous metal, mine,
Formula
StartDate 2003 2003
EndDate 2006 2006
OtherPeriodText
Period of public environmental report Period of public environmental report
Text Material requirements for gold concentration

Land use of US-mine, other inputs based on are estimated with Australian data, emissions
rough assumptions. to water and air are estimated with Australian
data
Text Assumptions for buildings and machinery
based on the data set "non-ferrous metal, The mining of ore from open pit and
mine, surface/underground with 50 years life underground mines is considered.
time.
Percent
Extrapolations
Buildings and machinery are based on global
See Geography, several values have been
assumptions. The figures are extrapolated on
scaled from one mine
the basis of the ore production.
Person 44 44
DataPublishedIn 2 2
Copyright 1 1
CompanyCode
CountryCode
PageNumbers
Validator 51 51

5.1.8 Gold production, South Africa (ZA)

Data stem from the environmental report of Western Areas Limited (2006) for the South Deep Mine.
The mine produces only gold and is an underground mine. Tab. 5.49 shows the production volume
2005 of the South Deep Mine. The recovery rate (proportion of valuable material physically recovered
in the processing of ore) is taken into account in order to define the resource extraction. The average
concentration (head grade, expressed in kg metal per kg ore) of metals in ore for South Africa is calcu-
lated with the quantity of ore milled, the corresponding production rate and the recovery rate.
Tab. 5.49: Production volume of the South Deep Mine in 2005 (Western Areas Limited, 2006).
South Deep
Head grade
Recovery rate 97%
Land use is based on the figures from the previous environmental report of Placer Dome Inc. (2002b)
see Tab. 5.50).
Tab. 5.50: Land use of the South Deep Mine in 2002 (Placer Dome Inc., 2002b)
Land use Remark

Area disturbed during life of mine 2 4.61E+06 Transformation, from unknown and Transformation, to mineral
m
extraction site,
Disturbed area in active use 2 4.61E+06 Occupation, mineral extraction site, during 50 years
m
Area available for revegetation 2 0.00E+00 Transformation, to unknown
m
Area revegetated 2 0.00E+00 Transformation, to heterogeneous, agricultural
m

Tab. 5.51: Unit process raw data of the gold mine in South Africa.
Infrastructur
StandardDe
Uncertainty
viation95%
eProcess
Location
Type
Unit
Location ZA
Unit unit
product mine, gold ZA 1 unit 1.00E+0
report, Placer-Dome 2002
Transformation, to heterogeneous, agricultural - - m2 0 1 2.01
Mining and refining

The ore is mined by conventional underground long wall mining methods (50%) and mechanized drift
and fill trackless methods (50%). The conventional stope ore is drilled, blasted and scraped or washed
into a series of gullies that feed an ore pass system. The trackless ore is drilled, blasted and loaded
with scoop trams to internal ore passes. The ore is then transported by rail to a shaft loading system
and hoisted to surface. During June 2002 the new gold extraction plant was commissioned. It includes
a two stage milling circuit (SAG Mill and Ball Mill), a gravity gold recovery circuit inclusive of leach
reactors and a pump cell CIP plant. It has a planned capacity of 7,200 tonnes per day. It will be capa-
ble of producing at low cost and high extraction efficiency after an increased steady state mill feed is
obtained. ESKOM is the supplier of electrical energy in the country; electricity at 132 kilovolts (kV)
voltage level is supplied to South Deep from the national grid. The annual power requirement of the
mine is approximately 610 GWh. The underground feed is presently 6.6kV with a plan for 11kV feed
down the new South Deep Main Shaft complex.
Since the module “blasting” does not include the CO2 emissions from explosives, they are calculated
with the factor of the mines of Newmont (2005b) (0.16 kg CO2 per kg explosive). Air emissions are
approximated with Australian data, based on produced gold. Water emissions are estimated on the
previous environmental report of Placer Dome Inc. (2002b) and assumed to be emitted to rivers.
Transport of materials is calculated on the basis of standard distances. Data about amounts of tailing
disposal is approximated with data from Papua New Guinea.
The following table (Tab. 5.52) sums up the total requirements and emissions of the South Deep Mine
in 2005.

Tab. 5.52: Requirements and emissions of the South Deep Mine in 2002 (Placer Dome Inc., 2002b).
Energy Remark
Electricity for mining kWh 1.68E+08 1)
South African electricity mix is approximated with the
CENTREL mix, because of ist high share of fossil energy
sources
Electricity for refining kWh 2.34E+04 assumed with Wohlwill-electrolysis
Petrol l 2.38E+03 converted in transport, van <3.5t (0.38l/tkm)
Materials .
Lime kg 5.01E+06 In LCI as Limestone, milled
Caustic kg 5.83E+05 in LCI as sodium hydroxide
Carbon kg 5.50E+04 from environmental report, in LCI as charcoal, at plant
kg 4.15E+04 Scaled from Steward (1999) by processed gold, accounted for
Zinc as zinc, primary, at regional storage
kg 6.60E+04 organic compounds as flocculent / flotation / magnafloc 351, in
Organic compounds LCI as fatty alcohol
Hydrochloric Acid kg 1.11E+03 from environmental report
kg 1.07E+06 estimated with 0.5kg per ton of ore, approximated with Tovex
Explosives
Water resources
3 2.19E+06 in LCI: Water, well, in ground
Groundwater Use m
Mine dewatering Use kg 0.00E+00 recycled Water, not accounted for in inventory
Emissions to Water South Deep
based on older environmental report, scaled by milled ore
Arsenic kg 7.94E+01 (Placer-Dome, 2002)
Cadmium kg 8.90E+00 (Placer-Dome, 2002)
Copper kg 4.40E+02 (Placer-Dome, 2002)
Cyanide kg 4.40E+03 (Placer-Dome, 2002)
Lead kg 5.29E+01 (Placer-Dome, 2002)
Mercury kg 8.90E+00 (Placer-Dome, 2002)
Nickel kg 1.01E+03 (Placer-Dome, 2002)
Zinc kg 6.61E+02 (Placer-Dome, 2002)
Emissions to Air
Antimony kg 2.31E+00 based on Australian value, scaled by milled ore
Arsenic kg 1.06E+02 based on Australian value, scaled by milled ore
Beryllium kg 6.22E+00 based on Australian value, scaled by milled ore
Cadmium kg 3.30E-01 based on Australian value, scaled by milled ore
Carbon Dioxide (from explosives) kg 1.71E+05 Calculation, based on data from Newmont (2005)
Chromium kg 5.39E+01 based on Australian value, scaled by milled ore
Cobalt kg 9.33E+00 based on Australian value, scaled by milled ore
Lead kg 2.62E+00 based on Australian value, scaled by milled ore
Manganese kg 2.92E+02 based on Australian value, scaled by milled ore
Mercury kg 1.50E+02 based on Australian value, scaled by milled ore
Selenium kg 1.78E+02 based on Australian value, scaled by milled ore
1
South African electricity mix in 2007 (Nationmaster, 2004): fossil fuel 93.5 %, hydro 1.1%, nuclear
5.5%.

Tab. 5.53: Unit process raw data of gold mining and refining in South Africa
Infrastructur
StandardDe
Uncertainty
viation95%
eProcess
Location
Unit
Location ZA
Unit kg
product gold, at refinery ZA 0 kg 1.00E+00
technosphere CENTREL 0 kWh 1.17E+04 1 1.29 (3,4,1,5,3,3); Environmental report
CENTREL, at grid
transport, van <3.5t RER 0 tkm 4.36E-01 1 2.03 (3,4,1,2,1,3); Environmental report
RER 0 kg 4.06E+01 1 1.17 (3,3,3,3,1,3); Environmental report
at plant
hydrochloric acid, 30% in H2O, at plant RER 0 kg 7.72E-02 1 1.14 (3,3,1,3,1,3); Environmental report
data for 1999
charcoal, at plant GLO 0 kg 3.83E+00 1 1.17 (3,3,3,2,1,3); Environmental report
fatty alcohol, petrochemical, at plant RER 0 kg 4.60E+00 1 1.14 (3,3,1,3,1,3); Environmental report
transport, lorry >16t, fleet average RER 0 tkm 4.36E-01 1 2.24 (5,3,1,3,1,3); Estimations
infrastructure facilities precious metal refinery SE 1 unit 1.65E-07 1 3.06 (1,3,1,5,3,3); Annual report
mine, gold ZA 1 unit 1.39E-06 1 3.05 (1,3,1,1,3,3); Annual report
emission air, low (4,3,1,5,1,3); Estimation, based on
Arsenic - - kg 7.40E-03 1 5.07
population density australian Data
(4,3,1,5,1,3); Estimation, based on
Carbon dioxide, fossil - - kg 1.19E+01 1 1.25
australian Data
Lead - - kg 1.83E-04 1 5.07
australian Data
Manganese - - kg 2.03E-02 1 5.04
Australian data
Cadmium - - kg 2.30E-05 1 5.04
Australian data
Beryllium - - kg 4.34E-04 1 5.04
Australian data
Chromium - - kg 3.76E-03 1 5.04
Australian data
Cobalt - - kg 6.51E-04 1 5.04
Australian data
Selenium - - kg 1.20E-02 1 5.04
Australian data
Mercury - - kg 1.05E-02 1 5.07
australian Data
emission water, river Arsenic, ion - - kg 5.53E-03 1 5.07 previous environmental report (Placer-
Dome, 2002)
Cadmium, ion - - kg 6.20E-04 1 3.07 previous environmental report (Placer-
Dome, 2002)
Copper, ion - - kg 3.07E-02 1 3.07 previous environmental report (Placer-
Dome, 2002)
Cyanide - - kg 3.07E-01 1 3.07 previous environmental report (Placer-
Dome, 2002)
Lead - - kg 3.69E-03 1 5.07 previous environmental report (Placer-
Dome, 2002)
Mercury - - kg 6.20E-04 1 5.07 previous environmental report (Placer-
Dome, 2002)
Nickel, ion - - kg 7.01E-02 1 5.07 previous environmental report (Placer-
Dome, 2002)
(Estimated with data from other
mine,3,2,1,3,3); Estimation, based on
Zinc, ion - - kg 4.61E-02 1 5.06
previous environmental report (Placer-
Dome, 2002)

Tab. 5.54 Ecospold Meta information of gold production in South Africa

Location ZA ZA
Unit unit kg
Type 1 1
Version 2.0 2.0
energyValues 0 0
LanguageCode en en
Person 44 44
QualityNetwork 1 1
IncludedProcesses
This data set describes the mining of gold
concentrate in South Africa. It includes energy and
water consumption, land use, transports, waste
disposal as well as emissions to air and water.
Amount 1 1
Synonyms
GeneralComment
This data set includes land use of the considered
mines ( underground) incl. revegetated areas as
This data set describes the beneficiation,
well as estimates of buildings and machinery of the
concentration and refinement of gold.
mine based on the data set non-ferrous metal,
mine, surface/underground.
Formula
StartDate 2002 2002
EndDate 2006 2006
Text Material requirements for gold concentration as
Land use of South-African mine, other inputs
well as emissions to air are estimated with
based on rough assumptions.
Australian data.
Text
Assumptions for buildings and machinery based on
The mining of ore from underground mines is
considered.
Percent
assumptions. The figures are extrapolated on the See Geography, electricity mix is extrapolated
Person 44 44
DataPublishedIn 2 2
Copyright 1 1
CompanyCode
CountryCode
PageNumbers
Validator 51 51
5.1.9 Gold production, Australia (AU)

Data stem from the environmental report of Newmont (2005b) for the Kalgoorlie, Jundee, Tanami and
Pajingo Mines. The mines produce only gold. Tanami and Pajingo are underground mines, Kalgoorlie
open pit, and Jundee open pit and underground. The average concentration (head grade, expressed in
kg metal per kg ore) of metals in ore for Australia is calculated with the quantity of ore milled, the cor-
responding production rate and the recovery rate.

Tab 5.55 shows the production volume in 2005 of the considered Australian mines. The recovery rate
(proportion of valuable material physically recovered in the processing of ore) is taken into account in
order to define the resource extraction. The average concentration (head grade, expressed in kg metal
per kg ore) of metals in ore for Australia is calculated with the quantity of ore milled, the correspond-
ing production rate and the recovery rate.
Tab 5.55: Production volume 2005 of the Kalgoorlie, Jundee, Tanami and Pajingo Mines (Newmont, 2005b).
Australia Kalgoorlie Jundee Tanami Pajingo Australia Remark

Production Gold kg 2.55E+04 1.06E+04 1.54E+04 5.97E+03 5.74E+04
Ore milled kg 9.84E+10 1.40E+10 4.77E+09 1.28E+09 1.18E+11
Head grade
Gold kg/kg 2.81E-07 8.25E-07 3.50E-06 5.06E-06 5.27E-07
ore
Recovery rate n.a. n.a. n.a. n.a. 92% average value for Australia, based on data
of Placer-Dome, (2002)
formula:

Tab. 5.56: Land use of the Kalgoorlie, Jundee, Tanami and Pajingo Mines (Newmont, 2005b)
Land use Kalgoorlie Jundee Tanami Pajingo Australia Remark

Area disturbed m
2 2.01E+07 1.97E+07 1.67E+07 3.52E+06 6.00E+07 Transformation, from unknown
during life of mine and Transformation, to mineral
extraction site,
Disturbed area in m
2 1.09E+07 1.01E+07 1.05E+07 1.76E+06 3.33E+07 Occupation, mineral extraction
active use site, during 50 years

2 4.01E+05 5.00E+06 2.67E+05 1.42E+05 5.81E+06 added in "Transformation, to
revegetation unknown"
Area revegetated m
2 8.77E+06 4.54E+06 5.89E+06 1.62E+06 2.08E+07 Transformation, to
heterogeneous, agricultural

Tab. 5.57: Unit process raw data of the gold mine in Australia
Infrastructur
StandardDe
Uncertainty
viation95%
eProcess
Location
Type
Unit
Location AU
Unit unit
product mine, gold AU 1 unit 1.00E+0
Mining and refining

Although the production of gold and silver was assessed in an older detailed study (Stewart, 1999), the
production of silver is neglected due to the low production (see Tab. 5.58) and the absence of actual
data. Nevertheless, the data from the mentioned study has been used as a proxy value for material re-
quirements, where actual data is lacking. The allocation for gold and silver (see Tab. 5.58) is done by
economical values with prices of Tab. 2.5.
Tab. 5.58: Production of gold / silver and material requirement in gold / silver mining in the year 1999 (Stewart, 1999)
Production Australia Allocation per kg gold per kg silver

Gold 3.00E+05 99.81%
Silver 3.40E+04 0.19%
Material Use
Lime 1.90E+7 6.32E+01 1.04E+00
Sodium hydroxide 3.50E+6 1.16E+01 1.92E-01
Hydrochloric acid 4.10E+6 1.36E+01 2.25E-01
Zinc 8.70E+5 2.89E+00 4.77E-02
Carbon 5.10E+6 1.70E+01 2.79E-01
Sulphuric acid 2.00E+5 6.65E-01 1.10E-02
The following table (Tab. 5.59) sums up the total requirements and emissions of the Kalgoorlie, Jun-
dee, Tanami and Pajingo Mines in 2005. Information stems from the environmental reports. Carbon
and zinc consumption is calculated with data from Stewart (1999).

Tab. 5.59: Energy and material requirements of the Kalgoorlie, Jundee, Tanami and Pajingo Mines (Newmont, 2005b)
Australia
Energy Kalgoorlie Jundee Tanami Pajingo Australia Remark
Electricity for kWh 2.00E+08 - 9.76E+07 5.40E+07 3.51E+08 1)
Australian electricity mix is
mining approximated with the CENTREL mix,
because of its high share of fossil
energy sources
Electricity for kWh n.a. n.a. n.a. n.a. 9.36E+04 assumed with Wohlwill-electrolysis
refining
Diesel (generators) MJ - - 6.98E+07 - 6.98E+07 Diesel, burned in diesel-electric
generating set
Diesel (mobile MJ 2.09E+09 4.77E+08 1.43E+09 1.21E+08 4.12E+09 Diesel, burned in building machine
equipment)
Petrol l 2.70E+03 - 1.12E+06 4.52E+01 1.13E+06 converted in transport, van <3.5t
(0.38l/tkm)
Oil MJ 3.23E+07 1.66E+07 1.44E+07 8.08E+06 7.14E+07 heavy fuel oil, burned in industrial
furnace 1MW, non-modulating
LPG MJ - 4.66E+08 - - 4.66E+08 natural gas, burned in industrial
furnace
Materials . . . .
Cyanide kg 3.94E+06 5.18E+05 7.85E+05 1.41E+05 5.38E+06 In LCI as sodium cyanide
kg 6.53E+07 8.20E+06 3.71E+06 6.02E+05 7.78E+07 in LCI as limestone, milled, packed,
Lime at plant
Caustic kg 5.11E+06 4.30E+05 6.70E+05 3.62E+05 6.57E+06 in LCI as sodium hydroxide
kg 4.32E+05 1.80E+05 2.61E+05 1.01E+05 9.74E+05 Scaled from Steward (1999) by
processed gold, accounted for as
Carbon Charcoal, at plant
kg 7.37E+04 3.07E+04 4.44E+04 1.73E+04 1.66E+05 Scaled from Steward (1999) by
processed gold, accounted for as
Zinc zinc, primary, at regional storage
kg 8.96E+05 2.86E+05 8.17E+04 6.91E+04 1.33E+06 Acids (hydrochloric, sulphamic and
sulphuric) consumption is divided in
half sulphuric, half hydrochloric
Sulphuric Acid
Hydrochloric Acid kg 8.96E+05 2.86E+05 8.17E+04 6.91E+04 1.33E+06 see Sulphuric Acid
kg 1.81E+07 2.10E+06 8.48E+05 9.25E+05 2.20E+07 Partly ANFO explosives, in LCI:
Explosives "blasting"
Water resources
3 1.80E+06 - 2.54E+06 3.21E+05 4.66E+06 in LCI: Water, river
m
Surface Water Use
3 4.59E+06 1.66E+06 - 2.57E+04 6.27E+06 in LCI: Water, well, in ground
m
Groundwater Use
Mine dewatering m
3 3.70E+05 1.87E+05 - 1.50E+04 5.71E+05 recycled Water, not accounted for in
Use inventory
1
Australian electricity mix (Nationmaster, 2004): fossil fuel 90.8%, hydro 8.3%, other 0.9%)
Water emissions are taken from data of the Granny Smith Mine (Placer Dome Inc., 2003a) for arsenic,
and from data of the Henty Mine (Placer Dome Inc., 2003a) for the other water emissions. They are
assumed to be released into rivers. Data about amounts of tailings disposal is approximated with data
from Papua New Guinea. Since the module “blasting” does not include the CO2 emissions from explo-
sives, they are calculated with the factor of the mines of Newmont (2005b) (0.16 kg CO2 per kg explo-
sive).
impact. The infrastructure is approximated with the facility of precious metal refinery.

Tab. 5.60 Emissions of the Kalgoorlie, Jundee, Tanami and Pajingo Mines (Newmont, 2005b)
Emissions to Remark
Water Kalgoorlie Jundee Tanami Pajingo Australia
Emission based on data for Mine "Granny
Smith", Placer Dome Inc. (2002), scaled
Arsenic kg 2.75E+03 3.90E+02 1.33E+02 3.58E+01 3.30E+03 by milled ore
Emission based on data for Mine "Henty",
Placer Dome Inc. (2002), scaled by milled
Cadmium kg 1.24E+03 1.75E+02 5.99E+01 1.61E+01 1.49E+03 ore
Copper kg 2.24E+04 3.18E+03 1.09E+03 2.92E+02 2.70E+04 ore
Cyanide kg 3.66E+04 5.20E+03 1.78E+03 4.77E+02 4.41E+04 ore
Lead kg 7.56E+04 1.07E+04 3.66E+03 9.84E+02 9.09E+04 ore
Mercury kg 1.37E+03 1.95E+02 6.66E+01 1.79E+01 1.65E+03 ore
Nickel kg 1.24E+04 1.75E+03 5.99E+02 1.61E+02 1.49E+04 ore
Zinc kg 1.37E+05 1.95E+04 6.66E+03 1.79E+03 1.65E+05 ore
Emissions to Air
Data for Pajingo estimated with data of
Kalgoorlie, Environmental Report,
Antimony kg 8.16E+01 4.54E+01 0.00E+00 1.06E+00 1.28E+02 Newmont 2005
Arsenic kg 2.83E+03 2.68E+03 3.54E+02 1.81E+01 5.88E+03 Environmental Report, Newmont 2005
Beryllium kg 1.81E+01 0.00E+00 0.00E+00 3.27E+02 3.45E+02 Newmont 2005
Cadmium kg 9.07E+00 9.07E+00 0.00E+00 1.18E-01 1.83E+01 Newmont 2005
environmental report of Newmont (2005),

not accounted for in LCI as emissions are
already included in fossil fuels
Carbon Monoxide kg 6.80E+05 2.02E+05 5.16E+05 3.73E+04 -
Carbon Dioxide
Environmental Report, Newmont 2005
(from explosives) kg 2.90E+06 3.37E+05 1.36E+05 1.48E+05 3.52E+06
Chromium kg 2.30E+03 4.81E+02 1.72E+02 2.72E+01 2.98E+03 Environmental Report, Newmont 2005
Cobalt kg 4.63E+02 3.63E+01 9.07E+00 9.07E+00 5.17E+02 Environmental Report, Newmont 2005
Lead kg 1.09E+02 9.07E+00 1.81E+01 9.07E+00 1.45E+02 Environmental Report, Newmont 2005
Manganese kg 1.32E+04 1.24E+03 1.22E+03 5.26E+02 1.62E+04 Environmental Report, Newmont 2005
Mercury kg 8.31E+03 0.00E+00 9.07E+00 0.00E+00 8.32E+03 Environmental Report, Newmont 2005
Nitrogen Oxide
(NOx) kg 1.70E+06 4.64E+05 1.18E+06 4.50E+04 -

Particulates kg 3.80E+06 1.33E+06 5.77E+05 2.61E+05 -
Data for Pajingo / Jundee estimated with
data of Kalgoorlie, Environmental Report,
Selenium kg 8.20E+03 1.16E+03 2.72E+01 1.07E+02 9.50E+03 Newmont 2005

Sulfur Dioxide
(SO2) kg 1.40E+08 1.82E+04 2.58E+04 3.44E+03 -
The unit process raw data are summarized in Tab. 5.61.

Tab. 5.61: Unit process raw data of gold mining and refining in Australia.
UncertaintyType
StandardDeviati
Infrastructure
Location
Process
on95%
Unit
Location AU
Unit kg
product gold, at refinery AU 0 kg 1.00E+00
technosphere CENTREL 0 kWh 6.12E+03 1 1.29 (3,4,1,5,3,3); Environmental report
CENTREL, at grid
heavy fuel oil, burned in industrial furnace 1MW,
non-modulating
natural gas, burned in industrial furnace
>100kW
RER 0 kg 1.14E+02 1 1.17 (3,3,3,3,1,3); Environmental report
at plant
data for 1999
data for 1999
infrastructure mine, gold AU 1 unit 3.48E-07 1 3.05 (1,3,1,1,3,3); Annual report
facilities precious metal refinery SE 1 unit 1.65E-07 1 3.05 (1,3,1,1,3,3); Annual report
emission air, low
Manganese - - kg 2.82E-01 1 5.04 (3,3,1,5,1,3); Environmental report
population density
Cadmium - - kg 3.18E-04 1 5.04 (3,3,1,5,1,3); Environmental report
Beryllium - - kg 6.01E-03 1 5.04 (3,3,1,5,1,3); Environmental report
Chromium - - kg 5.20E-02 1 5.04 (3,3,1,5,1,3); Environmental report
Cobalt - - kg 9.01E-03 1 5.04 (3,3,1,5,1,3); Environmental report
Selenium - - kg 1.65E-01 1 5.04 (3,3,1,5,1,3); Environmental report
Arsenic - - kg 1.02E-01 1 5.02 (3,3,1,1,1,3); Environmental report
Lead - - kg 2.53E-03 1 5.02 (3,3,1,1,1,3); Environmental report
emission water, river Arsenic, ion - - kg 5.76E-02 1 5.09 (3,4,2,1,3,3); Environmental report
Cadmium, ion - - kg 2.59E-02 1 3.07 (3,4,2,1,3,3); Environmental report
Copper, ion - - kg 4.70E-01 1 3.07 (3,4,2,1,3,3); Environmental report
Cyanide - - kg 7.68E-01 1 3.07 (3,4,2,1,3,3); Environmental report
Nickel, ion - - kg 2.59E-01 1 5.09 (3,4,2,1,3,3); Environmental report

Tab. 5.62 Ecospold Meta information of gold production in Australia

Location AU AU
Unit unit kg
Type 1 1
Version 2.0 2.0
energyValues 0 0
LanguageCode en en
Person 44 44
QualityNetwork 1 1
IncludedProcesses This data set describes the mining of gold
concentrate in Australia. It includes energy and water
Land use, buildings and machinery of mining
consumption, land use, transports, waste disposal as
well as emissions to air and water.
Amount 1 1
Synonyms
GeneralComment
This data set includes land use of the considered
mines (open pit and underground) incl. revegetated
This data set describes the beneficiation,
areas as well as estimates of buildings and
concentration and refinement of gold.
machinery of the mine based on the data set non-
ferrous metal, mine, surface/underground.
Formula
StartDate 2001 2001
EndDate 2006 2006
Text Land use of Australian mine, other inputs based on
the electricity mix has been estimated
rough assumptions.
Text Assumptions for buildings and machinery based on
The mining of ore from open pit and underground
mines is considered.
Percent
Extrapolations
Buildings and machinery are based on global Material requirements for gold concentration are
assumptions. The figures are extrapolated on the estimated with older data from an Australian thesis,
basis of the ore production. several values are scaled from one mine
Person 44 44
DataPublishedIn 2 2
Copyright 1 1
CompanyCode
CountryCode
PageNumbers
Validator 51 51
5.1.10 Gold production, Tanzania (TZ)

Data stem from the environmental report of Placer Dome Inc. (2005) and Barrick (2006b) for the
North Mara Mine. The mine produces only gold and is an open pit mine. Tab 5.63 shows the produc-
tion volume in 2005 of the North Mara Mine. The recovery rate (proportion of valuable material
physically recovered in the processing of ore) is taken into account in order to define the resource ex-
traction. The average concentration (head grade, expressed in kg metal per kg ore) of metals in ore for
Tanzania is calculated with the quantity of ore milled, the corresponding production rate and the re-
covery rate.
.

Tab 5.63: Production volume in 2004 of the North Mara Mine (Placer Dome Inc., 2005).
Production North Mara

Head grade
Recovery rate 92% based on reports of Placer-Dome Inc. (2005)
Tab, 5.64: Land use of North Mara Mine in 2004 (Placer Dome Inc., 2005).
Land use North Mara Remark

Area disturbed during life m 2 3.72E+06 Transformation, from unknown and
of mine Transformation, to mineral extraction site,
Disturbed area in active m 2 2.92E+06 Occupation, mineral extraction site, during 50
use years
Area revegetated m
2 3.00E+04 added in "Transformation, to heterogeneous,
agricultural"
formula:
Disturbed area in active use 2.92 *106

Area disturbed during life of mine 3.72 *106

Tab. 5.65: Unit process raw data of the gold mine in Tanzania.
Infrastructur
StandardDe
Uncertainty
viation95%
eProcess
Location
Type
Unit
Location TZ
Unit unit
product mine, gold TZ 1 unit 1.00E+0
report, Placer Dome 2003
Mining and refining

Ore is fed via a single jaw crusher into a secondary crusher and then onto a crushed ore stockpile. The
crushed ore is reclaimed from the stockpile and conveyed into an open circuit mill before passing to a
closed circuit ball mill regrind. A gravity separation circuit is used to scalp free gold from regrind cir-
cuit. The ore is then processed via cyanidation and carbon in leach, followed by electrowinning and
gold refining to doré on site.
Water for processing is pumped from the Mara River, which is located three kilometres from the proc-
essing plant site. The water is pumped into a raw water dam which can store up to three months water
demand. A plant at the mine treats water for own use. Placer Dome Tanzania has a license to extract
4,200 cubic metres of water daily from the river. The following table (Tab. 5.66) sums up the total re-
quirements and emissions of the North Mara Mine in 2005.
Material use (except for cyanide and explosives) is approximated with figures from Australian data for
gold mining (Stewart, 1999), air emissions with Australian Data. Since the module “blasting” does not
include the CO2 emissions from explosives, they are calculated with the factor of the mines of
Newmont (2005b) (0.16 kg CO2 per kg explosive). Water emissions are assumed to be emitted to riv-
ers and are estimated on the base of data for South Africa. Estimation of transport is based on standard
distances for Europe. Data about amounts of tailings disposal are approximated with data of mines in
Papua New Guinea.

Tab. 5.66: Requirements and emissions of the North Mara Mine in 2005 (Placer Dome Inc., 2005).
Energy North Mara Remark

Petrol l 2.90E+05 calculate with the specific fuel consumption of delivery van, converted
in transport, van <3.5t (0.38l/tkm)
Electricity for refining kWh 1.89E+04 assumed with Wohlwill-electrolysis, electricity demand for refining is
1)
approximated with CENTREL-mix
Materials
Lime kg 7.31E+05 Scaled from Steward (1999) by processed gold
kg 1.35E+05 Scaled from Steward (1999) by processed gold, accounted for sodium
Caustic hydroxide
kg 1.96E+05 Scaled from Steward (1999) by processed gold, accounted for as
Carbon charcoal
kg 3.35E+04 Scaled from Steward (1999) by processed gold, accounted for as zinc,
Zinc primary, at regional storage
Sulphuric Acid kg 7.70E+03 Scaled from Steward (1999) by processed gold
Hydrochloric Acid kg 1.58E+05 Scaled from Steward (1999) by processed gold
Explosives kg 1.55E+06 Partly ANFO explosives, in LCI: "blasting"
Water resources
3 in LCI: Water, unspecified natural origin
Water from river m 9.25E+05
2)
Emissions to Water
Arsenic kg 4.43E+01 estimated with ZA-data, scaled by milled ore
Cadmium kg 4.97E+00 estimated with ZA-data, scaled by milled ore
Copper kg 2.46E+02 estimated with ZA-data, scaled by milled ore
Cyanide kg 2.46E+03 estimated with ZA-data, scaled by milled ore
Lead kg 2.96E+01 estimated with ZA-data, scaled by milled ore
Mercury kg 4.97E+00 estimated with ZA-data, scaled by milled ore
Nickel kg 5.62E+02 estimated with ZA-data, scaled by milled ore
Zinc kg 3.70E+02 estimated with ZA-data, scaled by milled ore
3)
Emissions to Air
Carbon Dioxide (from Calculation, based on data from Newmont (2005)
explosives) kg 2.48E+05
Antimony kg 3.11E+00 estimated with AU-data, scaled by milled ore
Arsenic kg 1.43E+02 estimated with AU-data, scaled by milled ore
Beryllium kg 8.36E+00 estimated with AU-data, scaled by milled ore
Cadmium kg 4.43E-01 estimated with AU-data, scaled by milled ore
Chromium kg 7.24E+01 estimated with AU-data, scaled by milled ore
Cobalt kg 1.25E+01 estimated with AU-data, scaled by milled ore
Lead kg 3.52E+00 estimated with AU-data, scaled by milled ore
Manganese kg 3.92E+02 estimated with AU-data, scaled by milled ore
Mercury kg 2.02E+02 estimated with AU-data, scaled by milled ore
Selenium kg 2.30E+02 estimated with AU-data, scaled by milled ore
Disposal kg 2.04E+09 estimated with data of PG, in LCI as: Disposal, sulphidic tailings
1 Neither a Tanzanian electricity mix nor the exact place of refinement are available, therefore we
assumed the use of CENTREL-electricity mix for its high share of fossil electricity generation.
2 Estimation, based on emissions in South Africa, scaled by processed gold (Placer-Dome, 2002).
3 Estimation, based on emissions in Australia, scaled by processed gold.

Tab. 5.67: Unit process raw data of gold mining and refining in Tanzania
StandardDeviation95
UncertaintyType
Location
Unit
%
Location TZ
Unit kg
product gold, at refinery TZ 0 kg 1.00E+00
diesel, burned in diesel-electric
technosphere GLO 0 MJ 8.64E+04 1 1.16 (3,4,1,2,1,3); Environmental report
generating set
CENTREL 0 kWh 1.63E+00 1 1.16 (3,4,1,2,1,3); Environmental report
CENTREL, at grid
(3,3,3,5,1,3); Calculation, based on thesis on
limestone, milled, packed, at plant CH 0 kg 6.32E+01 1 1.20 Australian precious metal production, data for
1999
sodium hydroxide, 50% in H2O,
RER 0 kg 1.16E+01 1 1.17 Australian precious metal production, data for
1999
charcoal, at plant GLO 0 kg 1.70E+01 1 1.17 Australian precious metal production, data for
1999
zinc, primary, at regional storage RER 0 kg 2.89E+00 1 1.17 Australian precious metal production, data for
1999
sulphuric acid, liquid, at plant RER 0 kg 6.65E-01 1 1.17 Australian precious metal production, data for
1999
hydrochloric acid, 30% in H2O, at plant RER 0 kg 1.36E+01 1 1.17 Australian precious metal production, data for
1999
(4,5,na,na,na,na); Estimation, based on
standard distances
infrastructure facilities precious metal refinery SE 1 unit 1.65E-07 1 3.03 (3,4,1,2,1,3); Environmental report
mine, gold TZ 1 unit 1.73E-06 1 3.03 (3,4,1,2,1,3); Environmental report
emission air, low
Antimony - - kg 2.68E-04 1 5.05 (3,3,3,5,1,3); Estimated with data from AU
population density
Arsenic - - kg 1.23E-02 1 5.05 (3,3,3,5,1,3); Estimated with data from AU
Beryllium - - kg 7.22E-04 1 5.05 (3,3,3,5,1,3); Estimated with data from AU
Cadmium - - kg 3.83E-05 1 5.05 (3,3,3,5,1,3); Estimated with data from AU
Carbon dioxide, fossil - - kg 2.14E+01 1 1.20 (3,3,3,5,1,3); Estimated with data from AU
Chromium - - kg 1.08E-03 1 5.05 (3,3,3,5,1,3); Estimated with data from AU
Cobalt - - kg 3.04E-04 1 5.05 (3,3,3,5,1,3); Estimated with data from AU
Lead - - kg 3.39E-02 1 5.05 (3,3,3,5,1,3); Estimated with data from AU
Mercury - - kg 1.99E-02 1 5.05 (3,3,3,5,1,3); Estimated with data from AU
Manganese - - kg 1.74E-02 1 5.05 (3,3,3,5,1,3); Estimated with data from AU
Selenium - - kg 0.00E+00 1 5.05 (3,3,3,5,1,3); Estimated with data from AU
emission water, river Arsenic, ion - - kg 3.83E-03 1 5.10 (3,3,3,5,3,3); Estimated with data from ZA
Cadmium, ion - - kg 4.30E-04 1 3.08 (3,3,2,5,3,3); Estimated with data from ZA
Copper, ion - - kg 2.13E-02 1 3.08 (3,3,2,5,3,3); Estimated with data from ZA
Cyanide - - kg 2.13E-01 1 3.08 (3,3,2,5,3,3); Estimated with data from ZA
Lead - - kg 2.56E-03 1 5.09 (3,3,2,5,3,3); Estimated with data from ZA
Mercury - - kg 4.30E-04 1 5.09 (3,3,2,5,3,3); Estimated with data from ZA
Nickel, ion - - kg 4.85E-02 1 5.09 (3,3,2,5,3,3); Estimated with data from ZA
Zinc, ion - - kg 3.19E-02 1 5.09 (3,3,2,5,3,3); Estimated with data from ZA
resource, in water Water, river - - m3 7.99E+01 1 1.14 (3,3,1,1,1,3); Environmental report

Tab. 5.68 Ecospold Meta information of gold production in Tanzania

Location TZ TZ
Unit unit kg
Type 1 1
Version 2.0 2.0
energyValues 0 0
LanguageCode en en
Person 44 44
QualityNetwork 1 1
IncludedProcesses This data set describes the mining of gold concentrate
Land use, buildings and machinery of a gold and silver in Tanzania. It includes energy and water consumption,
mine land use, transports, waste disposal as well as
emissions to air and water.
Amount 1 1
Synonyms
GeneralComment
This data set includes land use of the considered mines
(open pit and underground) incl. revegetated areas as
This data set describes the beneficiation, concentration
well as estimates of buildings and machinery of the
and refinement of gold.
mine based on the data set non-ferrous metal, mine,
Formula
StartDate 2003 2003
EndDate 2006 2006
Text Material requirements for gold concentration are
Land use of Tanzanian mine, other inputs based on estimated with Australian data. Emissions to water have
rough assumptions. been estimated with South-African, emissions to air with
Australian data.
Text Assumptions for buildings and machinery based on the
The mining of ore from open pit and underground mines
data set "non-ferrous metal, mine, surface/underground
is considered.
with 50 years life time.
Percent
See Geography, electricity mix has been estimated with
assumptions. The figures are extrapolated on the basis
foreign mix
of the ore production.
Person 44 44
DataPublishedIn 2 2
Copyright 1 1
CompanyCode
CountryCode
PageNumbers
Validator 51 51
5.1.11 Data quality for primary gold / silver production

The datasets regarding the primary gold and silver production based mostly on environmental reports
of mine corporations. Concerning energy, cyanide and explosive consumption, the data quality is
good. Further material requirements have been estimated and scaled of Australian data by the produc-
tion of gold.
Only a few companies show their emissions into air and water, thus in several of the country-specific
datasets, emissions have been estimated by scaled ore. Information about tailings referred to Papua
New-Guinea, where the only available data stems from. Therefore the data quality for emissions into
air / water, as well as disposal of tailings is rather low.
A comparison of the datasets for gold and silver production on a country level can only be done with
respect to the above mentioned limitations: A better performance of a country can be explained with
better technology or with other assumptions on the emission level.

5.1.12 Primary gold, production mix (GLO)

These data sets describe an approximated average origin of gold. Shares of the different countries are
specified in Tab. 2.2. No transports are considered, since these are included in the dataset “gold, at re-
gional storage”.
Tab. 5.69: Unit process raw data for „gold, primary, at refinery“, Source: Tab. 2.2
Infrastructur
StandardDe
Uncertainty
viation95%
eProcess
Location
gold,
Type
Unit
Name primary, at GeneralComment
refinery
Location GLO
Unit kg
product gold, primary, at refinery GLO 0 kg 1.00E+0
(1,1,2,1,1,3); average gold production 2000-2004 &
technosphere gold, at refinery ZA 0 kg 1.86E-1 1 1.08
approximation for remaining countries
gold, at refinery US 0 kg 1.20E-1 1 1.08
gold, at refinery AU 0 kg 2.39E-1 1 2.00
gold, at refinery CA 0 kg 5.81E-2 1 1.08
gold, from combined gold-silver (1,1,2,1,1,3); average gold production 2000-2004 &
PG 0 kg 1.88E-1 1 1.51
production, at refinery approximation for remaining countries
CL 0 kg 5.28E-2 1 2.00
PE 0 kg 9.69E-2 1 2.00
gold, at refinery TZ 0 kg 4.75E-2 1 1.08
gold, from combined metal (1,1,2,1,1,3); average gold production 2000-2004 &
SE 0 kg 1.17E-2 1 2.00
5.1.13 Production of primary silver from auriferous ores, production mix

Shares of the different countries are specified for silver in Tab. 2.4. No transports are considered, since
this takes place in the dataset “silver, at regional storage”.
Tab. 5.70: Unit process raw data for „silver, from combined gold-silver production, at refinery”, Source: Tab. 2.4.
StandardDeviation95
UncertaintyType
Location
silver, from combined

Unit
Name GeneralComment
%
gold-silver production,
at regional storage
Location RER
Unit kg
silver, from combined gold-silver
product RER 0 kg 1.00E+0
production, at regional storage
silver, from combined gold-silver (1,1,2,1,1,3); average silver production 2000-2004 &
technosphere PE 0 kg 3.13E-1 1 1.08
CL 0 kg 3.20E-1 1 1.08
PG 0 kg 1.48E-1 1 2.00
SE 0 kg 2.19E-1 1 1.08

Tab. 5.71 Ecospold Meta information of gold & and silver mixes in ecoinvent v2.0
Name silver, from combined gold-silver production, at

gold, primary, at refinery
regional storage
Location GLO RER
Unit kg kg
Type 1 1
Version 2.0 2.0
energyValues 0 0
LanguageCode en en
Person 44 44
QualityNetwork 1 1
IncludedProcesses none none
Amount 1 1
LocalName Silber, aus kombinierter Gold-Silber-Produktion,
Gold, aus Primärproduktion, ab Raffinerie
ab Regionallager
Synonyms
GeneralComment
This data set describes the composition of primary This data set describes the composition of silver
gold production from at a global level. production from auriferous ores at a global level.
Formula
CASNumber 007440-57-5 007440-22-4
StartDate 2000 2000
EndDate 2004 2004
OtherPeriodText Period of public statistics Period of public statistics
Text global data global data
Text Assumption on classification of the production Assumption on classification of the production
countries has been made, based on global countries has been made, based on global
production statistics. production statistics.
Percent
ProductionVolume 2533 t gold 2362 t silver from auriferous ores
SamplingProcedure Production statistics, based on average Production statistics, based on average
production rate 2000-2004 production rate 2000-2004
Extrapolations The energy use, materials, emissions and land The energy use, materials, emissions and land
use of the mining and beneficiation are use of the mining and beneficiation are
considered in the corresponding datasets at considered in the corresponding datasets at
country level. country level.
Person 44 44
DataPublishedIn 2 2
Copyright 1 1
CompanyCode
CountryCode
PageNumbers
Validator 51 51

5.2 Secondary gold production process and infrastructure

Gold can be recovered using different technologies: pyrometallurgy, hydrometallurgy (Kinoshita et
al., 2003), electrochemistry, and biotechnology (Aylmore & Muir, 2001; Syed, 2006). Prior to these
gold recovery processes a mechanical treatment of the gold containing electronic scrap can be per-
formed. A comprehensive review of the various techniques and process combinations is compiled in
Syed (2006).
For the present dataset information on the pyrometallurgical technique applied by Boliden (Rönnskär
plant, Sweden) was used to represent the inventory for secondary gold production. The plant is an
copper-lead smelter that treats complex ores and to a large extent electronic scraps and integrates the
recovery of, among others, gold, silver, lead, palladium and other useful metals into one facility. The
electronic scrap however passes only a part of the facility, which is indicated by the red circles in Fig.
5.3. The scrap enters the process at the Kaldo plant. The valuable metals distribute into the copper
matte and which is further processed to copper. There again the precious metals go into the anode
slime, which is specifically treated in the precious metal recovery plant.
Fig. 5.3 Concept of the Rönnskär (Boliden) process. The processing steps relevant to the dataset are those the elec-
tronic scrap passes through. They are encircled in red colour (adopted from Boliden, 2000).
Pretreatment: Normally no pre-treatment (mechanical) is applied to the electronic scrap. Instead the
plastics fraction of the gold containing scrap is used as energy carrier in the pyrometallurgical ap-
proach (Boliden, 2006c).
Converting: The actual recovery process begins with feeding the electronic scrap into the KALDO
smelter unit (Boliden, 2000; Isaakson & Lehner, 2000; Ludvigsson & Larsson, 2003). This in 1977
patented Boliden-own technology Peterson & Lundquist (1977) comprises in principle a brick-lined
vessel made of stainless steel that has an effective volume of 2 m3 (Ludvigsson & Larsson, 2003). This
vessel is tiltable along its horizontal axis and can also be rotated during operation. For smelting, the
vessel is fed with oxygen (via a lance), coke breeze and heat (oil). Key products of this treatment step
are selenium-rich off-gas (which is then treated in the selenium plant) and black copper (74-80% Cu,
6-8% Sn, 5-6% Pb, 1-3% Zn, 1-3% Ni, 5-8% Fe). The latter is subsequently further processed in the
Aisle converter unit.
In the Aisle converter unit the black copper is converted into blister copper. Secondary lead and slag
are separated from the process stream. Technologically, the Aisle converter is a brick-lined (magne-

site-chromium), horizontally aligned converter unit fed with hydrocarbon fuel, industrial oxygen and
calcium oxide (Fig. 5.4, Davenport et al., 2002; Outokumpu, 2006).
Fig. 5.4 Detail of a converter unit (adopted from Davenport et al., 2002).
Refining. The processing of the electronic scrap in the Aisle converter is followed by a further re-
finement in a) the anode casting plant and b) the electrolytic refinery. At BOLIDEN’s Rönnskär facil-
ity a new Twin M-16 anode casting facility (Fig. 5.5) is deployed to cast the hot blister copper into
anodes. The facility consists of an inlet through which the hot blister copper is transported to the
loader. Increasing the processing efficiency, the loader alternately feeds the blister copper into two
wheel caster (Outokumpu, 2004).
Fig. 5.5 Sketch of a twin-wheel anode caster (adopted from Outokumpu, 2004) and the key infrastructure materials.
Once cast, the anodes are then finished and stored in the anode container. The anode sheets typically
have a size of 1.00 m * 1.40 m (approximated based upon Outokumpu, 2004) and consist of 99.5%
copper including, however, impurities of sulphur, oxygen and traces of precious metals (Davenport et
al., 2002). In the next step the copper anode is processed in the electrolytic refinery plant. This facility
is a large hall with a large number of cells.

Fig. 5.6 Sketch of an electrolytic cell (adopted from Corrosion Technology International, 2006) and the key infra-
structure materials
In 1999, the Rönnskär plant was equipped with 644 electrorefining cells and with 2 circulation tanks
(Boliden, 2006c; Corrosion Technology International, 2006). The cells are made of stainless steel and
reinforcement of glass fibre. The volume of the cells can vary considerably depending upon the proc-
ess it is intended to be applied. The cells are continuously circulated by water spiked with copper, sul-
phuric acid and chloride (Davenport et al., 2002) Typically, the circulation time is about 0.02 m3/min
and the power consumption ranges from less than 300 kWh/t material to nearly 400 kWh/t material
(Davenport et al., 2002). The final products of the electrolytic step are copper and nickel. As a resid-
ual, precious metal containing anode slime is generated, which is finally treated in the precious metals
plant in order to recover secondary gold, silver, palladium, and platinum.
Anode slime treatment. In a first step the precious metals containing anode slime is decopperised.
The remaining residues are dried and processed by pyrometallurgical means. There also exists a hy-
drometallurgical route (Pesl & Anzinger, 2002). Whereas the pyrometallurgical route is apt to process
large quantities of well defined anode slime from primary production, the hydrometallurgical route is
more flexible to changes in feed composition as it occurs in the processing of slimes from secondary
copper smelters. However, in Boliden the pyrometallurgical route is applied (Ludvigsson & Larsson,
2003). After de-copperising and precipitation of copper telluride the residue is dried and pyrolytically
refined to silver rich dorée metal, which is cast into anodes for a subsequent electrorefining of silver
(Moebius electrolysis) and Gold (Wohlwill electrolysis). The remaining slime consists mainly of PGM
(platinum group metal) that is recovered.

Fig. 5.7 Diagram of pyro- and hydrometallurgical process routes for anode slime treatment (Pesl & Anzinger, 2002).
Fig. 5.8 Sketch of the anode slimes treatment and precious metal recovery approach applied at Rönnskär
(Ludvigsson & Larsson, 2003).
5.2.1 Waste
The recovery of precious metals, such as gold, is associated with the generation of a number of by-
products off-gases, sludge, slag, waste heat and dust (Umweltbundesamt, 2001). Most of the by-
products are captured by appropriate filters (Umweltbundesamt, 2001). From an environmental per-
spective, most important, is the treatment of the residuals of the cyanide-containing sludge generated
while concentrating gold from doré-grade gold. This sludge is treated applying sulphur-dioxide and air
before it is dumped into tailing ponds (Boliden, 2002b). Concerning the other by-products and their
treatment, no utilisable information could be compiled.
5.2.2 Emissions
Despite of the application of modern filter technologies, the precious metals recovery process is also
associated with the generation of emissions to the environment (HELCOM, 2002). In particular the
following emissions to air are considered problematic: dust, copper, lead, zinc, cadmium, arsenic,
mercury, sulphur dioxide, nitrogen oxide, fluoride, chloride and PCDD/F. Emissions to water are in-
clude copper, lead, zinc, cadmium, arsenic, mercury (Mark & Lehner, 2002 and Umweltbundesamt,
2001)
5.2.2.1 Gold / Silver, secondary, from electronic scrap in ecoinvent

No explicit LCA study covering the extraction of gold as secondary material from electronic scrap has
been performed so far. As mentioned in the beginning of this chapter the recovery process of secon-
dary gold is represented by the recovery of electronic scrap by integrated smelters as conducted by
Boliden in the Rönneskär site. This inventory is solely based upon available industry information and
literature data. In cases where no explicit gold recovery information was available, information on the
extraction of copper was used as proxy.
The precious metals in the scrap traverse a lead smelting step (“Kaldo plant”) and a subsequent con-
version of the precious metals containing copper matte. The resulting blister copper is cast into anodes
that are further refined, whereas the precious metals are transferred into the anode slime. The precious

metals are extracted from the anode slime in the final step. The production of secondary gold and sil-
ver therefore consists of a series of three consecutive multi-output datasets:
Secondary Copper Conversion
- lead, secondary, from electronic and electric scrap recycling, at plant
- metal values from electric waste, in blister-copper, at converter
Secondary Copper Refining
- PM from electric waste, in anode slime, at refinery
- copper, secondary, from electronic and electric scrap recycling, at refinery
- nickel, secondary, from electronic and electric scrap recycling, at refinery
Secondary Copper Precious Metal Refining
- silver, secondary, at precious metal refinery
- gold, secondary, at precious metal refinery
- palladium, secondary, at precious metal refinery
The meta information for these processes is compiled in Tab. 5.74, the reference flows for the multi-
input processes are displayed in Fig. 5.10, Fig. 5.11 and Fig. 5.12.
Based on process descriptions and corporate reports a mass flow balance has herewith been estab-
lished that serves as skeleton for the three models of the three multi-output processes (Fig. 5.9). The
mixed feed of the copper refinery considered does not exclusively consist of electronic scrap. The in-
termediate reference flows “metal values from electric waste, in blister-copper, at converter” and “PM
from electric waste, in anode slime, at refinery” therefore relate only to that part of the metals con-
tained in the blister copper and the anode slime that is attributable to the electronic scrap in the input
feed (Tab. 5.72). The reference flows do not reflect the intermediary, but the intended metals con-
tained in the intermediary. This is reflected by the naming of these flows (in analogy to the naming of
elementary flows), denoting the content and not the container.

Electronic scrap, to secondary copper plant

5882 kg
secondary copper conversion
30.7% 1804 kg Epoxies (co-combusted)
31.0% 1826 kg Slag, silica, at converter
36.7% 1.6% 94 kg lead, secondary, from electronic and electric scrap

recycling, at plant
2159 kg metal values from electric waste, in blister-copper, at converter
secondary copper refining

24.66% 532 kg Others (Al, Ca, Fe, Sn)
9.1% 196 kg nickel, secondary, from electronic and electric scrap

recycling, at refinery
1.66% 65% 1394 kg copper, secondary, from electronic and electric

scrap recycling, at refinery
36 kg PM from electric waste, in anode slime, at refinery
precious metal refining, secondary copper

5.10% 1.82 kg palladium, secondary, at precious metal refinery
92.1% 33 kg silver, secondary, at precious metal refinery

2.8%
1 kg gold, secondary, at precious metal refinery
Fig. 5.9 Mass Balance of the tree consecutive multi-output processes for Gold / Silver, secondary from electronic
scrap. Mass is scaled to 1 kg gold output.
Tab. 5.72 Constituents of the intermediate reference flows in the system model (Fig. 5.9).
Scrap input into blister-copper To cathode copper PGM in anode sli-

me
Corre- Electronic scrap, to sec- metal values from electric copper, secondary, from PM from electric waste, in
sponding ondary copper plant waste, in blister-copper, electronic and electric anode slime, at refinery
flow at converter scrap recycling, at refin-
ery
Mass 5882 kg 2159 kg 1394kg 36 kg
Cu 23.67% 64.5% 99.9%
Pb 1.60%
Ni 3.34% 9.1%
Ag 0.560% 1.53% 92.1%
Au 0.049% 0.13% 0.1% 2.8%
Pd 0.030% 0.08% 5.1%
Epoxies 30.7%
Silica 31.0%
other 9.05% 25%
Allocation procedures
Based on the amount, the composition and the market prices of the materials involved, the burdens are
allocated to the coupled reference flows based on their value (Tab. 5.73). The values contained were
assessed according to the three years average from 2004 through 2006 based on USGS (2007).

Tab. 5.73 Allocation of the burdens to the coupled multi-output reference flows according to their values depending
on composition (Tab. 5.72) and the values (three years average from 2004 through 2006 based on USGS,
2007).
$ per kg % wt product % burden

lead, secondary 1.15 4.2% 0.1%
metal values from electric waste 36.1 95.8% 99.9%
PM from electric waste 1105 1.66% 81.5%

copper, secondary 4.47 65% 12.9%
nickel, secondary 13.8 9.1% 5.6%
silver, secondary 269 92.1% 22.4%

gold, secondary 15682 2.8% 39.7%
palladium, secondary 8214 5.10% 37.9%

Tab. 5.74 Meta information for the modelled production processes for secondary gold from electric scrap.
Name secondary copper secondary copper refi- precious metal refining,

conversion ning secondary copper
Location SE SE SE
Unit kg kg kg
Included Processes Inculdes the processing of e- Includes the refining of the Inculdes the refining of gold and
scrap in the Boliden process by copper anode to cathode cop- silver.
the Kaldo plant and the con- per.
verter aisle.
Amount 1 1 1
Local Name Sekundär Kupfer Verhüttung Sekundär Kupfer Raffinierung Sekundär Kupfer Hütte, Edelmet-
allgewinnung
Synonyms
General Comment to The multi-output-process "sec- The multi-output-process "sec- The multi-output-process "precious
reference function ondary copper conversion" de- ondary copper refining" deliv- metal refining, secondary copper"
livers the two co-products ers the three co-products "PM delivers the co-products "silver,
"metal values from electric from electric waste, in anode secondary, at precious metal refin-
waste, in blister-copper, at slime, at refinery", "copper, ery", "gold, secondary, at precious
converter" and "lead, secon- secondary, from electronic and metal refinery" and "palladium,
dary, from electronic and elec- electric scrap recycling, at re- secondary, at precious metal refin-
tric scrap recycling, at plant". finery", and "nickel, secondary, ery". The flows "silver, …" and
The product "metal values from from electronic and electric "gold, …" are part of the global
electric waste, …" is an inter- scrap recycling, at refinery". supply mix of gold and silver mod-
mediary in the production of The product "PM from electric elled in this part. The by-product
secondary gold from electric waste, …" is an intermediary in "palladium, …" receives its part of
scrap and is not to be used by the production of secondary the burdens, but it is not aimed to
the LCA practitioner. The flow gold from electric scrap and is be used by the data-user.
"lead, …" is a by-product of this not to be used by the LCA
process that receives part of practitioner. The flows "copper,
the burdens allocated. It must …" and "nickel, …" are by-
not be used to reflect lead. products of this process that
receive part of the burdens
allocated. They must not be
used.
Start Date 2002 2002 2002
End Date 2005 2005 2005
Data Valid For Entire 1 1 1
Period
Geography text Data describes a Swedish Data describes a Swedish Data describes a Swedish plant
plant that is singular in the re- plant that is singular in the re- that is singular in the region and
gion and ranges worldwide gion and ranges worldwide ranges worldwide among the big
among the big producer. It may among the big producer. It may producer. It may be taken as ap-
be taken as approximation for be taken as approximation for proximation for the global situation
the global situation the global situation
Technology text Conversion of Copper in a Production of cathode copper Anode slime treatment by pressure
Kaldo Converter and treatment by electrolytic refining. leaching and top blown rotary con-
in converter aisle. verter. Production of Silver by
Möbius Electrolysis, Gold by
Wohlwill electrolysis, Palladium to
further processing
Uncertainty Adjust- See Geography and Technology.
ments

Secondary Copper Conversion

The energy and auxiliary requirements have been calculated on a per year basis from the respective
patent and literature. 40’000 t annual electronic scrap treatment (Boliden, 2006c), 300 days actual an-
nual operation time and 50 year service time were assumed to calculate process inputs and emissions
per kg electronic scrap treated (Tab. 5.75). According to Boliden (2006c) and the assumed composi-
tion of the electronic scrap based on Motorola (2005), it can be calculated that 2.37% of the copper
produced stems from electronic scrap (640 t of 27’000 t Pb per year) and 6.58% of the copper produc-
tion, respectively (14’680 t of 223’000 t Cu per year). The Kaldo process uses the plastic in the scrap
as fuel and as a reducing agent at the same time as organic compounds are oxidised. The caloric value
in the electronic scrap is assumed to make up for the energy required to reduce the metals in the scrap.
Therefore no additional demand in energy and oxygen is accounted for melting. For rotation of the
furnace and ventilation in the converter electricity is required. No specific information was available.
The electricity demand is stated in Boliden (2006c) to be 603 GWh per year for the whole site. As-
suming an arbitrary share 5% that is attributable to the Kaldo plant, and disregarding the electricity for
ventilation, 1.12 kWh per kg lead produced in the Kaldo furnace is calculated.
Electronic scrap, to secondary copper plant
Off-gas, selenium rich, at plant

Kaldo smelter unit Kaldo furnace Sodium hydoxide, at plant
Building Epoxies (co-combusted) Filter, at plant
30.7% Sulphur dioxide, at plant
Black copper (74-80% Cu, 6-8% Sn, 5-6% Pb, 1-3% Zn, 1-3% Ni, 5-8% Fe)
Converter unit
Building Aisle converter Slag, silica, at converter
31.0%
lead, secondary, from electronic and electric

1.6%
37% scrap recycling, at plant
metal values from electric waste, in blister-
copper, at converter
Fig. 5.10 Unit process presentation of the converter step within the secondary copper plant. This multi-output proc-
ess yields the flows “secondary copper conversion”.
The arising slag is calculated from the content of ceramics in the scrap feed. Of the electronic scrap
31% consists of ceramics (Motorola, 2005). The yearly emissions for the converter aisle (“anode fur-
nace” in source) and for the Kaldo plant have been taken from Umweltbundesamt (2001) and Mark &
Lehner (2002), respectively. The distribution of differential PM classes have been adopted from cop-
per smelting and amount PM<2.5 : PM2.5-10 : PM>10 = 3 : 2 : 1 (CEPMEIP-Database, SNAP code
030304). The emissions have been allocated to electronic scrap processing according to their share in
the production of lead and copper respectively (Tab. 5.75). The infrastructure is housed on an area of
1’200 m2 (Boliden, 2000). The infrastructure was considered with the respective module “Facilities
anode production, secondary copper” (Tab. 5.78). A service life time of 50 years and yearly produc-
tion of 250’000 tons of combined copper and lead per year.

Tab. 5.75 Emssions to air and and water during the converter step. Respective Data for Kaldo plant taken from
Umweltbundesamt (2001) and Mark & Lehner (2002) for converter aisle. Calculation bases on 27’000 t of lead
and 223’000 t of copper, of which 2.37% and 6.58% originate from electronic scrap.
Kaldo Converter Aisle Total

Emissions to air per year per year per kg e-scrap proc-
essed
Attribution to e-scrap % 2.37% 6.58%
Dust kg 1780 1200 3.03E-06
PM >10 to air kg 5.05E-07
PM 2.5-10to air kg 1.01E-06
PM <2.5 to air kg 1.51E-06
Copper kg 30 34 7.37E-08
Lead kg 828 9 5.05E-07
Zinc kg 111 6.58E-08
Cadmium kg 2 1.19E-09
Arsenic kg 0.02 3.29E-11
Mercury kg 4 0 2.37E-09
Sulphur dioxide kg 9000 85600 1.46E-04
Nitrogen oxide kg 116 0 6.87E-06
Fluor kg 34 0 2.01E-08
Chlorid kg 4240 0 2.51E-06
PCDD/F kg 0.084 0 4.98E-11
Emissions to Water per year per year
Copper kg n.a. 8 1.32E-08
Lead kg n.a. 2.8 4.61E-09
Zinc kg n.a. 2.6 4.28E-09
Cadmium kg n.a. 0.02 3.29E-11
Arsenic kg n.a. 0.23 3.79E-10
Mercury kg n.a. 0.06 9.87E-11
Secondary Copper Refining

In this multi-output process the cast anodes are refined by electrolytic refining. Based on the assumed
constitution of the electric scrap a process flow balance was established (Fig. 5.11). Whereas the win-
ning of cathode copper may be considered as the intended process, in the context of the secondary
gold production the arising of anode slime is of importance. The process burdens are allocated to the
three products anode slime, nickel sulphate, and cathode copper as described above. The values for the
calculations are taken from Davenport et al. (2002).
metal values from electric waste, in blister-

copper, at converter
Anode casting unit Anode casting plant Others (Al, Ca, Fe, Sn)
Building 25%
nickel, secondary, from electronic and electric scrap

Anode copper (0.002% S, 0.15% O) 9.1%
Electrolyte cell
copper, secondary, from electronic and electric
Building Electrolytic refinery scrap recycling, at refinery
65%
1.66% PM from electric waste, in anode slime, at refinery

Fig. 5.11 Unit process representation of the electro refining step within the secondary copper plant. This multi-output
process yields the three (main) flows “Precious metals (PM) from electric waste”, “copper, secondary” and
“nickel, secondary”.
The electrorefining consumes 320 kWh per t of copper cathode. An amount of 0.2 m3 electrolyte per
tonne is bled and purified for recycling into electrolytic refining. It is assumed that all of this volume
is replenished with fresh electrolyte. The main constituent in the electrolyte is 180 kg sulphuric acid
per m3. Other additives like HCl, bone glue, thiourea, and avitone are disregarded, since they all range
from 20 to 100 g / m3. The 45 kg Copper per m3 electrolyte is assumed to originate from the anode
material or salts from within the operation. It is thus not inventoried. To heat the electrolyte from 20°C
to 65 °C no additional energy requirements are assumed, since the electrolysis heats up the electrolyte.
No heat recovery between circulations is assumed.
The data for the emissions (Tab. 5.76) originate from HELCOM (2002) and Umweltbundesamt
(2001).
The infrastructure is reflected with the specific inventory (see section Infrastructure for Secondary
gold production). There the lifetime production of the respective unit is said to amount to 6’900’000 t
cathode copper output. The infrastructure requirements in this dimensioned according to the copper
produced
Tab. 5.76 Emissions to air and water during the refining of the copper anode. Respective data taken from
Umweltbundesamt (2001) and Mark & Lehner (2002) Calculation bases on 223’000 t/a cathode copper and .
Emission unit annual release Per t cathode per kg anode

copper slime
Copper to water, lake kg 311.3 1.40E-03 5.44E-05
Lead to water, lake kg 16.8 7.53E-05 2.94E-06
Zinc to water, lake kg 16.8 7.53E-05 2.94E-06
Cadmium to water, lake kg 7.1 3.18E-05 1.24E-06
Arsenic to water, lake kg 8.04 3.61E-05 1.41E-06
Mercury to water, lake kg 0.09 4.04E-07 1.57E-08
Nickel to water kg 28.7 1.29E-04 5.02E-06
Dust to air kg 1300 5.83E-03 2.27E-04
Arsenic to air kg 14 6.28E-05 2.45E-06
Copper to air kg 24 1.08E-04 4.20E-06
Lead to air kg 53 2.38E-04 9.27E-06
Nickel to air kg 9 4.04E-05 1.57E-06
Precious metal refining, secondary copper

The processing of anode slimes from secondary copper production takes place in four steps:
- Pressure leaching and Drying
- Smelting and converting of the residue
- Electrowinning of Silver
- Electrowinning of Gold
A diagram of the multi-output process is given in Fig. 5.12

leaching vessel
kaldo oven Slime conversion
electrolysis cells
dorée metal palladium, secondary, at precious metal refinery

Building 5.10%
Precious metals extract.
2.8% 92.1%
gold, secondary, at precious
metal refinery silver, secondary, at precious
metal refinery
Fig. 5.12 Unit process representation of the precious metal refinery within the secondary copper plant. This multi-
output process yields the flows “gold”, “silver” and “palladium, secondary, from electronic and electric
scrap recycling”.
Pressure leaching. The slime is leached with electrolyte taken from the tankhouse under pressure (3-7
bar) and at elevated temperature of 115 °C. In Chen & Dutrizac (1990) we learn that the leaching can
occur in a single stage for 5 h at 100 psi O2 in a solution of 250 g/l H2SO4 and 10% NaCl to prevent
Ag from dissolution. This reduces the mass of the slime to 77% of the initial value. Assuming a vol-
ume of the leaching vessel of 10 m3 and a reaction mass of 5 m3 and a batch leaching procedure of
2’500 kg anode slime, 23 kg O2 per batch or 0.00925 kg O2 / kg slime are required. The sulphuric acid
is assumed to originate from the electrolyte bleed and the NaCl is regarded not to be consumed by pre-
cipitation reactions. To heat up the 5’000 l electrolyte from 65 °C to 115 °C an additional
0.17 kWh/kg slime is needed, assuming 70% heat conversion efficiency.
Drying. 0.1 kWh electricity per kg are assumed to be required for drying.
Smelting / Refining. For fuel oil and requirements of oxygen as reducing agent the specific values for
the conversion step in the Kaldo oven is taken as a proxy. Also the specific emission factors are taken
(Kaldo furnace, Tab. 5.75) that derive from dividing the yearly emissions of the Kaldo furnace by the
produced 27’000 t lead per year and scaling it to the metal produced by this process. According to
Peterson & Lundquist (1977), the Kaldo process consumes 52.9 l oil and 123.5 Nm3 of oxygen per
tonne of charge (45% lead) or 4 MJ light fuel oil and 0.369 kg oxygen per kg lead bullion produced.
These figures are multiplied with the total metal output of this inventory.
Möbius electrolysis. The silver concentration in the electrolyte is ca. 50 g/l. The free HNO3 is 10 g/l
and the anodic current density is 400 – 500 A/m2. The cell voltage is 2.0 – 2.5 V. The anodes and
cathodes are usually arranged in parallel in the cell, and the energy consumption is ca. 0.6 kWh per kg
of pure silver. To compensate for the anodically dissolved base metals that are not removed at the
cathode (mainly copper), silver nitrate solution is fed into the electrolyte to prevent depletion of silver
(Renner et al., 2002b). However, no additive consumption and replenishment of the electrolyte was
considered. Likewise no emissions were assumed because of the purity and the high value of the mate-
rial processed.
Wohlwill electrolysis. This process uses an electrolyte containing 2.5 mol/l of HCl and 2 mol/l of
HAuCl4 acid. Electrolysis is carried out with agitation at 65 – 75 °C. The raw gold is introduced as
cast anode plates. The cathodes, on which the pure gold is deposited, were for many years made of
fine gold of 0.25 mm thickness. These have now largely been replaced by sheet titanium or tantalum
cathodes, from which the thick layer of fine gold can be peeled off. In a typical electrolysis cell, gold
anodes weighing 12 kg and having dimensions 280×230×12 mm (0.138 m2 surface) are used. Opposite
to them are conductively connected cathode plates, arranged by two or three on a support rail. One cell
normally contains five or six cathode units and four or five anodes. The maximum cell voltage [V] is
1.5 V and the maximum anodic current density [A] 1500 A/m2 (Renner et al., 2002a). The South Afri-
can Rand refinery (Auerswald & Radcliffe, 2005) gives a specific gold production rate of 0.2 kg per
hour Wohlwill electrolysis. Assuming a current efficiency of 95% the energy consumption is [V] x [A]
/ 0.2 [kg/h] = 1.63 kWh per kg gold refined. No emissions were assumed because of the purity and the

high value of the material processed. The resulting sludge contains the PGM present in the electric
scrap and is sold for further processing.

General Flow information Representation in ecoinvent Uncertainty information

Infra-
Process Sub- Loca- Modul name in Mean Mean StDv General
Input Output Remarks Category struc- Unit Source Type
Name category tion ecoinvent value value 95% Comment
ture
light fuel oil, burned in
Light fuel oil, Not inventoried because of calorific value
burned
 contained in electronic scrap
oil heating systems No RER industrial furnace 1MW, non- 0.00E+00 MJ (Petterson, 1977) 1 1.51 (2,3,5,2,1,1,1)
modulating
Natural gas,
burned
 secondary copper conversion Not inventoried because of calorific value
contained in electronic scrap
natural gas heating systems No RER
0.00E+00 MJ (Davenport, 2002) 1 1.10 (2,3,1,2,1,3,1)
Pulverised Not inventoried because of calorific value
lignite, burned
 contained in electronic scrap
lignite fuels No DE pulverised lignite, at plant 0.00E+00 MJ (Petterson, 1977) 1 1.51 (2,3,5,2,1,1,1)
Oxygen, liquid, Not inventoried due to reducing abilities of
at plant
 electronic scrap.
chemicals inorganics No RER oxygen, liquid, at plant 0.00E+00 kg (Petterson, 1977) 1 1.53 (3,3,5,2,1,1,4)
Electricity, high to operate the Kaldo Smelter: 1.12 kWh per electricity, production mix
voltage
 kg lead produced.
electricity production mix No UCTE
UCTE
1.26E-01 kWh 1 1.14 (3,3,1,2,1,3,2)
Quicklime,
construction quicklime, milled, loose, at
milled, loose, at  to operate the converter Aisle
materials
additives No CH
plant
1.02E+00 kg (Davenport, 2002) 1 1.10 (2,3,1,2,1,3,4)
plant
Infrastructure Assumption: Service time 50 years, 250'000 t facilities blister-copper
Kaldo furnace,  metal produced per year; infrastructure metals extraction Yes SE conversion, secondary 8.35E-11 unit (BOLIDEN, 2006) 1 3.01 (2,3,1,2,1,3,9)
Aisle Converter scaled to Cu+Pb output of this process. copper
Slag, to further By product of Kaldo plant/Converter Aisle, waste residual
 treatment burdenfree management material landfill
No CH 0% water, to residual 1.68E+00 kg Mass balance 1 1.52 (2,3,1,2,4,3,6)
material landfill
APME 2000;
low population
 PM >10 to air See table and text in the respective section air
density
Particulates, > 10 um 1.38E-06 kg Questionaire 1999; 1 5.00 uncertainty estimated
CEPMEIP
APME 2000;
 PM 2.5-10to air See table and text in the respective section air
density 10um
2.75E-06 kg Questionaire 1999; 1 5.00 uncertainty estimated
CEPMEIP
APME 2000;
low population
 PM <2.5 to air See table and text in the respective section air
density
Particulates, < 2.5 um 4.13E-06 kg Questionaire 1999; 1 5.00 uncertainty estimated
CEPMEIP
Fig. 5.13 Flows for the multi-output process "secondary copper conversion" and its representation in the ecoinvent database (part I, continued on next page)


Infra-
ture
low population APME 2000;

▼ continued  Copper to air See table and text in the respective section air
density
Copper 2.01E-07 kg
Questionaire 1999
1 5.00 uncertainty estimated
Copper to HELCOM 2002;

 water, lake
See table and text in the respective section water lake Copper, ion 3.59E-08 kg
Questionaire 1999

 Lead to air See table and text in the respective section air
density
Lead 1.38E-06 kg
Questionaire 1999
Lead to water, HELCOM 2002;

 lake
See table and text in the respective section water lake Lead 1.26E-08 kg
Questionaire 1999
low population
 Zinc to air See table and text in the respective section air
density
Zinc 1.79E-07 kg APME 2000 1 5.00 uncertainty estimated
Zinc to water, HELCOM 2002;


See table and text in the respective section water lake Zinc, ion 1.17E-08 kg 1 5.00 uncertainty estimated
lake Questionaire 1999
low population
 Cadmium to air See table and text in the respective section air
density
Cadmium 3.23E-09 kg APME 2000 1 5.00 uncertainty estimated
Cadmium to HELCOM 2002;

 water, lake
See table and text in the respective section water lake Cadmium, ion 8.97E-11 kg
Questionaire 1999

 Arsen to air See table and text in the respective section air
density
Arsenic 8.97E-11 kg
Questionaire 1999
Arsen to water, HELCOM 2002;

 lake
See table and text in the respective section water lake Arsenic, ion 1.03E-09 kg
Questionaire 1999
low population
 Mercury to air See table and text in the respective section air
density
Mercury 6.46E-09 kg APME 2000 1 5.00 uncertainty estimated
Mercury to HELCOM 2002;

 water, lake
See table and text in the respective section water lake Mercury 2.69E-10 kg
Questionaire 1999
APME 2000;
Sulphur dioxide low population
 to air
See table and text in the respective section air
density
Sulfur dioxide 3.98E-04 kg Questionaire 1999; 1 5.00 uncertainty estimated
HELCOM 2002
Nitrogen oxide low population
 to air
See table and text in the respective section air
density
Nitrogen oxides 1.87E-05 kg HELCOM 2002 1 5.00 uncertainty estimated
low population
 Fluor to air See table and text in the respective section air
density
Fluorine 5.49E-08 kg APME 2000 1 5.00 uncertainty estimated
low population
 Chlorid to air See table and text in the respective section air
density
Chlorine 6.85E-06 kg APME 2000 1 5.00 uncertainty estimated

 PCDD/F to air See table and text in the respective section air
1.36E-10 kg APME 2000 1 5.00 uncertainty estimated
calculation linked
Waste, heat,
 emissions to air
air unspecified Heat, waste 4.53E-01 MJ to electricity 1 1.14 calculation
demand
lead, secondary, from
By-product of Kaldo plant/Converter Aisle
 Lead bullion
at 1.2 $ / kg, recieves 0.14% of the burden
metals extraction No SE electronic and electric scrap 4.36E-02 kg
recycling, at plant
Blister Copper, metal values from electric
Key output of the process at 36.1 $ / kg,
 to anode
recieves 99.86% of the burden
metals extraction No SE waste, in blister-copper, at 1.00E+00 kg
casting converter
Fig. 5. Flows for the multi-output process "secondary copper conversion" and its representation in the ecoinvent database (part II, continued from previous page)


Infra-
ture
Metal values 64% copper from electric waste contained, metal values from electric
from e-scrap, in  corresponds to a feed of 39 kg bilster metals extraction No SE waste, in blister-copper, at 6.04E+01 kg Calculations 1 3.02 (3,3,2,1,1,1,9)
blistercopper copper. converter
Infrastructure for Lifetime production: 230000 t per year

facilities anode refinery, Isaksson and
anode casting  during 30 years, normalised to 60.4 kg metals extraction Yes SE
secondary copper
5.65E-09 unit
Lehner, 2000
1 3.02 (3,3,2,1,1,1,9)
and refinining metal produced
Electricity, high electricity, production mix (300 to 400 kWh / t
voltage
 320 kWh per t of copper cathode required electricity production mix No UCTE
UCTE
1.25E+01 kWh Davenport 2002 2 1.95
cath Cu)
Water, Assumption: 100% of the electrolyte bled water, completely softened,
electrolyte
 (0.2kg / kg cathode copper) is refreshed
water supply production No RER
at plant
7.80E+00 kg 1 2.56 (5,5,5,5,5,5,4)
Concentration in the electrolyte 180 kg/m3.

Sulfuric acid, sulphuric acid, liquid, at
electrolyte
 Other additives (HCl, bone glue, thiourea chemicals inorganics No RER
plant
1.40E+00 kg 1 2.56 (5,5,5,5,5,5,4)
20 to 100 g/m3) disregarded .
0.2 m3 electrolyte per tonne is bled and
 Bleed stream 7.80E-03 m3 Davenport 2002
purified for recycling.

To heat the electrolyte from 20 °C to 65
Heat, unspecific, heat, natural gas, at
in chemical plant
 °C. Circulated electrolyte (bleed), no heat natural gas heating systems No RER
1.47E+00 MJ 1 1.40 (4,5,3,1,3,3,1)
recovery assumed.
 PM >10 to air air low population Particulates, > 10 um 3.79E-05 kg Questionaire 1999, 1 2.24 (1,2,1,3,5,3,25)
 PM 2.5-10to air air low population Particulates, > 2.5 um, and < 7.58E-05 kg Questionaire 1999, 1 2.67 (1,2,1,3,5,3,26)
 PM <2.5 to air air low population Particulates, < 2.5 um 1.14E-04 kg Questionaire 1999, 1 3.67 (1,2,1,3,5,3,27)
Copper to Calculated for Anode Caster/Electrolytic HELCOM 2002;
 water, lake Refinery Plant
water lake Copper, ion 5.44E-05 kg
Questionaire 1999
1 5.07 (4,4,4,5,5,2,12)
Lead to water, Calculated for Anode Caster/Electrolytic HELCOM 2002;
 lake Refinery Plant
water lake Lead 2.94E-06 kg
Questionaire 1999
1 5.07 (4,4,4,5,5,2,12)
Zinc to water, Calculated for Anode Caster/Electrolytic HELCOM 2002;
water lake Zinc, ion 2.94E-06 kg
Questionaire 1999
1 5.07 (4,4,4,5,5,2,12)
Cadmium to Calculated for Anode Caster/Electrolytic HELCOM 2002;
water lake Cadmium, ion 1.24E-06 kg
Questionaire 1999
1 5.07 (4,4,4,5,5,2,12)
Arsen to water, Calculated for Anode Caster/Electrolytic HELCOM 2002;
water lake Arsenic, ion 1.41E-06 kg
Questionaire 1999
1 5.07 (4,4,4,5,5,2,12)
Mercury to Calculated for Anode Caster/Electrolytic HELCOM 2002;
water lake Mercury 1.57E-08 kg
Questionaire 1999
1 5.07 (4,4,4,5,5,2,12)
Calculated for Anode Caster/Electrolytic
 Nickel to water
Refinery Plant
water lake Nickel, ion 5.02E-06 kg Questioaire 1999 1 5.07 (4,4,4,5,5,2,12)
Calculated for Anode Caster/Electrolytic low population HELCOM 2002;
 Arsen to air
Refinery Plant
air
density
Arsenic 2.45E-06 kg
Questionaire 1999
1 1.58 (4,4,4,5,5,2,12)
 Copper to air
Refinery Plant
air
density
Copper 4.20E-06 kg
Questionaire 1999
1 1.58 (4,4,4,5,5,2,12)
 Lead to air
Refinery Plant
air
density
Lead 9.27E-06 kg
Questionaire 1999
1 1.58 (4,4,4,5,5,2,12)
 Nickel to air
Refinery Plant
air
density
Nickel 1.57E-06 kg
Questionaire 1999
1 1.58 (4,4,4,5,5,2,12)
 Heat, waste, based on electricity consumption air unspecified Heat, waste 4.49E+01 MJ Calculation
Nickel, nickel, secondary, from
By-product at 4.5 $ / kg, recieves 12.86% of
 precipitated
the burden
metals extraction No SE electronic and electric scrap 5.49E+00 kg
from bleed recycling, at refinery
copper, secondary, from
Copper main at 13.8 $ / kg, recieves 5.60% of the
 cathodes burden
metals extraction No SE electronic and electric scrap 3.90E+01 kg
precious metals from
Anode slime, at By-product at 1105 $ / kg, recieves 81.54%
 refinery of the burden
metals extraction No SE electric waste, in anode 1.00E+00 kg
slime, at refinery
Fig. 5.14 Flows for the multi-output process "secondary copper refining" and its representation in the ecoinvent database


Infra-
ture
precious metals from

Precious metals 25% content of precous metals in slime,
in anode slime
 corresponds to 143.1 kg slime feed.
metals extraction No SE electric waste, in anode 3.58E+01 kg 1 1.09 (2,2,1,2,1,3,3)
slime, at refinery
facilities precious metal
Infrastructure  metals extraction Yes SE
refinery
5.89E-06 unit 1 3.01 (2,2,1,2,1,3,9)
10 m3 reaction vessel with 5 m3 reaction
electricity for mass. Leaching a 2500 kg slime batch at 100 Gürmen et al.,
leaching and  psi O2 and 115°C. 0.17 kWh/kg slime. 0.1 electricity production mix No UCTE 2.73E+01 kWh 2004; Davenport, 1 1.31 (4,2,1,2,3,3,2)
drying kWh per dry product assumed for 2002
subsequent drying.
oxygen  chemicals inorganics No RER oxygen, liquid, at plant 1.32E+00 kg Aylmore, 2001 1 1.31 (4,2,1,2,3,3,3)
Light fuel oil, 52.9 l oil per t charge, 45% lead content = 4
burned
 MJ per kg lead bullion
oil heating systems No RER industrial furnace 1MW, non- 6.44E+00 MJ Peterson 1977 1 5.00 estimation
modulating
Oxygen, liquid, 123.5 kg O2 per t charge, 45% lead content,
at plant
 1.34 kg/m3 O2 = 0..369 per kg lead bullion
chemicals inorganics No RER oxygen, liquid, at plant 5.50E-01 kg Peterson 1977 1 5.00 estimation
Electricity use electricity, high voltage,

Möbius (Silver)
 electricity production mix No UCTE
1.98E+01 kWh Kennethcott, 2004 1 2.10 (4,2,1,2,5,5,2)

Wohlwill (Gold)
1.63E+00 kWh Mech, 2005 1 2.10 (4,2,1,2,5,5,2)
specific for Kaldo, no emissions from low population Questionaire, 1999,

 PM >10
electrolysis. Particle distribution from
air
density
Particulates, > 10 um 3.93E-04 kg
CEPMEIP
specific values for Kaldo oven, no emissions low population
 Copper (air)
from electrolysis considered
air
density
Copper 3.97E-05 kg Questionaire, 1999 1 5.00 uncertainty estimated
 Lead (air)
air
density
Lead 1.10E-03 kg Questionaire, 1999 1 5.00 uncertainty estimated
 Zinc (air)
air
density
Zinc 1.47E-04 kg HELCOM 2002 1 5.00 uncertainty estimated
 Cadmium (air)
air
density
Cadmium 2.65E-06 kg Questionaire, 1999 1 5.00 uncertainty estimated
 Mercury (air)
air
density
Mercury 5.30E-06 kg HELCOM 2002 1 5.00 uncertainty estimated
Sulphur dioxide specific values for Kaldo oven, no emissions low population
 (air) from electrolysis considered
air
density
Sulfur dioxide 1.19E-02 kg HELCOM 2002 1 5.00 uncertainty estimated
Nitrogen oxide specific values for Kaldo oven, no emissions low population
air
density
 Fluor (air)
air
density
Fluorine 4.50E-05 kg HELCOM 2002 1 5.00 uncertainty estimated
 Chlorid (air)
air
density
Chlorine 5.62E-03 kg HELCOM 2002 1 5.00 uncertainty estimated
specific values for Kaldo oven, no emissions low population Dioxins, measured as 2,3,7,8- Lindström, 2001,
 PCDD/F (air)
air
1.11E-07 kg
HELCOM, 2002
By-product at 15682 $ / kg, recieves 39.7% gold, secondary, at precious
 Gold, secondary
of the burden
metals extraction No SE
metal refinery
1.00E+00 kg 1
Silver, By-product at 269 $ / kg, recieves 22.4% of silver, secondary, at
 secondary the burden
precious metal refinery
3.29E+01 kg 1
Palladium, By-product at 8214 $ / kg, recieves 37.9% palladium, secondary, at
 secondary of the burden
1.82E+00 kg 1
Fig. 5.15 Flows for the multi-output process "secondary copper, precious metal refinery" and its representation in the ecoinvent database

Infrastructure for Secondary gold production

This sub-chapter presents the datasets describing the facilities used in the production of secondary
gold. For each of the three steps conversion, refinery, and precious metal plant a corresponding infra-
structure inventory was established. The respective meta-data are shown in Tab. 5.77, whereas the
flows and representations can be found in Fig. 5.16, Fig. 5.17 and Fig. 5.18. For more detailed descrip-
tion and use of the facilities refer to the previous process description in this chapter.
Tab. 5.77 Meta information for the infrastructure in the production of secondary gold from electric waste.
Name facilities copper conversion, facilities anode refinery, sec- facilities precious metal refin-
secondary copper ondary copper ery
Location SE SE SE
Infrastructure Proc- 1 1 1
ess
Unit unit unit unit
Included Processes Construction of facilities and Construction of facilities and Construction of facilities and
housing, and their disposal. No housing, and their disposal. No housing, and their disposal. No
transports and construction work transports and construction work transports and construction work
considered. considered. considered.
Amount 1 1 1
Local Name Anlagen Blisterkupfer Verhüt- Anlagen Anoden Raffinierung, Anlagen Edelmetall Hütte
tung, sekundär Kupfer Sekundär Kupfer
Synonyms Kaldo Converter
General Comment to Production of Kaldo converter Production of twin anode casting The precious metal plant is
reference function and converter aisle: chromium wheel and electrolytic cells: equipped as follows: 1 Electro-
steel, refractory bricks and con- chromium steel, refractory bricks lyte cell, 1 Eluation cell, 5 Elec-
crete foundation. Building hall and concrete foundation. Build- trowinning cell, 1 Calcination
houses the facilities, covering ing hall houses the facilities, furnace, 1 Smelting furnace, 1
2
1200m2. Service lifetime 50 covering 6’883m . Service life- Electrorefining cell. A building
year, 250'000 t of metals pro- time 30 years, yearly capacity of hall houses the equipment cov-
2
duced per year. 230'000 t copper refined. ering 5'000m . The service life
time assumed is 25 years, the
yearly capacity of anode slime
treatment is 243 t.
Start Date 2002 2002 2002
End Date 2005 2005 2005
Data Valid For Entire 1 1 1
Period
Geography text Data describes a Swedish plant Data describes a Swedish plant Data describes a Swedish plant
that is singular in the region and that is singular in the region and that is singular in the region and
ranges worldwide among the big ranges worldwide among the big ranges worldwide among the big
producer. It may be taken as producer. It may be taken as producer. It may be taken as
approximation for the global approximation for the global approximation for the global
situation situation situation
Technology text Due to high temperature chro- Due to high temperature chro- Due to high temperature chro-
mium steel and refractory mate- mium steel and refractory mate- mium steel and refractory mate-
rials are used. rials are used. rials are used.
Uncertainty Adjust- see Geography and Technology see Geography and Technology see Geography and Technology
ments
Facilities copper conversion, secondary copper

For the furnaces a concrete foundation of 130 m3 was assumed. The converter aisle and the Kaldo
plant modelling was based on official commercial descriptions (Tab. 5.78). The housing was estimated
to be 1’200 m2 large, based on the corporate environmental report. The foundation was assessed with
product descriptions to amount 130 m3, of which one third consists of exacting concrete, the rest of
sole plate concrete. The disposal of the refractory lining is approximated with the disposal of nickel
smelter slag. The steel is assumed to be recycled completely at the end of life of the facilities. Follow-
ing the general modelling principles of ecoinvent, the steel therefore leaves the system burden-free and
no disposal processes for steel are considered. The complete inventory is shown in Fig. 5.16.
Tab. 5.78 Calculations for the infrastructure of the converter; 50 years service lifetime, 250’000 t production of total
metals per year.
Materialisation Converter Kaldo Converter Aisle Total

Chromium steel 18/8, at plant kg 1.99E+03 5.96E+04 6.16E+04
Section bar rolling, steel kg 1.99E+03 5.22E+04 5.42E+04
Drawing of pipes, steel kg 7.44E+03 7.44E+03
Refractory, basic, packed, at plant kg 1.02E+05 3.22E+06 3.32E+06
Disposal, nickel smelter slag, 0% water, to residual ma-
1.02E+05 3.22E+06
terial landfill kg 3.32E+06
Facilities anode refinery, secondary copper

This dataset comprises the anode casting unit and an array of 674 electrolytic cells with an estimated
lifetime of 50 years and an annual copper production capacity of 230’000 t (Isaakson & Lehner,
2000). The anode casting unit (Fig. 5.5) and the electrolytic cells (Fig. 5.6) were herewith modelled
based on official commercial descriptions (Tab. 5.79). The moulds of the anode casting wheels are
lined with 0.002 m graphite. The walls of the electrolytic cells are made of stainless steel and rein-
forced with glass fibres. The disposal of the cell is approximated with the disposal of refractory. The
steel is assumed to be recycled completely at the end of the facilities life time. Following the general
modelling principles of ecoinvent, the steel therefore leaves the system burden free and no disposal
processes for steel are considered. In the context of dataset’s use within the production of secondary
gold, 14’680 t blister copper from electronic scrap is yearly processed by the facilities.
Tab. 5.79 Calculations for infrastructure of the anode refinery. 30 years service lifetime, 230’000 t copper per year.
Casting Electrolyte
Materialisation anode refinery Total
unit cell
Requriements units 1 674
Chromium steel 18/8, at plant kg 1.47E+05 2.97E+03 2.15E+06
Section bar rolling, steel kg 4.40E+04 4.40E+04
Sheet rolling, chromium steel kg 1.03E+05 2.97E+03 2.10E+06
Glass fibre reinforced plastic, polyamide, injection mould-
kg 7.41E+02 4.99E+05
ing, at plant
Graphite, at plant kg 7.49E+03 7.49E+03
Electronics for control units kg 3.50E+01 3.50E+01
Disposal, refractory SPL, Al elec.lysis, 0% water, to re-
kg 3.71E+03 2.50E+06
sidual material landfill
Precious metal refinery, secondary copper

This dataset describes the infrastructure operated in the precious metal refinery. The precious metal
refinery stretches over 2’000 m2 and houses following equipment: 1 pressure leaching vessel, 5 elec-

trorefining / electrowinning cells, 1 Kaldo furnace. The material consumption and disposal of the
equipment is displayed in Tab. 5.80. The land use was assessed calculating with the 2’000 m2 and the
considered 25 year life time.
The electrowinning cell was reflected with the required steel for the vessel, the production of the steel
wool (electrode, approximated with wire drawing) and a lead electrode, according to a operation for
copper electrowinning described in Davenport et al. (2002). The other equipment parts have been ma-
terialised based on estimations and datasheets of the producer. The steel is assumed to be recycled
completely at the end of the facilities life time. Following the general modelling principles of ecoin-
vent, the steel therefore leaves the system burden free and no disposal processes for steel are consid-
ered. The facilities produce 243 tonnes precious metals – silver, gold and palladium – per year.
Tab. 5.80 Calculations for equipment of the precious metal plant. 25 years service lifetime, 243 t PM per year.
Pressure
leaching
winning
Electro-
Electro-
refining
vessel
Kaldo
oven
cell
cell
Total
Requirements unit 1 5 1 5
Chromium steel 18/8, at
3.45E+03 6.39E+01 1.99E+03 5.63E+02 8.58E+03
plant kg
Section bar rolling, steel kg 1.99E+03 1.11E+02 2.55E+03
Sheet rolling, chromium
3.45E+03 4.21E+02 5.63E+02 8.37E+03
steel kg
Wire drawing, steel kg 6.39E+01 3.19E+02
Glass fibre reinforced plastic kg 7.41E+02 2.09E+02 1.79E+03
Concrete, exacting, at plant m3 7.24E-01 3.62E+00
Refractory, basic, packed, at
1.02E+05 2.33E+03 1.14E+05
plant kg
Lead, at regional storage kg 5.41E+02 2.70E+03
Tin, at regional storage kg 5.29E+00 2.64E+01
Calcium carbide, technical
7.47E-02 3.74E-01
grade, at plant kg
Disposal, refractories kg 1.02E+05 1.02E+05
Disposal, El. Cells kg 3.71E+03 3.71E+03
Disposal, concrete kg 7.24E-01 3.62E+00

Infra-
ture
Chromium steel 18/8,

at plant
 metals extraction No RER chromium steel 18/8, at plant 6.16E+04 kg 1 1.07 (2,1,1,2,1,1,3)
facilities copper conversion, secondary copper

Section bar rolling,
steel
 metals processing No RER section bar rolling, steel 5.42E+04 kg 1 1.07 (2,1,1,2,1,1,3)
Drawing of pipes,
steel
 metals processing No RER drawing of pipes, steel 7.44E+03 kg 1 1.07 (2,1,1,3,1,1,3)
Refractory, basic, construction refractory, basic, packed, at
packed, at plant
 materials
bricks No DE
plant
3.32E+06 kg 1 1.07 (2,1,1,2,1,1,3)
Building, hall, steel Hall housing the kaldo oven and the construction building, hall, steel
construction
 converter aisle on 1'200 m2. processes
buildings Yes CH
construction
1.20E+03 m2 1 1.21 (4,1,1,3,1,1,3)
Only real industry facilities taken into
Transformation, from
unkown
 account, no other areas (e.g. parking resource land Transformation, from unknown 1.20E+03 m2 1 2.00 (1,1,1,1,1,1,8)
places) considered
Transformation, to the area on which the Kaldo oven and the
Transformation, to industrial
industrial area, built  converter aisle are build, calculated as resource land
area, built up
1.20E+03 m2 1 2.00 (1,1,1,1,1,1,8)
up see above
Occupation, industrial the area occupied by the Kaldo oven and Occupation, industrial area,
area, built up
 the converter aisle during 50 years
resource land
built up
6.01E+04 m2a 1 1.52 (3,1,1,1,1,1,7)
Concrete, sole plate

70% of the total 130 m3 concrete construction concrete, sole plate and
and foundation, at  foundation materials
concrete No CH
foundation, at plant
8.72E+01 m3 1 1.12 (3,1,1,3,1,1,3)
plant
Concrete, exacting, at 30% of the total 130 m3 concrete construction
plant
 foundation materials
concrete No CH concrete, exacting, at plant 4.36E+01 m3 1 1.12 (3,1,1,3,1,1,3)
Disposal, nickel
smelter slag, 0% disposal, nickel smelter slag,
waste residual
 water, to Proxy for the disposal of refractory bricks
No CH 0% water, to residual material 3.32E+06 kg 1 1.07 (2,1,1,3,1,1,6)
residual landfill
material landfill
Disposal, steel
 from converter
Recycled, not inventoied due to cut off. 6.16E+04 kg 1 1.07 (2,1,1,3,1,1,6)
Disposal, waste disposal, building, concrete,

 concrete management
recycling No CH
not reinforced, to sorting plant
1.31E+02 kg 1 1.12 (3,1,1,3,1,1,6)
Facilities facilities copper conversion,

 (conversion)
electronics Metals Yes SE
secondary copper
1.00E+00 unit
Fig. 5.16 Flows for "facilities copper conversion" and its representation in the ecoinvent database.


Infra-
ture

at plant
steel
facilities anode refinery, secondary copper
Sheet rolling,
chromium steel
 metals processing No RER sheet rolling, chromium steel 2.10E+06 kg 1 1.07 (2,1,1,2,1,1,3)
glass fibre reinforced plastic,
Glass fibre reinforced
plastic
 plastics others No RER polyamide, injection moulding, 4.99E+05 kg 1 1.07 (2,1,1,3,1,1,3)
at plant
Graphite, at plant  chemicals inorganics No RER graphite, at plant 7.49E+03 kg 1 1.07 (2,1,1,2,1,1,3)
Electronics for control construction
units
 processes
buildings No RER electronics for control units 3.50E+01 kg 1 1.07 (2,1,1,3,1,1,3)
Building, hall, steel To house the Anode casting plant and the construction building, hall, steel Davenport 2002, Boliden
construction
 Electrolytic Refinery Plant. processes
buildings Yes CH
construction
6.88E+03 m2
2004
1 1.21 (4,1,1,3,1,1,3)
unkown
places) considered
Transformation, to the area on which the Anode casting plant
industrial area, built  and the Electrolytic Refinery Plant are resource land
area, built up
6.88E+03 m2 1 2.00 (1,1,1,1,1,1,8)
up build, calculated as see above
the area occupied by the Anode casting
Occupation, industrial Occupation, industrial area,
area, built up
 plant and the Electrolytic Refinery Plant resource land
built up
2.06E+05 unit 1 1.52 (3,1,1,1,1,1,7)
during 30 years
Concrete, sole plate Foundation for Anode casting plant and
construction concrete, sole plate and
and foundation, at  Electrolytic Refinery Plant, assuming a
materials
concrete No CH
foundation, at plant
6.73E+03 m3 3 1.12 (3,1,1,3,1,1,3)
plant fundament thickness of 1.2 m
Foundation for Anode casting plant and
Concrete, exacting, at Electrolytic Refinery Plant, assuming an construction
plant
 additional concrete mass of 10% of sole materials
and plate.
disposal, refractory SPL, Al
Disposal, El. waste residual
 Cells
Approx with disposal Al Electrolysis
No CH elec.lysis, 0% water, to residual 2.50E+06 kg 1 1.51 (2,1,1,3,4,1,6)
material landfill
Disposal, steel,
 anode cast
Recycled, not inventoied due to cut off. 1.47E+05 kg 1 1.07 (2,1,1,3,1,1,6)

 concrete
Sum of concrete input
management
recycling No CH
6.88E+03 kg 1 1.07 (2,1,1,3,1,1,6)
Facilities facilities anode refinery,

 (refining)
electronics Metals Yes SE
secondary copper
1.00E+00 unit
Fig. 5.17 Flows for "facilities anode refinery" and its representation in the ecoinvent database.


Infra-
ture

at plant
steel
Sheet rolling,
chromium steel
 metals processing No RER sheet rolling, chromium steel 8.37E+03 kg 1 1.07 (2,1,1,2,1,1,3)
glass fibre reinforced plastic,
Glass fibre reinforced
plastic
 plastics others No RER polyamide, injection moulding, 1.79E+03 kg 1 1.07 (2,1,1,3,1,1,3)
facilities precious metal refinery
at plant
Concrete, exacting, at construction
plant materials
Refractory, basic, construction refractory, basic, packed, at
packed, at plant
 materials
bricks No DE
plant
1.14E+05 kg 1 1.07 (2,1,1,3,1,1,3)
Lead, at regional
storage
 metals extraction No RER lead, at regional storage 2.70E+03 kg 1 1.51 (2,1,1,3,4,1,3)
Tin, at regional
storage
 metals extraction No RER tin, at regional storage 2.64E+01 kg 1 1.51 (2,1,1,3,4,1,3)
Calcium carbide,
calcium carbide, technical
technical grade, at  chemicals inorganics No RER
grade, at plant
3.74E-01 kg 1 1.51 (2,1,1,3,4,1,3)
plant
Building, hall, steel construction building, hall, steel
construction
 To house the precious metal plant.
processes
buildings Yes CH
construction
2.00E+03 m2 1 1.21 (4,1,1,3,1,1,3)
unkown
places) considered
Transformation, to
industrial area, built  resource land
area, built up
2.00E+03 m2 1 2.00 (1,1,1,1,1,1,8)
up
Disposal, Approximated with Nickel smelter slag waste residual
 refractories disposal management material landfill
No CH 0% water, to residual material 1.02E+05 kg 1 1.78 (5,1,1,3,4,1,6)
landfill
disposal, refractory SPL, Al
Disposal, El. waste residual
 Cells management material landfill
No CH elec.lysis, 0% water, to residual 3.71E+03 kg 1 1.78 (5,1,1,3,4,1,6)
material landfill
 Disposal, steel Recycled, not inventoied due to cut off. 5.62E+03 kg 1 1.07 (2,1,1,3,1,1,6)

 concrete
Sum of concrete input
management
recycling No CH
3.62E+00 kg 1 1.07 (2,1,1,3,1,1,6)
Facilities
 (Precious electronics Metals Yes SE
refinery
1.00E+00 Unit
metal)
Fig. 5.18 Flows for "facilities precious metal refinery" and its representation in the ecoinvent database.

5.2.3 Data quality

Data uncertainty is derived using the Pedigree matrix as described in Frischknecht et al. (2004). The
uncertainty for the inventory items related to the Kaldo process and the conversion are underestimated
by the Pedigree approach, therefore the standard deviation of the flows is set to 5 to reflect their specu-
lative nature. As there was only a limited access to information of the industry and as data partly dated
back more than ten years ago, the reliability and correctness of the data set is limited. Due to limitation
in time for this LCI study, only the Boliden process for the production of secondary gold was ana-
lysed. That may further limit the explanatory power when it comes to LCA studies of products where
the production of secondary gold becomes important. However, since the concentration of precious
metals in the electronic scrap feed is high compared to natural ores, the resulting exchanges related to
the production of secondary gold and silver are low compared to gold and silver from ores. Therefore
the quality is sufficient for the use as background data in a supply mix of primary and secondary gold.
5.3 Production of Silver from Lead-Zinc ores

Today, the silver in bullion lead, together with gold and platinum metals, is concentrated by the Parkes
process. The process involves the production of a solid alloy of lead, zinc, and silver by stirring zinc
into molten impure lead. On cooling a crust is formed that floats on the surface of the molten metal
(450 °C) where it is skimmed off (Renner et al., 2002b). The composition of this crust is determined
by the laws of solid solution of the ternary alloy system Ag-Pb-Zn. The composition of the crust lies
between 5 to 10% Ag, 15 to 30% Zn, the remainder being lead with a small amount of oxygen and
other contaminants such as copper (Evans et al., 1993). Reuter (2005) indicates in the calculations
made a composition of 18% Ag, 43% Zn with 38% lead, and 1% antimony. This so-called parkes de-
silvering crust is processed to raw silver by following steps:
 The adhering liquid lead is removed by hot pressing.
 The zinc is distilled in a retort or vacuum furnace. The residue consists of a lead – noble-
metal alloy with up to 50 % silver, ca. 1 % zinc, and some copper, arsenic, antimony, and bis-
muth.
 It is oxidized by cupellation to obtain crude silver and oxidised lead (litharge) with remaining
contaminants.
 The resulting raw silver is electrolytically refined (electrorefined) by the Möbius electrolysis.
5.3.1 Silver from lead production in ecoinvent

No explicit study was available that covers the life cycle inventory of the production of silver from
lead. Thus this inventory is solely based upon available information from literature. In cases where no
explicit silver recovery information was available, information on the refining of zinc and lead was
used as proxy.
The meta information for these processes are compiled in Tab. 5.81, the reference flows for the multi-
input processes are displayed in Fig. 5.19.

Tab. 5.81 Meta information for the production processes of silver from lead production.
Name silver, from lead production, at plant

Location GLO
Unit kg
Included Processes Processing of raw material from lead desilvering, infrastructure, electricity and
process emissions to air.
Amount 1
Local Name Silber, aus Bleiproduktion, ab Werk
Synonyms
General Comment to reference
function Production of refined silver (99.9%) from Parkes crust from lead desilvering.
Start Date 2002
End Date 2005
Geography text Data plant on average global context.
Technology text Processing of Parkes desilvering crust by hot pressing, dezincing (vacuum dis-
tillation), cupellation of lead and moebius electrolysis (electrowinning)
Uncertainty Adjustments see Geography and Technology
Raw and auxiliary materials, waste

Assuming a silver concentration in the Parkes crust of 18% and an overall recovery yield of 97.3%
(Reuter, 2005) an amount of 5.77 kg crust is fed to obtain 1 kg refined silver.
To form a Parkes crust, zinc is stirred into the molten lead. This added zinc however goes to 97% to
the lead that is pressed out of the Parkes crust and is returned to the lead processing. The remainder is
recovered to 98.2% by vacuum distillation and is returned to form the Parkes crust. Only the remain-
ing 1.8% reports to the intermediate litharge (Reuter, 2005). All the intermediate products as well as
the litharge are returned into the process. No further auxiliary materials such as fluxing agents are con-
sidered. All employed materials are supposed to be recycled into the production process, therefore no
waste fraction is considered.
Energy demand
Electricity demand for the different process stages is reported in Reuter (2005) per kg silver as fol-
lows: Vacuum Distillation: 1.44 kWh; Cuppellation: 4.53 kWh; Moebius Electrolysis: 0.60 kWh.
Process emissions
Rentz et al. (1999) report process emissions related to different processes in the lead and zinc industry
in Germany. Since no emission factors for the silver production are available, values for comparable
best available technologies (Remelting plant and New Jersey Retort) are taken as approximation. The
distribution of differential PM classes have been adopted from copper smelting and amount PM<2.5 :
PM2.5-10 : PM>10 = 3 : 2 : 1 (CEPMEIP-Database, SNAP code 030304). No emissions to water are
considered.
Transport and infrastructure

Since the parkes crust is supposed to be processed on site and only a small part of the feed is assumed
to be acquired from supply external to the facility, no transports of the feed are considered. The infra-
structure is reflected with the generic dataset for a metal smelter that is based on inventories from alu-
minium and iron industry. Lifetime: 50a, output: 2'650'000 t metal per year, reporting to the total
amount of silver, zinc and lead intermediate produced.
Tab. 5.82 Emissions to air from parkes crust processing. Approximation from BAT processes in German Zinc industry
(Rentz et al., 1999), geometric mean of the lowest and highest value.
Emission to CBAT Remel- CBAT New low high geometric

air ting plant Jersey Retort mean (this
study)
g / t Zinc pro- g / t Zinc pro- g / t Zinc pro- g / t Zinc pro- g / t Zinc pro-
duced duced duced duced duced
Dust 25 - 90 190 25 190 68.9
Zn 15 - 50 80 15 80 34.6
Cd 0.03 0.05 0.03 0.05 0.04
Pb 2-8 6 2 8 4
5.3.2 Data quality

The emissions are approximated with generic processes in zinc and lead production. Data uncertainty
is derived using the Pedigree matrix as described in Frischknecht et al. (2004). Quality is sufficient to
figure as part in the supplymix of silver.


Infra-
ture
Variability based on
18.1% Silver contained, overall parkes process crust, from
Parkes crust feed  recovery yield of 97.4%
metals extraction No GLO
desilverising of lead
5.77E+00 kg Reuter 2005 1 1.10 values reported in
literature.
Hot pressing: 0.00 kWh; Vacuum
Total electricity Distillation: 1.44 kWh; Cuppellation: electricity, medium voltage,
demand
 4.53 kWh; Moebius Electrolysis: 0.60
electricity production mix No UCTE
6.57E+00 kWh Reuter 2005 1 1.24 (3,1,2,1,1,5,2)
kWh;
silver, from lead production, at plant
Based on inventories from aluminium

Infrastructure  and iron industry. Lifetime: 50a, metals extraction Yes GLO non-ferrous metal, smelter 3.84E-11 unit assumptions 1 3.32 (4,1,2,1,4,5,9)
output: 2'650'000 t metal per year
17% share in the total dust emissions low population Rentz et al 1999,
 PM >10 to air
of 69 g per tonne crust processed
air
density
CEPMEIP
1 2.24 (1,2,1,3,5,3,25)
33% share in the total dust emissions low population Particulates, > 2.5 um, and < Rentz et al 1999,
 PM 2.5-10to air
air
density 10um
1.33E-04 kg
CEPMEIP
1 2.67 (1,2,1,3,5,3,26)
50% share in the total dust emissions low population Rentz et al 1999,
 PM <2.5 to air
air
density
Particulates, < 2.5 um 1.99E-04 kg
CEPMEIP
1 3.67 (1,2,1,3,5,3,27)
Zinc emission to low population

 air
air
density
Zinc 3.46E-05 kg Rentz et al 1999 1 2.24 (1,2,1,3,5,3,31)
Cadmium low population

 emission to air
air
density
Cadmium 3.87E-08 kg Rentz et al 1999 1 2.24 (1,2,1,3,5,3,31)
Lead emission low population

 to air
air
density
Lead 3.46E-06 kg Rentz et al 1999 1 2.24 (1,2,1,3,5,3,31)
low population
 Waste heat air
density
Heat, waste 2.37E+01 MJ Calculation 1 1.06 (1,2,1,3,1,1,13)
Internally returned to metallurgical

Zinc
 intermediate
productio process for recovery, not 2.28E+00 kg Reuter 2005 1 1.24 (3,1,1,1,1,5,3)
inventoried
Lead
 intermediate
productio process for recovery, not 1.81E+00 kg Reuter 2005 1 1.24 (3,1,1,1,1,5,3)
inventoried
 Litharge productio process for recovery, not 6.86E-01 kg Reuter 2005 1 1.24 (3,1,1,1,1,5,3)
inventoried
silver, from lead production, at
 Silver metals extraction No GLO
plant
1.00E+00 kg
Fig. 5.19 Flows for "silver, from lead production, at plant" and its representation in the ecoinvent database

5.4 Production of silver and raw copper telluride from copper

ores
No explicit study covering the extraction of silver as by-product from copper production could be
found. This inventory is based on available industry information and literature data.
The processing of anode slimes from copper production takes place in four steps:
- Pressure leaching of slimes
- Precipitating of copper telluride from solution
- Drying, smelting and converting of the residue
- Electrowinning of silver
In the decopperising leaching process of copper refinery slimes, a considerable amount of tellurium
dissolves (70-80% and 95% under atmospheric pressure leaching and pressurized leaching, respec-
tively). The precipitation of telluride is done by copper choppings that are arranged in a column. The
leach solution is led through the column and telluride precipitate on the surface (Shibasaki et al.,
1992). Fig. 5.20 shows the process as done in Mitsubishi’s Naoshima refinery.
Fig. 5.20 Processing of anode slimes to retract silver and tellurium at Naoshima refinery, Mitsubishi Materials Corp.
(Shibasaki et al., 1992).
5.4.1 Processing of anode slime, primary copper production, in ecoinvent

This module consists of a multi-output dataset (see Fig. 5.22 and Tab. 5.83). The multi-output-process
"processing of anode slime, primary copper production" delivers the two co-products "copper-telluride

cement, from copper production" and "silver, from copper production, at refinery". The product "cop-
per-telluride cement, …" is an intermediary in the production of telluride and is not intended to be
used by the LCA practitioner. The flow "silver, …" is a by-product of this process that receives part of
the burdens. It must not be used to reflect silver. For silver and telluride, please refer to the respective
adequate datasets.
The raw anode slime is inventoried by the by-product dataset “anode slime, silver and tellurium con-
taining, primary copper production” (Fig. 5.21). The slime represented contains 20% Silver and 3%
Tellurium which in turn reflects the coupled production ratio of silver and tellurium from copper pro-
duction industry on a global average (see Part III “Copper” in this report).
Copper production Silver production

anode slime processing of
generation, primary anode slime, silver, from copper
anode slime, silver and tellurium
copper production primary copper production, at refinery
containing, primary copper production
production
cathode, copper, primary copper production copper-telluride cement, from copper production
Fig. 5.21 Model overview and interrelation of silver production with copper extraction. For detailed information see
Part III “Copper” in this report.
The anode slime processing corresponds to a part of similar processes described in the chapter 5.2.2.1
Gold / Silver, secondary, from electronic scrap in ecoinvent. Hence, the respective values are adapted
for the processing of silver and telluride containing slimes.
Tab. 5.83 Meta information for the production processes of silver from copper production.
Name processing of anode slime, primary copper production

Location GLO
Unit kg
Included Processes Generation of anode slime from primary copper production (input), leaching and precipita-
tion of copper telluride, melting and electrolysis of silver
Amount 1
Local Name Anodensumpf-Aufbereitung, primäre Kupferproduktion
General Comment to The multi-output-process "processing of anode slime, primary copper production" delivers
reference function the two co-products "copper-telluride cement, from copper production" and "silver, from
copper production, at refinery". The product "copper-telluride cement, …" is an intermedi-
ary in the production of telluride and is not to be used by the LCA practitioner. The flow
"silver, …" is a by-product of this process that receives part of the burdens allocated. The
data set must not be used to reflect silver! For tellurium and silver the respective, ade-
quate end-user datasets have to be chosen.
Geography text Data on average global context.
Technology text Leaching of anode slime from primary copper production and precipitation of copper tellu-
ride on copper choppings, melting of residues with subsequent Moebius electrolysis.
Uncertainty Adjust- see Geography and Technology
ments
Pressure leaching. The slime (5 kg) is leached with electrolyte taken from the tankhouse under pres-
sure (3-7 bar) and at elevated temperature of 115 °C. From Chen & Dutrizac (1990) we learn that the

leaching can occur in a single stage for 5 h at 100 psi O2 in a solution of 250 g / l H2SO4 and 10%
NaCl to prevent Ag from dissolution. This reduces the mass of the slime to 77% of the initial value.
Assuming a volume of the leaching vessel of 10 m3 and a reaction volume of 5 m3 and a batch leach-
ing procedure of 2’500 kg anode slime, 23 kg O2 per batch or 0.00925 kg O2 / kg slime are required
(see calculations in 5.2.2.1 Gold / Silver, secondary, from electronic scrap in ecoinvent). The sulphuric
acid is assumed to originate from the electrolyte bleed and the NaCl is regarded not to be consumed by
precipitation reactions. To heat up the 5’000 l electrolyte from 65 °C to 115 °C an additional
0.17 kWh/kg slime is needed, assuming 70% heat conversion efficiency. A process efficiency of 90%
is assumed, the rest of the Tellurium reports to the residual solution.
Precipitation. The copper telluride is precipitated from the leach solution by solid copper. This pre-
cipitation – or stripping – is spontaneous given the optimal reaction time. Hence only the metallic
copper choppings are accounted for, that are used to strip the raw copper telluride. They are contained
in the produced raw copper telluride by 50%wt.
Drying. 0.1 kWh electricity per kg dry product is assumed to be required for drying.
Smelting / Refining. For fuel oil and requirements of oxygen as reducing agent the same specific val-
ues as for the conversion step in the Kaldo oven is taken as a proxy. Also the specific emission factors
are taken (Kaldo furnace, Tab. 5.75) that derive from dividing the yearly emissions of the Kaldo fur-
nace by the produced 27’000 t lead per year and scaling it to the metal produced by this process. Ac-
cording to Peterson & Lundquist (1977), the Kaldo process consumes 52.9 l oil and 123.5 Nm3 of
oxygen per tonne of charge (45% lead) or 4 MJ light fuel oil and 0.369 kg oxygen per kg lead bullion
produced. These figures are multiplied with the total metal output of this inventory. Eventual electric-
ity consumption for furnace operation is assumed to be low and therefore is disregarded in this step.
Möbius electrolysis. The silver concentration in the electrolyte is ca. 50 g/L. The free HNO3 is 10 g/L
and the anodic current density is 400 – 500 A/m2. The cell voltage is 2.0 – 2.5 V. The anodes and cath-
odes are usually arranged in parallel in the cell, and the energy consumption is ca. 0.6 kWh per kg of
pure silver. To compensate for the anodically dissolved base metals that are not removed at the cath-
ode (mainly copper), silver nitrate solution is fed into the electrolyte to prevent depletion of silver
(Renner et al., 2002b). However, no additive consumption and replenishment of the electrolyte was
considered. Likewise no emissions were assumed because of the purity and the high value of the mate-
rial processed.
Allocation procedure. The allocation of the subprocesses copper telluride and silver formation is not
evident to solve, since only the final commodity prices are known but no in-firm price for the CuTe
cement, on which the allocation would rely. Also the share of the products in total costs (gross sales
method as proposed in Guinée et al. (2004) and Guinée et al. (2001)) is not easy to estimate. The by-
product allocation problem – the joint production of silver and copper telluride – is reduced here by a
subdivision of the sub-processes (according to Ekvall & Finnveden, 2001). The tellurium production
from the raw copper telluride is treated in a distinct inventory. Hence it is the task to allocate the bur-
dens between the silver and the CuTe cementate, whereas for latter no published price is available.
As starting point for an estimation it is assumed that the profit margin for the production of copper tel-
luride cement corresponds to the company’s performance. Taking an arbitrary performance value of
10% - which reflects the general profit expectations of major enterprises – the value of the cement cor-
responds consequently to 10% of the proceeds from sale of copper telluride. Therefore 10 % of the
copper telluride price is taken to estimate the value of the cement.
The composition and the corresponding values in the by-products result in a split of 8% to the copper
telluride cementate and 92% to the silver (Tab. 5.84). The values contained were assessed according to
the three years average from 2004 through 2006 based on USGS (2007).

Tab. 5.84 Allocation factors for copper telluride and silver using economic criteria.
Reference flows Amount $ / kg Value Split

Copper Telluride, by-product kg 0.0489 7.57 0.370 8%
Silver, from copper production kg 1.0 4.48 4.48 92%
5.4.2 Data quality

The dataset bases to large parts on extrapolations and approximations. Data uncertainty is derived us-
ing the Pedigree matrix as described in Frischknecht et al. (2004). The uncertainty for the inventory
items related to the Kaldo process and the conversion are underestimated by the Pedigree approach,
therefore the standard deviation of the flows is set to 5 to reflect their speculative nature. However, the
dataset is sufficient to serve as upstream process of silver and telluride used in background data. The
dataset must not be used to make assertions regarding different refining processes.


Infra-
ture
Anode Slime anode slime, silver and

25% content of precous metals in slime,
from Copper  corresponds to 143.1 kg slime feed.
metals extraction No GLO tellurium containing, 5.00E+00 kg 1 1.09 (2,2,1,2,1,3,3)
refining primary copper production
Infrastructure  metals extraction Yes SE
refinery
1.73E-07 unit 1 3.01 (2,2,1,2,1,3,9)
10 m3 reaction vessel with 5 m3 reaction
electricity for mass. Leaching a 2500 kg slime batch at 100 Gürmen et al.,
leaching and  psi O2 and 115°C. 0.17 kWh/kg slime. 0.1 electricity production mix No UCTE
9.35E-01 kWh 2004; Davenport, 1 2.06 (4,2,1,2,5,3,2)
drying processing of anode slime, primary copper production kWh per dry product assumed for 2002
subsequent drying.
oxygen  chemicals inorganics No RER oxygen, liquid, at plant 4.63E-02 kg Aylmore, 2001 1 2.06 (4,2,1,2,5,3,3)
Light fuel oil, 52.9 l oil per t charge, 45% lead content = 4
burned
 MJ per kg lead bullion
oil heating systems No RER industrial furnace 1MW, non- 1.89E-01 MJ Peterson 1977 1 5.00 uncertainty estimated
modulating
Oxygen, liquid, 123.5 kg O2 per t charge, 45% lead content,
at plant
 1.34 kg/m3 O2 = 0..369 per kg lead bullion
chemicals inorganics No RER oxygen, liquid, at plant 1.61E-02 kg Peterson 1977 1 5.00 uncertainty estimated

Möbius (Silver)
6.00E-01 kWh Kennethcott, 2004 1 2.10 (4,2,1,2,5,5,2)
Copper A RER specific mix of secondary copper,

Choppings
 primary pyro and SXEW copper
metals extraction No RER copper, at regional storage 2.45E-02 kg Kennethcott, 2004 1 1.24 (3,2,1,1,1,5,3)
low population Questionaire, 1999,
 PM >10 air
density
CEPMEIP
specific for Kaldo, no emissions from
low population Particulates, > 2.5 um, and < Questionaire, 1999,
 PM 2.5-10 electrolysis. Particle distribution from air
density 10um
2.31E-05 kg
CEPMEIP
CEPMEIP Database for copper dust.
low population Questionaire, 1999,
 PM <2.5 air
density
Particulates, < 2.5 um 3.46E-05 kg
CEPMEIP

 Copper (air)
air
density

 Lead (air)
air
density

 Copper (air)
air
density

 Lead (air)
air
density

 Zinc (air)
air
density
Zinc 4.31E-06 kg HELCOM 2002 1 5.00 uncertainty estimated

 Cadmium (air)
air
density
Cadmium 7.77E-08 kg Questionaire, 1999 1 5.00 uncertainty estimated

 Mercury (air)
air
density
Mercury 1.55E-07 kg HELCOM 2002 1 5.00 uncertainty estimated
Sulphur dioxide specific values for Kaldo oven, no emissions low population
air
density
Sulfur dioxide 3.50E-04 kg HELCOM 2002 1 5.00 uncertainty estimated
Nitrogen oxide specific values for Kaldo oven, no emissions low population
air
density

 Fluor (air)
air
density
Fluorine 1.32E-06 kg HELCOM 2002 1 5.00 uncertainty estimated

 Chlorid (air)
air
density
Chlorine 1.65E-04 kg HELCOM 2002 1 5.00 uncertainty estimated
specific values for Kaldo oven, no emissions low population Dioxins, measured as 2,3,7,8- Lindström, 2001,
 PCDD/F (air)
air
3.26E-09 kg
HELCOM, 2002
Silver, from
Main Product at 269 $ / kg, recieves 99.9% silver, from copper
 copper
of the burden
production, at refinery
1.00E+00 kg
production
Copper
Main Product at 8 $ / kg, recieves 0.1% of copper telluride cement,
 Telluride, by-
the burden
from copper production
4.89E-02 kg
product.
Fig. 5.22 Flows for "Silver, from copper production, at plant" and its representation in the ecoinvent database

5.5 Gold, at regional storage

A large amount of gold is obtained from secondary materials. Renner et al. (2002a) estimates that
around 50% of the world gold supply originates from secondary production. The world mine produc-
tion in 2005 has been 2’518 tonnes, the supply of gold from scrap was reported to be 0.9 tonnes, or
26% 10 . GFSM reports for 2006 an increased scrap supply (1’100 tonnes) while the mine production
dropped to 2.44 tonnes. (minus 3%), which corresponds to a secondary supply of 31%. In this inven-
tory a share of 30 % secondary production is assumed.
Today, small-scale mining might account for one-quarter of the world gold output 11 . However, these
processes were not inventoried in this study due to lack of information.
Fig. 5.23 World mine production (Data: GFMS).
A default distance of 250 km from the refinery to the airport and 250 km further to the regional dealer
is assumed. Transport distances are estimated on bee-line measures (see Tab. 5.85) and reflect the
mine production mix in Tab. 5.69. The resulting inventory is displayed in Fig. 5.25
5.6 Silver, at regional storage

This dataset represents the use of silver in Europe. The total mine production in 2005 amounted to
70% of the demand, whereas 21% of the silver was obtained from secondary sources (scrap), 7% from
governmental sales and 2% from hedging (Klapwijk et al., 2006).
For primary production, only silver production as by-product from lead / zinc, copper, and gold ex-
traction is available in the ecoinvent Database. These supply sources contribute for 76% of the total
primary supply which is regarded sufficient to reflect the production of silver on a global scale. The
shares of the available sources (lead/zinc, copper, gold) are scaled for the respective dataset to 100%
(see Tab. 2.3). No detailed statistics on silver import to Europe was available, the transports of the
primary silver are estimated on the supply statistics provided by the silver institute on a global average
10
http://www.gold.org/value/markets/supply_demand/recycled.html, accessed in july 2007.
11
Christian Beinhoff (Unido, 2005) http://www.unido.org/file-storage/download/?file_id=10644, accessed in may 2007.

(Tab. 5.85). A default distance of 250 km from the refinery to the airport and 250 km further to the re-
gional dealer is assumed. The resulting inventory is displayed in Fig. 5.24.
12 13
Tab. 5.85 Silver mine production in 2005, country share on total of 20’000 t and distance to Europe .
Country Mine production Share Distance, bee-line

Tonnes km
Peru 3206 16% 11000
Mexico 2884 14% 10000
Australia 2419 12% 18000
China 2022 10% 8000
Chile 1384 7% 12000
Russia 1319 7% 2000
Poland 1266 6% 250
United States 1225 6% 8000
Canada 1066 5% 8000
Kazakhstan 809 4% 5000
Bolivia 400 2% 10000
Indonesia 309 2% 11000
Sweden 284 1% 1300
Morocco 231 1% 1500
Argentina 163 1% 12000
Turkey 163 1% 1500
South Africa 88 0% 5000
Iran 81 0% 4000
Uzbekistan 69 0% 5000
India 66 0% 6000
12
http://www.silverinstitute.org/, accessed in may 2007
13
http://www.nucleus2000.de/, accessed in may 2007


Infra-
ture
Silver from Lead 31% share in world mine production, which silver, from lead production, Silver institute
/ Zinc
 in turn accounts for 70%
at refinery
3.21E-01 kg
2005
1 1.05 (1,1,1,1,1,1,3)
Silver from
silver, at regional 25% share in world mine production, which silver, from copper Silver institute
Copper
 in turn accounts for 70%
production, at refinery
2.59E-01 kg
2005
1 1.05 (1,1,1,1,1,1,3)
14% share in world mine production, which silver, from combined gold- Silver institute
Silver from Gold  metals extraction No GLO 1.45E-01 kg 1 1.05 (1,1,1,1,1,1,3)
storage
in turn accounts for 70% silver production, at refinery 2005
silver, secondary, at
Secondary Silver Accounts for 21% of the world supply metals extraction No SE 2.75E-01 kg Klapwijk 2006 1 1.05 (1,1,1,1,1,1,3)
Based on share of countries supplying
primary silver from Pb/Zn, Cu and Au transport
Plane airplane No RER transport, aircraft, freight 8.89E+00 tkm calculations 1 2.00 (2,1,1,1,1,2,5)
production, beeline. 50% additional weight systems
due to packaging.
2 x 250 km; 50% additional weight due to transport transport, lorry >16t, fleet
lorry road No RER 7.50E-01 tkm calculations 1 2.00 (2,1,1,1,1,2,5)
packaging. systems average
 Silver metals extraction No RER silver, at regional storage 1.00E+00 kg
Fig. 5.24 Flows for "Silver, at regional storage" and its representation in the ecoinvent database.

Infra-
ture
Primary gold,
production mix
 According to the producting countries metals extraction No GLO gold, primary, at refinery 6.89E-01 kg Estimation 1 1.22 (4,3,1,1,1,1,3)
Assumption: from within Europe (Boliden, gold, secondary, at precious

regional
gold, at

storage
Secondary gold metals extraction No SE 3.11E-01 kg Estimation 1 1.22 (4,3,1,1,1,1,3)

Umicore, Norddeutsche Affinerie, etc.) metal refinery
Based on share of supplying countries,
transport
Air freight beeline. 50% additional weight due to airplane No RER transport, aircraft, freight 1.43E+01 tkm calculations 1 2.00 (2,1,1,1,1,2,5)
systems
packaging.
2 x 250 km; 50% additional weight due to transport transport, lorry >16t, fleet
Lorry road No RER 7.50E-01 tkm calculations 1 2.00 (2,1,1,1,1,2,5)
packaging. systems average
 Gold metals extraction No RER gold, at regional storage 1.00E+00 kg
Fig. 5.25 Flows for "Gold, at regional storage" and its representation in the ecoinvent database.

6. Literature
6 Literature
Arana (2002) Arana M. A. (2002) Resolución de Conflictos Medioambientales en la Micro-
cuenca del Río Porcón, Cajamarca 1993-2002. Doctor. Pontífica Universidad
Católica del Perú, Lima.
Auerswald & Radcliffe (2005) Auerswald D. A. and Radcliffe P. H. (2005) Process technology development at
Rand Refinery. In: Minerals Engineering, 18(8), pp. 748-753, Online-Version
under: http://dx.doi.org/10.1016/j.mineng.2005.03.011.
Aylmore & Muir (2001) Aylmore M. G. and Muir D. M. (2001) Thiosulfate Leaching of Gold - A Re-
view. In: Minerals Engineering, 14(2), pp. 135-174, Online-Version under:
http://dx.doi.org/10.1016/S0892-6875(01)00057-7.
Barrick (2006a) Barrick (2006a) Barrick Gold Corporation: Porgera JV from http://www.barrick.
com/Default.aspx?SectionID=AE16ED96-78D3-4451-AB11-281B502746FB&
LanguageID=1&ProjectId=e84dc542-4437-4761-af82-35d7a603d457
Barrick (2006b) Barrick (2006b) Environment: Performance Tables from http://www.barrick.
com/Default.aspx?SectionID=8906c4bd-4ee4-4f15-bf1b-565e357c01e1&
LanguageId=1
Boliden (2000) Boliden (2000) New Rönnskär. Boliden A.B., Skelleftehamn.
Boliden (2002a) Boliden (2002a) Boliden - World Class Recycler. BOLIDEN Mineral AB, Skel-
leftehamn.
Boliden (2002b) Boliden (2002b) Gold Leaching. In: Smältdegeln 2, pp. 1-4.
Boliden (2004) Boliden (2004) Energy and emissions for PGM production at Rönnskär. New
Boliden.
Boliden (2006a) Boliden (2006a) Environmental Facts 2006 - statistics from 2005, Mines: Aitik,
Boliden, Garpenberg, Tara. Boliden Ltd, Online-Version under: http://www.
boliden.se/.
Boliden (2006b) Boliden (2006b) Environmental Health & Safety 2006 - Statistics from 2005:
Rönnskär. Boliden Ltd, Online-Version under: http://www.boliden.se/.
Boliden (2006c) Boliden (2006c) Rönnskär. Hälsa, Miliö & Säkerhet 2006. Statistik fran 2005.
Boliden (2006d) Boliden (2006d) Sustainability Report 2005. Boliden AB, Stockholm, Online-
Version under: http://www.boliden.se/.
Boliden Contech A.B. (1995) Boliden Contech A.B. (1995) Boliden Contech KGHM Polska Miedz S.A. Bo-
liden Contech A.B., Stockholm.
Butterman & Hilliard (2004) Butterman W. C. and Hilliard H. E. (2004) Silver. In: Mineral Commodity Pro-
file. U. S. Geological Survey, Online-Version under: http://minerals.usgs.gov/
minerals/pubs/commodity/.
Chen & Dutrizac (1990) Chen T. T. and Dutrizac J. E. (1990) Mineralogical Characterization of Anode
Slimes: Part 6 - Pressure Leached Slimes from the CCR Division of Noranda
Minerals INC. In: Canadian Metallurgical Qarterly, 29(4), pp. 293-305.
Corrosion Technology International (2006) Corrosion Technology International (2006) UNICELL® Feautures
and References Worldwide. Corrosion Technology International. Retrieved from
http://www.ctiunicell.com/
Online-Version under:

6. Literature
http://www.engineeringvillage2.org/controller/servlet/OpenURL?genre=book&i
sbn=0080440290.
Ekvall & Finnveden (2001) Ekvall T. and Finnveden G. (2001) Allocation in ISO 14041- A Critical Review.
In: Journal of Cleaner Production, 9(3), pp. 197-208.
Evans et al. (1993) Evans P. G., Hancock P. J., Harris C. L. and Harris R. L. (1993) One step proc-
ess for the treatment of Parkes desilvering crust to recover zinc and produce a
suitable feed for cupellation. Patent Number U. S. 5232486 held by Brunswick
Mining and Smelting Corporation Limited (Bathurst, CA)
Gasparrini (1993) Gasparrini C. (1993) Gold and other precious metals. From ore to market.
Springer-Verlag, Berlin Heidelberg New York. ISBN 3-540-54976-5.
GENI (2004) GENI (2004) Canada: grid summary. Global Energy Network Institute. Re-
trieved from
http://www.geni.org/globalenergy/library/national_energy_grid/canada/
index.shtml
Geoffroy & Cardarelli (2005) Geoffroy N. and Cardarelli F. (2005) A Method for Leaching or Dissolving Gold
from Ores or Precious Metal Scrap. In: JOM, 57(8),.
George (2006) George M. W. (2006) Gold. U.S. Geogolical Survey, Washington, D.C., Online-
Version under: http://minerals.usgs.gov/minerals/pubs/commodity/gold/
gold_myb04.pdf.
Goosey & Kellner (2002) Goosey M. and Kellner R. (2002) A Scoping Study - End-of-Life Printed Circuit
Boards. Department of Trade and Industry, Yorkshire.
Guinée et al. (2001) Guinée J. B., Gorrée M., Heijungs R., Huppes G., Kleijn R., de Koning A., van
Oers L., Wegener Sleesewijk A., Suh S., Udo de Haes H. A., de Brujin H., van
Duin R., Huijbregts M. A. J., Lindeijer E. W., Roorda A. A. H., van der Ven B.
L. and Weidema B. P. (2001) Life Cycle Assessment; An Operational Guide to
the ISO Standards; Final Report, May 2001. Ministry of Housing, Spatial Plan-
ning and Environment (VROM) and Centrum voor Milieukunde (CML),
Rijksuniversiteit, Den Haag and Leiden, Online-Version under:
http://www.leidenuniv.nl/cml/lca2/index.html.
HELCOM (2002) HELCOM (2002) Land-based Pollution Group - 6h Meeting. In: Matters related
to HELCOM, HELCOM HOD, other HELCOM Groups and HELCOM PITF.
Number. HELCOM, Vilnius.
Holliday & Goodman (2002) Holliday R. and Goodman P. (2002) Going for Gold. In: IEE Review(May),
Online-Version under: http://www.gold.org/pr_archive/.
Isaakson & Lehner (2000) Isaakson Ö. and Lehner T. (2000) The Rönnskär Smelter Project: Production,
Expansion and Start-up. In: Journal of Minerals, Metals & Material, 52(8), pp.
26-29.
JMB (2002) JMB (2002) Spezialausgabe: Silber - November 2002. In: JMB News. Johnson
Matthey & Brandenberger AG, Zürich, Online-Version under:
http://www.johnson-matthey.ch/download/2002JMBNewsSilver241002.pdf.

6. Literature
Kelly & Matos (2007) Kelly T. D. and Matos G. R. (2007) Historical Statistics for Mineral and Mate-
rial Commodities in the United States. In: Data Series 140. U. S. Geological
Survey, Online-Version under: http://minerals.usgs.gov/ds/2005/140/.
Kinoshita et al. (2003) Kinoshita T., Akita S., Kobayashi N., Nii S., Kawaizumi F. and Takahashi K.
(2003) Metal recovery from non-mounted printed wiring boards via hydrometal-
lurgical processing. In: Hydrometallurgy, 69, pp. 73-79.
Ludvigsson & Larsson (2003) Ludvigsson B. M. and Larsson S. R. (2003) Anode Slimes Treatment: The Bo-
liden Experience. In: Journal of Minerals, Metals & Material, 55(4), pp. 41-44.
Mark & Lehner (2002) Mark F. E. and Lehner T. (2002) Plastics Recovery from Waste Electrical &
Electronic Equipment in Non-Ferrous Metal Processes. Association of Plastics
Manufacturers in Europe (APME), Brussels.
Motorola (2005) Motorola (2005) Material Assay of the GSM 900 Handy PWB and GSM 900
BTS rack PWB. Rapid Environmental Assessment Lab, Taunusstein.
Nationmaster (2004) Nationmaster (2004) Electricity production by source. Nationmaster. Retrieved
from http://www.nationmaster.com/country/sf/Energy
Newmont (2004) Newmont (2004) How gold is mined. Newmont. Retrieved from
http://www.newmont.com/en/gold/howmined/index.asp
Newmont (2005a) Newmont (2005a) Minera Yanacocha, Peru- Now & Beyond 2005: Sustainabil-
ity Report. Newmont. Retrieved from
http://www.newmont.com/en/social/reporting/
index.asp
Newmont (2005b) Newmont (2005b) Now & Beyond: Sustainability Reports. Newmont Mining
Corporation. Retrieved from http://www.newmont.com/en/social/reporting/
index.asp
Outokumpu (2004) Outokumpu (2004) Outokumpu: Anode casting technology. Outokumpu Tech-
nology Oyj, Espoo.
Outokumpu (2006) Outokumpu (2006) Outokumpu Technology. Outokumpu Technology Oyi. Re-
trieved from http://www.outokumputechnology.com/pages/Page____6907.aspx
Pesl & Anzinger (2002) Pesl J. and Anzinger A. (2002) Treatment of Anode Slimes. Mineralogical and
Process Distinctions of Secondary Copper Smelters. In: Erzmetall, 55(5/6), pp.
305316.
Peterson & Lundquist (1977) Peterson S. A. and Lundquist S. A. (1977) Smelting and Reduction of oxidic and
sulphated Lead Material. Patent Number Sweden 4,017,308, held by Boliden
Aktiebolag
Placer Dome Inc. (2000) Placer Dome Inc. (2000) Misima Mines - Sustainability Report. Place Dome Inc,
Papua New Guinea.
Placer Dome Inc. (2001) Placer Dome Inc. (2001) Misima Mine - Sustainability Report 2000. Place Dome
Inc, Papua New Guinea.
Placer Dome Inc. (2002a) Placer Dome Inc. (2002a) Misima Mines - Sustainability Report. Place Dome
Placer Dome Inc. (2002b) Placer Dome Inc. (2002b) Sustainability - environmental charts 2002. Placer
Dome Inc. Retrieved from
http://www.placerdome.com/sustainability/env_charts/
greendata02.html

6. Literature
Placer Dome Inc. (2003a) Placer Dome Inc. (2003a) Annual report 2002. Placer Dome Inc., Vancouver.
Placer Dome Inc. (2003b) Placer Dome Inc. (2003b) Porgera Mine - Sustainability Report. Place Dome
Placer Dome Inc. (2004) Placer Dome Inc. (2004) Sustainability - environmental charts 2001-2003. Placer
Dome Inc. Retrieved from http://www.placerdome.com/sustainability/
environment/reports/index.html
Placer Dome Inc. (2005) Placer Dome Inc. (2005) Sustainability - environmental charts 2004. Placer
Dome Inc. Retrieved from http://www.placerdome.com/sustainability/
performance/environment.htm
Renner et al. (2002a) Renner H., Schlamp G., Hollmann D., Lüschow H. M., Rothaut J., Knödler A.,
Hecht C., Schlott M., Drieselmann R., Peter C. and Schiele R. (2002a) Gold,
Gold Alloys, and Gold Compounds. In: Ullmann's Encyclopedia of Industrial
499.
Renner et al. (2002b) Renner H., Schlamp G., Hollmann D., Lüschow H. M., Rothaut J., Knödler A.,
Hecht C., Schlott M., Drieselmann R., Peter C. and Schiele R. (2002b) Silver,
Silver Compounds, and Silver Alloys. In: Ullmann's Encyclopedia of Industrial
107.
Rentz et al. (1999) Rentz O., Hähre S. and Schultmann F. (1999) Report on Best Available Tech-
niques (BAT) in German Zinc and Lead Production. French-German Institute for
Environmental Research (DFIU), University of Karlsruhe (TH), Karlsruhe.
Reuter (2005) Reuter M. A. (2005) The metrics of material and metal ecology harmonizing the
resource, technology and environmental cycles. Elsevier, Amsterdam. ISBN 0-
444-51137-7.
Shibasaki et al. (1992) Shibasaki T., Abe K. and Takeuchi H. (1992) Recovery of Tellurium from De-
copperizing Leach Solution of Copper Refinery Slimes by a Fixed-Bed Reactor.
In: Hydrometallurgy, 29(1-3), pp. 399-412.
Stewart (1999) Stewart M. (1999) Environmental Life Cycle Considerations for Design-Related
Decision Making in Minerals Processing. PhD Thesis. University of Cape Town,
Cape Town.
Syed (2006) Syed S. (2006) A green technology for recovery of gold from non-metallic sec-
ondary sources In: Hydrometallurgy, 82, pp. 48-53.
Umweltbundesamt (2001) Umweltbundesamt (2001) Data and Information with respect to the Primary
Non-ferrous Metal Industry (Zinc, Copper, Lead and Nickel Works). Umwelt-
bundesamt, Berlin.
Western Areas Limited (2006) Western Areas Limited (2006) Annual report 2005. Online-Version under:
http://www.westernareas.co.za/financials/financial_reports/annual_reports/annua
l_report_2005/print/print.htm.
Zhang et al. (2000) Zhang S., Forssberg E., Houwligen J. v., Rem P. and Wei L.-Y. (2000) End-of-
life electric and electronic equipment management towards the 21st century. In:
Waste Management & Research, 18, pp. 73-85, Online-Version under:

6. Literature
http://www.blackwell-synergy.com/links/doi/10.1034/j.1399-
3070.2000.00090.x/abs/.

Part X
Semiconductors:
Tellurium, Gallium and Cadmium
Compounds
Data v2.1 (2009)
Authors: Matthias Tuchschmid, esu-services

Mischa Classen, Empa Dübendorf
Review Mischa Classen, Empa Dübendorf
This version corresponds to version 2.0 with changes of the errors detected in the meantime.
The cadmium emissions to air were too high due to a mistake regarding the filtration of airborne emissions in the
cadmium purification process. The correct values are given in Tab. 1.
Tab. 1 Corrected values of cadmium emissions ( to air/high population density) of the “cadmium, semiconductor
grade” datasets
Dataset ID Name Value in Correct

ecoinvent value Unit
Data v2.01
DS-ID cadmium, semiconductor-grade, at plant, US, [kg]; 1.62E-04 6.0E-06 Kg/kg
6902
DS-ID cadmium chloride, semiconductor-grade, at plant, 4.4572E-5 3.876E-06 Kg(kg
6905 US, [kg]
DS-ID cadmium telluride, semiconductor-grade, at plant, 1.5E-5 6.0E-06 Kg/kg
6906 US, [kg]
The waterborne cadmium emissions in the dataset “cadmium sulphide, semiconductor-grade, at plant,
US, [k]”are mismatched as “from nature”. However, they are emissions “to nature”.

Table of Contents
Table of Contents
1 INTRODUCTION ....................................................................................................... 4
2 TELLURIUM ............................................................................................................ 5
2.1 Introduction ........................................................................................................................... 5
2.2 Reserves and resources of tellurium...................................................................................... 5
2.3 Characterisation of tellurium................................................................................................. 5
2.4 Use and application of material............................................................................................. 6
2.5 Production process ................................................................................................................ 7
2.6 Life cycle inventory .............................................................................................................. 9
3 GALLIUM, SEMICONDUCTOR-GRADE ....................................................................... 13

3.1 Introduction ......................................................................................................................... 13
3.2 Resources and production of gallium.................................................................................. 13
3.3 Characterisation of gallium ................................................................................................. 13
3.4 Use and application of material........................................................................................... 14
3.5 Production process .............................................................................................................. 14
3.5.1 Extraction from aluminium oxide liquor............................................................................. 14
3.5.2 Purification.......................................................................................................................... 16
3.6 Life cycle inventories .......................................................................................................... 16
3.6.1 Gallium, in Bayer liquor from aluminium production, at plant........................................... 16
3.6.2 Gallium, semiconductor-grade ............................................................................................ 16
3.6.3 Gallium, at regional storage ................................................................................................ 18
3.6.4 Data Quality ........................................................................................................................ 19
4 CADMIUM PURIFICATION AND CADMIUM COMPOUNDS................................................ 23

4.1 Introduction and Overview.................................................................................................. 23
4.2 Characterisation of cadmium compounds ........................................................................... 23
4.3 Production and Use of cadmium compounds...................................................................... 24
4.3.1 Cadmium, semi-conductor grade ........................................................................................ 24
4.3.2 Cadmium telluride (CdTe) .................................................................................................. 24
4.3.3 Cadmium chloride (CdCl2).................................................................................................. 25
4.3.4 Cadmium sulphide (CdS) .................................................................................................... 25
4.4 Life cycle inventories .......................................................................................................... 25
4.4.1 Cadmium, semi-conductor grade ........................................................................................ 25
4.4.2 Cadmium telluride, semi-conductor grade .......................................................................... 26
4.4.3 Cadmium chloride, semi-conductor grade .......................................................................... 28
4.4.4 Cadmium sulphide, semi-conductor grade .......................................................................... 29
5 LITERATURE ......................................................................................................... 32

1. Introduction
1 Introduction
Several metals feature semi conductive behaviour. For this they have to be purified to a high degree and ar-
ranged in layers onto a carrier. This part describes the extraction and purification of some of the most important
semiconductor materials:
 Tellurium raw material from copper production (part III)
 Gallium, semiconductor grade raw material from aluminium production (part I)
 Cadmium, semiconductor grade raw material of lead-zinc production (part VIII)
 Cadmium Telluride, semiconductor grade raw material of lead-zinc production (part VIII)
 Cadmium Chloride, semiconductor grade raw material of lead-zinc production (part VIII)
 Cadmium Sulphide, semiconductor grade raw material of lead-zinc production (part VIII)
The raw materials for these metals often are by-products of base metal production. For information on these,
please refer to the respective part within this report.

2. Tellurium
2 Tellurium
Acknowledgement
We thank Vasilis Fthenakis and Olliver Krahl for the information provided in a personal communica-
tion.
2.1 Introduction
Tellurium (CAS-Number: 13494-80-9) is a metallic element that is often extracted as a by-product of
processing copper, lead, gold, and bismuth ores. Tellurium has 8 natural stable isotopes and some 21
additional artificial unstable isotopes are known (Knockaert (2002)). A Hungarian scientist discovered
tellurium in 1782. For a long time, only minor amounts were extracted. One reason for the increasing
demand of tellurium is the application in the new field of thin film solar cells. Other common applica-
tions are thermoelectric devices and as an alloy for the production of free-machining steel.
2.2 Reserves and resources of tellurium

The main source of tellurium is the copper anode slime, which is a by-product of the electro winning-
process from copper-treatment (Fthenakis (2004)).
The principal source of tellurium is anode sludge, produced during the electrolytic refining of blister
copper. Tellurium is produced mainly in the United States, Canada, Peru and Japan. The recovery of
tellurium from secondary materials accounts for 10-20% of world tellurium production. As copper
ores are the main sources for tellurium, public statistics are based on the copper industry. The follow-
ing numbers are from the former “United States Bureau of Mines“ and are an estimation about the re-
serves, based on a fixed recovery factor of 0.065 kg Te per tonne of copper (Knockaert (2002)).
According to Fthenakis (2000) a recycling of the metals from the thin-film solar cell can be done, but
is not accounted here.
Tab. 2.1 Reserves and resources of tellurium (t), Source: Guilinger (1999d))
North America 11’000 United States 6’000

Canada 2’000
Others 3’000
South America 9’000 Chile 6’000
Peru 2’000
Others 1’000
Europe 5’000
Africa 5’000
Asia 2’000
Oceania 2’000
World 34’000
2.3 Characterisation of tellurium

Tellurium is a relatively rare element, in the same chemical family as oxygen, sulphur, and selenium.
Tellurium has a metallic glance in its pure state, and is relatively soft (Knockaert (2002)). For a long
time, it was more a waste-product of metallurgy, than a raw material for industrial purposes.

2. Tellurium
Tab. 2.2 Chemical Characterisation of tellurium, Source: Knockaert (2002)
Chemical Symbol Te
Atomic mass 127.60 g/mol
Density 6.24 g·cm−3
Melting point 449.5° C
Boiling point 988 °C
−1 −1
Heat capacity 25.73 J·mol ·K (25 °C)
Mohs hardness 2.25
2.4 Use and application of material

According to Georg (2004)) there are two main producers of refined tellurium. One is ASARCO from
the United States and the other is Umicore in Belgium. Guilinger (1999d) states a list of 29 producers
all around the world that fabricate three different categories of tellurium:
 tellurium dioxides – sludge and precipitates of variable purity around 35%
 commercial grade tellurium (99.7% purity)
 high purity tellurium (99.9% to 99.9999%)
The world production per year is in the range of 200 to 300 metric tons. Knockaert (2002) states a
broad trade of concentrate, blister and anode copper. This complicates a detailed estimation about the
global refinery production. The different market sectors are given in Georg (2004). Some further de-
tails are described below:
 Improvement of steel by addition of small quantities of tellurium (up to 0.1%), It helps to con-
trol the chill depth of rapidly cooled surfaces and produces hard, wear-resistant top layers.
 Tellurium is used in combination with other components in many catalysts (chemical reac-
tions: oxidation, ammonisation, (de-) hydrogenation, (de-) halogenations, etc.)
 In electronics, tellurium is a well-known additive in selenium-photoreceptors. Tellurium in-
creases the sensitivity and broadens the spectral response of the photoreceptor.
 Polycrystalline cadmium telluride is used in photovoltaic solar cells; CdTe has a direct band
gap of 1.5eV and a high optical absorption coefficient.
Tab. 2.3 Market share of tellurium (data for United States), Source: Georg (2004)
Alloying element to improve the properties or iron and steel 40%

Catalyst and chemicals 24%
Photoreceptors and thermoelectric devices 22%
Additive to non-ferrous-metals 8%
Digital video discs and other minor uses (incl. thin-film solar cells) 6%
Total 100%
The price for tellurium has increased rapidly in the last 2 years, mainly due to the increased demands
from China and solar cell manufacturers worldwide. In November 2004 the price was about US$ 50
per kilogram (Georg (2004). In 2005 the price ranged from US$ 110 to US$ 150 per kg tellurium
(Metal Bulletin (2006), respectively US$ 100 to 140 US$ according to PPM Pure metals. 1 Knockaert
1
Personal communication: Olliver Krahl, PPM Pure Metals GmbH, Germany (24th of January 2007)

2. Tellurium
(2002) reports that tellurium 99.99% (4N) is normally 40-50% more expensive than tellurium 99.9%
(3N) and tellurium 99.999% (5N) is more than double price of the standard-grad tellurium (3N).
2.5 Production process

This part is mainly based on the information of Knockaert (2002) and Fthenakis (2004). After py-
rometallurgic separations and electrolytic refining, the copper can be collected on the cathode with a
purity of 99.95%. The metallic impurities form the anode slime, which is the main source for the pro-
duction of tellurium (Fthenakis (2004). Further details and information about the copper process can
be found in the corresponding section of this report.
The slime contains tellurium, beside various metals like Bi, Sb, Se, Pb, Au, Ag, As or Ni (see Tab.
2.4.). Therefore, the separation process is usually done for all metals at one place. Fig. 2.1 shows an
example of a flow sheet for anode slime treatment.
Tab. 2.4 Composition of several copper anode slimes, Source: Knockaert (2002)
Canadian Copper refiners Nippon Mining Inco-Mine Geometric mean

Silver 21.3% 20.6% 6.0% 13.80%
Copper 20.3% 4.7% 25.0% 13.39%
Selenium 10.9% 15.2% 7.0% 10.51%
Lead 8.5% 6.5% 1.0% 3.82%
Tellurium 3.2% 3.6% 2.0% 2.85%
Arsenic 1.8% 1.6% 0.8% 1.33%
Nickel 0.5% 0.0% 26.0% 0.74%
Bismuth 0.4% 1.6% 0.1% 0.39%
Antimony 1.0% 1.0% 0.1% 0.36%
Gold 0.0% 1.0% 0.1% 0.05%
Total 67.9% 55.9% 68.1%
Residual 32.10% 44.10% 31.90%

2. Tellurium
Fig. 2.1 Example flow sheet for anode slime treatment, Source: IPPC (2001)
The tellurium content in the copper varies. The mentioned “United States Bureau of Mines” estimates
an average tellurium content of about 0.065 kg per ton of copper (Knockaert (2002). The tellurium
content in the sludge varies from 0.5% to 10%. Three copper-producers stated a typical tellurium con-
centration of 2%, 3.2% and 3.64%, with a geometrical mean of 2.85% (Tab. 2.4).
As the tellurium is separated from the slime as copper-telluride-cementate, the life cycle assessment
starts with the separation of tellurium from the cementate, the Pre-treatment is not assessed in this
dataset. The following processes are considered (see also Fig. 2.2):
(i) Dissolving of copper-cementate with sulphuric acid
(ii) Electrolysis of tellurium from the solution
(iii) Purification of cathode-tellurium with Zone melting

2. Tellurium
Fig. 2.2 Production process of tellurium
Dissolving cementate and electrolysis: The cementate of copper-telluride is treated with sulphuric
acid. The next step is the electrolysis of the solution to extract the tellurium at the cathode. Further de-
tails and chemical reactions can be found in Knockaert (2002).
Purification: According to Fthenakis (2004), high purity tellurium is produced by electrolytic purifi-
cation and subsequent melting and atomization or by vacuum-distillation. According to industrial
sources in the mentioned report, electrolytic purification does not produce any emissions and all waste
is recycled. On the other hand, it is mentioned that for the production of 0.6 kg tellurium (99.999% or
5N) 0.75 kg tellurium (99.99% or 4N) is required. 2 This represents a tellurium loss of 20% during this
phase of purification.
2.6 Life cycle inventory

No LCA-study on tellurium production is available 3 . This module is based on reported values from
Knockaert (2002) and (Fthenakis & Kim (2005); Fthenakis (2004)). The functional unit of this process
is 1 kilogram of tellurium with a purity of 99.999% (5N). This purity allows a direct use for the pro-
duction of solar cells and electronic devices. This dataset describes the production and purification of
tellurium in Europe, starting from the anode sludge from copper purification.
Transports: Knockaert (2002) reported a wide trade with concentrate, blister and anode copper, so an
average transport distance of 4000 km by ship was assumed for the anode slime with a 3% concentra-
tion of tellurium. The allocation for this transport was done by weight (see Tab. 2.4). The transport to
the plant was assumed with standard transport distances.
Demand in materials and energy: The input feed is copper-telluride cementate, that is derived from
the primary copper production cycle (See Part “Copper” in the corresponding chapter). The telluride
2
Personal communication: Email from V. Fthenakis to N. Jungbluth, Nov. 2006
3
Steinberger (1997) analysed the life cycle of CdTe-solar cells, though the assumptions for the production and purification of
tellurium were to rough. Therefore no data from the mentioned thesis has been used.

2. Tellurium
content in the cementate is 50% (Shibasaki et al. 1992), an overall process yield of 90% is assumed.
The arising copper rich solution is recycled to the copper electrowinning cells and leaves the system
burden free.
During the process of leaching, for each tonne of copper-telluride 150 kg of H2SO4 (25%) is used
(Davenport et al. 2002). The demand of energy is reported by Fthenakis (2004). It is assumed, that the
energy carrier is electricity. Since there is no specific process-data available for the tellurium produc-
tion, the data from the corresponding chapters about the hydrometallurgical zinc-production was taken
instead for water-use. However, the uncertainty is fairly high.
Emissions: Fthenakis (2004) states that “emissions in hydrometallurgical / electrolytic plants are
likely to be negligible unless the sulphuric-acid tanks are open to the atmosphere.” For emissions,
proxy-data of zinc-production has been used: Assuming a production with 50% open tanks, a value of
9.55g sulphuric acid per kg tellurium was chosen (Xiao et al. (2003), cited in the chapter about zinc-
production)
Infrastructure: The infrastructure of a chemical plant, organics has been taken as an assumption with
the mentioned output of 50’000 tons per year and a lifespan of 50 years.
Tab. 2.5 Demands in Materials and Energy / Emissions, according to literature
Unit Minimal / Maximal Used Value in this

Value study
(geometrical mean)
Input
cadmium-telluride cement, from copper production kg 1 / 0.5 / 90% = 2.22 2.22
cadmium-telluride cement, from copper production, in-
kg 2.22/0.8 = 2.775 2.775
cluding a loss of 20% during purification
electrolysis of tellurium, Electricity
kWh 6.04 - 8.22 7.13
Personal communication V. Fthenakis (2006)
Purification of tellurium, Electricity 3.84
kWh 3.47 - 4.22
Personal communication V. Fthenakis (2006)
Water, Xiao et al. (2003), cited in the chapter for zinc pro- kg
16.2 16.2
duction
Emissions to air
Xiao et al. (2003), cited in the chapter for g
SO2 9.55 9.55
zinc production

2. Tellurium
Tab. 2.6 Unit process raw data for tellurium
StandardDeviation
InfrastructureProc
UncertaintyType
Location
tellurium,
95%
Unit
ess
Name semiconductor- GeneralComment
grade, at plant
Location GLO
Unit kg
product tellurium, semiconductor-grade, at plant GLO 0 kg 1.00E+0
technosphere copper-telluride cementate, from copper production GLO 0 kg 2.78E+0 1 1.17 (2,3,3,1,1,4); Literature (Shibasaki, 1992)
(4,3,1,3,3,5); Literature, personal communication

electricity, medium voltage, production UCTE, at grid UCTE 0 kWh 1.09E+1 1 1.38
(Fthenakis)
sulphuric acid, liquid, at plant RER 0 kg 3.75E-2 1 2.15 (4,5,2,3,5,5); Estimation
chemical plant, organics RER 1 unit 4.00E-10 1 3.81 (4,5,2,3,5,5); Estimation
(4,5,na,na,na,na); standard distances (100km),

transport, lorry >16t, fleet average RER 0 tkm 1.39E-1 1 2.09
allocation of copper-cementate by weight
(4,5,na,na,na,na); standard distances (200km),
transport, freight, rail RER 0 tkm 2.78E-1 1 2.09
(4,5,na,na,na,na); estimated distances (4000km),
transport, transoceanic freight ship OCE 0 tkm 5.55E+0 1 2.09
(2,4,3,3,4,5); Literature, Xiao in Althaus 2004
treatment, sewage, to wastewater treatment, class 5 CH 0 m3 1.62E-2 1 1.60
(extrapolation from zinc)
(2,4,3,1,4,5); Literature, Xiao in Althaus 2004
resource, in water Water, river - - m3 1.62E-2 1 3.30
(extrapolation from zinc)
emission air, low (2,4,3,1,4,5); Literature, Xiao in Althaus 2004
Sulfur dioxide - - kg 9.55E-3 1 1.60
population density (extrapolation from zinc)

2. Tellurium
Tab. 2.7 Ecospold Meta information of "tellurium, semi-conductor-grade, at plant, GLO”
ReferenceFunction Name tellurium, semiconductor-grade, at plant

Geography Location GLO
ReferenceFunction InfrastructureProcess 0
ReferenceFunction Unit kg
DataSetInformation Type 1
Version 2
energyValues 0
LanguageCode en
LocalLanguageCode de
DataEntryBy Person 44
QualityNetwork 1
ReferenceFunction DataSetRelatesToProduct 1
This datasets describes the production of high-
grade tellurium (99.999%) from copper-telluride
IncludedProcesses cementate by hydrometallurgical processes.
The energy consumption, emissions as well as
transports are considered.
Amount 1
LocalName Tellur, Halbleiter, ab Werk
Synonyms
This datasets describes the production of high-
grade tellurium (99.999%) from copper-telluride
cementate by hydrometallurgical processes.
These hydrometallurgical processes are the
GeneralComment dissolving of copper-telluride cementate with
sulphuric acid, the following electrolysis and
subsequent purification with zone-melting. This
purity of tellurium can be used in the production
of thin film solar cells.
InfrastructureIncluded 1
Category metals
SubCategory extraction
LocalCategory Metalle
LocalSubCategory Gewinnung
Formula Te
TimePeriod StartDate 1992
EndDate 2005
DataValidForEntirePeriod 1
OtherPeriodText based on literature and patents
Geography Text Production in GLO.
Production of tellurium from copper-telluride
Technology Text cementate (from copper production), purification
to 99.999% (5N)
Representativeness Percent
ProductionVolume 200 - 300 t in 2005
SamplingProcedure Analysis based on literature.
Some data available only for zinc are used as
Extrapolations
proxy
UncertaintyAdjustments none
DataGenerator Person 44
AndPublication DataPublishedIn 2
ReferenceToPublishedSource 10
Copyright 1
AccessRestrictedTo 0
CompanyCode
CountryCode
PageNumbers Tellurium
ProofReading Validator 58
Details automatic validation in Excel
OtherDetails none

3. Gallium, semiconductor-grade
3 Gallium, semiconductor-grade
3.1 Introduction
Gallium (CAS-Number: 7440-55-3) is a metallic element, similar in many characteristics to alumin-
ium. Gallium is an excellent conductor of heat and electricity and has a melting point of 29.8° C. It can
be extracted from bauxite ore by several different processes. Gallium is widely used in the field of op-
toelectronics (e.g. LEDs), telecommunication, aerospace and many commercial and household items
such as alloys, computers and DVDs (Moskalyk 2003).
3.2 Resources and production of gallium

Gallium is widely spread on Earth, but is rarely found in concentrations greater than 0.1%. Gallium-
containing minerals do not have an economic significance. Gallium can be extracted as a trace compo-
nent in bauxite, coal or several minerals. Gutiérrez et al. (1997) investigated the extraction of gallium
from coal fly ash from power plants. Occasionally, high concentrations of gallium are found in asso-
ciation with polymetallic deposits (e.g. copper, zinc, silver).
However, the dominant world production of gallium is as a by-product from the production of alumin-
ium oxide. Most gallium is recovered from the crude aluminium hydroxide (Guilinger 1999c). The
U.S. Geological Survey estimates the world gallium reserves in known reserves of bauxite (concentra-
tion of 50 ppm) and zinc ores of about 1 million-ton gallium. France is the largest producer of refined
gallium in the world (Moskalyk 2003).
Tab. 3.1 Production capacity and production in 2005, Source: Guilinger (1999c)
4
Production capacity Production 2005 Market Share
Country t t
Australia 50 21.3 34%
Germany 20 8.5 14%
Kazakhstan 20 8.5 14%
Russia 19 8.1 13%
China 8 3.4 5%
Hungary 8 3.4 5%
Slovakia 8 3.4 5%
Japan 7 3.0 5%
Ukraine 5 2.1 3%
United States 3 1.3 2%
Worldwide 148 63 100%
Company average 16 companies 3.9 t / a
3.3 Characterisation of gallium

Before its discovery, most of the properties of gallium have been predicted and described by Men-
deleyev on the basis of its position in his periodic table. The French Lecoq de Boisbaudran discovered
4
Estimation (based on the world production in 2005 of 63 tons of gallium, assuming that each enterprise has the same ra-
tio of production / production capacity)

gallium spectroscopically in 1875 (Latin Gallia meaning Gaul, essentially modern France) (Guilinger
1999c).
Tab. 3.2 Physical properties of gallium, Source: Greber (2005) and Moskalyk (2003)
Chemical Symbol Ga
Atomic mass 69.723 g/mol
Density 5.91 g·cm−3
Melting point 29.76°C
Boiling point 2204°C
Heat capacity 25.86 J·mol−1·K−1 = 370 J·kg−1·K−1 (25°C)
Mohs hardness 1.5
Heat of fusion 5.59 kJ/mol
3.4 Use and application of material

About 95% of the gallium consumed annually (57 t) is in the form of gallium arsenide (GaAs) com-
pounds, which are used in optoelectronic devices (e.g. LEDs = light emitting diodes, laser diodes,
photo detectors and solar cells) or in integrated circuits. The current price for 1 kg gallium of
99.9999% purity was at the end of the year 2005 about 512 US$ (Kramer 2005).
Tab. 3.3 Market share of gallium in United States, Source: Kramer (2005)
Gallium (kg) %
Optoelectronic devices
Laser diodes and light-emitting diodes 3880 20.6%
Photo detectors and solar cells 219 1.2%
Integrated circuits:
Analogue 11700 62.3%
Digital 115 0.6%
Research and development 2800 14.9%
Other 76 0.4%
Total 18700 100%
3.5 Production process

3.5.1 Extraction from aluminium oxide liquor
This part is mainly based on Greber (2005) and Kramer (2005). The most important process for gal-
lium production is the extraction of the metal from the Bayer liquor in aluminium oxide plants. The
liquor contains 70 – 150 mg / litre of gallium, depending on the bauxite and the concentration of the
caustic liquor. Selvi et al. (2004) reported a typical Bayer liquor as shown in Tab. 3.4.

Tab. 3.4. Composition of Bayer liquor, according to Selvi et al. (2004)
Sodium Hydroxide (NaOH) 220 g / litre

Aluminium 60 g / litre
Gallium 145 mg / litre
Vanadium 130 mg / litre
Viscosity 5.75 Cp (at 28°C)
3
Specific Gravity 1.35 g/cm
The gallium extraction plant is usually located within or very near to the aluminium oxide plants. Fig.
3.1. shows the gallium extraction within the chain of aluminium production.
Fig. 3.1 Flow diagram showing gallium extraction by Rhõne-Poulenc-Process, Source: Greber (2005)
Three types of processes are mainly used to extract the gallium. Below is a short summary of each
process. A description in more details can be found in Greber (2005) and Power Porto (1990).
(1) extraction with chelating agents (the Rhône-Poulenc-Process)
In 1974, it was discovered, that hydroxyquinolines could be used to extract gallium from so-
dium aluminates liquors. The aluminate liquor is brought into contact with a solution of
Kelex 100 5 in kerosene. Gallium is thereby extracted from the liquor along with aluminium
and sodium. Most of the aluminium and sodium is scrubbed from the extraction by a treatment
with dilute acid. The gallium itself is then stripped with concentrated sulphuric acid. The
metal can be isolated from this concentrate by direct electrolysis. Although the process solu-
tions are recycled, considerable quantities of waste acid is generated (Greber 2005). 35 to 40
tons of gallium is produced via this route.
(2) electrolytic processes (the Alusuisse-Process)
The electrolytic process depends on the fact that mercury forms an amalgam with gallium
5
Kelex100 is a well-known solvent extraction product, chemical name: 7-(4-ethyl-1-methyloctyl)quinolin-8-ol CAS-
Number: 29171-27-5

from which the gallium may be extracted with a caustic soda solution. The mercury is recy-
cled to produce again sodium amalgam. Hence the mercury circulates in a closed system.
This process requires special conditions to reach economically acceptable values (like > 150
g/l sodium and very low concentrations of vanadium, tungsten and molybdenum). The envi-
ronmental problems associated with mercury further discourage the use of electrolytic proc-
esses.
(3) fractional precipitation (the Péchiney-Process)
The gallium-containing sodium aluminate solution is treated with carbon dioxide to precipitate
its pure aluminium hydroxide. This increases the gallium content of the liquor. Further treat-
ment wit fractional precipitation is used to separate the gallium. Finally metallic gallium is ob-
tained by electrolysis. This process is labour and energy-intensive. China produces gallium by
this process.
Izumi (2004) describes some laboratory-scale processes about the re-extraction and recycling of gal-
lium from electronic scrap. However, no recycling processes are taken into account since there is no
data available.
Since the Rhône-Poulenc-Process dominates the production and is the least-cost effective, the focus of
the following life-cycle inventory was set on this process (Moskalyk 2003).
3.5.2 Purification
Different procedures can be used to remove the impurities such as mercury and zinc:
 vacuum distillation
 washing with aqueous acids and alkalis
 fractional distillation of liquid gallium compounds
 fractional crystallization
 single crystal growth
The purest gallium is obtained by fractional crystallization (Meijer et al. 2003) or single crystal growth
with a purity of 99.999999% (Power Porto 1990). The fractional crystallization process purifies a raw
gallium material utilizing segregation. Since solid gallium is lighter than liquid gallium, the solid gal-
lium floats in the form of a small mass on the surface of the liquid gallium. Seed crystals are present in
a melted raw gallium material and at surface. The melted raw gallium material is cooled with a gas to a
temperature less than its melting point. The gallium material is then solidified on the seed crystals as
nuclei with unidirect solidification (Maeda & Kadokura 1993). Bautista (2003) gives an overview
about the different purification methods.
3.6 Life cycle inventories

3.6.1 Gallium, in Bayer liquor from aluminium production, at plant
The production process of gallium starts from the liquid and crystallized Bayer-Liquor. Although there
is a production capacity of 160 t available, only 60 t are produced annually. This indicates a low eco-
nomic interest of gallium production at the moment. Therefore, no upstream processes are taken into
account. This is in line with the proposed methodology of (Guinée et al. 2004).
3.6.2 Gallium, semiconductor-grade

Production (Accumulation): The following data from Bautista (2003) is from a laboratory-scale in-
stallation. Selvi et al. (2004) reports that chelating agents can be reused more than 30 times in labora-
tory scale installations. In other settings (different types of chelating agents) more than 10 cycles have
been found. This study assumes, that the chelating agent (Kelex 100) can be used 30 times and other

organic solvents can be re-used for one year. Since the average production is about 3.9 t per company
and year, the calculation has been made with a recycling-rate of 3900 times at a production rate of
1 kg per cycle. It is assumed, that sulphuric acid is taken for stripping.
Krauss et al. (1999) reports for the “mixing and pumping of diluted leach, extracting agent and old
electrolyte” in the copper-extraction an energy consumption of 350 kWh per t. Therefore this value
has been taken as a proxy.
Electrolysis (Reduction): The resulting liquor after solvent extraction has an average gallium concen-
tration of 145 g / litre. From this solvent, the gallium can be directly extracted via electrolysis. Since
of missing data for electro winning, the related value for copper has been used: 2.22 kWh per kilogram
of Gallium (Dresher 2001).
Meijer et al. (2003) assumed a material efficiency of 90% for electrolysis and purification (including
partial reuse of waste), therefore an overall efficiency of 81% was assumed. However, the uncertainty
is relatively high.
Tab. 3.5 Key-figures of solvent extraction
Unit Amount Remark

Bayer Liquor
Gallium concentration g Ga/l 0.145 Selvi et al. (2004)
Process efficiency % 81% Meijer et al. (2003)
Bayer Liquor consumption l/kg Ga 8514
Solvents
Bayer Liquor to solvent ratio 1:1 Selvi et al. (2004)
Solvent consumption l/kg Ga 8514
Production
Average production per company kg/a 3900 Guilinger (1999c)
Reuse of solvents cycles 3900 Solvents are reused for one year (ex-
cept Kelex 100)
Reuse of Kelex 100 cycles 30 Selvi et al. (2004)
Average production per cycle kg Ga/Cycle 1 Assumption
Tab. 3.6 Consumption of solvent and acid, Source: Selvi et al. (2004)
Re-use per kg Gallium

Total of solvent 100 vol% 8514 litre - -
Kelex 100 10 vol% (851 litre) 30-times 28.4 litre
Versatic 10 5 vol% (426) litre 3900 times 0.11 litre
n-decanol 8 vol% (681) litre 3900 times 0.17 litre
Kerosene (as 77 vol% (6556) litre 3900 times 1.68 litre
diluent)
Acid: H2SO4 100 vol% (8514) litre 3900 times 2.18 litre
Transports: As the installations for gallium extraction are within the aluminium-plants, no transport
of the Bayer liquor has been accounted for. Since it is assumed, that the production takes place in
Europe, the transport of solvents and limestone is calculated with the standard distance from ecoinvent
(see Tab. 3.7).

Tab. 3.7 Standard distances for transports of solvents and acid
Type of transport Distance Source:

Railway 200 km in Europe (production) Frischknecht et al. (2004)
Lorry 100 km in Europe (production) Frischknecht et al. (2004)
Purification: Meijer et al. (2003) describes the process of fractional crystallization for producing
semiconductor grade gallium. The theoretical values for fractional crystallization are estimated with
the energy needed for heating the metal from room temperature to melting-temperature and the heat of
fusion. It is assumed that this process has an energy efficiency of 50%. No other purification methods
have been taken into account.
Tab. 3.8 Physical properties of gallium to calculate the process of purification
Heat of fusion per mol 5,59 kJ/mol

Heat of fusion per kg 80.2 kJ/kg
−1 −1
Heat Capacity per mol 25.86 J·mol ·K (25°C)
Weight 69.7 g/mol
−1 −1 −1 −1
Heat Capacity per mol 371 J·kg ·K = 0.371 kJ·kg ·K
Melting point 29.7° C
Room temperature 20° C
Calculation (9.7° C . 0.371 kJ·kg−1·K−1 + 80.2 kJ·kg−1) / 0.5 = 0.0465 kWh
Infrastructure: As there is no specific data about the gallium production available, proxy-data has
been used: For the processes of enrichment and extraction, the infrastructure of “chemical plant, or-
ganics” is considered, the electrolysis takes place in the plant of “aluminium electrolysis”.
Waste and Emissions: No specific data is available. Richardson (2002) reports a typical loss of
0.05% of feedstock on refinery storages. Therefore a yearly loss of 0.05% * 8520 l = 4.26 litre of or-
ganic solvent is emitted to the air. It is assumed, that 50% of this stream consists of NMVOC. The sul-
phuric acid is neutralized with calcium carbonate, each Mol of calcium carbonate neutralize 1 mol of
sulphuric acid. The worst-case scenario represents the fact, that carbonate and sulphate are emitted as
ions into the water. It is more likely, that they will be precipitated as salt.
It is assumed, that solvents as Kelex 100, Versatic 10, n-decanol or Kerosene, which have to be re-
placed, are sent to hazardous waste incineration.
3.6.3 Gallium, at regional storage

Since the production of gallium take place all around the world, the transport distance to regional stor-
age has been estimated on the base of Kazakhstan (Russia and Kazakhstan produce 27% of worldwide
gallium production) to Europe. Gallium is with an average price of 500US$ per kilogram a precious
metal, therefore the transport was considered with airplane over a distance of 4300km (Distance: Ka-
zachstan-Moskau-Germany)
Tab. 3.9 Standard distances for transports of solvents and acid
Type of transport Distance Source:

Lorry 300 km in Europe (to regional storage) Adapted from Frischknecht et al.
(2004)
Airplane 4300km from Kazakhstan to Germany Estimation

3.6.4 Data Quality

Data for the steps in the gallium production are roughly estimated. The values for the production and
purification steps stem from older experimental sources. The proxy chosen to represent the electrolysis
however is decreasing the overall quality.

Tab. 3.10 Unit process raw data for gallium, in Bayer liquor from aluminium production, at plant
gallium, in Bayer
UncertaintyTy
StandardDevi
Infrastructure
SubCategory
ation95%
Category
Location
liquor from
Unit
Name aluminium GeneralComment
pe
production, at
plant
Location GLO
Unit kg
product gallium, in Bayer liquor from aluminium GLO - - 0 kg 1
resource, in
gallium, 0.014% in bauxite, in ground - resource in ground - kg 1.00E+0 1 1.05 (1,1,1,1,1,1); data from literature (Selvi, 2004)
ground
Tab. 3.11 Unit process raw data for gallium, semiconductor-grade, at plant
StandardDeviation
InfrastructureProc
UncertaintyType
SubCategory
Category
gallium,
Location
95%
Unit
ess
Name semiconductor- GeneralComment
grade, at plant
Location GLO
Unit kg
product gallium, semiconductor-grade, at plant GLO - - 0 kg 1.00E+0
gallium, in Bayer liquor from aluminium
technosphere GLO - - 0 kg 1.23E+0 1 1.59 (2,4,2,1,4,5); Literature (Selvi, 2004)
production, at plant
electricity, medium voltage, production UCTE, at (4,4,2,2,4,5); Literature (Dresher 2001) &
UCTE - - 0 kWh 2.63E+0 1 1.64
grid (Meijer 2003)
sulphuric acid, liquid, at plant RER - - 0 kg 2.18E+0 1 1.84 (5,4,2,2,4,5); Literature (Bautista 2003)
(3,3,2,2,3,5); Calculation (for neutralization of
lime, hydrated, packed, at plant CH - - 0 kg 2.24E+0 1 1.33
sulphuric acid)
kerosene, at regional storage RER - - 0 kg 1.68E+0 1 1.84 (5,4,2,2,4,5); Literature (Bautista 2003)
(4,4,2,2,4,5); Literature (Bautista 2003), similar
fatty alcohol, petrochemical, at plant RER - - 0 kg 1.70E-1 1 1.64
to n-decanol
(5,4,2,2,4,5); Literature (Bautista 2003), sas a
EDTA, ethylenediaminetetraacetic acid, at plant RER - - 0 kg 2.84E+1 1 1.84
proxy value to Kelex 100
(5,4,2,2,4,5); Literature (Bautista 2003), as a
adipic acid, at plant RER - - 0 kg 1.10E-1 1 1.84
proxy value to Versatic 10
chemical plant, organics RER - - 1 unit 4.00E-10 1 3.76 (4,3,2,2,5,5); Estimation
aluminium electrolysis, plant RER - - 1 unit 7.69E-9 1 3.76 (4,3,2,2,5,5); Estimation
transport, freight, rail RER - - 0 tkm 6.96E+0 1 2.09 (4,5,na,na,na,na); standard distances (200km)
transport, lorry >16t, fleet average RER - - 0 tkm 3.48E+0 1 2.09 (4,5,na,na,na,na); standard distances (100km)
disposal, solvents mixture, 16.5% water, to
CH - - 0 kg 3.04E+1 1 1.33 (3,3,2,1,3,5); Calculation
hazardous waste incineration
emission air, high
Heat, waste - air high population density - MJ 9.46E+0 1 1.33 (3,3,2,1,3,5); Calculation
population density
NMVOC, non-methane volatile organic (3,3,2,1,3,5); Estimation, Literature (Richardson
- air high population density - kg 1.09E-3 1 1.33
compounds, unspecified origin 2002)
emission water, river Calcium, ion - water river - kg 1.12E+0 1 1.33 (3,3,2,1,3,5); Calculation
Sulfate - water river - kg 2.18E+0 1 1.33 (3,3,2,1,3,5); Calculation

Tab. 3.12 Unit process raw data for gallium, semiconductor-grade, at regional storage
StandardDeviation
InfrastructureProc
UncertaintyType
SubCategory
gallium,
Category
Location
semiconductor-
95%
Unit
ess
Name GeneralComment
grade, at regional
storage
Location RER
Unit kg
gallium, semiconductor-grade, at regional
product RER - - 0 kg 1.00E+0
storage
technosphere gallium, semiconductor-grade, at plant GLO - - 0 kg 1.00E+0 1 1.21 (1,1,2,2,1,5); Calculation
(4,5,na,na,na,na); estimated distances, based
transport, lorry >16t, fleet average RER - - 0 tkm 3.00E-1 1 2.09
on standard distances
transport, aircraft, freight RER - - 0 tkm 4.30E+0 1 2.28 (5,5,na,na,na,na); Estimation (4300km)

Tab. 3.13 EcoSpold Meta Information of gallium production
gallium, in Bayer liquor from gallium, semiconductor-grade, at gallium, semiconductor-grade, at

ReferenceFunction Name
aluminium production, at plant plant regional storage
Geography Location GLO GLO RER
ReferenceFunction InfrastructureProcess 0 0 0
ReferenceFunction Unit kg kg kg
DataSetInformation Type 1 1 1
Version 2.0 2.0 2
energyValues 0 0 0
LanguageCode en en en
LocalLanguageCode de de de
DataEntryBy Person 44 44 44
QualityNetwork 1 1 1
ReferenceFunction DataSetRelatesToProduct 1 1 1
This data set describes the
The module includes the production
simultaneous extraction of gallium The energy consumption and
from gallium as well as the transport
IncludedProcesses together with aluminium in the auxiliary materials are considered
of gallium from global production
Bayer process of the aluminium for, emissions are estimated.
places to europe.
production.
Amount 1 1 1
Gallium, in Bayer-Liquor während Gallium, Halbleiter, ab
LocalName Gallium, Halbleiter, ab Werk
Aluminiumherstellung, ab Werk Regionallager
Synonyms
This dataset is designed for
accounting for resource depletion in
the gallium production. As gallium is
Transport of gallium to european
assumed to be a by-product of the
Production of gallium from Bayer- regional storage, this module calls
GeneralComment aluminium production, no upstream
liquor, purification to 99.9999% (6N) the module "gallium, semiconductor-
burden is allocated to the gallium
grade, at plant".
solved in the bayer liquor. Do not
use this dataset for the production
of gallium.
InfrastructureIncluded 1 1 1
Category metals metals metals
SubCategory extraction extraction extraction
LocalCategory Metalle Metalle Metalle
LocalSubCategory Gewinnung Gewinnung Gewinnung
Formula Ga Ga Ga
CASNumber 7440-55-3 7440-55-3 7440-55-3
TimePeriod StartDate 2005 1999 2005
EndDate 2005 2005 2005
DataValidForEntirePeriod 1 1 1
OtherPeriodText patents and literatur
Geography Text Data from Literature Production in GLO. RER
The module includes the production
dissolution of gallium during the and purification of high grade
Technology Text Bayer-Process for aluminium gallium (99.9999%) by Transport
production hydrometallurgical processes
(Rhõne-Poulenc-Process).
Representativeness Percent 63.00
ProductionVolume unkown 60 t in 2005 60 t in 2005
SamplingProcedure estimation Analysis based on literature. Analysis based on literature.
For the electrolysis, the proxy-value
of copper is used, the infrastructure
Extrapolations none none
is estimated with related chemical
plants
DataGenerator Person 44 44 44
AndPublication DataPublishedIn 2 2 2
ReferenceToPublishedSource 10 10 10
Copyright 1 1 1
AccessRestrictedTo 0 0 0
CompanyCode
CountryCode
PageNumbers Gallium Gallium Gallium
ProofReading Validator 58 58 58
Details automatic validation in Excel automatic validation in Excel automatic validation in Excel
OtherDetails none none none

4. Cadmium purification and cadmium compounds
4 Cadmium purification and cadmium compounds
Acknowledgement
We thank Vasilis Fthenakis for the information provided in a personal communication.
4.1 Introduction and Overview

In thin-film solar cells, cadmium compounds like cadmium telluride, cadmium sulphide or cadmium
chloride are used for different tasks (photovoltaic junction, buffer-layer or back contact). For a more
in-depth explanation about the use of cadmium compounds and the physical principles in a thin-film
solar cell, see Karpov et al. (2005). Cadmium compounds are also used in other fields, mainly for elec-
tronical devices, detectors and pigments for painting.
Fig. 4.1 shows an overview of the inventoried cadmium compounds for solar industry, including the
purification of cadmium as an intermediate step. All compounds are of high-purity (99.99 % to
99.9999 %), according to the requirements of the thin-film-solar cell manufacturers.
Fig. 4.1 Overview of the inventoried cadmium compounds
The cadmium compounds consist of the elementary materials of cadmium, tellurium, sulphur and
chlor. Therefore no reserves or resources are mentioned here, estimations about the resources can be
found in the corresponding chapters.
4.2 Characterisation of cadmium compounds

The appearance of cadmium compounds is in the range of dark black, grey-metallic up to yellow-
orange, depending on the bounded element to the cadmium. All cadmium compounds are toxic.

Tab. 4.1 Physical properties of Cadmium compounds, Source: ESPI Metals (2007) and Schulte-Schrepping & Pisca-
tor (2002)
Cadmium Cadmium Cadmium sul- Cadmium chlo-

Name
telluride phide ride
Molecular formula Cd CdTe CdS CdCl2

Black cubic crystal- Yellow-orange white solid hygro-
Appearance silvery grey metallic
line powder powder scopic powder
Melting Point 321 °C 1041 °C 1750°C (100 bar) 564°C
Boiling Point 767 °C 1130 °C 980°C 960°C
3 3 3 3
Specific Gravity 8.65 g/cm 6.2 g/cm 4.82 g/cm 4.08 g/cm
CAS-Number 744-43-9 1306-25-8 1306-23-6 10108-64-2
Molecular Weight 112.4 g/mol 240 g/mol 144.5 g/mol 183.3 g/mol
Chemical Nature metallic inorganic inorganic inorganic
Solubility in Water insoluble insoluble insoluble in hydrochloric acid
-1 -1 -1 -1
213 J kg K 210 J kg K - -
Specific heat capacity
(at 293 K) (at 293 K)
−1
Enthalpy 99.87 kJ·mol
4.3 Production and Use of cadmium compounds

4.3.1 Cadmium, semi-conductor grade
High purity cadmium metal is essential for the preparation of compound semiconductors such as
CdTe, HgCdTe, CdZnTe, CdS or CdSe. These semiconductor devices are primarily used as solar cells,
IR detectors, imaging devices, electro-optic modulators, fluorescence and thermoelectric coolers. The
production of semiconductor compounds requires high-purified (6N and above) cadmium metal (Ali et
al. 2004).
4.3.2 Cadmium telluride (CdTe)

CdTe is a basic component of solar cells and is used as a several micrometers thin coating (Guilinger
(1999b)). Another application is the use of CdTe for optical infrared devices (alloyed with mercury) or
a solid-state x-ray and gamma ray detector (alloyed with zinc).
Menezes (2001), Izumi (2004) and Fthenakis (2000) describe the recycling of CdTe from photo-
voltaic-waste. However, cadmium compounds are not recycled on an industrial-scale and are, there-
fore, not accounted for here.
Four companies in the United States, two in Europe and one in Japan produce CdTe currently. Their
monthly output varies between 45 kg to 270 kg (Guilinger 1999b). The shape of the produced CdTe is
either in form of fragments (<1.5mm) or powder (<0.5mm). The typical purity is 99.999% CdTe with
some ppm-parts of Copper, Iron, Lead, Bismuth, Nickel and Silicon.

4.3.3 Cadmium chloride (CdCl2)

Cadmium chloride is used as "Cadmium Yellow" (a brilliant yellow pigment) and for electroplating in
the industry. In the production of thin-film-solar cells, CdCl2 is used as an auxiliary material for elec-
troplating of the CdS-buffer-layer.
4.3.4 Cadmium sulphide (CdS)

CdS is used in thin film solar cells as a junction partner in CIS 6 and Cd/Te solar cells. The CdS layer
itself is not intended to absorb light but instead allows light to pass through to either the CIS or CdTe
substrate. Therefore, the CdS layer must be thin (30 to 100 nanometre) in order to allow sufficient
light to pass through. A purity of minimal 99.99% (4N) is required.
CdS is also used as pigment, that is valued for its good thermal stability in many polymers, for exam-
ple in engineering plastics. By adding selenium in its formula it is possible to obtain colours ranging
from a greenish yellow to red violet. Pigments do not necessarily need a purity as high as for solar
cells.
4.4 Life cycle inventories

4.4.1 Cadmium, semi-conductor grade
This section is mainly based on the report of Ali et al. (2004) , additional information was taken
fromFthenakis & Kim (2005); Fthenakis (2004). The high purity cadmium metal (99.9999% or 6N)
can be produced either by zone refining process techniques or vacuum distillation. Although both
techniques have their distinct purification profiles, vacuum distillation is more appropriate since the
yield is higher in less time (Ali et al. 2004).
Vacuum distillation reduces the atmospheric pressure above the cadmium metal in order to decrease
the boiling point. The following setting is described at laboratory-scale (Ali et al. 2004):
 The system consists of two cylindrical crucibles. The evaporator and the condenser are made
with high-density graphite (150 mm OD x 120 mm ID x 300 mm Ht / 200 mm Ht). The two
cylinders are enclosed in a reactor of stainless steel, its vacuum is 2.1*10-3 Torr.
 The condensation temperature in the inner wall of the collector is maintained at around 270°C,
i.e. below cadmium freezing point.
 The cadmium metal in the evaporator is heated up to a temperature of 450°C.
 On average, it takes 5 hours for the entire deposition of 2 kg cadmium. Three stages of purifi-
cation are necessary to obtain a purity of 99.9999%. The reported yield varies within 70% and
86%, with a geometrical mean of 77%.
Purification of cadmium: For an industrial production. A larger reactor is assumed for a continuous
production. Due to the better ability of control, the energy source is electricity. Fthenakis 7 reports an
energy consumption from 12’200 kJ/kg to 15’800 kJ/kg. For this project, the geometric mean of
13’880 kJ/kg was used.
Three stages of purification are necessary to obtain a purity of 99.9999%. Each stage has an average
yield of 77%. The total yield is 46% for a purification setting of 3 cycles. The loss of 54% is captured
6
CIS stands for Copper-Indium-Diselenid
7
Personal communication of V. Fthenakis to N. Jungbluth, November 2006

in HEPA-Filters 8 , as described in Fthenakis (2004). The efficiency of HEPA filters in collecting par-
ticulates is reported with 99.97%, the rest is emitted to the air. It is assumed, that the filtrate can be
used again for a new purification cycle.
Infrastructure: The infrastructure has been roughly assessed with the dataset of the organic chemical
plant.
Transports: The transport has been calculated with the standard distances of ecoinvent (see Tab. 4.2).
Tab. 4.2 Standard distances for transports
Typo of transport Distance Source:

Railway 200 km in USA (production) Frischknecht et al. (2004)
Lorry 100 km in USA (production) Frischknecht et al. (2004)
Emissions and Waste: As there are no other materials used in the processes, only the emissions of
cadmium into the air has been taken into account: (1-99.97 %) * 2 % * 1 kg Cd = 6*10-6 kg per kilo-
gram purified cadmium.
Tab. 4.3 Unit process raw data for cadmium, semi-conductor grade, semi-conductor grade
StandardDeviation
uncertaintyType
Infrastructure-
cadmium,
Location
Process
95%
Unit
Explanations Name semiconductor- GeneralComment

grade, at plant
Location US
Unit kg
(5,4,2,1,5,5); calculated based on
Technosphere cadmium, primary, at plant GLO 0 kg 1.00E+0 1 2.30
formula and loss of air emissions
(5,4,2,1,5,5); personal communication
electricity, medium voltage, at grid US 0 kWh 3.86E+0 1 2.30
from Fthenakis
(4,5,na,na,na,na); standard distances

in Europe and USA (each 200km)

(5,4,2,1,5,5); assumption, based on

chemical plant, organics RER 1 unit 4.00E-10 1 3.96
annual production of 50'000 t
air, high population (5,4,2,1,5,5); calculation based on
Heat, waste MJ 1.39E+1 1 2.30
density used energy
Cadmium kg 6.00E-6 1 6.11 (5,4,2,1,5,5); literature and estimation
Outputs cadmium, semiconductor-grade, at plant US 0 kg 1.00E+0
4.4.2 Cadmium telluride, semi-conductor grade

Pinheiro (2006) reports different forms of CdTe used in close spaced sublimation (CSS)-techniques:
single-crystal or polycrystalline wafers, powders, pellets or pieces, CdTe film deposited onto glass,
molybdenum substrate (CdTe source-plate), sintered CdTe powder. Therefore, various techniques are
used to produce CdTe in one or another form.
Guilinger (1999b) and Triboulet (2004) describes the crystal precipitation: High purity (5N to 6N)
cadmium and tellurium are synthesized by melting the elements at a temperature of 1092°C in quartz
tubing. The reaction is exothermic. The cadmium and tellurium must be of high purity because con-
8
HEPA stand for High Efficiency-Particulate Air

taminates can cause an explosion. CdTe can only be produced in small batches of around 1.3 kg, be-
cause of the volatility of the chemical reaction.
Harman (1985) reports the preparation of CdTe for substrates: “Elements are sealed in evacuated, car-
bon-coated quartz tubes. The capsule is placed in a single furnace which is initially in a nearly hori-
zontal position. A slight tilt in the furnace keeps the liquid and solid material at the small diameter end
of the tube. The temperature of the furnace is raised to its final temperature of about 1110°C, at the
conical tip end approximately 100°C lower. The temperature of the growth tube is held for about 16 to
20hrs, then the furnace is rotated about 90° to a vertical position. After the temperature is stabilized,
the tube is lowered at a rate of about 1.2mm/hr until the solution has completely solidified.”
Production and purification of cadmium telluride: Since no production details about the use of re-
sources are available, we assume the following production:
 Resources: The reaction is according “the stoichiometric quantities of high purity elements “
(Harman (1985)): 1 Mol Te (= 127.6 g) + 1 Mol Cd (112.4g) → 1 Mol CdTe (240 g)
According to Frischknecht et al. (2004), a yield of 95% was assumed.
 Energy: Fthenakis (see Footnote 7) report the needed energy for different processes and/or
producers: Minimal 37’400 kJ/kg produced CdTe have to be used, the maximal value is
53’600 kJ/kg. In this project, the geometrical mean of 44’770 kJ/kg is used for an average
production process. As the process needs a precise controlling, electricity is used as energy
source.
Infrastructure: The infrastructure has been roughly assessed with the dataset of the organic chemical
plant.
Transports: see Tab. 4.2 in Section 4.4.1
Emission and Waste: With a yield of 95% and the mentioned HEPA-filters, the emissions are:
(1-99.97 %) * 2 % * 1 kg CdTe=6*10-6 kg of CdTe. The emissions are accounted as “cadmium emis-
sion into air, high population density”.

Tab. 4.4 Unit process raw data for cadmium telluride, semi-conductor grade
StandardDeviation
uncertaintyType
Infrastructure-
cadmium
Location
Process
telluride,
95%
Unit
Explanations Name GeneralComment
semiconductor-
grade, at plant
Location US
Unit kg
(5,3,1,3,3,5); personal communication
electricity, medium voltage, at grid US 0 kWh 1.24E+1 1 1.63
(Fthenakis)

for US (200km)

for US (100km)
chemical plant, organics RER 1 unit 4.00E-10 1 3.82 (4,5,2,5,5,5); assumption

cadmium, semiconductor-grade, at plant US 0 kg 4.93E-1 1 2.34
formula
tellurium, semiconductor-grade, at plant GLO 0 kg 5.60E-1 1 2.34
formula
air, high population
Heat, waste MJ 7.58E-1 1 1.68 (5,5,2,3,3,5); estimation
density
Cadmium kg 6.00E-6 1 5.42 (5,5,2,3,3,5); estimation
cadmium telluride, semiconductor-grade, at plant US 0 kg 1.00E+0
4.4.3 Cadmium chloride, semi-conductor grade

Cadmium chloride can be produced either by reaction of molten cadmium and chlorine gas at 600°C
or by dissolving cadmium metal (or the oxide) in hydrochloric acid (Schulte-Schrepping & Piscator
2002). As data from literature is not available, the process and material consumption has been roughly
estimated.
Production of cadmium chloride: Anhydrous cadmium chloride can be prepared by the action of an-
hydrous chlorine or hydrogen chloride gas on heated cadmium metal. The following reaction has been
assumed:
1 Mol Cd(s) (112.4 g) + 2 Mol HCl(g) (36.46 g) → 1 Mol CdCl2(s)(183.16g) + 1 Mol H2(g) (2.01g)
or with an efficiency of 95% for synthesis, see Frischknecht et al. (2004):
0.645 kg Cd(s) + 0.419 kg HCl(g) → 1 kg CdCl2(s) + 0.011 kg H2(g)
As no further information about the energy consumption are available, an estimation based on the en-
thalpy has been made: The heat capacity of cadmium is 26.02 J·mol−1·K−., as the melting temperature
of cadmium is about 320°C, we consider a difference of temperature of 300°C for heating the cad-
mium. The heat of fusion of cadmium is 6’210 J·mol−1 , we estimate an overall efficiency of 50%.
Heat capacity: 300 K * 26.02 J·mol−1·K− = 7.8 kJ
Heat of Fusion: 6.2 kJ
Sum: 14 kJ = 0.014 MJ = 0.0039 kWh → 0.0078 kWh/mol (incl. efficiency of 50%)
As no information about purification is available, presumably the environmental impacts are underes-
timated.
Infrastructure: The infrastructure has been roughly assessed with the available dataset of the organic
chemical plant.
Emission and Waste: From the molten cadmium, some emissions are considered. We assume the
same efficiency of production process as for the purification of cadmium (see section 4.3.1) Therefore,

we assume that from 0.645 kg Cadmium 77% are converted into cadmium chloride, the rest is cap-
tured in HEPA-filters with an efficiency of 99.97%: (1-99.97 %) * 2% * 0.645 kg Cd = 3.876*10-6 kg
per kilogram cadmium chloride. The generated hydrogen is also emitted into the air.
Tab. 4.5 Unit process raw data for cadmium chloride
StandardD eviation
uncertaintyType
Infrastructure-
cadmium
Location
Process
chloride,
95%
Unit
Explanations Name GeneralComment
semiconductor-
grade, at plant
Location US
Unit kg
electricity, medium voltage, at grid US 0 kWh 4.26E-2 1 2.30 (5,4,2,1,5,5); assumption,


chemical plant, organics RER 1 unit 4.00E-10 1 3.81 (4,5,2,1,5,5); assumption

cadmium, semiconductor-grade, at plant US 0 kg 6.46E-1 1 2.30
formula
hydrochloric acid, 30% in H2O, at plant RER 0 kg 4.19E-1 1 2.30
formula
air, high population
Heat, waste MJ 1.53E-1 1 2.33 (5,5,2,1,5,5); estimation
density
Cadmium kg 3.88E-6 1 6.16 (5,5,2,1,5,5); estimation
Hydrogen kg 1.15E-2 1 2.55 (5,5,2,1,5,5); estimation

cadmium chloride, semiconductor-grade, at plant US 0 kg 1.00E+0
4.4.4 Cadmium sulphide, semi-conductor grade

This part is mainly based on the reports of Guilinger (1999a) and Faria & Chiola (1981). Guilinger
(1999a) reported that in 1999 Asarco in the United States was the only producer of semi-conductor
grade cadmium sulphide. It is assumed, that the production today also takes place in the US. Cadmium
sulphide is being produced commercially by two methods:
 Hydrometallurgical:
According to Guilinger (1999a), this aqueous technique of producing CdS involves a number
of handling steps and is a complex procedure. According to Faria & Chiola (1981), is the
normal procedure “dissolve a cadmium salt in water, preferably the chloride or sulphate, and
then bubble hydrogen sulfide through the solution to precipitate essentially all of the cadmium
in the form of cadmium sulfide.” The reported yield is quite high, anionic and cationic impuri-
ties co-precipitates with the cadmium sulfide. The following purification involves extensive
washing and leaching, until a high purity of the output product can be reached.
 Thermal:
This technique involves the fusing of cadmium and sulphide material. The production rate can
be high, but size distribution and purity of the final product can be a problem. Therefore re-
processing is necessary to achieve a consistent product. This process is generally used to pro-
duce lower grade CdS for use in pigments.
According to Faria & Chiola (1981), the purification of thermal cadmium sulphide can be
done with hydrogen sulfide: “luminescent grade cadmium sulphide is placed in a quartz tube
reactor and a flow of nitrogen in maintained while the temperature is raised to about 500°C.
Hydrogen sulfide is used to replace nitrogen and the temperature in the reactor is raise to
about 1000°C. The temperature is held for 0.5h and then cooled down to 500°C under the hy-

drogen sulfide flow. The hydrogen sulfide is shut off and nitrogen is introduced while cooling
the material to ambient temperature.”
Production of cadmium sulphide: Since no specific information about the production process is
available, this part is mainly based on assumptions. It is assumed that the reaction is according the
stoichiometric quantities: 1 Mol CdCl2 (= 183.3 g) + 1 Mol H2S (34.1) → 1 Mol CdS (144.5 g) + 1
Mol HCl (112.4 g) According to Frischknecht et al. (2004) , a yield of 95% was assumed. As no in-
formation about the energy consumption is available, the proxy-value of cadmium chloride (see sec-
tion 4.4.3) is used instead (14 kJ per kg produced cadmium sulphide). As no information about a puri-
fication are available, presumably the environmental impacts are underestimated.
Infrastructure: The infrastructure has been roughly assessed with the available dataset of the organic
chemical plant.
Emission and Waste: Guilinger (1999a) stated that “(The production) does generate a cadmium con-
taminated liquid waste stream that is environmentally difficult to handle and neutralize.” Therefore, a
cadmium stream into the water is assumed. As a first estimation, the related value from zinc produc-
tion has been chosen (2.60*10-6 kg cadmium per kg cadmium sulphide). However, this value is fairly
uncertain.
Tab. 4.6 Unit process raw data for cadmium sulphide
Kommentar [hja2931]: er-

StandardDeviation95%
UncertaintyType
setzen
SubCategory
cadmium
Category
Location
sulphide,
Unit
Name GeneralComment
semiconductor-
grade, at plant
Location US
Unit kg
cadmium sulphide,
product US - - 0 kg 1
semiconductor-grade, at plant
cadmium chloride,
technosphere US - - 0 kg 1.34E+0 1 2.30 (5,4,2,1,5,5); calculated based on formula
semiconductor-grade, at plant
hydrogen sulphide, H2S, at
RER - - 0 kg 2.48E-1 1 2.31 (5,4,2,5,5,5); calculated based on formula
plant
chemical plant, organics RER - - 1 unit 4.00E-10 1 3.82 (4,5,2,5,5,5); assumption
US - - 0 kWh 3.89E-3 1 1.93 (4,5,5,1,4,5); assumption
production US, at grid
(4,5,na,na,na,na); standard distances in US
transport, freight, rail RER - - 0 tkm 3.17E-1 1 2.09
(200km)
transport, lorry >16t, fleet (4,5,na,na,na,na); standard distances in US
RER - - 0 tkm 1.58E-1 1 2.09
average (100km)
water, unspecified Cadmium, ion - water unspecified - kg 2.60E-6 1 4.00 (5,5,2,1,5,5); estimation
emission water,
Chloride - water unspecified - kg 5.16E-1 1 4.00 (5,5,2,1,5,5); estimation
unspecified
emission air, low
Heat, waste - air low population dens - MJ 1.40E-2 1 2.33 (5,5,2,1,5,5); estimation
population density

Tab. 4.7 EcoSpold Meta Information of cadmium purification and cadmium compounds
cadmium telluride, cadmium chloride, cadmium sulphide,

cadmium, semiconductor-
ReferenceFunction 401 Name semiconductor-grade, at semiconductor-grade, at semiconductor-grade, at
grade, at plant
plant plant plant
Geography 662 Location US US US US
ReferenceFunction 493 InfrastructureProcess 0 0 0 0
ReferenceFunction 403 Unit kg kg kg kg
DataSetInformation 201 Type 1 1 1 1
202 Version 2.0 2.0 2.0 2.0
203 energyValues 0 0 0 0
205 LanguageCode en en en en
206 LocalLanguageCode de de de de
DataEntryBy 302 Person 44 44 44 44
304 QualityNetwork 1 1 1 1
ReferenceFunction 400 DataSetRelatesToProduct 1 1 1 1
The energy consumption The energy consumption
The energy consumption, The energy consumption,
and transports are and transports are
402 IncludedProcesses emissions and transports emissions and transports
considered for, emissions considered for, emissions
are estimated. are estimated.
are estimated. are estimated.
404 Amount 1 1 1 1
Cadmium, Halbleiter, ab Cadmiumtellurid, Halbleiter, Cadmiumchlorid, Halbleiter, Cadmiumsulfid, Halbleiter,
490 LocalName
Werk ab Werk ab Werk ab Werk
491 Synonyms
This datasets describes the This datasets describes the This datasets describes the This datasets describes the
production of high-grade production of high-grade production of high-grade production of high-grade
cadmium (99.9999%) from cadmium telluride (99.999%) cadmium chloride (99.999%) cadmium sulphide (99.99%)
metalurgical cadmium. This from cadmium and tellurium. from cadmium and from cadmium and sulfuric
492 GeneralComment
purity of cadmium can be This purity of cadmium hydrochloric acid. This purity acid. This purity of cadmium
used in the production of telluride can be used in the of cadmium chloride can be sulphide can be used in the
various cadmium production of thin film solar used in the production of production of thin film solar
compounds. cells. thin film solar cells. cells.
494 InfrastructureIncluded 1 1 1 1
495 Category metals metals metals metals
496 SubCategory extraction extraction extraction extraction
497 LocalCategory Metalle Metalle Metalle Metalle
498 LocalSubCategory Gewinnung Gewinnung Gewinnung Gewinnung
499 Formula Cd CdTe CdCl2 CdS
501 StatisticalClassification 0 0 0
502 CASNumber 000744-43-9 001306-25-8 010108-64-2 001306-23-6
TimePeriod 601 StartDate 2004 1985 2004 1981
602 EndDate 2004 2005 2005 1999
603 DataValidForEntirePeriod 1 1 1 1
611 OtherPeriodText patents patents
Geography 663 Text Production in GLO. Production in US Production in US Production in US
Production of cadmium
Production of high-purity Production of cadmium Production of cadmium
chloride by reaction of
Technology 692 Text cadmium by vacuum telluride from cadmium and sulphide by
cadmium with hydrogen
distillation tellurium hydrometallurgical process
chloride
Representativeness 722 Percent
724 ProductionVolume unknown unknown unknown unknown
literature and own literature and own literature and own literature and own
725 SamplingProcedure
estimations estimations estimations estimations
Energy consumption is
Emissions of cadmium into Cadmium emissions are
extrapolated from CdCl due
726 Extrapolations none the air are extrapolated from extrapolated from cadmium
to lack of data and similarity
cadmium purification purification.
of processes
727 UncertaintyAdjustments none none none none
DataGeneratorAndPublication 751 Person 44 44 44 44
756 DataPublishedIn 2 2 2 2
757 ReferenceToPublishedSource 10 10 10 10
758 Copyright 1 1 1 1
759 AccessRestrictedTo
760 CompanyCode
761 CountryCode
762 PageNumbers cadmium compounds cadmium compounds cadmium compounds cadmium compounds
ProofReading 5616 Validator 58 58 58 58
5615 Details automatic validation in Excel automatic validation in Excel automatic validation in Excel automatic validation in Excel

5. Literature
5 Literature
Ali et al. (2004) Ali S. T., Munirathnam N. R., Sudheer C., Reddy R. C. and Prakash T. L. (2004)
Purification of cadmium by vacuum distillation and its analysis. In: Materials
Letters, 58, pp. 1638-1641.
Bautista (2003) Bautista R. G. (2003) Processing to Obtain High-Purity Gallium. In: JOM, 55,
pp. 23-26 as pdf-File under: TY - JOUR, Online-Version under: http://
www.ingentaconnect.com/content/tms/jom/2003/00000055/00000003/art00006.
Online-Version under: http://www.engineeringvillage2.org/controller/servlet/
OpenURL?genre=book&isbn=0080440290.
Dresher (2001) Dresher W. H. (2001) Copper Applications in Mining & Metallurgy: How Hy-
drometallurgy and the SX/EW Process Made Copper the "Green" Metal. Copper
Development Association, Online-Version under: http://www.copper.org/inno-
vations/2001/08/hydrometallurgy.html.
ESPI Metals (2007) ESPI Metals (2007) Material Safety Data Sheet: Cadmium telluride from
http://www.espimetals.com/msds's/Cadmium%20Telluride.htm
Faria & Chiola (1981) Faria S. and Chiola V. (1981) Solar energy grade cadmium sulfide: Nr. 4 245
093. Patent Number Unites States of America held by
Fthenakis (2000) Fthenakis V. (2000) End-of-life management and recycling of PV modules. In:
Energy Policy, 28(14), pp. 1051-1058.
Fthenakis & Kim (2005) Fthenakis V. and Kim H. C. (2005) Energy use and greenhouse gas emissions in
the life cycle of CdTe photovoltaics. In proceedings from: Materials Research
Society Fall Meeting, Boston.
Fthenakis (2004) Fthenakis V. M. (2004) Life cycle impact analysis of cadmium in CdTePV pro-
duction. In: Renewable & Sustainable Energy Reviews, 8(4), pp. 303-334,
Online-Version under: http://dx.doi.org/10.1016/j.rser.2003.12.001.
Georg (2004) Georg M. W. (2004) U.S. Geological Survey Mineral Yearbook 2004 - Sele-
nium and Tellurium. USGS, Online-Version under:
http://minerals.usgs.gov/minerals/ pubs/commodity/selenium/selenmyb04.pdf.
Greber (2005) Greber J. F. (2005) Gallium and Gallium compounds. In: Ullmann's encyclope-
dia of industrial chemistry: Electronic release (ed. Arpe et al.). 6th Edition.
Wiley-VCH, Weinheim, Germany. Online-Version under:
http://www.mrw.interscience.wiley.com/ueic/index.html.
Guilinger (1999a) Guilinger J. (1999a) Assessment of critical thin film resources: Cadmium sulfide
(RAF-9-29609): Extension. World Industrial Minerals, Golden.
Guilinger (1999b) Guilinger J. (1999b) Assessment of critical thin film resources: Cadmium tellu-
ride (RAF-9-29609). World Industrial Minerals, Golden.
Guilinger (1999c) Guilinger J. (1999c) Assessment of critical thin film resources: Gallium (RAF-9-
29609). World Industrial Minerals, Golden.
Guilinger (1999d) Guilinger J. (1999d) Assessment of critical thin film resources: Tellurium (RAF-
9-29609). World Industrial Minerals, Golden.

5. Literature
Gutiérrez et al. (1997) Gutiérrez B., Pazos C. and Coca J. (1997) Recovery of Gallium From Coal Fly
Ash By a Dual Reactive Extraction Process. In: Waste Management & Research,
15, pp. 371-382.
Harman (1985) Harman T. (1985) Process for making mercury cadmium telluride. Patent Num-
ber U.S. 4642142, held by
European Commission. Retrieved from http://www.jrc.es/pub/english.cgi/0/
733169
Izumi (2004) Izumi S. (2004) Environmental safety issues for semiconductors (research on
scarce materials recycling). In: Thin Solid Films, 461, pp. 7-12.
Karpov et al. (2005) Karpov V. G., Shvydka D. and Roussillon Y. (2005) Physics of CdTe Photovol-
taics: from Front to Back. In proceedings from: MRS Spring Meeting 2005, San
Francisco, United States.
Knockaert (2002) Knockaert G. (2002) Tellurium. In: Ullmann's Encyclopedia of Industrial
177.
Kramer (2005) Kramer D. A. (2005) U.S. Geological Survey Mineral Yearbook 2005 - Gallium.
USGS, Online-Version under: http://minerals.usgs.gov/minerals/pubs/commod-
ity/gallium.pdf.
Maeda & Kadokura (1993) Maeda H. and Kadokura H. (1993) Process for purification of gallium material.
Patent Number United States of America
Meijer et al. (2003) Meijer A., Huijbregts M. A. J., Schermer J. J. and Reijnders L. (2003) Life-cycle
Assessment of Photovoltaic Modules: Comparison of mc-Si, InGaP and In-
GaP/mc-Si Solar Modules. In: Prog. Photovolt. Res. Appl., 11(2003), pp. 275-
287, Online-Version under: http://dx.doi.org/10.1002/pip.489.
Menezes (2001) Menezes S. (2001) Electrochemical approach for removal, separation and re-
trieval of CdTe and CdS films from PV module waste. In: Thin Solid Films, 387,
pp. 175-178.
Metal Bulletin (2006) Metal Bulletin (2006) Prices & Archive: Various metals from http://www.metal-
bulletin.com
Moskalyk (2003) Moskalyk R. R. (2003) Gallium: the backbone of the electronics industry. In:
Minerals Engineering, 16(921-929).
Pinheiro (2006) Pinheiro W. (2006) Comparative Study of CdTe Sources Used for Deposition of
CdTe Thing Films by Close Spaced Sublimation Technique. In: Materials Re-
search, 9(1), pp. 47-49.
Power Porto (1990) Power Porto G. F. (1990) Abtrennung von Gallium, Germanium und Indium im
Rahmen der hydrometallurgischen Zinkgewinnungsprozesse. Aachen.
Richardson (2002) Richardson S. (2002) Atmospheric Emission Inventory Guidebook. Third Edi-
tion. EMEP: Co-operative Programme for Monitoring and Evaluation of the

5. Literature
Long Range Transmission of Air Pollutants in Europe, CORINAIR: The Core

Inventory of Air Emissions in Europe, EEA: European Environment Agency,
Copenhagen, DK, Online-Version under: http://reports.eea.eu.int/EMEPCORIN
AIR3/en/tab_content_RLR.
Schulte-Schrepping & Piscator (2002) Schulte-Schrepping K.-H. and Piscator M. (2002) Cadmium and
Cadmium Compounds. In: Ullmann's Encyclopedia of Industrial Chemistry.
Online version, posting date: September 15, 2000 Edition. Wiley-Interscience,
Online-Version under: http://dx.doi.org/10.1002/14356007.a04_499.
Selvi et al. (2004) Selvi P., Ramasami M., Samuel M. H. P., Sripriya R., Senthilkumar K., Adaik-
kalam P. and Srinivasan G. N. (2004) Gallium recovery from Bayer's liquor us-
ing hydroxamic acid resin. In: Journal of Applied Polymer Science, 92(2), pp.
847-855, Online-Version under: http://dx.doi.org/10.1002/app.20047.
Shibasaki et al. (1992) Shibasaki T., Abe K. and Takeuchi H. (1992) Recovery of Tellurium from De-
copperizing Leach Solution of Copper Refinery Slimes by a Fixed-Bed Reactor.
In: Hydrometallurgy, 29(1-3), pp. 399-412.
Steinberger (1997) Steinberger H. (1997) Lebenszyklusanalyse von Dünnschichtsolarmodulen auf
der Basis der Verbindungshalbleiter Cadmiumtellurid und Kupferindiumdiesele-
nid. Ph.D. TU München.
Triboulet (2004) Triboulet R. (2004) Fundamentals of the CdTe synthesis. In: Journal of Alloys
and Compounds, 371, pp. 67-71.
Xiao et al. (2003) Xiao X., Songwen X., Xueyi G., Kelong H. and Ryoichi Y. (2003) LCA Case
Study of Zinc Hydro and Pyro-Metallurgical Process in China. In: International
Journal for Life Cycle Assessment, Online first, pp. 1 - 5.

Part XI
Tantalum Powder
Data v2.1 (2009)
Author: Mischa Classen, Empa Dübendorf
Review Roberto Dones, PSI
Summary
Comments to version 2.1

This part and all data correspond to the part and data reported in version 2.0. No changes were made.
Summary
The inventories of mining of tantalum containing ores and the production of capacitor grade tantalum powder
are documented in this report. The inventory is bases on the environmental reporting of the most important pro-
ducer. The production of capacitor grade tantalum powder is based on theoretical models, similarity assump-
tions, and estimations. No recycling (secondary tantalum) is modelled. Hence, this dataset represents primary
tantalum powder.
The production of tantalum metal, as it is used in the steel industry for alloying, follows a different process path.
It includes the processing of K7TaF7 with subsequent remelting in electron beam furnaces, which gives much
higher burdens than the process described in this chapter. Therefore, the tantalum powder dataset described in
this report is not recommended for use to represent tantalum metal manufacturing

Table of Contents
Table of Contents
TANTALUM POWDER .......................................................................................................I

Summary........................................................................................................................................... i
Table of Contents............................................................................................................................ iii
1 INTRODUCTION ....................................................................................................... 5
2 MATERIAL CHARACTERISATION ............................................................................... 6
3 RESERVES AND PRODUCTION .................................................................................. 6

3.1 Primary Resources................................................................................................................. 6
3.1.1 Ores ....................................................................................................................................... 7
3.1.2 Tantalum containing slags..................................................................................................... 7
3.2 Secondary Resources............................................................................................................. 7
3.3 Production ............................................................................................................................. 7
4 USE ....................................................................................................................... 9

5.1.1 Production of Tantalum powder in ecoinvent ..................................................................... 11
5.1.2 Mining of Tantalumpentoxide............................................................................................. 12
5.1.3 Production of Ta2O5 ............................................................................................................ 14
5.1.4 Reduction to Tantalum powder ........................................................................................... 16
5.1.5 Transport and Infrastructure................................................................................................ 17
6 DATA QUALITY CONSIDERATIONS .......................................................................... 20
7 REFERENCES ....................................................................................................... 20

1. Introduction
1 Introduction
The main uses of Tantalum are in the alloying of high temperature steels and as metal powder in ca-
pacitors used in the microelectronic industry. Today the biggest share of Tantalum goes into the ca-
pacitors used in mobile applications such as mobile phones, handheld and laptops. Due to its easy
processing and favourable electric properties it is widely used. Although the price for Tantalum has
risen drastically, discussed alternatives – alumina or niobium based capacitors – are in their infancy.
The price boom has led to a proliferation of small scale mining activities mainly in the Republic
Congo during the Civil War in 1998 – 2004 that led to uncontrolled exploitation (Behrendt et al.,
2007).
Since the production of capacitor grade Tantalum is extensive and the material itself is used in microe-
lectronics in comparatively large amounts, it plays a key role in the environmental assessment.

2. Material Characterisation
2 Material Characterisation
Tantalum is dark, dense, ductile, very hard, easily fabricated, and highly conductive of heat and elec-
tricity. The metal is renowned for its resistance to corrosion by acids; in fact, at temperatures below
150 °C tantalum is almost completely inert to attack by the normally aggressive aqua regia. It can be
dissolved with hydrofluoric acid or acidic solutions containing the fluoride ion and sulphur trioxide, as
well as with a solution of potassium hydroxide. Tantalum's high melting point of 3017°C (boiling
point 5458°C) is exceeded only by tungsten and rhenium (Wikipedia Contributors, 2006.).
3 Reserves and Production

Tantalum is 54th in order of abundance of the elements in the Earth's crust (2.1 g/t). In many deposits,
it occurs in association with the much more abundant niobium (24 g/t). The most important tantalum-
containing minerals are tantalite, wodginite, microlite (the tantalum-rich end member of the pyro-
chlore series), and columbite (Tab. 3.1).
In the past, most of commercial tantalum came as a by-product of tin smelting especially in Thailand.
Since these deposits are now largely depleted (Forrest, 2004) the tin production in this area has been
decreased and is today of minor importance for tantalum production. Therefore in future the tantalum
from tin slag will be of reduced importance.
Tab. 3.1 Composition of important tantalum-containing minerals (Andersson et al., 1996).
Mineral Composition Ta2O5 content, % Nb2O5 content, %

Tantalite (Fe, Mn)Ta2O6 42 – 84 2 – 40
Microlite (Na, Ca)(Ta, Nb) O F 60 – 70 5 – 10
2 6
Columbite (Fe, Mn)(Nb, Ta)2O6 1 – 40 40 – 75
Wodginite (Ta, Nb, Sn, Mn, Fe, Mn)16O32 45 – 56 3 – 15
3+
Yttrotantalite (Y, U, Ca)(Ta, Fe )2O6 14 – 27 41 – 56
3+
Fergusonite (RE )(Nb, Ta)O * 4 – 43 14 – 46
4
Strueverite (Ti, Ta, Nb, Fe)2O4 7 – 13 9 – 14
Tapiolite (Fe, Mn)(Nb, Ta)2O6 40 – 85 8 – 15
Euxenite (Y, Ca, Ce, U, Th)(Nb, Ta, Ti)2O6 1 – 6 22 – 30
Samarskite (Fe, Ca, U, Y, Ce)2(Nb, Ta)2O6 15 – 30 40 – 55
* RE = rare earth element.
3.1 Primary Resources

The principal raw material with the highest potential for extracting tantalum is found in the pegmatitic
deposits in Australia, The typical Ta2O5 concentration in processible pegmatitic deposits suitable to
processing, is less than ca. 0.1%. The uranium- and thorium- containing minerals yttrotantalite,
strueverite, euxenite, and samarskite are sometimes relatively high radioactive, which limits their
workability (Andersson et al., 1996). Today ores with contents of 200 ppm (0.02%) of Ta2O5 are
workable by means of open pit and bulk hydrometallurgical technology (Gille et al., 2006).

3. Reserves and Production
3.1.1 Ores
Tantalum is mainly gained from Pegmatite as Tantalite. The tantalum production rates from tin slags
(25%) and from secondary resources (20%) imply a global share of primary production from Tantalite
ores of 55%. The main producer of tantalum is the company Sons of Gwalia (SGW) that mines today
50% of the known primary Tantalite reserves (Polyakov & Polyakova, 2003). The company has been
producing around 55-60% of the world’s tantalum from two mines in Western Australia, and in 2003
this figure reached almost 70% (Forrest, 2004). The reserves at Greenbushes and Wodgina mines to-
gether amount to 49’000 t Ta2O5 at 240 ppm and an additional 47’000 t at 162 ppm (Sons of Gwalia,
2003).
The USGS data shows the high dynamic of the mine production during the last ten years. Today the
production of ores is estimated to amount 1’500 t and 1’900 tantalum (contained in various intermedi-
ates and commodities) for the years 2004 and 2005 respectively (USGS, 2006).
3.1.2 Tantalum containing slags

An important potential tantalum reserve lies in the so called Southeast Asian Tin Belt, stretching from
China to Indonesia through Thailand and Malaysia, and consists of cassiterite (SnO2) that contains
some tantalum and niobium. During tin smelting, the tantalum and niobium become concentrated in
the slag (Andersson et al., 1996). This slag can be processed directly for tantalum and niobium recov-
ery. Roskill Information Services (Gille et al., 2006) estimated that 25% of the 2002 worldwide de-
mand in tantalum was covered by the recovery of tin slags from the Southeast Asian Tin Belt. Since
many of these operations are closing down due to depletion and accompanying decreased profits in the
tin market, this source today is of minor importance.
3.2 Secondary Resources

Due to the high price of tantalum, the market of secondary resources remains attractive. In 2002
Roskill Information Services (Gille et al., 2006) estimated that 20% of the tantalum originates from
tantalum bearing scrap.
3.3 Production
The Tantalum Niobium International Study Centre TIC used to keep track of the production and de-
mand of worldwide tantalum market until 2001. Data gathered from various sources show the devel-
opment in the last ten years (Tab. 3.2). The supply of primary tantalum is largely monopolised by two
large producers: HC Stark/Bayer GP (Germany) and the Cabot Corp. (U.S.). These two companies ac-
count for 85% of the world production of tantalum powder, tantalum wire, and tantalum oxides
(Polyakov & Polyakova (2003)).

3. Reserves and Production
Tab. 3.2 Compilation of Production, supply and demand of tantalum. Data has been drawn from different sources indicated
below.
5
YEAR Tantalum Supply, Tantalum Demand Mine Reserve Reserve Ba- Reserve
5 5 6
Production from tan- production se Reach
talite, slags & scrap
Year t Ta contained t Ta contained t Ta contained ’000 t Ta cont. ‘000 t Ta cont. years
1 1
1991 549 907
1 1
1992 596 905
1 1
1993 644 1035
1 1
1994 800 1017 325
1 1
1995 1281 1303 356
1 1
1996 1532 1227 383 22 35 57
1 1
1997 1274 1495 413 15 26 36
1 1
1998 1437 1478 454 14 24 31
1 1
1999 1639 1736 495 19 24 38
1 1
2000 2143 1766 836 12 36 14
2 3
2001 2026 (1664) 1300 28 60 22
2 3
2002 1219 (1563) 1540 39 110 25
2 4
2003 1147 1461 1210 39 110 32
4
2004 n. a. 1633 1270 43 150 34
4
2005 n. a. 1742 1910 43 150 23
1 2 3
dme (2002); Anonymous (2004), comprise production from slag and tantalite only; Interpolation between 2000 and 2003;
4 5 6
Forrest (2004); USGS (2006) and preceding; Obtained dividing the number in column “Reserve” by the corresponding
number in column “Mine production"
2500 60
Years (reach)
Tonnes per year
Reach in years
50
2000
40
1500
Demand (incl. Slags & 30

scrap) TIC
1000
supply Processor 20
receits, TIC
500
10
Mine production (USGS)
0 0
1990 1992 1994 1996 1998 2000 2002 2004 2006
Fig. 3.1 Compilation of data on production and demand of tantalum. Different sources are used (see Tab. 3.2).

4. Use
4 Use
The Brussels-based Tantalum Niobium International Study Centre (TIC) estimates that in 2003 50% of
the tantalum went into powder for capacitors, 15% into metal for chemical plants in corrosion resistant
alloys, 10% into temperature resistant superalloys, 10% into cutting tools as TaC and 15% into speci-
ality applications (Forrest, 2004). Most probably the special applications correspond to a great extent
to the application in the military sector as liner material and in anti-armour warheads (Muller, 1996).
TIC estimated that tantalum demand in 2003 reached 1’900 tonnes Ta2O5 and increases at a rate of
7 to 8% to 2’200 tonnes in 2005 (Forrest, 2004).
Fig. 4.1 Applications of tantalum in 2003 (Sons of Gwalia, 2006).
The major use for tantalum, as metal powder, is in the production of electronic components, mainly
capacitors and some high-end audio-grade resistors. Tantalum electrolytic capacitors exploit the char-
acteristic of tantalum to form a protective oxide surface layer, using tantalum foil as one plate of the
capacitor, the oxide as the dielectric, and an electrolytic solution as the other plate
(Wikipedia Contributors, 2006).
Today, the tantalum electrolytic capacitor industry requires approximately 68 percent of the world's
tantalum production. Tantalum capacitors are characterized by a very high volumetric efficiency of
capacitance and very high reliability under extreme conditions. Compared with aluminum or ceramic
capacitors, tantalum capacitors show a distinctly lower temperature and frequency dependence of their
electrical properties. The typical application area of application of tantalum capacitors is shown in Fig.
4.2 (H.C. Starck, 2006).
Beside electronics, tantalum is used as alloying element in metals and superalloys as well as carbide
for hard metals (Fig. 5.1). However, these applications are not covered by this dataset and are not dis-
cussed further.
Fig. 4.2 Application of tantalum in capacitors (H.C. Starck, 2006).

Depending on the intended product and company, a multitude of processes are applied to extract the
tantalum and to manufacture products from it. Different technologies exist to produce metallic tanta-
lum products:
 From its heptafluoride by means of Electrochemic reduction (KF/KCl,) or metallothermic re-
duction with potassium (Yoon et al., 2005).
 From its pentoxide by means of a reduction with gaseous magnesium.
Both processes are to be found in the schematic in Fig. 4. As feedstock, slag from tin production is
used apart from ores. Low grade tin slags are either enriched with scrap to form synthetic ores that can
be subjected to the liquid-liquid extraction with MIBK (methyl isobutyl ketone, 4-methyl pentan-2-
one) or else pass, together with tantalum containing scrap from electronic scrap, the calcination route.
Secondary tantalum raw materials are always extracted by hydrometallurgical processes via tantalum
pentachloride or oxide (Fig. 5.1).
Fig. 5.1 General overview of the production of tantalum products from different feeds (Andersson et al., 1996) and
selection for the inventoried system (red).
Tantalum powders or semi finished products such as wires and sheets for applications in alloys and
steel industry are mainly processed by the reduction of K2TaF7 with sodium. The powder obtained is
then pressed to bars, that are formed to ingots by electron beam melting and shaped to the desired
form. This process therefore differs much from the capacitor powder production. Further information
on the respective process is given in Albrecht & Ingalls (1983) and is not further discussed here.

In this report no overview of all processes can be given, but only the technologies relevant to the pro-
duction of capacitor grade tantalum powder is regarded. Three different processes are applied in ca-
pacitor grade tantalum powder production (H.C. Starck, 2006), namely:
 Electron beam melted process (via K2TaF7)
 Na-reduced process (via K2TaF7)
 Mg-reduced process (via Ta2O5)
For capacitor powders, e.g. from H.C. Starck, all these three different processes are applied in depend-
ence of the desired application. To cover the needs for higher volumetric capacitance and lower rated
voltage, the magnesium reduced process is employed (Fig. 5.2). Since the miniaturisation of capacitors
that are used in microelectronics require ever higher specific electric capacitance, the magnesium re-
duced process via the tantalum oxide is considered relevant in the context of this inventory.
Fig. 5.2 Tantalum powder production processes according to the application fields (H.C. Starck, 2006).
5.1.1 Production of Tantalum powder in ecoinvent

This inventory describes the production of tantalum powder for the use in capacitors in electric appli-
cations. The tantalum powder is produced from virgin tantalite ore to tantalumpentoxide which is re-
duced with magnesium vapour to tantalum powder. No secondary production from consumer scrap
and also no by-product tantalum from tin production is considered. This dataset is exclusively meant
to be used in the context of electronic devices, since the production of tantalum for alloying purposes
follows a different route (Fig. 5.1).
The metadata of the established inventory is given in (Tab. 5.1), their exchanges are shown in (Fig.
5.4). The unit process data for the individual steps are presented in separate data for tantalite mining
(Tab. 5.3), extraction (Tab. 5.5) and reduction (Tab. 5.7).

Tab. 5.1 Ecoinvent meta information for the processes for the production of 1 kg tantalum powder for capacitors.
Name Tantalum, powder, capacitor-grade, at regional storage

Location GLO
Unit kg
Dataset Version 2.0
Included Processes Includes the mining of tantalite, its transport to the metallurgical plant, purification
and reduction steps. Infrastructure is assessed on related metallurgical and mining
operations.
Amount 1
Local Name Tantal, Pulver, für Kondensatoren, ab Regionallager
Synonyms tantalium
General Comment This dataset is functional for the use of tantalum as powder in capacitors of port-
able electronic devices. It is not valid for the use as an alloying element in steel.
Only primary production from tantalite / pegmatite ore is considered. Tantalum as
by-product from tin production is not considered.
Start Date 2000
End Date 2003
Data valid for entire pe- 1
riod
Geography Text
Technology Text Open pit mining and beneficiation at Wodgina and Greenbushes in Australia, re-
duction from pentoxide at HC Starck, Germany. Relates to state of the art technol-
ogy, where purified Ta2O5 powder is reduced by magnesium vapour in a rotary kiln
or a multiple hearth furnace. Production from secondary materials such as post
consumer scrap is disregarded, as is the by product tantalum from tin production.
Representativeness 80%
Production Volume 1045.2 t per year
Sampling Procedure Literature
Uncertainty Adjustments This dataset is established on for the use of tantalum in capacitors. Any other ap-
plication is not covered by this dataset.
5.1.2 Mining of Tantalumpentoxide

Most of the tantalum is mined in the two operations Greenbushes and Wodgina in Western Australia.
The resulting concentrates contain around 40% Ta2O5. An estimated worldwide 10% of Niobium sup-
ply originates from joint extraction with tantalite deposits (Polyakov & Polyakova (2003)). However,
this coupled product is not modelled within this dataset.
The overall recovery of tantalum from ore ranges between 72-74% (73% is assumed in this study),
upgrading (concentrating) the ore from 0.1% to 38% Ta2O5 (Burt, 1996). Assuming an average ore
concentration of 194 ppm (weighted mean of Greensbushes and Wodgina deposits, see Section 3.1
Primary Resources) 7067 t ore is required to produce 2.63 t concentrate containing 1 t Ta2O5.
The tantalite is currently mined in open pits – one underground pit in Greenbushes is suspended but
may resume operation in the future. The dust emissions are approximated in both amount and particle

distribution approximated with specific values from copper mining from the CEPMEIP-Database 1 .
The infrastructure has been approximated with the respective dataset for a generic non ferrous metal
mine with 20’000 Mt ore throughput in 50 year lifetime. The use of conveyor belts has been assumed
to be of the same specific value per tonne ore mined as in the copper production as approximation (see
Tab. 5.2). Land use and occupation has been taken from Sons of Gwalia (2003). The remaining mass
after separating the tantalite is the tailings that must be disposed of. The transport services reflect the
transportation of tantalum oxide to the harbour by train (60 km) and further by ship to Germany
(14’000 km) and train inland to the manufacturer (400 km).
Tab. 5.2 Specific values for infrastructure employed for mining and of tantalite concentrate
Specific Values Unit Value Source & Comment

2
Land use m / t ore 7.85E-02 Sons of Gwalia (2003)
Land occupation m2a / t ore 7.85E-02 Sons of Gwalia (2003)
Conveyor belts m / t ore 7.57E-05 Proxy with copper production
Dust emissions (Total solid particles, TSP) g / t ore 101.72 CEPMEIP 1
Dust emissions. PM10 g / t ore 50 CEPMEIP 1
Dust emissions, PM2.5 g / t ore 5 CEPMEIP 1
Tab. 5.3 Process data of mining and beneficiation per 1 t Ta2O5 contained in the resulting tantalite concentrate.
Mining and Beneficiation of Tantalite 1 t Ta2O5 contained Source & Comment

in concentrate
Inputs
Ore t 7067 Calculation
Ta2O5 in ore kg 1370 Calculation
Electricity kWh 155466 Sons of Gwalia (2003)
Water Consumption m3 5212 Sons of Gwalia (2003)
Fuel Consumption MJ 828720 Sons of Gwalia (2003)
exploitation infrastructrue, conveyor belts m 0.535 Approximation
infrastructure of mining (open pit) unit 1.99E-04 Mine
Land transformation m2 554
Land occupation m2 554
Concentrate transport by train tkm 1.21
Concentrate transport by ship tkm 36.8
Outputs
Tailings disposed t 7064 Calculation
PM<2.5 kg 35.3
PM2.5-10 kg 318
PM>10 kg 365
3
Water effluents m 5.21
Concentrate at 38% Ta2O5 kg 2632
Ta2O5 contained in concentrate kg 1000
1
http://www.air.sk/tno/cepmeip/em_factors.php, accessed in 2003
5.1.3 Production of Ta2O5

As explained in the introduction of 5 System Characterisation, the adequate process for reducing the
tantalite is assumed to make use of magnesium (see Fig. 5.2). HC Stark/Bayer patented a process with
gaseous magnesium (Shekhter et al., 2001) is taken here as the method to produce tantalum powder.
The process considered for the production of capacitor grade tantalum powder consists of the follow-
ing steps (Gille et al., 2006, Fig. 5.3).
 Breakdown with HF
 Extraction with MIBK
 Selective extraction of Niobium
 Stripping of tantalum from MIBK
 Precipitation of oxide hydrates
 Calcination to Ta2O5
 Reduction of Ta2O5 with magnesium vapour
Fig. 5.3 Hydrometallurgical Processing of Niobium and Tantalum (Eckert et al., 2001)

The ore concentrate, up to 40% Ta2O5 (in this study 38%, see 5.1.2 Mining of Tantalumpentoxide), is
dissolved in concentrated hydrofluoric acid (Andersson et al., 1996). Petrova et al. (1997) report ex-
traction efficiency for artificial tantalum concentrate (including TaC bearing scrap) of 80-85% in tradi-
tional industries in 1996 and 95% with optimised parameters.
Gupta (2003) gives a stoichiometry in the leaching process of 16 moles of HF per 2 moles H2TaO6 in
solution. Eckert (1996) indicates large amounts of HF (2 t), H2SO4 (3 t) and NH3 (700 kg) required in
1990 to produce one tonne pure Ta2O5. Although the process advanced since then regarding regenerat-
ing the acids, no quantitative indication on the specific net requirements of the process chemicals can
be made. In this study therefore the values of Eckert (1996) (also cited in Polyakov & Polyakova,
2003) are used.
The following reaction was assumed for subsequent calcination (dehydration):
(Ta2O5) 2 H2O → Ta2O5 + 2 H2O vapour (40.65 kJ per mole evaporation enthalpy)
This results in a theoretical energy requirement of 0.23 MJ heat per kg oxide, assuming an efficiency
of 80%. The electricity demand for pumping was disregarded, because no information on this was
available. Ammonia recovery is assumed to amount to 99.9% of the used ammonia in a 28% NH3 so-
lution (Eckert, 1996), the rest being emitted into the air. Also the spent HF is recovered as diluted so-
lution. No process information for recovery of the spent reactants is available. Both recovered solu-
tions are therefore regarded here to be by-products and leave the system burden free without substitu-
tion for process chemical inputs.
Tab. 5.4 Assumptions Extraction of tantalum oxide form Tantalite concentrate
Specific Values Unit Value Remarks / Source

Extraction efficiency % 95% Assumption
Infrastructure facility unit 4.00E-10 approximated with module "chemical plant, organ-
ics": 4.2 ha, output of 50'000 t/a
Tab. 5.5 Extraction of tantalum oxide form Tantalite concentrate
Extraction 1 kg Ta2O5 Remarks / Source

Inputs
Concentrate fed kg 2.77 Tantalite obtained from mining and beneficiation, see
Tab. 5.3
HF kg 2 Polyakov & Polyakova (2003)
H2SO4 kg 3 Polyakov & Polyakova (2003)
NH3 kg 0.7 Polyakov & Polyakova (2003)
Energy for Calcination MJ 0.23 Evaporation of Ta2O5 x 2 H2O with 40,651 kJ per
mole evaporation enthalpy
Electricity for pumping kWh - Not available
sludge to disposal kg 14 solids, precipitated salts: CaF2, CaSO4, TiO2, Fe2O3.
assumed to be sold (Eckert et al., 2001)
HF recovered kg 2.00 Cut-off, considered as by-product
Ammonia recovered kg 0.70 Cut-off, considered as by-product
Outputs
Ammonia emitted kg 0.00070 0.1 % of the ammonia employed.
Ta2O5, purified kg 1 Contains 0.82 kg tantalum

5.1.4 Reduction to Tantalum powder

Only little details are known from this reductive step, therefore the reasoning is based on the respec-
tive patent. The reduction is effectuated by means of magnesium vapour that is brought into contact
with Ta2O5 powder in a continuous process, e.g. in a rotary kiln or a multiple hearth furnace. The heat
required is assumed to be supplied by electricity. The resulting product contains impurities in the order
of 0.2% to 0.25%, most of it being oxygen (Andersson et al., 1996). In the calculation for this inven-
tory therefore a process efficiency of 100% was assumed. An overview of the specific values and as-
sumptions used is given in Tab. 5.6.
The energy required is calculated upon the reaction stoichiometry:
Ta2O5 + 5 Mg → 5 MgO + 2 Ta
Normalised to 1 kg Ta output an amount of 1.22 kg Ta2O5 and 0.336 kg Mg is needed as input materi-
als. Assuming that Magnesium is added 10% in excess (Shekhter et al. (2001)), 0.369 kg Magnesium
required is calculated (= 0.336 kg * 110%). The unreacted magnesium is supposed to be recovered.
Hence no burdens are allocated to it.
The reducing agent magnesium passes two phase transitions from solid state to liquid (8.95 kJ / mol)
and from liquid to gas (127.6 kJ / mol). The total theoretical energy requirement therefore amounts to
136.5 kJ / mol with a stoichiometry of 13.81 moles Mg per kg Ta.
The magnesium oxide is removed from the tantalum powder with diluted sulphuric acid and cleaned
with ultra pure water. Since no quantitative information on this last cleaning step is available, 3 kg of
sulphuric acid and 2 kg of ultra pure water per kg Ta product were assumed. Furthermore it is assumed
that the resulting magnesium sulphate in solution can be processed further. Considered as by-product
it leaves the system burden free without substituting raw materials.
Tab. 5.6 Specific values and assumptions used for calculation in the reduction step.
Reduction 1 kg Ta powder Remarks / sources

Specific values
Evaporation enthalpy of Mg kJ / mol 127.6
Enthalpy of fusion of Mg kJ / mol 8.95
Magnesium stoichiometry % 110% Ta2O5 + 5 Mg --> 5 MgO + 2 Ta
Density Argon g/l 1.784 norm pressure
Density Oxygen g/l 1.429 norm pressure
reduction efficiency % 100%

Tab. 5.7 Process data for the reduction of 1 kg tantalumpentoxide with magnesium vapour.
Reduction 1 kg Ta powder Remarks / sources

Input
Ta2O5, purified kg 1.22 See. Tab. 5.5. Assuming reduction efficiency of
100%
Magnesium kg 0.369 with Mg 10% in excess to stoichiometry
Electricity for reduction kWh 0.720 Assuming 80% efficiency in heat production and
a total theoretical energy of 136.5 kJ / mol Mg for
evaporation
Argon kg 0.0892 Assuming 5 m3 Volume per 100 kg tantalum
powder
Oxygen kg
Dilute sulphuric acid kg 3 estimation
ultrapure water kg 2 estimation
Output
Wastewater kg 5.594 Loaded with SO4, solved 0.557 kg MgO, 0.0369
kg solved unreacted Mg, by-product, not
counted, estimation
Pure Tantalum Powder kg 1
5.1.5 Transport and Infrastructure

The concentrate is assumed to be transported by train to the nearest port (60 km), by ship to Germany
(14’000 km), then by train to the facility (400 km). Based on the ecoinvent quality-guidelines, devel-
oped for Europe, the chemicals are assumed to be transported 600 km by train and 100 km by lorry
also in this case. An overview over the resulting transport service is given in Tab. 5.8. In this study,
the infrastructure for extraction and reduction of tantalum pentoxide is estimated based on the module
"chemical plant, organics", that has a built area of about 4.2 ha and an average output of 50'000 t/a
(Althaus et al., 2003).
Tab. 5.8 Inventory and calculations for transport and infrastructure.
Unit Value remarks

Chemicals by train tkm 7.7 12.7 kg over 600 km, estimation using ecoinvent
standard distance for Europe
Chemicals by lorry tkm 1.3 12.7 kg over 100 km, estimation using ecoinvent
standard distance for Europe
Infrastructure facility Unit 4.00E-10 approximated with module "chemical plant, organ-
ics": 4.2 ha total surface, output of 50'000 t/a


Infra-
ture
9.08 t ore at 194 ppm Tantalite, Tantalum, 39.5% in tantalite,

Ta2O5 in ore  resource in ground 1.76E+00 kg 0 1 1.13 (2,2,2,1,1,4,12)
73% extraction yield 7.7E-5% in crude ore, in ground

Electricity  Environmental Report 2003 electricity production mix No UCTE 2.00E+02 kWh SGW (2003) 1 1.14 (2,3,2,1,1,4,2)
Water Consumption  Environmental Report 2003 resource in water Water, river 6.70E+00 m3 SGW (2003) 1 1.13 (2,2,1,1,1,4,12)

Fuel Consumption  Environmental Report 2003 machinery No GLO 1.07E+03 MJ SGW (2003) 1 1.13 (2,2,2,1,1,4,4)
processes machine
exploitation
same values as in copper construction
infrastructrue,  machinery Yes RER conveyor belt, at plant 6.87E-04 m Althaus (2003) 1 3.36 (5,5,1,1,3,5,9)
exploitation (7.6E-5 m / t ore) processes
conveyor belts
infrastructure of non-ferrous metal mine,
 no underground mining metals extraction Yes GLO 0.00E+00 unit SGW (2003) 1 3.36 (5,5,1,1,3,5,9)
mining (underground) underground
infrastructure of env. 20'000'000 t during 50 years non-ferrous metal mine,

 metals extraction Yes GLO 2.56E-07 unit estimation 1 3.36 (5,5,1,1,3,5,9)
mining (surface) assumed surface
Land transformation meanvalue of 1999 through 2003
 resource land Transformation, from unknown 7.13E-01 m2 SGW (2003) 1 2.05 (2,2,2,1,3,2,8)
Mining
from nature (7.8E-2 m2 / t ore)

Land transrormation meanvalue of 1999 through 2003 Transformation, to mineral
 resource land 7.13E-01 m2 SGW (2003) 1 2.05 (2,2,2,1,3,2,8)
to mine (7.8E-2 m2 / t ore) extraction site
Land use for mining  steady state assumed resource land 7.13E-01 m2a estimation 1 1.57 (2,2,2,1,3,2,7)
site
Land transrormation meanvalue of 1999 through 2003 Transformation, from mineral
 resource land 7.13E-01 m2 SGW (2003) 1 2.05 (2,2,2,1,3,2,8)
from mine (7.8E-2 m2 / t ore) extraction site
Land transformation meanvalue of 1999 through 2003
 resource land Transformation, to unknown 7.13E-01 m2 SGW (2003) 1 2.05 (2,2,2,1,3,2,8)
to nature (7.8E-2 m2 / t ore)
Concentrate transport mine->port (AUS): 60 km / port transport
 train No RER transport, freight, rail 1.56E-03 tkm estimation 1 2.02 (2,1,1,1,1,4,5)
train (GER)->facility: 400 km systems
Concentrate transport port (AUS)->port (GER): 14000 transport
 ship No RER transport, barge 4.74E-02 tkm estimation 1 2.02 (2,1,1,1,1,4,5)
ship km systems
Tailings waste residual disposal, non-sulfidic tailings,

 No GLO 9.08E+03 kg 1 1.13 (2,2,2,1,1,4,6)
disposed management material landfill off-site
4.9% of 0.102 kg dust per tonne low population
 PM<2.5 air Particulates, < 2.5 um 4.54E-02 kg CEPMEIP 1 2.10 (2,2,2,2,3,5,26)
ore mined density
44.2% of 0.102 kg dust per low population Particulates, > 2.5 um, and <
 PM2.5-10 air 4.09E-01 kg CEPMEIP 1 1.62 (2,2,2,2,3,5,25)
tonne ore mined density 10um
50.8% of 0.102 kg dust per low population
 PM>10 air Particulates, > 10 um 4.70E-01 kg CEPMEIP 1 3.10 (2,2,2,2,3,5,27)
tonne ore mined density
 Water effluents not accounted for 6.70E+00 m3 1 1.13 (2,2,2,1,1,4,6)
Fig. 5.4 Flows for "tantalum, powder, capacitor-grade, at regional storage" and their representation in the ecoinvent database.


Infra-
ture
HF  chemicals inorganics No GLO hydrogen fluoride, at plant 2.44E+00 kg Polyakov (2003) 1 1.16 (2,2,3,1,1,4,4)
H2SO4  chemicals inorganics No RER sulphuric acid, liquid, at plant 3.66E+00 kg Polyakov (2003) 1 1.16 (2,2,3,1,1,4,4)
ammonia, liquid, at regional

Extraction
NH3  chemicals inorganics No RER 8.55E-01 kg Polyakov (2003) 1 1.16 (2,2,3,1,1,4,4)
storehouse
Dehydration approximated with
heat, natural gas, at industrial
Energy for Calcination  40,6 kJ per mole natural gas heating systems No RER 2.80E-01 MJ 1 1.88 (5,5,2,1,4,5,2)
furnace >100kW
evaporationenthalpy for water.
solids, precipitated salts: CaF2,
sludge to
 CaSO4, TiO2, Fe2O3. assumed 1.71E+01 kg Eckert (1996) 1 1.13 (2,2,2,1,1,4,6)
disposal
to be sold
HF recovered
 used as by-product 2.44E+00 kg 1
(40%)
Ammonia
 used as by-product 8.54E-01 kg 1
recovered
Ammonia low population
 air Ammonia 8.55E-04 kg 1 1.52 (2,2,2,1,1,4,16)
emitted density
Magnesium  with Mg 10% in excess metals extraction No RER magnesium, at plant 3.69E-01 kg 1 1.38 (4,5,1,1,1,5,4)
Assuming 80% efficiency in heat

production and a total of 136.5
Electricity for electricity, medium voltage,
Reduction
 MJ / mol Mg for evaporation. electricity production mix No UCTE 7.20E-01 kWh 0 1 1.38 (4,5,1,1,1,5,2)
reduction production UCTE, at grid
Heating-up considered by heat
exchange from reaction.
Assuming 5 m3 Volume per 100
Argon  chemicals inorganics No RER argon, liquid, at plant 8.92E-02 kg 1 1.38 (4,5,1,1,1,5,4)
kg Tantalum powder
Dilute sulphuric acid  estimations chemicals inorganics No RER sulphuric acid, liquid, at plant 3.00E+00 kg estimation 1 1.38 (4,5,1,1,1,5,4)
ultrapure water  estimations water supply production No GLO water, ultrapure, at plant 2.00E+00 kg estimation 1 1.38 (4,5,1,1,1,5,4)
Loaded with SO4 and Mg,

 Wastewater 5.59E+00 kg calculation 1
byproduct, not counted
approximated with module
Infrastructure facility  "chemical plant, organics": 4.2 chemicals organics Yes RER chemical plant, organics 4.00E-10 unit 1 3.55 (5,5,2,1,4,5,9)
ha, output of 50'000 t/a
transport
Chemicals by train  12.4 kg over 600 km train No RER transport, freight, rail 7.45E+00 tkm estimation 1 2.02 (2,1,1,1,1,4,5)
systems
transport transport, lorry >16t, fleet
Chemicals by lorry  12.4 kg over 100 km road No RER 1.24E+00 tkm estimation 1 2.02 (2,1,1,1,1,4,5)
systems average
tantalum,
powder,
tantalum, powder, capacitor-
 capacitor- metals extraction No GLO 1.00E+00 kg
grade, at regional storage
grade, at
regional storage
Fig. 5.4 (cont.) Flows for "tantalum, powder, capacitor-grade, at regional storage" and their representation in the ecoinvent database

6. Data Quality Considerations
6 Data Quality Considerations

The production process is rather new and not much is known from public sources about quantitative
process measures. Therefore the data has been established based on reasoning and rough assumptions.
Direct emissions others than dust and ammonia have been omitted due to lack of data. This might re-
strain the quality of this inventory, if omitted emissions prove to be important in the overall assess-
ment. However, no estimation of the relative importance of this omission can be performed on the ba-
sis of the current information. Infrastructure requirement has been approximated with the related tech-
nology of copper mining and chemical plants. The uncertainty has been calculated with the pedigree
approach. The data describe the production process supplying the demands for the use in capacitors for
electronic devices. This dataset should not be used when tantalum is a substantial part of the system
modelled (e.g. comparing different capacitor materials). It has to be stressed that this dataset is solely
valid for tantalum powder that is used in capacitors. For all other uses of tantalum – namely: tantalum
based superalloys in high temperature machinery and tantalum carbides in hard metals – this dataset is
not applicable.
7 References
Albrecht & Ingalls (1983) Albrecht W.-W. and Ingalls D. P. (1983) E4 Hochtemperaturreinigung von Tan-
tal und Niob. In: Raffinationsverfahren in der Metallurgie internationales Sym-
posium; Hamburg, 20.-22.Oktober 1983. pp. 302. Gesellschaft Deutscher Met-
allhütten- und Bergleute, Institution of Mining and Metallurgy, Verlag Chemie,
Weinheim u.a.
Andersson et al. (1996) Andersson K., Reichert K. and Wolf R. (1996) Tantalum and Tantalum Com-
pounds. In: Ullmann's Encyclopedia of Industrial Chemistry. Fifth Edition Edi-
tion. Wiley-Interscience.
Anonymous (2004) Anonymous (2004) Tantalum Production: Stability Returns, Increases Predicted.
In: Eng Min J, 205(4), pp. 10.
Behrendt et al. (2007) Behrendt S., Kahlenborn W., Feil M., Dereje C., Bleischwitz R., Delzeit R. and
Scharp M. (2007) Rare metals - Measures and concepts for the solution of the
problem of conflict-aggravating raw material extraction - the example of coltan.
In: Texte 23. Umweltbundesamt.
Burt (1996) Burt R. (1996) Beneficiation of Tantalum ore - how is it Achieved and Could it
be better. In proceedings from: Tantalum. 125th TMS Annual Meeting and Ex-
hibition, Anaheim, Ca.
dme (2002) dme (2002) The Tantalum Market. A Micro-Economic Analysis. Report
R37/2002. The dme. Minerals and Energy Department of the Republic South Af-
rica.
Eckert (1996) Eckert J. (1996) The Industrial Application of Pyrometallurgical, Chlorination
and Hydrometallurgy for Producing Tantalum Compounds. In proceedings from:
Tantalum. 125th TMS Annual Meeting and Exhibition, Anaheim, Ca.

7. References
Eckert et al. (2001) Eckert J., Reichert K., Schnitter C. and Seyeda H. (2001) Processing of Colum-
bite-Tantalite Ores and Concentrates for Niobium and Niobium Compounds in
Electronic Applications. In proceedings from: International Symposium Nio-
bium 2001, Orlando, Florida, USA, December 2-5, 2001.
Forrest (2004) Forrest M. (2004) Tantalum tantrums? In: Materials World, 12(11), pp. 27-29.
Gille et al. (2006) Gille G., Gutknecht W., Haas H., Schnitter C. and Olbrich A. (2006) Die Re-
fraktärmetalle Niob, Tantal, Wolfram, Molybdän und Rhenium. In: Winna-
cker/Küchler. Chemische Technik: Prozesse und Produkte. Band 6b Metalle,
Vol. 6b. 2006 Edition. Wiley-VCH, Weinheim.
Gupta (2003) Gupta C. K. (2003) Chemical Metallurgy. Principles and Practice. Wiley-VCH
Verlag GmbH & Co., Weinheim. ISBN 3-527-30376-6.
H.C. Starck (2006) H.C. Starck (2006) Tantal Kondensatoren. H. C. Starck. Retrieved from
http://www.hcstarck.de/
Muller (1996) Muller J. F. (1996) An Overview of the Application of Tanalum and its Alloys
as Warhead Liner Materials. In proceedings from: Tantalum. 125th TMS Annual
Meeting and Exhibition, Anaheim, Ca.
Petrova et al. (1997) Petrova V. A., Palant A. A. and Reznichenko V. A. (1997) Breakdown of artifi-
cial tantalum concentrate by hydrofluoric acid. In: Russian Metallurgy(5), pp.
32-35, Online-Version under: <Go to ISI>://000073839400007
Polyakov & Polyakova (2003) Polyakov E. G. and Polyakova L. P. (2003) Current trends in the production of
tantalum and niobium. In: Metallurgist, 47(1-2), pp. 33-41.
Shekhter et al. (2001) Shekhter L. N., Tripp T. B. and Lanin L. L. (2001) Method for Producing Tan-
tallum/Niobium Metal Powders by the reduction of their Oxides with gaseous
Magnesium. Patent Number U.S. US 6,171,363, held by H.C. Stark, Inc.
Sons of Gwalia (2003) Sons of Gwalia (2003) Environmental Report. Online-Version under:
http://www.sog.com.au/.
Sons of Gwalia (2006) Sons of Gwalia (2006) Corporate Website at November 2006 from
http://www.sog.com.au/
USGS (2006) USGS (2006) Tantalum. In: Mineral Commodity Summaries:. U. S. Geological
Survey, Online-Version under: http://minerals.usgs.gov/minerals/pubs/commod-
ity/niobium/tantamcs06.pdf.
Wikipedia Contributors (2006) Wikipedia Contributors (2006) Tantalum. Wikipedia, The Free Encyclopedia.
Retrieved at 31 October 2006 from http://en.wikipedia.org/w/index.php?title=
Tantalum&oldid=84645326. Date of last revision: 30 October 2006.
Yoon et al. (2005) Yoon J. S., Park H. H., Bae I. S., Goto S. and Kim B. I. (2005) Production of
tantalum powder by external continuous supply of feed materials and reductant
(Reprinted by Japan Inst. Metals, vol 68, pg 247-251, 2004). In: Materials
Transactions, 46(2), pp. 272-276.

Part XI
Tin
Data v2.1 (2009)
Author: Hans-Jörg Althaus, EMPA Dübendorf
Review: Mischa Classen, EMPA Dübendorf
Update: Mischa Classen, Empa Dübendorf
Summary

Summary
The production of tin is inventoried in the modules described in this report. Data reported are mainly based on
theoretical models, similarity assumptions and estimations. Secondary sources of tin (recycling) is not consid-
ered, thus the overall data quality is poor.
Within ecoinvent, these data are used for representation of materials of minor importance. If in an LCA study
one of these materials is of importance, the data should not be used.
This part is the excerpt from “Part VIII Other metals” in Althaus et al. (2004) that has been reorganised for the
actual version 2.0. No changes from the original dataset v 1.1 are reported, hence no adjustments have been ac-
complished for the update to version 2.0. Note that the flow names in the tables displaying unit processes have
not been updated to reflect the name changes with version 2.0.

Table of Contents
Table of Contents
SUMMARY ......................................................................................................................I
1 INTRODUCTION ....................................................................................................... 5
2 RESERVES AND RESOURCES TIN ............................................................................. 5
3 CHARACTERISATION OF TIN ..................................................................................... 6
4 USE / APPLICATION OF TIN ...................................................................................... 8

5.1 Overview ............................................................................................................................... 9
5.2 Mining and beneficiation ...................................................................................................... 9
5.2.1 Underground mining ............................................................................................................. 9
5.2.2 Open pit mining................................................................................................................... 11
5.2.3 Offshore mining .................................................................................................................. 11
5.3 Production of crude tin ........................................................................................................ 12
5.4 Refining............................................................................................................................... 12
5.5 Emissions ............................................................................................................................ 13
5.5.1 Air ....................................................................................................................................... 13
5.5.2 Water................................................................................................................................... 13
5.6 Waste................................................................................................................................... 13
5.7 Tin production in ecoinvent ................................................................................................ 13
5.8 Data quality ......................................................................................................................... 13
6 LITERATURE ......................................................................................................... 20

1. Introduction
1 Introduction
The information available for these inventories was rather sparse. Thus the data quality must be de-
scribed as poor. One important shortcoming is the neglecting of tin recycling. Within ecoinvent, these
data are used for representation of materials of minor importance. If in an LCA study one of these ma-
terials is of importance, the data should not be used.
2 Reserves and Resources Tin

The most important tin ore is cassiterite (SnO2), although small amounts of tin are recovered from sul-
phide minerals such as stannite (Cu2FeSnS4). Although most tin is obtained from mining tin ores, cas-
siterite is also found in association with ores of tungsten, tantalum and lead and minor quantities of tin
are recovered as by-products of mining these metals.
Tin occurs in both primary (typically acid igneous intrusive rocks) and secondary deposits such as al-
luvial and eluvial deposits. Only about 20 percent of the world's tin deposits occur as primary hard-
rock veins or lodes. These ores contain predominantly high temperature minerals and commonly occur
in close association with silicic granites. About 80 % of the world's tin deposits occur not as primary
lodes, but as unconsolidated secondary or placer deposits in river beds and valleys or on the sea
ground. The largest concentration of both onshore and offshore placers is in the extensive tin belt of
South-East Asia, stretching from China in the north, through Thailand, Burma and Malaysia, to the is-
lands of Indonesia in the south.
Compared to commercially viable deposits of copper, lead, zinc, nickel and bauxite, tin deposits are
generally small. Further, tin is almost always found closely allied to the granite from which it origi-
nates. 1
The tin production given in Tab. 2.1 is taken from the itri homepage 2 that refers to the bi-monthly 'Tin
Monitor' published by CRU International Ltd.
1
http://www.miningtrading.com/tin/index.html,
http://www.tintechnology.com/library/detail/library_/Tin%20Mining.htm, access date 14.03.03
2
http://www.itri.co.uk/market.htm, access date 14.03.03

3. Characterisation of Tin
2
Tab. 2.1 Western World Tin Production by Country (tonnes)
Country 1995 1996

Refined Tin Tin-in- Refined Tin Tin-in-
Concentrates Concentrates
Malaysia 41,000 6,402 38,100 5,175
Indonesia 44,200 46,058 48,900 51,024
Thailand 8,200 1,784 11,000 1,299
Brazil 17,000 19,500 17,600 19,200
Bolivia 18,000 16,300 16,700 14,802
Peru - 22,020 500 27,002
Australia - 7,750 - 8,496
Portugal - 4,616 - 4,625
UK - 2,000 - 2,000
Africa - 3,200 - 3,200
Rest of West 14,000 2,000 13,000 2,000
TOTAL 142,400 131,630 145,800 138,823
Note: China and Russia are major tin producers but production data is not currently verifiable.
Tin is also gained from recycling. On one hand there is the recycling of "new scrap" that can be done
easily. On the other hand there is the recovery of tin from tinplate which is becoming increasingly dif-
ficult, as the change to electrolytic methods of tinning is leading to very thin coatings which some-
times diffuse into the steel sheet. The recycling of this material will continue to be a technically and
economically difficult task. In principle, two processes are used today for recovering tin from tinplate:
the alkaline electrolytic method and the alkaline chemical method. Processing tin-containing alloys is
easier than recovering tin from scrap tinplate. However, the tin content of many alloys has decreased
over recent years, and cheaper materials have in some cases replaced alloys. Both these developments
have tended to limit the potential for recovering secondary tin, and also explain why the amount of re-
covered tin in Europe has decreased from 15 800 t in 1980 to 13 300 t in 1990. (Graf (1997))
3 Characterisation of Tin
This information is based on Graf (1997).
Tin (Sn) exists in two crystalline modifications, the - and -forms. -tin is also called grey tin while
-tin is referred to as white tin. Some physical properties of - and -tin are listed in Tab. 3.1.
In the periodic table, tin lies on the boundary between metals and non-metals. The transformation of
the - to the -modification is accompanied by a complete change of lattice structure, affecting the
physical, chemical, and mechanical properties. The transformation of -tin (white tin) into -tin (grey
tin) is of practical importance, as it involves a volume increase of 21 %. The transformation process
requires a high energy of activation, and can be very strongly hindered. White -tin can therefore exist
for many years at –30 °C. The presence of -tin seed crystals is important for the transformation proc-
ess, and these are formed by repeated phase transitions. Foreign "elements" also affect the transforma-
tion temperature and rate. These can consist of impurities and deformations. Tin vapour consists of Sn2
molecules.

3. Characterisation of Tin
Tab. 3.1 Physical properties of - and -tin (Graf (1997))
Natural isotopes 10
Relative atomic mass 118.69
Transformation temperature -Sn  -Sn 286.2 K
Enthalpy of transformation 1966 J/mol
Density of -tin (20 °C) 7.286 g/cm3
Density of -tin 5.765 g/cm3
Density of liquid tin (240 °C) 6.992 g/cm3
Molar heat capacity of -tin (25 °C) 27.0 J mol–1 K–1
Molar heat capacity of liquid tin 28.5 J mol–1 K–1
Melting point 505.06 K
Enthalpy of fusion 7029 J/mol
Boiling point 2876 K
Enthalpy of vaporization 295'763 J/mol
Coefficient of thermal conductivity of -tin at 0 °C 0.63 W cm–1 K–1
Specific electrical resistivity of -tin (0 °C) 5×10–6 Wm
Specific electrical resistivity of -tin (25 °C) 11.15×10–6 Wm
Tin is stable in dry air, but is considerably more rapidly oxidised at a relative humidity of 80 %. Bright
metallic tin becomes dull within 100 d even in indoor atmospheres. Oxygen is rapidly and irreversibly
chemisorbed, and the oxide layer formed grows at an exponentially increasing rate. Typical impurities
present after metallurgical production (e.g., Sb, Tl, Bi, and Fe) promote oxidation. Treatment with
carbonate or chromate solutions leads to passivation. For the use of tin in food packaging it is impor-
tant that lactic, malic, citric, tartaric, and acetic acids do not react at all at normal temperatures or do so
to a negligible extent, especially in the absence of atmospheric oxygen. This is also true for alcohols
and hydrocarbons.

4. Use / Application of Tin
4 Use / Application of Tin

Important applications for Tin are: 3
- Food packaging and culinary uses: Tinplate (sheet steel coated with a thin layer of tin) is the pri-
mary material for food canning and shares the beverage can market with aluminium.
- Electronics: The principal use of tin in electronics is in the use of solders for the joining of elec-
tronic components. But there are many other varied uses in the general electronics field.
- Tin chemicals: The biggest single use of tin chemicals is as PVC stabilisers, but besides this they
are used in a very wide range of applications, for example as polymerisation catalysts in silicone
resins and polyurethane foam manufacture, and in ceramic pigments.
- Plumbing solders: Plumbing solders containing lead are being replaced by safe tin-silver (usually
96.5 percent tin/3.5 percent silver) or tin-copper (97 to 99 percent tin/1 to 3 percent copper) alloys,
particularly for drinking water systems.
- Engineering alloys: Tin is used as an alloy with other metals.
- Pewter and bronze: Pewter (an alloy containing about 92 percent tin with small amounts of other
metals added for strength) has long been known for its decorative appearance and ease of working
as a craft metal. Tin bronzes (alloys of copper with 5-20% tin) are renowned for their use in art
castings and as bell metals.
- Dental amalgams: Dental amalgam contains about 13 percent tin by weight together with silver
and mercury.
- Anti-corrosion and engineering coatings: Tin and tin-alloy coatings are widely used in the manu-
facture of bearings and in many kinds of machinery and fabricated parts, both for their anti-
corrosion and lubricant properties.
- Wine capsules: A tin coating acted as a barrier layer in traditional lead-based capsules to prevent
contact between the lead and the wine. Tin-lead capsules are now being replaced with other mate-
rials but the most prestigious and closest in feel to the traditional capsules, are made from almost
pure tin.
Consumption of tin in the western world is around 200’000 tonnes per year. The dominant consumers
are Europe (33%), USA (17%) and Japan (12%). The distribution of tin consumption is unusually
weighted toward Europe where 46% of all beverage containers are tinplate in contrast to the USA
where tinplate is not used for beverage containers. Consumption can be divided into four segments,
tinplate (30%), solder (29%), alloys and coatings (25%) and chemicals (16%). 4
3
http://www.tintechnology.com/library/detail/library_/Applications%20for%20Tin.htm, access date 14.03.03
4
http://www.murchisonunited.com.au/operations/renison.asp#production, access date 28.03.03

This chapter is mainly based on Graf (1997).
5.1 Overview
Tin is mined, beneficiated, smelted and refined. Fig. 5.1 gives an overview of the processes.
underground mining open pit mining offshore mining
sorting (flotation)
sorting (flotation)
thermal ore
beneficiation
smelting
(electro-)refining
tin, 99% pure
Fig. 5.1 Overview of tin production
5.2 Mining and beneficiation

No precise information is found on the amount of tin ore mined in the different types of mines. Based
on descriptive information 5 it is therefore assumed that 20% of the tin ore is gained from underground
mines and 40% each from open pit mines and from offshore deposits.
5.2.1 Underground mining
Drilling and blasting is only needed in underground mining where depths can reach 1000 meters.
Common technology for non-ferrous metal mining is applied. Tin ores are separated from the gangue
using sorting methods based on density. The large variation in achievable tin concentration of about
6% to 60% in the ore is due to different sizes and intergrowth of the cassiterite with other minerals.
5
http://www.tintechnology.com/library/detail/library_/Tin%20Mining.htm, access date 14.03.03 and
http://www.murchisonunited.com.au/operations/renison.asp#production, access date 28.03.03

Recoveries of around 75% of the tin are achieved by this beneficiation. Unacceptably high losses of
material would occur if ore concentrates with higher tin contents were produced. Therefore, a py-
rometallurgical "thermal ore beneficiation" stage is necessary prior to the actual reduction process.
This is a volatilisation process that exploits the fact that the iron compounds and other slag formers
have low vapour pressures at 1000 – 1500 °C, while SnO and especially SnS volatilise very readily.
This technique can also be used to treat tin-containing slags. In the process where tin ore or slag is
heated together with pyrite (FeS2) the SnO reacts to SnS and the FeS2 to FeO. With careful process
control, 90 – 95 % of the tin can be recovered as an oxidised product containing 40 – 60 % Sn.
Although this is at the lower end of the commercial viability we assume an average tin content of 0.1%
in the crude ore that is mined underground and an average recovery of the tin minerals of 70%. Thus
1.43 tons of material has to be mined to gain 1 kg of tin from underground mines. With the assumed
average density of 1.5 tons/m3 crude ore this corresponds to 0.95 m3 crude ore.
For a lack of specific information the energy consumption and the amount of explosives used for min-
ing are calculated based on the values derived for platinum group metal (PGM) mining. The assump-
tion underlying the calculation is that the respective consumption is proportional to the mass of crude
ore extracted. The electricity needed for mechanical beneficiation is taken from Richter et al. (1996)
and the water consumption is estimated. The in- and outputs for the thermal beneficiation step given in
Tab. 5.2 are calculated with the values given in Tab. 5.1.
Tab. 5.1 Values used for calculation of in- and output flows of the thermal beneficiation step.
Values used for calculation of heat for thermal beneficiation

Tin content after mechanical beneficiation [%] 30 assumption
Thermal capacity of input [J/kg/K] 235.91 Value for Sn
Enthalpy of fusion [J/kg] 59221 Value for Sn
Enthalpy of vaporisation [J/kg] 2491894 Value for Sn
Efficiency of kiln 0.25 assumption
Heating to [K] 1250 assumption
Molar weight of Sn [kg/mol] 0.11869
Values used for calculation of SO2 emissions to air of thermal beneficiation
Part of SnO evaporating as SnS 0.9 assumption
Overall efficiency of abatement 0.6 assumption
Molar weight of SO2 [kg/mol] 0.064058
Values used for calculation of FeS2 input for thermal beneficiation
Molar weight of FeS2 [kg/mol] 0.119967
Utilisation of S in FeS2 0.7 assumption

Tab. 5.2 In- and outputs for underground mining and beneficiation of tin ore per kg of tin
Underground mining of tin Flows per kg tin Source

Crude ore [kg] 1430 calculation based on ore composition
Diesel, burned in building machines [MJ] 98.96 calculation based on PGM
Electricity [kWh] 64.66 calculation based on PGM
blasting [kg] 0.66 calculation based on PGM
mine infrastructure [unit] 5.34*10-14 calculation based on PGM
mechanical beneficiation of tin ore from underground mining
Electricity [kWh] 1.43 Richter et al. 1996
3
Water [m ] 1.43 assumption
thermal ore beneficiation of tin ore from underground mining
heat from heavy fuel oil furnace [MJ] 37.95 calculation
Electricity [kWh] 4 calculation based on Richter et al. 1996
FeS2 input [kg] 1.179 calculation based on assumptions
SO2 emissions to air [kg] 0.9715 calculation based on assumptions
5.2.2 Open pit mining

For open pit mining it is assumed that no cap of the deposit exists. Water cannons operating at pres-
sures of up to 1.5 MPa are used to separate the ore from the overburden. The mixture of water and
heavy sand with about 70% tin content is then fed to the treatment plant.
The tin content in placer deposits lies in the range of 0.01 to 3%. We assume an average tin content of
0.1% in the crude ore that is mined and an average recovery of the tin minerals of 90%. Thus 1.11 tons
of material has to be mined to gain 1 kg of tin from open pit mines. With the assumed average density
of 1.5 tons/m3 crude ore this corresponds to 0.74 m3 crude ore. The thickness of the recoverable open
deposit is assumed to be 5 meters. Thus 0.148 m2 of land is transformed to mining area per kg of tin
recovered. It is assumed that a mining site is depleted within 10 years and left for natural recultivation
afterwards.
For a lack of specific information the energy consumption used for mining, the infrastructure needed
therefore and the dust emissions generated by it are calculated based on the average values derived for
iron ore and bauxite mining. The assumption underlying the calculation is that the respective con-
sumption is proportional to the mass of crude ore extracted. No explosives are assumed to be used.
The water consumption is estimated.
5.2.3 Offshore mining

Marine deposits are mined several kilometres offshore in depths of around 50 meters below sea level
using bucket line dredges. Thus a similar mixture of water and heavy sand as in the open pit mining
with water jets is won.
The tin content in placer deposits lies in the range of 0.01 to 3%. We assume an average tin content of
0.1% in the crude ore that is mined and an average recovery of the tin minerals of 90%. Thus 1.11 tons
of material has to be mined to gain 1 kg of tin from offshore deposits. With the assumed average den-
sity of 1.5 tons/m3 crude ore this corresponds to 0.74 m3 crude ore. The thickness of the recoverable
offshore deposit is assumed to be 5 meters. Thus 0.148 m2 of benthos is transformed to mining area
per kg of tin recovered. It is assumed that a mining site is depleted within 10 years and left for natural
recultivation afterwards.
For a lack of specific information the energy consumption of open pit mining of tin ore is used assum-
ing that the electricity used in open pit mining is replaced by diesel, burned in building machines with
5 times lower efficiency.

5.3 Production of crude tin

The beneficiatied ores with the average tin content of 68% (20% with 50% Sn from underground min-
ing, 80% with 70% Sn from open pit and offshore mining) are heated together with powdered coal in a
reverbarory furnace at 1200 °C. Thus a tin containing slag is formed. By a second melting in a sloping
hearth the metallic tin runs off (97% purity) and a slag with about 8% Sn content is formed. This slag
is recycled to the beneficiation step.
According to Richter et al. (1996), 15% to 20% carbon is added to the ore for smelting and 75 to 100
kg heating oil per tonne of ore is burned for heating. About 10-14% of the input ore mass is turned to
slag that is used in a secondary smelting. This process uses 550 to 1000 kWh electricity (in an electric
furnace) and 1 to 10 kg of electrodes per ton of slag. Beside that, 122 kg coal, 67 kg of gypsum and 39
kg of limestone are needed per ton of slag. 136 kg of secondary slag has to be landfilled per ton of
primary slag input. These values are used to calculate the in- and outputs of the smelting of tin (see
Tab. 5.3).
Tab. 5.3 In- and outputs for smelting per kg pure tin (based on Richter et al. (1996))
Sn concentration in Input 0.68

Sn concentration in output 1
Sn concentration in slag 0.08
fraction of ore concentrate input to slag output 0.12
fraction of Sn that can be extracted of slag 0.7
carbon content in coal and electrode 0.9
Ore smelting per kg ore per kg Sn per kg Sn
input input output
ore input 1 1.47 1.481
coal input [kg] 0.175 0.257 0.259
heating oil [kg] 87.5 129 130
slag to slag smelting output [kg] 0.12 0.176 0.178
carbon dioxide to air [kg] 0.578 0.849 0.855
Slag smelting per kg slag per kg ore per kg Sn per kg Sn
input input2 input3 output4
electrical power [kWh] 0.775 0.0930 0.137 0.138
electrodes input [kg] 0.0055 0.000660 0.000971 0.000977
coal input [kg] 0.122 0.0146 0.0215 0.0217
gypsum input [kg] 0.067 0.00804 0.0118 0.0119
limestone input [kg] 0.039 0.00468 0.00688 0.00693
slag to landfill output [kg] 0.1365
carbon dioxide to air [kg] 0.421 0.050 0.074 0.075
1 calculated from value per input, tin fraction in slag, fraction of ore input to slag and fraction of extraction
from slag
2 calculated from input per kg slag and slag to smelting per kg ore input
3 calculated from input per kg slag and slag to smelting per kg Sn input
4 calculated from input per kg slag and slag to smelting per kg Sn output
5 value given in Richter et al. (1996)
5.4 Refining
The crude tin obtained by the reduction process is insufficiently pure for most applications. Refining
can be done pyrometallurgically or by electro-refining.

Electro-refining can only be operated at low current densities, leading to low process rates and ineffi-
cient utilisation of energy. Thus electro-refining is only worthwhile if the tin contains high concentra-
tions of noble metals. It is therefore not considered in this study.
There are different types of pyrometallurgical refining depending on the impurities that have to be re-
moved. All the processes have in common that liquid tin is treated in a way that the impurities form
intermetallic compounds that can be removed. For removing arsenic it is even necessary to have a
vacuum over the liquid metal. Graf (1997) gives an energy demand of 400 to 700 kWh per ton tin for
heating and providing the vacuum. The average of this value is used as electricity input for lack of
other information.
5.5 Emissions
5.5.1 Air
Beside the emissions from heat generation there are dust emissions from mining, SO2 emissions from
beneficiation and CO2 emissions from the reduction of SnO with carbon.
5.5.2 Water
No information about emissions to water is available. The beneficiation process being predominantly
mechanical is not believed to be a relevant source of water emissions. Some emissions are likely to oc-
cur from waste gas treatment in the thermal beneficiation of the ore from underground mining.
5.6 Waste
The overburden of mining is refilled in the mine and not treated as waste. Acid rock drainage (ARD)
that is possibly occurring where sulfidic ores are present is thus not modelled for this waste fraction.
The only waste considered as such is the secondary slag that is landfilled. No detailed information
about the composition of this slag was found. From the composition of the crude ore it is assumed that
the slag contains mainly Sn, SiO2, Fe, Ta2O5, TiO2, Al2O3, CaO, Nb2O5 and As.
For a lack of better data the disposal of slag form unalloyed steel making in the electric arc furnace
(EAF) is inventoried as a proxy.
5.7 Tin production in ecoinvent

The background information to the process "tin, at regional storage" are given in Tab. 5.4, flows are
shown in Fig. 5.2.
5.8 Data quality

Most data for the tin production are assumptions based on qualitative information and theoretical
models. Thus uncertainty is rather high. These data are of insufficient quality for LCA where tin or tin
derived products (e.g. solders) play an important role.

Tab. 5.4 Meta information for the tin production processes
Name tin, at regional storage

Location RER
Unit kg
The cradle to gate inventory of world-wide primary tin production in-
cludes mining, beneficiation, smelting and refining of tin ore and tin
Included Processes
respectively. Also included are the transports from the major produc-
ers to Europe.
Amount 1
Local Name Zinn, ab Regionallager
Synonyms
General Comment to refer- Most data are calculated based on assumptions and theoretical mod-
ence function els. Their overall quality is poor.
Start Date 1996
End Date 2003
Other Period Text
Production data relate to the global average. Transports relate to the
Geography text
consumption in Europe
Data relate to an assumed average technology for mining, benefici-
Technology text
ation, smelting and refining.
Production Volume > 280'000 tonnes per year (worldwide)
Sampling Procedure based on literature and internet
Extrapolations none

ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
for mean
Remarks
Output
General
Process
name in
Source
Modul
Name
value
value
Input
Infra-
Sub-
Mean
Type
Unit
95%
recovery of tin f rom ore = 70%; 20% of tin
Tin, 79% in
production; cassiterite = 0.127% in mined
cassiterite, 0.1% in
Underground mining and beneficiation (20% of tin
tin ore ore
 material, Sn = 79 % in cassiterite --> 1 kg resource in ground 2.86E-01 kg calculation 1 1.415 (4,5,3,5,1,5,12)
(resource) crude ore, in
Sn in ore corresponds to 790kg of ore and
ground
to 1t of mined material
land use
no land use inventoried f or underground
(mine) 
mining
[m2a]
diesel,
burned in
recovery of tin f rom ore = 70%; 20% of tin construction diesel, burned in calculation based
building  machinery No GLO 2.83E+01 M J 1 2.382 (5,5,1,5,3,5,5)
production)
production processes building machine on PGM

machine
[MJ]
electricity, medium
calculation based
electricity  electricity production mix No UCTE voltage, production 1.85E+01 k Wh 1 1.415 (3,5,1,5,3,5,2)
on PGM
UCTE, at grid
blasting [kg construction civil calculation based

 No RER blasting 1.90E-01 kg 1 1.415 (3,5,1,5,3,5,1)
explosive] processes engineering on PGM
inf rastruct input includes inf rastructure needed f or

non f errous metal calculation based
ure (mine)  mining and benef iciation (buildings, access metals extraction Yes GLO 5.34E-14 unit 1 3.161 (3,5,1,5,3,5,9)
mine, underground on PGM
[unite] roads,...) (exclusive building machines)
low population
 w aste heat f rom electricity use air Heat, w aste 6.65E+01 M J calculation 1 1.415 (3,5,1,5,3,5,13)
density
Fig. 5.2 Flows for "tin, at regional storage" and its representation in the ecoinvent database

ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
for mean
Remarks
Output
General
Process
name in
Source
Modul
Name
value
value
Input
Infra-
Sub-
Mean
Type
Unit
95%
electricity, medium
beneficiation of recovery of tin f rom ore = 70%; 20% of tin Richter et al.
underground
electricity  electricity production mix No UCTE voltage, production 4.09E-01 k Wh 1 1.415 (3,5,1,5,3,5,2)
Mechanical
production 1996
tin from UCTE, at grid
Water, unspecif ied
w ater  1m3 / ton crude ore assumed resource in w ater 4.09E-01 m 3 assumption 1 1.69 (5,5,1,5,3,5,12)
natural origin
low population
density
heat, heavy f uel oil,

recovery of tin f rom ore = 70%; 20% of tin heating
from underground mining (20%
heat  oil No RER at industrial f urnace 1.08E+01 M J calculation 1 1.415 (3,5,1,5,3,5,1)

production systems
Thermal beneficiation of tin
1MW
electricity, medium calculation based
electricity  electricity production mix No UCTE voltage, production 1.14E+00 k Wh on Richter et al. 1 1.415 (3,5,1,5,3,5,2)
of tin production)
UCTE, at grid 1996

production; limestone as proxy f or mining construction calculation based
Pyrite  additives No CH limestone, at mine 3.37E-01 kg 1 1.462 (4,5,1,5,3,5,4)
of pyrite --> acid rock drainage (ARD) is materials on assumptions
neglected
low population
density
recovery of tin f rom ore = 70%; 20% of tin low population calculation based
 SO2 air Sulf ur dioxide 2.78E-01 kg 1 1.938 (5,5,1,5,3,5,31)
production density on assumptions
Fig. 5.2 Flows for "tin, at regional storage" and its representation in the ecoinvent database (cont.)

ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
for mean
Remarks
Output
General
Process
name in
Source
Modul
Name
value
value
Input
Infra-
Sub-
Mean
Type
Unit
95%
Tin, 79% in
tin ore ore cassiterite, 0.1% in
 material, Sn = 79 % in cassiterite --> 1 kg resource in ground 4.44E-01 kg calculation 1 1.415 (4,5,3,5,1,5,12)
ground
Mining and beneficiation of tin from open pit mining (40% of tin production)
land use
recovery of tin f rom ore = 90%; 40% of tin Occupation, mineral calculation based
(mine)  resource land 5.92E-01 m 2a 1 1.938 (5,5,1,5,3,5,7)
production extraction site on assumptions
[m2a]
land
transforma Transf ormation, calculation based
 resource land 5.92E-02 m 2 1 2.382 (5,5,1,5,3,5,8)
tion (mine) from unknow n on assumptions
[m2]
land
Transf ormation, to
transforma calculation based
 resource land mineral extraction 5.92E-02 m 2 1 2.382 (5,5,1,5,3,5,8)
tion (mine) on assumptions
site
[m2]
diesel,
burned in calculation based
building  machinery No GLO 4.06E+00 M J on Iron and 1 2.382 (5,5,1,5,3,5,5)
machine bauxite
[MJ]
electricity  electricity production mix No UCTE voltage, production 2.46E-01 k Wh on Iron and 1 1.69 (5,5,1,5,3,5,2)
UCTE, at grid bauxite
Water, unspecif ied
w ater  resource in w ater 6.36E-01 m 3 assumption 1 1.69 (5,5,1,5,3,5,12)
natural origin
input includes inf rastructure needed f or
infrastruct calculation based
mining and benef iciation (buildings, access
ure (mine)  metals extraction Y es GLO mine, iron 9.85E-11 unit on Iron and 1 3.376 (5,5,1,5,3,5,9)
roads, machines...) (exclusive building
[unite] bauxite
machines); iron mine as proxy
low population
 w aste heat f rom electricity use air Heat, w aste 8.86E-01 M J calculation 1 1.69 (5,5,1,5,3,5,13)
density
calculation based
dust, recovery of tin f rom ore = 90%; 40% of tin low population Particulates, > 10
 air 1.79E-01 kg on Iron and 1 1.938 (5,5,1,5,3,5,25)
coarse production density um
bauxite
calculation based
dust, low population Particulates, > 2.5
 air 1.61E-01 kg on Iron and 1 2.382 (5,5,1,5,3,5,26)
medium density um, and < 10um
bauxite
calculation based
low population Particulates, < 2.5
 dust, f ine air 1.79E-02 kg on Iron and 1 3.376 (5,5,1,5,3,5,27)
density um
bauxite

ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
for mean
Remarks
Output
General
Process
name in
Source
Modul
Name
value
value
Input
Infra-
Sub-
Mean
Type
Unit
95%
Tin, 79% in
tin ore ore cassiterite, 0.1% in
 offshore mining (40% of tin production) material, Sn = 79 % in cassiterite --> 1 kg resource in ground 4.44E-01 kg calculation 1 1.415 (4,5,3,5,1,5,12)
Mining and beneficiation of tin from
ground
land use
recovery of tin f rom ore = 90%; 40% of tin Occupation, mineral calculation based
(mine)  resource land 5.92E-01 m 2a 1 1.938 (5,5,1,5,3,5,7)
production extraction site on assumptions
[m2a]
land
Transf ormation,
transf orma calculation based
 resource land f rom sea and 5.92E-02 m 2 1 2.382 (5,5,1,5,3,5,8)
ocean
[m2]
land
Transf ormation, to
transf orma calculation based
 resource land industrial area, 5.92E-02 m 2 1 2.382 (5,5,1,5,3,5,8)
benthos
[m2]
diesel,
burned in calculation based
building  machinery No GLO 8.49E+00 M J on Iron and 1 2.382 (5,5,1,5,3,5,5)
machine bauxite
[MJ]
calculation based
hard coal mix, at
coal  31.4 MJ/kg hard coal f uels No UCTE on Richter et al.
2.59E-01 kg 1 1.415 (3,5,1,5,3,5,3)
regional storage
1996
Ore smelting
heat, heavy f uel oil, calculation based

heating
heat  41.2 MJ/kg oil No RER at industrial f urnace 5.34E+00 M J on Richter et al. 1 1.415 (3,5,1,5,3,5,1)
systems
1MW 1996
low population
 w aste heat f rom burning of coal air Heat, w aste 8.39E+00 M J calculation 1 1.415 (3,5,1,5,3,5,13)
density
calculation based
low population Carbon dioxide,
 CO2 f rom burning of coal air 8.55E-01 kg on Richter et al. 1 1.415 (3,5,1,5,3,5,14)
density f ossil
1997

ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
for mean
Remarks
Output
General
Process
name in
Source
Modul
Name
value
value
Input
Infra-
Sub-
Mean
Type
Unit
95%
electricity  electricity production mix No UCTE voltage, production 1.38E-01 k Wh on Richter et al. 1 1.415 (3,5,1,5,3,5,2)
UCTE, at grid 1996
calculation based
anode, aluminium
electrodes  31.4 MJ/kg metals extraction No RER 9.77E-04 k g on Richter et al. 1 1.415 (3,5,1,5,3,5,3)
electrolysis
1996
calculation based
hard coal mix, at
coal  31.4 MJ/kg hard coal f uels No UCTE 2.17E-02 k g on Richter et al. 1 1.415 (3,5,1,5,3,5,3)
Secondary slag smelting
regional storage
1996
calculation based
construction gypsum, mineral, at
gypsum  others No CH 1.19E-02 k g on Richter et al. 1 1.415 (3,5,1,5,3,5,3)
materials mine
1996
calculation based
construction
limestone  additives No CH limestone, at mine 6.93E-03 k g on Richter et al. 1 1.415 (3,5,1,5,3,5,3)
materials
1996
f rom electricity use and burning of coal low population

 w aste heat air Heat, w aste 1.21E+00 M J calculation 1 1.415 (3,5,1,5,3,5,13)
and electrodes density
calculation based
low population Carbon dioxide,
 CO2 f rom burning of coal air 7.47E-02 kg on Richter et al. 1 1.415 (3,5,1,5,3,5,24)
density f ossil
1997
disposal, dust,
unalloyed EA F
dust f rom unalloyed electric steel w aste residual Richter et al.
 w aste No CH steel, 15.4% w ater, 1.36E-01 k g 1 1.415 (3,5,1,5,3,5,6)
production to landf ill as proxy management material landf ill 1996
to residual material
landf ill
electricity, medium assumption
Refining
electricity  electricity production mix No UCTE voltage, production 5.50E-01 k Wh based on Graf 1 1.415 (3,5,1,5,3,5,2)
UCTE, at grid 1997
low population
density
Transport transport,
transport
Transport to
ocean  10'000 km, 1.2 kg per kg tin ship No OCE transoceanic 1.20E+01 tk m assumption 1 2.382 (5,5,1,5,3,5,5)
systems
Europe
f reighter f reight ship

Transport transport
 500 km, 1.2 kg per kg tin road No RER transport, lorry 32t 6.00E-01 tk m assumption 1 2.382 (5,5,1,5,3,5,5)
lorry systems
ref ined tin tin, at regional
 metals extraction No RER 1.00E+00 kg
(99%) storage

6. Literature
6 Literature
www.ecoinvent.ch.
Graf (1997) Graf G. G. (1997) Tin, Tin Alloys, and Tin Compounds. In: Ullmann's encyclo-
pedia of industrial chemistry (ed. Anonymous). 5th edition on CD-ROM Edition.
Wiley & Sons, London.
Richter et al. (1996) Richter K., Brunner K. and Bertschinger H. (1996) Ökologische Bewertung von
Wärmeschutzgläsern, Integraler Vergleich verschiedener Verglasungsvarianten.
Eidg. Materialprüfungs- und Forschungsanstalt (EMPA), Dübendorf.

Part VIII
Magnesium and
Magnesium Alloys
Data v2.1 (2009)
Author: Niels Jungbluth, ESU-services
Review Mischa Classen,

Hans-Joerg Althaus, EMPA Dübendorf
Rolf Frischknecht, ESU-services
Summary

Acknowledgement
The authors thank Johanna Öster-Haagensen and Liv Kari Sivertsen from the research centre of Norsk
Hydro in Porsgrunn, Norway for her support while interpreting and using the environmental data pro-
vided by Norsk Hydro.

This part corresponds to the original part “Other Metals” in Althaus et al. (2004) that described the metals lead,
zinc and magnesium and has been suspended for this version 2.0. The original content has been distributed over
the new part VII on lead, zinc, cadmium and indium as well as this part that now exclusively is designated to the
description of magnesium and its alloys AZ91 and AlMg3. Chapter “Cumulative Results and Interpretation” of
the original report has been deleted for the current update.
Summary
Magnesium, metal alloys containing magnesium, i.e. AZ91, AlMg3, and die castings are important
construction materials, e.g. in the automobile sector or for electric equipment. The environmental im-
pacts of magnesium production are investigated as a unit process over the life cycle from raw material
extraction until the gate of the factory. The investigation is based on environmental data provided by a
producer and on a literature research. High uncertainties still exist for the average energy and SF6 use
in magnesium industry.
Keywords
AlMg3, Aluminium alloys, AZ91, Magnesium, Magnesium alloys, Magnesium die casting
ecoinvent report v2.1 No. 10 - - i

Table of Contents
Table of Contents
ACKNOWLEDGEMENT ......................................................................................................I
COMMENTS ON THIS PART ................................................................................................I
SUMMARY ......................................................................................................................I
KEYWORDS ....................................................................................................................I
1 INTRODUCTION ....................................................................................................... 1
4 PRODUCTION AND USE OF MAGNESIUM .................................................................... 1

5.1 Life Cycle Inventories ........................................................................................................... 3
5.1.1 Magnesium............................................................................................................................ 3
5.1.2 Magnesium Plant................................................................................................................... 7
5.1.3 Magnesium-Aluminium Alloys............................................................................................. 7
5.1.4 Magnesium Die Castings ...................................................................................................... 8
5.1.5 Recycling of Magnesium ...................................................................................................... 9
5.2 Data Quality Considerations ................................................................................................. 9
5.3 Conclusions ........................................................................................................................... 9
ABBREVIATIONS.......................................................................................................... 10
APPENDICES: ECOSPOLD META INFORMATION ............................................................. 10
6 LITERATURE ......................................................................................................... 11
ecoinvent report v2.1 No. 10 - - iii

1. Introduction
1 Introduction
The metal magnesium is getting more and more important as a construction material for different
technical purposes. It is used as a lightweight material and as an addition to metal alloys such as
MgAl3. This inventory for magnesium and related materials is meant to be used as a background unit
process for the life cycle assessments of technical products such as e.g. equipment for electricity
transmission or photovoltaic systems.

Magnesium can be extracted from seawater, from minerals (dolomite, magnesite and brucite) or can be
gained by recycling materials. It is the eighth most abundant element in the earth’s crust. Seawater
contains about 0.13wt% Magnesium (Hoy-Petersen et al. (1999)).
Norsk Hydro uses about 207 000 tonnes of dolomite with an approximate magnesium content of
26 330 tonnes and seawater to produce 106 000t of magnesium (Hydro Magnesium (1999b)).

Magnesium is a silvery white, light metal. It burns in air with an intense white flame. Pure magnesium
has a high resistance to corrosion because its galvanic activity is low (Hoy-Petersen et al. (1999)).
4 Production and Use of Magnesium

The most important producer in Europe is Norsk Hydro with an annual production of 105’600 tonnes
in 1998 while the total annual production was about 372’000 tonnes. Tab. 4.1 shows an overview of
the production places of magnesium. It is important to know that the import of magnesium from the
eastern world plays an important role for meeting the western worlds demand. But there are not many
shipments between Europe and America. Various investment projects are going on to increase the pro-
duction capacities due to the increasing demand. Due to the possibility to produce magnesium from
seawater there is no shortcoming of resources for production.
Tab. 4.1 Magnesium production and consumption in 1996-97 (Aroule (1998))
Countries and companies Production Type of production process Im-/ Exports

1’000t/a 1’000t/a
Western world (8 producers) 254 Mainly use of SF6 for oxidation protection. 52 Import
Norsk Hydro 106 Electrolytic
Others 148
Eastern world 128 Mainly use of SO2 for oxidation protection.
CIS countries (3 producers) 48
-80 Export
China (200 producers) 80
Total 372 80% electrolytic, 20% thermal
Magnesium is used for different technical purposes. The consumption of magnesium in the western
world is summarized in Fig. 4.1 (Avedesian & Baker (1999)). Aluminium alloys and production of
die-castings are the main purposes for consumption.
ecoinvent report v2.1 No. 10 -1-

Fig. 4.1 Consumption of magnesium in Western world in 1997 (total 333’700 tonnes). Amount used for magnesium
alloys (die castings) is shown shaded (Avedesian & Baker (1999)).
The main production processes are electrolysis of magnesium chloride melts (made from resources
like seawater and dolomite) and the metallothermic reduction of magnesium oxide with a share of 80%
and 20%, respectively for the total production (Hoy-Petersen et al. (1999)). The different producers in
the world use different processes. Fig. 5.1 shows a block diagram for the process used by Norsk Hydro
that is investigated here. Magnesium is produced from seawater and dolomite. The production of
Mg(OH)2 includes calcination of dolomite, mixing with sea water and precipitation and filtration of
Mg(OH)2. Thereafter Mg(OH)2 filter cake is calcinated to MgO. Water free MgCl2 is prepared through
pelletisation and chlorination of MgO. MgCl2 is electrolysed and ingots are produced in the foundry
(Haagensen et al. (1996)).

Fig. 5.1 Block diagram of the dolomite and seawater process for magnesium production in Porsgrunn, NO
(Haagensen et al. (1996))
5.1 Life Cycle Inventories

The inventory is split up into the primary production of magnesium and the further production of
magnesium alloys. Than the manufacturing of diecastings is investigated. The infrastructure for mag-
nesium production is assumed roughly.
5.1.1 Magnesium
The life cycle inventory investigates the production of magnesium from seawater for a modern pro-
duction plant based on data from a Norwegian producer. All processes from seawater uptake until the
magnesium leaves the plant are included in the inventory.
Energy is used for the preparation of dolomite lime, making MgCl2 cell feed from seawater and acid
generation. Tab. 5.1 shows an overview for the energy requirements of magnesium production investi-
gated by different authors. Different methodologies are used for the reporting of the energy require-
ments. Large differences exist for the reported energy requirements for the production of one kilogram
magnesium. But, it has to be considered that some of the data have been investigated in the seventies.
Thus a higher energy efficiency (a lower specific energy consumption) can be assumed for today’s
production.

Tab. 5.1 Energy requirement of magnesium production as investigated in different studies. It has to be noted that
assumptions for the calculation of the total energy requirements vary among different studies
Different
Further remarks about type of energy use and references,
Products and processes types of
>year of first publication
energy use
MJ/kg
46.8
Electrolysis of magnesium Electricity use (Hoy-Petersen et al. (1999:568,>1977-87)).
-50.4
36
Electrolysis of magnesium Electricity use (Celik et al. (1996)).
-66.6
Silicothermic reduction of magne- 61
Electricity use (Celik et al. (1996)).
sium -72
Metallothermic (Pidgeon process)
108 Total energy requirement (Hoy-Petersen et al. (1999,>1980)).
reduction of magnesium
Metallothermic (Bolzano process) 25
Energy requirement of furnace (Wilson et al. (1995,>1980-81)).
reduction of magnesium -26
Recycling of magnesium 10.8 Electricity use (Kammer et al. (2000))
Total energy requirement for recycling (Wilson et al.
Recycling of magnesium 6.7
(1995,>1974)).
Magnesium 267 Total energy requirement (Wilson et al. (1995,>1974)).
Total energy requirement at Dow in 1978. Reduction of 15%-
Magnesium 325
20% over ten years (Hoy-Petersen et al. (1999,>1980)).
Total energy requirement, Split up: 5%, 23%, 62%, 10% for
Magnesium 378 dolomite preparation, MgCl2 cell feed from seawater, electroly-
sis, and acid regeneration, respectively (Barclay (1977)).
Total energy requirement, Split up: 6%, 45%, 45%, 4% for sea-
Magnesium 358 water treatment, evaporation, electrolysis, and acid recovery,
respectively (Barclay (1977)).
329 Total energy requirement (Boustead & Hancock
Magnesium
-487 (1979:345,>1972-75)).
Total energy requirement 1 (Albright & Haagensen (1997); Haa-
Magnesium 144
gensen et al. (1996)).
Total energy requirement (Albright & Haagensen (1997); Haa-
Magnesium alloy AZ91 151
gensen et al. (1996)).
Total energy requirement for the production stage (Barclay
Magnesium sheet manufacture 75
(1977); Wilson et al. (1995)).
Magnesium extrusions manufac- Total energy requirement for the production stage (Barclay
83
ture (1977); Wilson et al. (1995)).
Magnesium die-casting produc- 21 Total energy requirement for the production stage (Barclay
tion -45 (1977); Wilson et al. (1995)).
Total energy requirement. 50% / 60% / 70% yield from die-
Magnesium (from magnesite) al- 229 / 220 /
caster. Including recycling of diecaster returns (Hydro Magne-
loy AZ91 cast part from diecaster 214
sium (1999a)).
Magnesium (from seawater and Total energy requirement. 40% / 60% / 80% yield from die-
205 / 182 /
dolomite) alloy AZ91 cast part caster. Including recycling of diecaster returns (Hydro Magne-
171
from diecaster sium (1999b)).
1
Johanna Öster Haagensen (Norsk Hydro, email communication 10.8.2000): The total energy demand for the production of
pure Mg, including hydro power electricity, fossil fuel and other sources such as steam. The system boundary is from cradle
to gate for "all" raw materials (life cycle perspective). This means that production of hydropower electricity is included in
the 144MJ/kg, as well as extraction, production and transport of fossil fuels and so forth.

The inventory for the production of magnesium is mainly based on the environmental reports of Norsk
Hydro for the year 1998 2 and on an LCA for the production of magnesium and magnesium diecastings
by Norsk Hydro. Tab. 5.2 shows the results of the cumulative life cycle inventory published by Norsk
Hydro for magnesium and magnesium die-castings (Albright & Haagensen (1997); Haagensen et al.
(1996); Hydro Magnesium (1999a); Hydro Magnesium (1999b); Norsk Hydro (1999)). There is no
explanation for the origin of some of the emissions, e.g. dioxins.
Production data are proprietary information. Therefore, these reports present only aggregated figures
of the life cycle inventory. Thus it was unavoidable to disaggregate some data with uncertain assump-
tions or use cumulative data in order to achieve an inventory table for this study.
Tab. 5.2 Cumulative energy consumption, direct emissions and waste data for the production of 1kg magnesium and
AZ91 part from die-casting at Norsk Hydro. All stages in the life cycle from cradle to gate are considered in
the inventory data for the alloys (Albright & Haagensen (1997); Hydro Magnesium (1999a); Hydro Magnesium
(1999b))
LCI AZ91, LCI AZ91,

LCI
Type of energy, emission or waste 60% yield, 60% yield,
Magnesium
Porsgrunn Becancour
Unit kg kg kg
Energy use Total energy consumption MJ 144.0 182.0 220.0
Hydro power electricity kWh 21.5 27.2 30.0
Energy from fossil fuels MJ 49.1 62.0 90.0
Emissions Chlorinated hydrocarbons to air kg 1.37E-05 1.24E-05 -
CO2 kg - 7.0 8.2
Dioxins to air kg 2.33E-11 2.10E-10 -
Hydrogen chloride kg 8.0E-4 - 4.4E-3 - -
NOx kg - 0.013 0.022
Sulphur dioxide (SF6 cast operations) kg 2.42E-03 24.0 24.0
Sulphur hexafluoride kg ca. 0.0007 0.0018 0.0012
Characterisation Global warming effect incl. SF6 kg 19.0 50.00 37.00
results Global warming effect excl. SF6 kg 6.0 8.0 9.0
Acidification kg 0.02 0.025 -
Winter smog kg 0.015 0.017 -
Solid waste kg 0.5 2.0 1.3
The newest figure for the cumulative energy use of total 144MJ/kg (Albright & Haagensen (1997)) is
used. This figure has to be disaggregated to the type of energy uses. It is assumed that about 60% of
the energy is delivered by electricity produced in Norway with a conversion efficiency of 82% (ac-
cording to the ratios shown in Tab. 5.2). Other energy uses are steam consumption, the use of gas and
due to the transports. This is estimated with 40% or 49MJ cumulative energy demand and a conver-
sion efficiency of 67%. Thus 39MJ/kg of natural gas are used. The uncertainty of the data for the en-
ergy use is high as the summary in Tab. 5.1 shows because a full disaggregating of the found data is
not possible.
An important factor for the environmental impacts is the use of SF6. It is used to protect molten mag-
nesium from oxidation in the cast house, in the production of ingots and at the diecaster. Sulphur diox-
ide has been known to fulfil this function for a long time, but SO2 is toxic in higher concentrations and
in contact with humidity it will enhance corrosion of steel equipment. The quantities used in different
2
Environmental information on http://www.magnesium.hydro.com/ (26.7.2000). The facility in Porsgrunn was closed down
1st of April 2002. A new activity has been established in a part of the old factory. This new remelting facility is called Hydro
Magnesium Alloys. Process scrap from magnesium users and pure magnesium from China are used to produce magnesium
alloys (Personal communication, Liv Kari Sivertsen, 14.5.02).

plants varies between 0.1kg and 11kg SF6 per ton produced magnesium components. The use of SF6
for die-casting is not included in this figure (Gjestland et al. (1996)). Norsk Hydro has reduced the
emissions from formerly 4kg/t to 0.7kg/t in 1998. 3 Estimates for the worldwide average specific con-
sumption vary between 1.3kg/t and 7kg/t (Maiss & Brenninkmeijer (1998:3084)). We assume an aver-
age emission figure of 2.9kg/t for the magnesium production with data for the Norsk Hydro plants
(0.7kg/t, 34% production share) and other plants (4kg/t).
The inventory for other emissions (e.g. SO2, dioxin, chlorinated hydrocarbons to water) is based on the
data published in the environmental reports of Norsk Hydro (Tab. 5.2, Norsk Hydro (1999)). Charcoal
is used for the chlorination of magnesia in an amount of 0.45kg/t (Hoy-Petersen et al. (1999:562)).
The import from Asia by ship has been considered for this material. The theoretical emission of
1.32kg CO2/kg magnesium has been calculated from the charcoal use which is used as a reduction
agent. The unit process raw data and data quality indicators for magnesium production are shown
inTab. 5.3. Transport of raw materials have been estimated with standard distances.
Large uncertainties exist for data on energy use and SF6 emissions during magnesium production. For
the energy use and the SF6 emission a literature review has been used to estimate the standard devia-
tion. The highest uncertainties exist for the SF6 emissions which might be as high as 11kg/t. Emission
data are used for cumulative emissions. Thus they overestimate the direct emissions in an unknown
range.
Tab. 5.3 Unit process raw data and data quality indicators for the production of magnesium. References and as-
sumptions for the inventory are presented in the text.
Standard
magnesium, Uncertainity
Name Deviation General Comment
at plant Type
95%
Location RER
Unit kg
Tecnosphere charcoal, at plant 4.50E-01 1 1.26 (3,4,1,3,1,5); Literature
sulphur hexafluoride, liquid, at plant 2.90E-03 1 1.22 (1,3,1,3,1,5); Estimation based on literature
tap water, at user 2.48E+02 1 1.22 (1,3,1,3,1,5); Data from env. report
disposal, decarbonising waste, 30% water, to residual material landfill 1.86E-01 1 1.57 (1,3,1,3,4,5); Environmental report, type of
waste not clear
disposal, separator sludge, 90% water, to hazardous waste incineration 5.74E-02 1 1.57 (1,3,1,3,4,5); Data from environmental report,
type of sludges not clear
treatment, sewage, from residence, to wastewater treatment, class 2 2.48E-01 1 1.31 (1,3,1,3,3,5); Data from env. report
transport, transoceanic freight ship 6.20E+00 1 2.05 (4,na,na,1,na,na); Imports of Mg from China,
imports of charcoal from overseas 10000km
transport, freight, rail 2.72E-01 1 2.10 (4,5,na,na,na,na); Standard distance 600km
electricity, medium voltage, production NORDEL, at grid 2.15E+01 1 1.25 desaggregated from cumulative energy
natural gas, burned in industrial furnace >100kW 3.87E+01 1 2.00 desaggregated from cumulative energy
demand with assumption of 67% conversion
transport, lorry >16t, fleet average 1.43E-01 1 2.10 (4,5,na,na,na,na); Standard distance
magnesium plant 2.00E-10 1 3.10 (4,4,1,3,1,5); 50a of uses with a production of
100000t/a
limestone, milled, packed, at plant 1.95E+00 1 1.31 (1,3,1,3,3,5); Estimation for dolomite use
Resource use Magnesium, 0.13% in water 7.52E-01 1 1.22 (1,3,1,3,3,3); Data from env. report
Emissions Carbon dioxide, biogenic 1.32E+00 1 1.24 (4,1,3,1,1,3); Calculation from charcoal use,
2.93kg/kg
Dioxins, measured as 2,3,7,8-tetrachlorodibenzo-p-dioxin 2.33E-11 1 3.01 (1,3,1,3,1,3); Data from env. report
Heat, waste 4.97E+01 1 1.22 (4,1,1,1,1,3); Calculated from electricity
Hydrogen chloride 2.00E-03 1 2.01 (1,3,1,3,1,3); Data from env. report
Sulfur dioxide 2.42E-03 1 2.01 (1,3,1,3,1,3); Data from env. report
Sulfur hexafluoride 2.90E-03 1 4.10 Estimation, large variations for different
producers
Chlorinated solvents, unspecified 2.48E-08 1 2.01 (1,3,1,3,1,3); Data from env. report
Reference flow magnesium, at plant 1.00E+00 - - -
3
Johanna Öster Haagensen (Norsk Hydro, email communication 8.8.2000): The total amount of SF6 is depending on the total
weight of magnesium handled. Norsk Hydro is in the process of changing production routines and SF6 will be phased out.
Hydro is also selling SO2 diecast equipment and competence to diecasters and is offering service. Some diecasters have ne-
ver used SF6 and are still using SO2.

5.1.2 Magnesium Plant

The land use area and building size for a magnesium production plant has been estimated roughly with
1km2 from a photo of the Norsk Hydro factory in Porsgrun. 4 Land transformation has been calculated
according to standard methodology. The unit process raw data and data quality indicators for the mag-
nesium production plant with an annual production capacity of 100 000t and a life time of 50 years are
shown in Tab. 5.4. Data for the necessary equipment (likely also metals and bricks) were not available.
They can be considered to be not important in comparison to the amount of produced magnesium.
Tab. 5.4 Unit process raw data and data quality indicators for the magnesium production plant and data quality con-
siderations.
Standard
magnesium Uncertainity
plant Type
95%
Location RER
Unit unit
Tecnosphere Occupation, industrial area 5.00E+05 1 1.60 (4,4,1,3,1,5); Estimatimation from map
Resource use Transformation, from unknown 5.00E+07 1 2.10 (4,4,1,3,1,5); Application of standard
procedure
Transformation, to industrial area 1.00E+06 1 2.10 (4,4,1,3,1,5); Application of standard
procedure
Reference flow magnesium plant 1.00E+06 - - -
5.1.3 Magnesium-Aluminium Alloys

AZ91D
The magnesium alloy AZ91D (also known as M11916 or DIN GD-MgAl9Zn1) is the most common
one for the production of die-castings. It contains 9.1% aluminium and small amounts of copper, zinc
and manganese. Environmental impacts of production are only known from a summarized inventory
(Albright & Haagensen (1997)). The inventory data derived from this investigation is shown in Tab.
5.5. The energy use has been disaggregated in the same way as for magnesium. Inventory data for the
infrastructure are not considered as it is assumed that the production will take place in the magnesium
production plant. Specific data for the invert gas use were not available. It can be assumed that it is not
much more than for the pure magnesium (which is already included in the inventory).
Tab. 5.5 Unit process raw data and data quality indicators for the production of magnesium alloy AZ91. References
and assumptions for the inventory are presented in the text.
magnesium- Standard
Uncertainity
Name alloy, AZ91, Deviation General Comment
Type
at plant 95%
Location RER
Unit kg
Tecnosphere magnesium, at plant 9.09E-01 1 1.20 (2,2,1,1,1,5); Literature
copper, at regional storage 1.02E-03 1 1.20 (2,2,1,1,1,5); Literature
manganese, at regional storage 3.05E-03 1 1.20 (2,2,1,1,1,5); Literature
zinc, primary, at regional storage 7.11E-03 1 1.20 (2,2,1,1,1,5); Literature
MG-silicon, at plant 2.03E-03 1 1.20 (2,2,1,1,1,5); Literature
transport, freight, rail 2.11E-02 1 2.10 (4,5,na,na,na,na); Standard distances
electricity, medium voltage, production NORDEL, at grid 1.51E+00 1 1.30 (4,2,1,1,1,5); Literature disaggregated
transport, lorry >16t, fleet average 1.06E-02 1 2.10 (4,5,na,na,na,na); Standard distances
aluminium, production mix, cast alloy, at plant 9.24E-02 1 1.30 (4,2,1,1,1,5); Literature disaggregated
Emissions Heat, waste 5.43E+00 1 1.30 (4,2,1,1,1,5); Literature disaggregated
Reference flow magnesium-alloy, AZ91, at plant 1.00E+00 - - -
4
Confirmed by Liv Kari Sivertsen, Norsk Hydro, NO, 14.5.02 in a personal communication.

AlMg3
The production of AlMg3 (AA 5754 or EN AW-5754) is also investigated with the content of differ-
ent metals used for the alloy. This aluminium alloy contains 3% of magnesium and little amounts of
other metals (Lang (1999:9)).
The additional cumulative energy for producing the alloy is 7MJ/kg (conversion efficiency estimated
with 82%) and it is assumed that it is delivered by the electricity mix of the aluminium industry.
Transports are considered for the alloying metals but not for the main ingredients, as it is assumed that
the alloy production is close to the magnesium or aluminium production plant. A material loss of 1.5%
during production (assumed by Albright & Haagensen (1997 for magnesium alloys)) is considered for
the inventory. It is assumed that the production takes place at the magnesium (or aluminium) produc-
tion plant. Thus an additional infrastructure is not considered. The inventory is shown in Tab. 5.6. In-
formation for further magnesium alloys can be found in the handbook of KAMMER et al. ( (2000)).
Tab. 5.6 Unit process raw data and data quality indicators for the production of magnesium aluminium alloy AlMg3.
References and assumptions for the inventory are presented in the text.
aluminium Standard
Uncertainity
Name alloy, AlMg3, Deviation General Comment
Type
at plant 95%
Location RER
Unit kg
Tecnosphere magnesium, at plant 3.05E-02 1 1.20 (2,2,1,1,1,5); Literature
cast iron, at plant 4.06E-03 1 1.20 (2,2,1,1,1,5); Literature
chromium, at regional storage 3.05E-03 1 1.20 (2,2,1,1,1,5); Literature
copper, at regional storage 1.02E-03 1 1.20 (2,2,1,1,1,5); Literature
manganese, at regional storage 5.08E-03 1 1.20 (2,2,1,1,1,5); Literature
zinc, primary, at regional storage 2.03E-03 1 1.20 (2,2,1,1,1,5); Literature
MG-silicon, at plant 4.06E-03 1 1.20 (2,2,1,1,1,5); Literature
(4,5,na,na,na,na); Standard distance
transport, freight, rail 3.86E-03 1 2.10
200km
electricity mix, aluminium industry 1.59E+00 1 1.30 (4,2,1,1,1,5); Literature disaggregated
(4,5,na,na,na,na); Standard distance
transport, lorry >16t, fleet average 1.93E-03 1 2.10
100km
aluminium, production mix, cast alloy, at plant 9.65E-01 1 1.20 (2,2,1,1,1,5); Literature
Emissions Heat, waste 5.74E+00 1 1.20 (2,2,1,1,1,5); Literature
Reference flow aluminium alloy, AlMg3, at plant 1.00E+00 - - -
5.1.4 Magnesium Die Castings

The diecasting process is used for the production of metal parts in a predefined form. SF6 is also used
during the die casting process in order to prevent molten magnesium from oxidation. Different authors
have investigated the SF6 use. GJESTLAND et al. ( (1996:39)) reported that up to 1kg SF6 per ton of Mg
component might be used and emitted during the die casting process. In Germany, magnesium die
casting producers used between 0.25 and 10 kg SF6 per ton of magnesium or in some cases only SO2.
The average use in 1995 and 1996 was 4.3 and 2.4 kg/t, respectively. Due to the decreasing amounts
used, a prognosis assumed that until 2000 the average use will decrease to 1 kg/t (Schwarz & Le-
isewitz (1996:41)). For Swiss die casters the amount of SF6 is reported to be 4kg SF6/t Mg-die cast-
ings. This is equal to an annual amount of 120kg (Klaus & Dinkel (1999:41)). A producer of die cast-
ing equipment 5 quantifies much smaller values for the usage of SF6 and SO2 with 0.012 to 0.018kg
SF6 and 0.049 to 0.075kg SO2 per tonne of magnesium. These data could not be verified. For our in-
ventory we assume based on the literature data 1kg SF6/t.
The cumulative energy use required to produce a finished die casting, including normal processing ef-
ficiencies, amounts to 21-45MJ/kg or 1.9-4.1kWh electricity/kg (Wilson et al. (1995:660)). It is as-
5
RAUCH Fertigungstechnik, Austria, 22.8.2000, http://www.rauch-ft.com/englisch/magnes/gas.html.

sumed that only electricity is used (2.22kWh/kg). The inventory for die castings is shown in Tab. 5.7.
Data for the infrastructure were not available.
Tab. 5.7 Unit process raw data and data quality indicators for magnesium die-casting. References and assumptions
for the inventory are presented in the text
magnesium-
Standard
alloy, AZ91, Uncertainity
diecasting, at Type
95%
plant
Location RER
Unit kg
Tecnosphere magnesium-alloy, AZ91, at plant 1.67E+00 1 1.20 (2,2,1,1,1,5); Literature
sulphur hexafluoride, liquid, at plant 1.00E-03 1 1.20 (2,2,1,1,1,5); Literature
transport, freight, rail 3.35E-01 1 2.10 (4,5,na,na,na,na); Standard distance
electricity, medium voltage, production UCTE, at grid 2.22E+00 1 1.50 200k /600k
Literature estimation
transport, lorry >16t, fleet average 1.67E-01 1 2.10 (4,5,na,na,na,na); Standard distance 100km
Emissions Heat, waste 8.00E+00 1 1.30 (4,2,1,1,1,5); Literature
Sulfur hexafluoride 1.00E-03 1 1.60 (2,2,1,1,1,5); Literature
5.1.5 Recycling of Magnesium

The possibilities for the recycling of magnesium are described by KAMMER et al. ( (2000:49ff)). To-
day about 17% of the magnesium supply is produced by recycling of magnesium wastes. Mainly pure
wastes with a known composition are used for the recycling. The recycled magnesium is used mainly
during the production of aluminium, iron and steel as a desulphurisation or reduction agent.
The use of electricity for the recycling is much lower than for the production of primary magnesium.
KAMMER et al. ( (2000)) estimated it to be only 3 kWh/kg in comparison to about 20 kWh/kg for the
primary production. Different processes can be used depending on the type of input material. The
processes need normally also a protecting gas like sulphur hexafluoride, but the tendency is to use SO2
to reduce the greenhouse gas emission.
Due to substantial lack of data we refrain from presenting unit process raw data for the recycling of
magnesium. One might use the figures for magnesium and reduce the energy use to one-tenth in a first
assumption. Other impacts especially for the SF6 emissions can be expected to be about the same.
5.2 Data Quality Considerations

The simplified approach with a pedigree matrix has been used for calculating most of the standard de-
viations. Large uncertainties exist for data on energy use and SF6 emissions during magnesium pro-
duction. For the energy use and the SF6 emission a literature review has been used to estimate the
standard deviation. The highest uncertainties exist for the SF6 emissions which might be as high as
11 kg/t. Also the value for transport distances of materials is very uncertain.
5.3 Conclusions
The environmental impacts of magnesium production have been investigated as a unit process over the
life cycle from raw material extraction until the gate of the factory. The investigation is based on envi-
ronmental data provided by a producer and on a literature research. High uncertainties still exist for
the average energy and SF6 use in magnesium industry. The inventory can be used in life cycle as-
sessment for complex products where magnesium or metal alloys are used.

Abbreviations
English
AZ91 Magnesium-aluminium (9.1%) alloy
AlMg3 Aluminium-magnesium (3%) alloy
Appendices: EcoSpold Meta Information

Type ID Field name // ID 67 2994 481 491 492
magnesium-alloy, AZ91, at magnesium-alloy, AZ91,
aluminium alloy, AlMg3, at plant magnesium plant magnesium, at plant
ReferenceFunction 401 Name plant diecasting, at plant
Geography 662 Location RER RER RER RER RER
ReferenceFunction 493 InfrastructureProcess 0 1 0 0 0
ReferenceFunction 403 Unit kg unit kg kg kg
DataSetInformation 201 Type 1 1 1 1 1
DataSetInformation 202 Version 2.0 2.0 2.0 2.0 2.0
DataSetInformation 203 energyValues 0 0 0 0 0
DataSetInformation 205 LanguageCode en en en en en
DataSetInformation 206 LocalLanguageCode de de de de de
DataEntryBy 302 Person 41 41 41 41 41
DataEntryBy 304 QualityNetwork 1 1 1 1 1
ReferenceFunction 400 DataSetRelatesToProduct 1 1 1 1 1
Resource extraction until gate
Production of alloy from primary Building infrastructure and land Production of alloy from primary Diecasting process, no extra
of the factory including
metals, no extra infrastructure. use, no machinery. metals, no extra infrastructure. infrastructure.
ReferenceFunction 402 IncludedProcesses infrastructure.
ReferenceFunction 404 Amount 1 1 1 1 1
Aluminium-Legierung, AlMg3, ab Magnesium-Legierung, AZ91, Magnesium-Legierung, AZ91,
Magnesiumwerk Magnesium, ab Werk
ReferenceFunction 490 LocalName Werk ab Werk Druckguss, ab Werk
ReferenceFunction 491 Synonyms AA 5754//EN AW-5754 M11916//DIN GD-MgAl9Zn1
The magnesium alloy AZ91D is

the most common one for the
production of die-castings. It
Large uncertainties exist for contains 9.1% aluminium and Large uncertainties exist for
The production of AlMg3 is
Production plant for magnesium data on energy use and SF6 small amounts of zinc and data on energy use and SF6
investigated with the content of
with an annual capacity of emissions during magnesium manganese. Environmental emissions during magnesium
different metals used for the
100'000t and a life time of 50 production. This inventory impacts of production are only diecasting. This inventory
alloy. This aluminium alloy
years. describes an assumption for known from a summarized describes an assumption for
contains 3% of magnesium.
the average production. inventory. It is assumed that the average production.
magnesium for this alloy has
been produced in a new
production plant.
ReferenceFunction 492 GeneralComment
ReferenceFunction 494 InfrastructureIncluded 1 1 1 1 1
ReferenceFunction 495 Category metals metals metals metals metals
ReferenceFunction 496 SubCategory extraction extraction extraction extraction extraction
ReferenceFunction 497 LocalCategory Metalle Metalle Metalle Metalle Metalle
ReferenceFunction 498 LocalSubCategory Gewinnung Gewinnung Gewinnung Gewinnung Gewinnung
ReferenceFunction 499 Formula Mg
ReferenceFunction 501 StatisticalClassification
ReferenceFunction 502 CASNumber 007439-95-4
TimePeriod 601 StartDate 1998 1998 1998 1998 1998
TimePeriod 602 EndDate 1998 1998 1998 1998 1998
TimePeriod 603 DataValidForEntirePeriod 1 1 1 1 1
TimePeriod 611 OtherPeriodText Environmental report Environmental report Environmental report Environmental report Environmental report
Data from literature and for 1
Data for 1 producer Data for 1 producer Data for 1 producer Data for 1 producer
Geography 663 Text producer
Production of magnesium from
Production of aluminium- Infrastructure (land use and
seawater and dolomite in a Production of magnesium alloy
magnesium alloy from primary buildings) for the production of Diecasting of AZ91, no
modern production plant based from primary metals, no
metals, no recycling of input magnesium not including recycling
on data from a Norwegian recycling of input materials.
materials. equipment.
Technology 692 Text producer.
Representativeness 722 Percent 30 30
Representativeness 724 ProductionVolume ca. 146000t/a 105600 tonnes in 1998 372000 tonnes in 1998 Not known ca. 100000 t/a
Environmental data for an Environmental data for an Environmental data from an Environmental data from an
Representativeness Estimation with map.
725 SamplingProcedure important producer important producer important producer important producer
From 1 producer to average
Europe, energy uses
From 1 producer to average From 1 producer to average From 1 producer to average From 1 producer to average
Representativeness disaggregated from cumulated
Europe Europe Europe Europe
figures. Type of electricity mix
726 Extrapolations not known.
Representativeness 727 UncertaintyAdjustments none none none none none
DataGeneratorAndPubli 751 Person 41 41 41 41 41
DataGeneratorAndPubli 756 DataPublishedIn 2 2 2 2 2
DataGeneratorAndPubli 757 ReferenceToPublishedSource 10 10 10 10 10
DataGeneratorAndPubli 758 Copyright 1 1 1 1 1
DataGeneratorAndPubli 759 AccessRestrictedTo 0 0 0 0 0
DataGeneratorAndPubli 760 CompanyCode
DataGeneratorAndPubli 761 CountryCode
DataGeneratorAndPubli 762 PageNumbers chapter magnesium chapter magnesium chapter magnesium chapter magnesium chapter magnesium
ecoinvent report v2.1 No. 10 - 10 -

6. Literature
6 Literature
Albright & Haagensen (1997) Albright D. L. and Haagensen J. Ö. (1997) Life Cycle Inventory of Magnesium.
In proceedings from: Magnesium Trends, Toronto, CA.
www.ecoinvent.ch.
Aroule (1998) Aroule P. (1998) Magnesium demand and supply. In proceedings from: Fifth
International Magnesium Conference.
Avedesian & Baker (1999) Avedesian M. M. and Baker H. (1999) Magnesium and Magnesium Alloys. In:
ASM Specialty Handbook. The Materials Information Society (ASM), US.
ISBN 0-87170-657-1.
Barclay (1977) Barclay J. A. (1977) Energy consumption in primary magnesium production. In
proceedings from: International conference on energy conservation in produc-
tion and utilization of magnesium, Massachusetts, USA, 25-27.5.
Boustead & Hancock (1979) Boustead I. and Hancock G. F. (1979) Handbook of Industrial Energy Analysis.
Ellis Horwood Ltd.
Celik et al. (1996) Celik C., Peacey J., Bishop G., White C., giasson E. and Avedesian M. (1996)
Magnola: An Innovative Prcess for Magnesium Production. In proceedings
from: Third International Magnesium Conference, Manchester, UK.
Gjestland et al. (1996) Gjestland H., Westengen H. and Plahte S. (1996) Use of SF6 in the Magnesium
Industry: An Environmental Challenge. In proceedings from: Third International
Magnesium Conference, Manchester, UK.
Haagensen et al. (1996) Haagensen J. Ö., Opheim B. and Westengen H. (1996) Life Cycle Inventory of
Magnesium - Results Reliability and How to Use them in the Automotive Indus-
try. In: Ganzheitliche Betrachtung im Automobilbau: Rohstoffe - Produktion -
Nutzung - Verwertung, Vol. VDI-Bericht 1307 (ed. VDI (Verein Deutscher In-
genieure)). pp. 175-182. VDI-Verlag GmbH, Düsseldorf.
Hoy-Petersen et al. (1999) Hoy-Petersen N., Aune T., Vralstad T., Andreassen K., Oymo D., Haugerod T.
and Skane O. (1999) Magnesium. In: Ullmann's encyclopedia of industrial
chemistry, Vol. A 15 (ed. Elvers B., Hawkins S. and Schulz G.). 6 Edition. pp.
559ff. VCH.
Hydro Magnesium (1999a) Hydro Magnesium (1999a) Life Cycle Environmental Data for Production of
Magnesium Diecastings: Hydro's Process using Magnesite as Raw Material. Be-
cancour, CAN, Online-Version under: www.magnesium.hydro.com.
Hydro Magnesium (1999b) Hydro Magnesium (1999b) Life Cycle Environmental Data for Production of
Magnesium Diecastings: Hydro's Process using Seawater and Dolomite as Raw
Materials. Porsgrunn, NO, Online-Version under: www.magnesium.hydro.com.
Kammer et al. (2000) Kammer C., Kammer U., Ditze A., Scharf C., Wasserboden J., Schmid-Fetzer
R., von Buch F. and Riehmann W. (2000) Magnesium Taschenbuch. Alumin-
ium-Verlag, Düssedorf, D. ISBN 3-87017-264-9.
Klaus & Dinkel (1999) Klaus T. and Dinkel F. (1999) Stoffflussanalyse Schweiz: Fluorkohlenwass-
erstoffe und Schwefelhexafluorid. Carbotech, Basel.
Lang (1999) Lang S. (1999) Entwicklung einer neuen AlMg-Legierung für Aussenhaut-
Karosseriebleche mit verbesserter Umformbarkeit. Dissertation. TU München.

6. Literature
Maiss & Brenninkmeijer (1998) Maiss M. and Brenninkmeijer C. a. M. (1998) Atmospheric SF6: Trends,
Sources, and Prospects. In: Environ. Sci. Technol., 32(20), pp. 3077-3086.
Norsk Hydro (1999) Norsk Hydro (1999) Environmental Report 1998. Oslo, NO, Online-Version un-
der: www.hydro.com.
Schwarz & Leisewitz (1996) Schwarz W. and Leisewitz A. (1996) Aktuelle und künftige Emissionen treib-
hauswirksamer fluorierter Verbindungen in Deutschland. 106 01 074/01. Büro
für Umweltforschung und -beratung GmbH im Auftrag des Umweltbundesam-
tes, Frankfurt am Main.
Wilson et al. (1995) Wilson G., Claus K. G., Earlam M. R. and Hillis J. E. (1995) Magnesium and
Magnesium Alloys. In: Encyclopedia of Chemical Technology, Vol. 15 (ed.
Kroschwitz J. I. and Howe-Grant M.). 4 Edition. pp. 622ff. John Wiley & Sons,
New York.

Part XIV
Non Ferrous Metal Winning

Auxiliary Processes
Data v2.1 (2009)

Gabor Doka, Doka Ökobilanzen, Zürich
report ecoinvent data v2.1, No 10. EMPA Dübendorf, Swiss Centre for Life Cycle
Summary

This part corresponds to the original report in Althaus et al. (2004). Chapter “6 Cumulative Results and
Interpretation” of the original report has been deleted for the current update.
Besides of these reported changes no alteration of the original text took place for version 2.0. Some of the
reference flow names have been changed compared to the preceding versions. They have been updated with the
actual names of version 2.0.
For version 2.1, a new model for emissions of sulfidic tailing disposal is introduced. The dataset “disposal
sulfidic tailings, offsite” is changed and the report adapted accordingly.
Summary
In this part, the modelling of disposal of sulphidic and non sulphidic tailings and overburdens and that of mine
and smelter infrastructure is described. These modules are used in the production of different non ferrous metals.
The long term emissions from sulfidic tailings due to acid rock drainage are believed to be of major importance
for the LCI of metals from sulphidic ores. For version 2.1 a dedicated tailings disposal model was taken up.

Table of Contents
Table of Contents
SUMMARY ......................................................................................................................I
1 INTRODUCTION ....................................................................................................... 1
2 SYSTEM CHARACTERISATION................................................................................... 2
2.1 Land Transformation and Land Use in Mining and Beneficiation........................................ 3
2.1.1 Facilities ................................................................................................................................ 4
2.1.2 Direct land use....................................................................................................................... 4
2.1.3 Land use for waste rock disposal .......................................................................................... 5
2.2 Infrastructure in the non-ferrous metal industry.................................................................... 5
2.2.1 Process description................................................................................................................ 7
2.2.2 Non-ferrous metal, mine, underground in Ecoinvent............................................................ 8
2.2.3 Non-ferrous metal, mine, surface in Ecoinvent..................................................................... 8
2.2.4 Non-ferrous metal, smelter in Ecoinvent .............................................................................. 9
2.3 Disposal of Overburden and Tailings.................................................................................. 16
2.3.1 Process and infrastructure ................................................................................................... 16
2.3.2 Emissions ............................................................................................................................ 16
2.3.3 Disposal of Sulphidic Tailings in ecoinvent ....................................................................... 17
2.3.4 Disposal, non-sulphidic tailings, off-site in ecoinvent ........................................................ 23
2.3.5 Disposal of overburden in ecoinvent................................................................................... 23
3 LITERATUR ........................................................................................................... 27

1. Introduction
1 Introduction
Several production processes of the non ferrous metal industry are similar to each other. Namely the
winning of the metal bearing ore and the beneficiation processes depend mainly on throughput of
minerals. Process specific differences are rather linked to the type of metal carrying ore than to the
desired metals present in the ore. Most of the non ferrous metals originate from sulphidic ores which
present different implication for processing than e.g. oxidic ores. Different metals, e.g. nickel and
copper, often are coexistent in the sulphidic ore and are won jointly by the same metallurgical
processes. Also infrastructure of mines, beneficiation facilities, metallurgical plants and refineries do
not differ greatly from one metal winning process to another. Due to these similarities several
auxiliary processes and infrastructure are not differentiated further in ecoinvent. The modelled
disposal-processes for overburden and tailings are valid for all non-ferrous-metal industries. The only
difference is made for sulphidic tailings, which undergo acid rock drainage – reactions with enhanced
leaching and considerable acidification. In v2.1 of ecoinvent a dedicated model for sulfidic tailings
was introduced.

In this chapter several auxiliary processes of the non-ferrous metal industry are discussed:
- disposal processes,
- land use and
- infrastructure.
Fig. 2.1 gives an overview over these impacts discussed in this chapter and the modules in ecoinvent
they are inventoried in. Basically three types of activities can be distinguished when inventorying
mining operations.
Operation: Extraction of ore effects a direct type of impact. For each tonne concentrate excavated a
certain amount of land is transformed to become an extraction site for a certain time. Additionally
infrastructure like hauling trucks, bucket wheel excavator and conveyor belts are required. The
demand in these resources (land) and processes (infrastructure) are inventoried directly within the
respective mining module, they are direct proportional to the mass 1 of ore processed. Also the
leaching of metal from ore is considered as extraction, the land use for leaching – which is
considerable – therefore also is where appropriate inventoried within the mining module. In the case of
froth flotation the ore is ground in mills. The mill heads and jaws are severely strained, the
consumption in these parts is also inventoried as direct impact in terms of demand in austenitic
manganese steel (2.1.2 Direct land use).
Infrastructure: The infrastructure is inventoried as separate modules, each for underground mine,
open pit mine, beneficiation and smelter. These are pure infrastructure modules and comprise the
building of the mining, beneficiation, etc. facilities, roads, docks etc inclusively the disposal. These
modules do not refer to the direct impact in the mineral extraction site where the ore is mined, but to
the facilities for ore-processing, maintenance buildings, staff accommodation, etc. The functional unit
is infrastructure units per kilogram raw-material processed, thus the functional unit refers to the input
(2.2 Infrastructure in the non-ferrous metal industry).
Tailings / overburden: The extraction of metal ore is actually a concentration process. First, the ore
has to be separated from rock, be it to access the veins, be it to clean the extracted material from
unwanted fractions. This unwanted material is referred to as overburden. In a second step, the metal
bearing ore is concentrated further using different mechanical and/or chemical beneficiation methods.
The unwanted mineral fraction is referred to as tailings. Tailings consist of ground mineral and often
form a slurry after separation, which is disposed in large ponds lined with dams. Apart from the
considerable land use, also emissions arise from disposing the material. It is assumed, that the
disposed material is alien to the place where it is disposed (off-site) and therefore virtually all of the
substances leached over time have to be regarded as emissions. The functional unit of these modules is
the mass of tailings or overburden disposed (c.f. 2.3 Disposal of Overburden and Tailings).
Rehabilitation of tailings: In this study no rehabilitation of the tailings and overburden disposal site
is considered. It is assumed, that the disposal site never will turn to its natural state and consequently Kommentar [gd1]: d.h.
no energy and materials are used for rehabilitation. This situation probably presents the norm in a infinite land use?
global view.
1
This assumption might be biased in the case of infrastructure like processing plants. The choice of the material’s volume
would probably more correct. Since the density is not exactly known in most cases and the difference in densities is expected
to be small from ore to ore, the mass is chosen as functional unit.
mining datasets in respective metals part

operation
open cut mining

heap leaching
transformation to
mineral extration
site
Mining and benefication of metal containing
ores
occupation
mineral
extraction site
infrastructure:
howling trucks,
conveyor belts
infrastructure tailings / rehabilitation

overburden tailings
rehabilitation transformation transformation
infrastructure to building site to dump site transformation
to unknown
transformation to
unknown
occupation occupation input: energy,

input: energy, building site dump site material
material
infrastructure: emissions: acid

disposal of long term
mining facilities, rock drainage,
facilities emissions
flotation mill leaching
- non-ferrous metal smelter - disposal, non-sulfidic tailings, off-site

- non-ferrous metal mine, underground - disposal, non-sulfidic overburden, off-site
- non-ferrous metal mine, surface - disposal, sulfidic tailings, off-site
Fig. 2.1 Overview over infrastructure, land use and rehabilitation of the mineral extraction site. Rehabilitation of
infrastructure (production site) and tailings are not considered in this study.
2.1 Land Transformation and Land Use in Mining and

Beneficiation
The effects of mining and beneficiation to land resources can be differentiated into three types:
- Land used by facilities (roads, buildings, stationary equipment)
- Direct land use trough exploration and beneficiation (excavation, eventually heap leaching)
- Land used for disposal purposes (stockpiles, tailings)

These three effects are treated in ecoinvent in different modules. As no detailed data for the working
of different ore was available, the same values were used for all the modules dealing with non-ferrous
metals.

2.1.1 Facilities
The land use of the facilities comprise roads, buildings, stationary equipment like screens, feeder,
conveyor belts, docks, crusher, mills, cyclones, etc. A good dependency of the Area covered with the
yearly production capacity of the operation is found in Martens et al. (2002). Results and assumptions
are shown in Tab. 2.1. This land use is inventoried in the infrastructure modules
- “non-ferrous metal mine, surface”
- “non-ferrous metal mine, underground”
- “non-ferrous metal smelter”
Tab. 2.1 Specific land use by facilities in open cut and underground mining. These figures comprise Beneficiation
2
facilities. Data for land use is a function of the production capacity as described in (Martens et al. (2002)).
Infrastructure Unit Open cut Underground

Yield, capacity t/a 400'000 400'000
Lifetime a 50 50
2
Land use m 122'011 184'830
Total yield, lifetime t 20'000'000 20’000’000
Specific land use m2/t 0.0061 0.0092
land occupation a m2/t 0.305 0.462
2.1.2 Direct land use

According to Martens et al. (2002) a total land transformation of 20 m2 per thousand tonnes mined
copper ore is effected. This figure of 0.02 m2/t raw ore is taken as proxy for all non ferrous metals
except for aluminium, tin and magnesium. It includes the effects of tailing ponds (42% of the land
transformation), leaching heaps (17%), stockpiles (13%), open cuts (10%), facilities (16%) and
lowering of the ground through underground mining (2%), covering world-wide mining activities
including open cut (80% of the ore) and underground (20%). In this study the lowering of the ground
is assumed not to be ecologically effective. Where heap-leaching is effectuated, the value for land
transformation adds up to 0.0059 m2/t ore, whereas no direct land use for underground mining is
considered (Tab. 2.2).
Tab. 2.2 Land use through direct exploration of the mine. A lifetime of 30 years was assumed. The lowering of the
ground through underground mining is not regarded as ecologically relevant and thus is not inventoried.
Specific land use open pit

transformation occupation
2 2
Unit m /t ore a m /t ore
Leaching heaps 0.00340 0.102
Open cut 0.00250 0.075
Total 0.00590 0.177
2
Log-log dependency: Area open cut[m2] = 23.869 Yield open cut[t/a]0.662 and Area underground[m2] = 0.322 Yield underground[t/a]1.028
2.1.3 Land use for waste rock disposal

In mining large fractions of overburden has prior to extraction to be disposed on stockpiles. For open
cut operations the ratio between ore and overburden ranges e. g. for nickel typically from 0.3 to 2
(Hilbrans & Hinrichs (1999)). Tailings consist of the unwanted part of the ground ore (gangue) which
is separated and usually is disposed as slurry in large ponds.
According to reported values for tailing ponds in IPPC (2002), an average land use of 0.051 m2/t and
6.62 m2a/t tailings disposed is calculated (Tab. 2.3) assuming an average mine operation of 30 years.
The land transformation is assumed to be from “unknown” to “disposal site” for the duration of 100
years after which it is rehabilitated in a natural way. Assuming that the whole disposal area for the
tailings and overburdens has to be cleared prior to any disposal activities, i. e. at the very start of the
mine operation, the duration of the whole land use is 130 years.
Tab. 2.3 Assumptions and calculations regarding land use and eventual rehabilitation from tailings. Duration of mine
operation phase is 30 years. The tailings are covered with 0.8 m of earth and a density of 0.8 kg / l. After 100
year the disposal site is assumed to be rehabilitated naturally.
Dimensio Tailings Lifetime Land Land Coverage Coverage

ns Production transformat occupation
ion
2 2
[ha] [t / a] [Mt] [m / t] [m ·a / t] [t] [kg / t]
Grapengerg 35 455’000 13'650’000 0.026 3.33 2268 0.17
Aitik 1450 17’700’000 531’000’000 0.027 3.55 93960 0.18
Pyhäsalmi 100 180’000 5’400’000 0.19 24.1 6480 1.2
Zinkgruvan 66 425’000 12’750’000 0.052 6.73 4276.8 0.34
Total
geometric 26’580'000 0.051 6.62 8766 0.33
mean
2.2 Infrastructure in the non-ferrous metal industry

For the inventory of the aluminium production a detailed estimation of the sintering and electrolysis
has been made in this study. These figures together with the inventories of the blast furnace, electric
arc and blast oxygen converter of the inventory of the iron production built the base for infrastructure
estimation of the non-ferrous metal production.
Since it can be assumed that the environmental impact due to infrastructure is very small compared to
the energetic intensive, high-throughput production processes in concern, a non-specific module for
mining, beneficiation, smelter and refinery is set up. Tab. 2.4 gives an overview over the process-steps
involved in the non-ferrous metal industry. Depending on the prevailing of a process step and raw
material input the factor is calculated.

Tab. 2.4 Process-steps involved in the non-ferrous metal modules.
mining Hydro- electric pyrometal

concen- mining Benefi- Metallo-
metal / ore Raw material yield under- sintering metallurgi- arc lurgical refinery
tration surface ciation thermal
ground cal furnace blast
Mn – ore Rock 35.7%
69.2% 30% 70% no - - - - - -
(14.2%)
Cr – ore Rock 25.5%
79% 79% 21%
(11.6%)
Pb/Zn – ore Rock 5 / 9% 68 / 85% 100
Pb – metal Concentrate 53% 90% yes 100% yes
Zn – metal Concentrate 53% 90% no 80% 20% no
PGM -ore Rock 3.E-04% 90% 100% no yes
PGM – metal Concentrate 5.1E-3% yes yes yes yes
Cu/Mo – ore Rock 0.99% 81% 30% 70% yes yes
Cu – primary Concentrate 29.7% 96.8% Yes 10% yes yes no
Cu – SX/EW Rock 0.99% 70% no 100% no yes yes Yes
Cu – second. Scrap / blister 78%* 90* yes Yes
Cu – blister Concentrate 24.7% 97.8% yes yes
Ni – ore Rock 1.13%
82% 100% yes
(0.76%)
Ni – metal concentrate 8.13% 97% yes yes, div. no
Ni – ferronickel Rock 1.98%
57.5% yes yes yes
(1.04%)
Cr – ferro-Cr Concentrate 30.7% 91.2% no 100% no
Mn – ferro-Mn Ore 42.4% 76.1% yes 83% 17% No
Cr – metal Diverse Electrolysis Alumino-
Diverse 91%
25% thermic 75%
Mn – metal Ferro-Mn / ore 74.5% 60% 25% 75% No

2.2.1 Process description

The following paragraphs discuss the main process steps in the non-ferrous metal industry.
Mining: Mining is differentiated into open-pit mining and underground mining. Depending on the
hardness, the ore can be mined in some cases without explosives. Such soft ore is directly extracted
with bucket loader, whereas hard material has to be blasted. Techniques in underground mining do not
differ vastly from those in surface mining, especially if the layer mined is high enough for big
vehicles. Usually vehicles used underground tend to be smaller with adaptation to low height and low
off-gas emissions. Another difference between surface and underground mining is the need for air
circulation and cleaning. Hence, beside a few specific features – like hoists in underground mining –
no major differences exist.
Beneficiation: Most ores are subjected to a first concentration step after being mined. In some cases
the only beneficiation possible is a mere classification by size and/or gravity, which yields only a little
concentration effect. The beneficiation step of these ores is not considered in this study. In most other
cases a beneficiation including a froth flotation is carried out: After mining, the ore is first ground. In a
next step it is subjected to gravity concentration to separate the metallic particles from the metal
bearing minerals. After this first concentration step, flotation is carried out to remove the gangue from
the sulphidic minerals. For this lime and several organic chemicals are added. Tailings usually are led
to tailing heaps or ponds. Flotation mills consist of several crushers and screens, ball and rod mills,
belt feeder, pumps, filters, float cells, thickener and usually a magnet, dust collector and buildings –
truck shop, crusher building, concentrator building and reagent building. In this setting it resembles
the hydroxide plant in the aluminium production where the ore also is ground and cleaned form
gangue. An example of a medium sized beneficiation plant (flotation mill) is shown in Fig. 2.2.
The beneficiation infrastructure in this study is not covered with a specific module, but approximated
with the inventory of the aluminium hydroxide plant, which itself is a beneficiation (specifications:
Medium sized plant. Output: 1'220'000 t/a, life time: 50 years). If the inventory of this module is
related to its input rather than to its output, the demand has to be corrected by a factor of 1.40, i. e. the
production of 1 kg aluminium hydroxide requires 1.4 kg bauxite (ore in the aluminium production).
The beneficiation of 1 kg ore therefore requires 8.40 10-12 units of the module “aluminium hydroxide,
plant”. IPPC (2001) reports a demand of 300 g steel due to abrasion of the mill jaws per tonne ground
ore, the steel being inventoried as “chromium steel 18/8, at plant”.
Metallurgy: The process routes to produce different non-ferrous metals are very divers. As seen in
Tab. 2.4 the metallurgical steps usually consist of a sintering / roasting step followed either by a
pyrometallurgical or an electrolytic step. The pyrolytical step usually consists of a smelter followed by
a converter. In some cases an additional refining step is necessary to yield the grade necessary for
trade. The products usually leave the plant after re-melting, casting or preparation steps as powder,
ingots, etc. An example of a metallurgical process is given in Fig. 2.4 (copper production).
Refining: Some metals like the platinum group metals are won from residues from electrolysis and
other residues like dross or from scrap containing some metal values. These Refineries resemble
chemical plants where metals subsequently are separated from each other using techniques as
solvent—solvent extraction or precipitation/filtration.
Refineries are approximated with the module of “chemical plant, organics”. 4 • 10-10 units of this
infrastructure module are afforded per kg chemicals. This value is taken also in this study for all those
processes, where the refined metal is won by a cascade of purification processes. This is true for the
Platinum group metals. For reasons of transparency, no specific module is made for the non-ferrous
metal industry. If appropriate in a module, the refinery infrastructure is inventoried as “chemical plant,
organics” and clearly is marked as proxy.

Fig. 2.2 Picture of a flotation mill with crusher, concentrator, thickener and several flotation cells. The output of the
mill is stated to be 6’000 – 10’000 t per day (around 2.5 Mt per year).
2.2.2 Non-ferrous metal mine, underground in Ecoinvent

The infrastructure use for a mine is estimated in Tab. 2.5. It comprises land use for mining facilities,
production of buildings and equipment including their disposal. No rehabilitation of the industrial site
is considered. The values for the surface mine differ only in specific points from those for the
underground mine. The data represent a rough estimate that is based on qualitative information like
pictures and process descriptions of actual mines.
Land use and occupation is chosen as explained in 2.1.1 Facilities, it thus comprises also the
beneficiation facilities.
2.2.3 Non-ferrous metal mine, surface in Ecoinvent

The infrastructure use for a mine is estimated in Tab. 2.6. It comprises land use for mining facilities,
production of buildings and equipment including their disposal. No rehabilitation of the industrial site
is considered. The values for the surface mine differ only in specific points from those for the
underground mine. The data represent a rough estimate that is based on qualitative information like
pictures and process descriptions of actual mines.
Land use and occupation is chosen as explained in 2.1.1 Facilities, it thus comprises also the
beneficiation facilities.

Tab. 2.5 Infrastructure mine, underground. Assumptions made and inventory chosen.
Mine, underground Demand Unit Material / Module Life time Life time demand
Yearly production capacity 400'000'000 kg Yearly production capacity 50 a 400'000'000
Area needed total [m2] 184'830 m2 Transformed area 50 a 184'830
Administration building 25'000 m3 Building, multi-storey 50 a 25'000
3
(50x50x10 m )
conveyor to storage [m] 300 m conveyor belt, at plant 25 a 600
2
Storage building (hall for 5'000 m Building, hall 50 a 5'000
100'000 t = 50'000 m3;
2
average height 10m) [m ]
10 Hoists [kg] 100'000 kg Industrial machine, heavy 25 a 200'000
2
3 Truck shop (50x50) 7'500 m Building, hall 50 a 7'500
3
2 Warehouses 50'000 m Building, multi-storey 50 a 50'000
(50x50x10 m3)
Accommodation building 100'000 m3 Building, multi-storey 25 a 200'000
(100x100x10 m3)
Fix Equipment 25'000 kg Industrial machine, heavy, 25 a 50'000
(cranes, screens)
Mobile Equipment 120'000 kg Industrial machine, heavy 10 a 600'000
(vehicles)
Tab. 2.6 Infrastructure mine, surface. Assumptions made and inventory chosen.
Mine, surface Demand Unit Material / Module Life time Life time demand
Yearly production capacity 400'000'000 kg Yearly production capacity 50 a 400'000'000
2 2
Area needed total [m ] 122'011 m Transformed area 50 a 122'011
3
Administration building 25'000 m Building, multi-storey 50 a 25'000
(50x50x10 m3)
conveyor to storage [m] 3'000 m conveyor belt, at plant 25 a 6'000
Storage building (hall for 5'000 m2 Building, hall 50 a 5'000
3
100'000 t = 50'000 m ;
2
average height 10m) [m ]
4 Truck shop (50x50) 10'000 m2 Building, hall 50 a 10'000
2 Warehouse 50'000 m3 Building, multi-storey 50 a 50'000
(50x50x10 m3)
3
Accommodation 100'000 m Building, multi-storey 25 a 200'000
3
(100x100x10 m )
Fix Equipment 5 x 5t 25'000 kg Industrial machine, heavy 25 a 50'000
(cranes, screens)
Mobile Equipment 12 x 10t 120'000 kg Industrial machine, heavy 10 a 600'000
(vehicles)
2.2.4 Non-ferrous metal, smelter in Ecoinvent

The infrastructure for the metallurgical step is inventoried as one unspecific module valid for a wide
range of metal industries, including hydro- and pyrometallurgical processes. Its functional unit is [kg]
ore concentrate in input. In this study the infrastructure of the metallurgical step is approximated
roughly with:

1. sintering / roasting furnace (rotary kiln)
2. electrolysis plant (50% of the cases) and
3. blast furnace, followed by a electric arc converter (50% of the cases)

These steps build up the system of metallurgical facilities to process a certain amount of ore
concentrate. This is shown in Fig. 2.3. With an average input of 6’775'000 t ore concentrate this
module is required in the inventories with a factor 1.48 10-10 per kg concentrate processed.
Step 1) is inventoried using the infrastructure of the aluminium oxide plant, which itself represents a
roasting (oxidising) step. Therefore the module “aluminium oxide, plant” (Medium sized plant.
Output: 800'000 t/a, life time: 50 years) is chosen, which has to be scaled by factor 8.5 to produce the
roasted concentrate required.
Step 2) represents the hydrometallurgical process route for which a fraction of 50% of the concentrate
roasted is chosen, i. e. as an approximation it is assumed that one half of the metal is produced
hydrometallurgically. The required infrastructure is approximated with the electrolysis of aluminium,
“aluminium electrolysis, plant” (Medium sized plant. Output: 130'000 t/a, life time: 50 years. Site is
shared with casting plant. Total land use is allocated 50% to electrolysis, 50% to casting). Also this
module has to be scaled with a factor of 17.
Step 3) is also weighted with 50% and includes two steps, the smelting and a subsequent converting.
Here too, existing infrastructure modules from the iron industry are chosen as proxy for the smelting
furnace and the converter. In this case it is the module “blast furnace” (Assumptions for medium sized
plant. Output: 1'500'000 t/a, life time: 50 years. Site is shared with BOF converter.) and “blast oxygen
furnace converter” (Output: 1'500'000 t/a, life time: 50 a). The yield for the blast furnace is 91%
(including sintering and smelting), yielding a factor of 1.22.
In most cases – as shown in Fig. 2.4 – the off-gas is captured and the sulphur dioxide from the roasting
/ smelting is used in the production of sulphidic acid. It is assumed that 0.7 kg H2SO4 is produced from
the off-gas of 1 kg concentrate roasted (Ayres et al. (2002)). Since most of the ores are of the sulphidic
type, and also non-sulphidic ore smelters have elaborate abatement technologies for off-gas, additional
1.90 units “chemical plant, organics” (lifetime: 50 a; production: 50'000 t / a).

52% yield
17.0 x
Electrolysis (130'000
1'153'281 t / a
t / a)
65% yield 50%
ore input: 8.47 x 2'214'300 t / a metal output
sintering /
6'775'758 t / a roasting 3'660'000 t / a pure metal 2'653'281 t / a
(800'000 t / a)
39% conc. 4'428'600 t / a 1.22 x 1x
50%
Blast furnace converter
in: 2'013'000 t / a 1'830'000 t / a

out: 1'830'000 t / a 1'500'000 t / a
yield: 91% 82%
Fig. 2.3 Scheme of the infrastructure modules building up the metallurgical step. The modules are scaled to a
specific throughput, which is set to be 1'500'000 t output per year at the converter. The modules for
sintering/roasting, electrolysis and blast furnace are scaled to meet the requirements. The concentration in
the ore is calculated to be 39%, which is true for many metals.
Fig. 2.4 Example of a pyrometallurgical process step in the copper industry 3 . Off-gas is captured in the sulphuric
acid plant. The smelting furnace is followed by a flash converter, the metal leaves after a final refining step.
This final refining is part of the metallurgical step and is not to be confused with the process denoted as
refining in this study, which is more of a chemical plant than a smelter.
3
Picture provided by Outokumpu Inc., http://www.outokumpu.com, accessed at 01.03.2003
Tab. 2.7 Ecoinvent meta information for non-ferrous-metal mines and smelters.
non-ferrous metal non-ferrous metal non-ferrous metal

Name
mine, surface mine, underground smelter
Location GLO GLO GLO
Unit unit unit unit
Includes the mine facilities. Includes the mine facilities. Includes facilities for
Does not include the land Does not include the land roasting, electrolysis and for
Included Processes
used for the actual mining used for the actual mining the blast furnace and
operations operations converter processes.
Amount 1 1 1
Buntmetall Mine,
Local Name
Buntmetall Mine, Tagbau Untertagbau Buntmetall Hüttenwerk
Synonyms
The module is designed for
unspecific use. It is based
on inventories from
General Comment to The module is designed for The module is designed for aluminium and iron industry.
reference function unspecific use. It is based unspecific use. It is based Lifetime: 50a, input:
on rough estimations. on rough estimations. 6'750'000 t ore concentrate
Lifetime: 50a, output: Lifetime: 50a, output: (39%), output: 2'650'000 t
400'000 t ore per year 400'000 t ore per year metal per year
CAS Number
Start Date 2003 2003 2003
End Date 2003 2003 2003
Period 1 1 1
Other Period Text
No geographical context, No geographical context, No geographical context,
Geography text inventory based on inventory based on inventory based on
calculations calculations calculations
Infrastructure modules from
aluminium and iron
production are chosen as
proxy: aluminium oxide,
plant / aluminium hydroxide,
Technology text
plant / electrolysis, plant /
blast furnace / blast furnace
converter. Their outputs
have been scaled to fit
Assumed technology Assumed technology average operations.
Production Volume nA nA nA
Sampling Procedure Literature Literature Literature
see Geography and see Geography and see Geography and
Extrapolations
Technology Technology Technology

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
name in
General
Output
Modul
Name
Input
Infra-
Type
Sub-
95%
Unit
Land trans-
Transformation, from (nA,nA,nA,nA,nA,n
formation, initial  resource land 1.85E+05 m2 rough estimate 1 5
non-ferrous metal mine, underground

unknown A,8); estimate
state
Land trans-
Transformation, to (nA,nA,nA,nA,nA,n
formation, to  resource land 1.85E+05 m2 rough estimate 1 2
industrial area A,8)
mining facilities
Occupation, (nA,nA,nA,nA,nA,n
Land Occupation  resource land 9.24E+06 m2a rough estimate 1 2
construction (nA,nA,nA,nA,nA,n
Bulidings  buildings building, multi-storey 2.75E+05 m3 rough estimate 1 3
processes A,9)
Halls  buildings building, hall 1.25E+04 m2 rough estimate 1 3
processes A,9)
industrial machine,
Machines  machinery heavy, unspecified, 8.50E+05 kg rough estimate 1 3
processes A,9)
at plant
construction conveyor belt, at (nA,nA,nA,nA,nA,n

conveyor belts  machinery 6.00E+02 m rough estimate 1 3
processes plant A,9)
underground mining Lifetime: 50a, output: 400'000 t ore non-ferrous metal

 metals extraction 1.00E+00 unit
facilities per year mine, underground
Fig. 2.5 Flows for “non-ferrous metal mine, underground” and its representation in the ecoinvent database.

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
name in
General
Output
Modul
Name
Input
Infra-
Type
Sub-
95%
Unit
Land trans-
Transformation, from (nA,nA,nA,nA,nA,n
formation, initial  resource land 1.22E+05 m2 rough estimate 1 5
unknown A,8); estimate
state
non-ferrous metal mine, surface

Land trans-
Transformation, to (nA,nA,nA,nA,nA,n
formation, to  resource land 1.22E+05 m2 rough estimate 1 2
mining facilities
Occupation, (nA,nA,nA,nA,nA,n
Land Occupation  resource land 6.10E+06 m2a rough estimate 1 2
Bulidings  buildings building, multi-storey 2.75E+05 m3 rough estimate 1 3
processes A,9)
Halls  buildings building, hall 1.50E+04 m2 rough estimate 1 3
processes A,9)
industrial machine,
Machines  machinery heavy, unspecified, 6.50E+05 kg rough estimate 1 3
processes A,9)
at plant
construction conveyor belt, at (nA,nA,nA,nA,nA,n

conveyor belts  machinery 6.00E+03 m rough estimate 1 3
processes plant A,9)
open-pit mining Lifetime: 50a, output: 400'000 t ore non-ferrous metal

 metals extraction 1.00E+00 unit
facilities per year mine, surface
Fig. 2.6 Flows for “non-ferrous metal mine, surface” and its representation in the ecoinvent database.

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
name in
General
Output
Modul
Name
Input
Infra-
Type
Sub-
95%
Unit
Roasting / aluminium oxide,
 metals extraction 8.47E+00 unit rough estimate 1 5 estimate
Sintering plant
non-ferrous metal
Electrolytical aluminium
 metals extraction 1.70E+01 unit rough estimate 1 5 estimate
plant (50%) electrolysis, plant
smelter
Smelter (50%)  metals extraction blast furnace 1.22E+00 unit rough estimate 1 5 estimate
blast oxygen furnace

Converter (50%)  metals extraction 1.00E+00 unit rough estimate 1 5 estimate
converter
chemical plant,
Off-gas treatment  chemicals organics 1.90E+00 unit rough estimate 1 5 estimate
organics
Lifetime: 50a, input: 6'750'000 t ore
non-ferrous metal
 smeltery facilities concentrate (39%), output: metals extraction 1.00E+00 unit
smelter
2'650'000 t metal per year
Fig. 2.7 Flows for “non-ferrous metal smelter” and its representation in the ecoinvent database.

2.3 Disposal of Overburden and Tailings

This chapter deals with ecological burdens due to the disposal of sulphidic overburden and tailings
from mining and Beneficiation. For modelling these deposits, three separate modules are generated:
- “disposal, sulfidic tailings, off-site”
- “disposal, non-sulfidic tailings, off-site”
- “disposal, non-sulfidic overburden, off-site”

Major impacts are emissions to water, land transformation and land use. Because of the acidic
environment in the tailings, a considerable leaching occurs, whereas the overburden remains more or
less inert. Overburden, which is refilled, is considered to have no ecological burden.
2.3.1 Process and infrastructure

To access an ore deposit, overburden has to be removed first. This material does not contain ores,
therefore is not processed and is usually deposited as crushed rock or finely ground meal but
chemically unchanged, either in the mining pit (backfill) or on piles. Overburden often is used as
building material for the lining of the tailings ponds.
The mined ore then is separated from the gangue, i.e. from the unwanted minerals. This usually is
done by grinding and flotation. The resulting slurry then is transported to so-called tailings stacks.
This disposed material covers a sometimes considerable area, it often contains sulphur, ore-borne
metals and different reagents from the beneficiation step. The sulphide minerals in the tailings are
oxidised to sulphuric acid, which leads to a low pH and high degree of leaching of metals. Tailings
stacks are about 20 m in thickness. Its borders are lined by dams and the excess water is led to a
cascade of ponds for further settling of solids. In the tailings ponds the slurry consequently sediments
and thickens. If the dam height is reached, the dams usually get enhanced, finally up to several tens of
metres high (IPPC (2002)).
After the mines operation phase, these ponds are usually rehabilitated. Often the tailings pond is
covered with earth, while the cleaning ponds stay to form natural surface waters.
2.3.2 Emissions
Water
The overburden is assumed to emit no relevant pollutants. The low pH in sulphidic tailings however is
a distinctly different chemical environment than the overburdens original location. Also tailings are a
finely ground rock meal, mixed with water and exposed to atmospheric oxygen. This leads to
additional emissions from formerly inert minerals.
The tailings are often subject to acid rock drainage (ARD), where sulphur oxidises in the presence of
metal, water and oxygen promoted by bacterial activity to form sulphuric acid. The reactions are
complex and persistent. Mines worked by the Romans in Spain and the Vikings in Scandinavia
continue to generate ARD today. This makes it an even bigger issue than cyanide leakage, which is
chemically unstable and does not persist in the environment. Once exposed to oxygen ARD reactions
on mineral surfaces will continue for thousands of years.
Air
No considerable emissions to air are reported. In arid climates wind-blown erosion can be a relevant
emission pathway for tailings.

2.3.3 Disposal of Sulphidic Tailings in ecoinvent

Tailings are a slurry waste produced during metal ore refining (beneficiation). Tailings are usually
poured into tailings impoundments, where water is removed and the slurry solids are stacked. Tailings
can contain sulfides, which are biochemically oxidised to sulfuric acid. The acidic pH then promotes
washout of many metals (Acid Rock Drainage, ARD). Large masses of tailings are produced per
kilogram metal especially if ore grades are low. Tailings impoundments represent a considerable
environmental burden in the primary metal process chain, because tailings contain large amounts of
toxic metals and large masses of tailings are produced: the current annual worldwide production of all
types of tailings is estimated to be several hundred million metric tonnes per year, which surpasses the
mass naturally eroded from the earths surface by rivers (Gardner & Sampat 1998). In the ecoinvent
database v2.0 typically sulfidic tailings are produced in production of primary copper, nickel, zinc,
lead, cobalt, molybdenum, and platinum group metals. Tailings from aluminium ore beneficiation
(redmud) are heeded in the respective chapter. Tailings from uranium ore beneficiation is heeded in
the nuclear process chain (ecoinvent report 6, part VII).
A dataset for sulfidic tailings was created in ecoinvent v1 in 2003. However, this dataset only included
land use and land transformation aspects, but no emissions for a lack of reliable composition data and
tailings deposit models. For ecoinvent v2 this data gap is filled with emissions data from a specifically
designed tailings model. This complemented dataset should provide a first informed estimate of the
full magnitude of sulfidic tailings burdens for non-ferrous metal products and encourage further
research for more detailed models for the ecoinvent v3. In order to avoid overestimations, the
modelling assumptions – within the ecoinvent framework – were chosen to be conservative. An
extended report on the sulfidic tailings model is freely available online (Doka 2008).
Tailings landfill model
The tailings landfill model was based on the modelling framework applied in other ecoinvent landfill
process models (ecoinvent report 13: Doka 2003). These models are based on a top-down approach.
Literature data on measurements of deposited compositions and currently measured concentrations in
landfill leachate are used to calculate the currently occurring emissions of the landfill. Heeding chief
parameters like hydrology and pH development these emissions are then projected into the future.
Thus these models incorporate physical and chemical conditions, while being rooted in the currently
observed emission behaviour. More details on general modelling assumptions of these models can be
taken from (Doka 2003). The resulting tailings model and its transfer coefficients are not applicable to
other waste types.
A large database with literature data on tailings composition and measured tailings leachate
concentrations mainly from copper, zinc, lead, nickel and molybdenum mining sites is assembled 4 .
The geometric mean of established data is taken to represent a global average sulfidic tailings
composition and leachate composition. The relation between leachate concentration and tailings
composition quantifies how easily this particular element is transferred to the aqueous phase and thus
how high its mobility is. To make comparisons with other landfill models, elemental transfer
coefficients for the tailings landfill are calculated, although this is currently not necessary, as only one
tailings material composition is considered here. Roughly 2600 individual data points are used to
create the model of tailings emissions for 39 chemical elements.
4
Data was taken from ATSDR 1997; ATSDR 2006; Boulet & Larocque 1998; Dorronsoro et al. 2002; Downing & Gravel
2006; EPA 1994; Garcia-Meza et al. 2006; Hadjibiros et al. 2005; Hernandez et al. 2007; IAEA 2003; IPPC 2004; Koerth
2002; Laliberte & Tremblay 2002; Lazar 2002; Matheis et al. 1998; McLemore et al. 2006; MIMI 2004; Muwanga 1997;
Nash 2000; OK Tedi 2002; Robinson et al. 2004; Thoms 2006; Tyler et al. 2004; Van Zyl 2002; Wels et al. 2000; Ye et al.
2002
Based on acid-base balances, it is assumed that no pH-induced changes in leachate concentrations

occur. I.e. the currently observed emission behaviour is taken to represent the average long-term
situation. This is a conservative approach, since the literature data includes data from impoundments
that have not, or not yet developed an acidic pH. The long-term mobility of metals tends therefore to
be underestimated.
Chemical weathering of tailings depends on climate conditions. In wet climates large volumes of
downward leaching through the tailings body are possible. In arid climates leaching is slow or in
reverse direction, because water evaporation transports soluble tailings materials to the surface.
Surface erosion by wind is a relevant emission pathway in arid climates (cf. Fig. 2.8). In this study,
however, only emissions via bottom leaching are considered.
While the landfill models in (Doka 2003) were based on Swiss climate, for the tailings model the
climate and precipitation that is relevant to tailings sites is heeded. Based on the location of the world's
largest producers of sulfidic tailings, a weighted average precipitation rate of 770 mm/a was calculated
(Spiess 2002, FAO 2000). In a global average situation, tailings deposits are rarely vegetated due to
the extreme type of ground. Evapotranspiration of precipitated water is therefore neglected. It is
assumed that tailings deposits are ideally constructed in such a way that mechanical erosion by surface
runoff is excluded. All precipitation water therefore infiltrates the landfill body. All major mining sites
are outside permafrost regions, so that a normal water circulation is possible. The height of the tailings
landfill is assumed to be 50 meters. With a tailings density of 2200 kg/m3 a mass of 110'000 kg
tailings is deposited per square meter tailings landfill surface. Thus, an amount of 0.007 litres
precipitation water per kilogram tailings infiltrates annually.
Fig. 2.8 Wind blown tailings in Clayton mine, Idaho, USA Fig. 2.9 Nickel Tailings in Sudbury Ontario, Canada
(Hammarstrom et al. 2002) (Burtynsky 1996)
A part of the infiltrated water flows preferentially through the tailings. Due to the fine grain size of
tailings, the amount of preferentially flowing water is assumed to be only 5%, and the residence time
of preferential water is assumed to be 18 weeks (=Tp). The water content of stabilised tailings is
assumed to be 30 w-%. It is assumed that preferentially flowing water has little contact with the
tailings materials and carries no pollutants from the landfill 5 . Based on model formulas given in Doka
5
If landfills dry out periodically, then preferential water might wash out evaporites and be significantly burdened. This effect
is not heeded here.
(2003:III.30ff.) it follows that the amount of effective, i.e. polluted annual leachate volume is
0.0067 litres per kilogram tailings (=Veff).
Transfer coefficients are calculated for two time horizons, 100 years (short-term) and 60'000 years
(long-term). In Doka (2003) a long-term time frame of 60'000 years was chosen to represent an
'ecological planning horizon': by that time the Swiss midlands will probably be covered again with
glaciers, that remodel the Swiss landscape. At that time the Swiss ecosphere will be redefined, creating
new boundary conditions for environmental goals. The emissions over 60'000 years therefore
represent a coarse mean expectation value of the burden inflicted on the ecosphere as we know it now
and are concerned about. In other regions this ecological planning horizon might be dissimilar.
Especially in tropical regions very long periods of climatic stability can be expected 6 . For simplicity a
long-term time frame of 60'000 years for all tailings sites was chosen.
Due to lack of data, certain transfer coefficients are estimated by chemical similarity. Transfer
coefficients for halogens are based on data from fluorine. Transfer coefficients for nitrogen are based
on sulfur. Transfer coefficients for scandium are based on the arithmetic mean of antimony and
selenium. Transfer coefficients for tungsten are based on the arithmetic mean of molybdenum and
selenium. All these proxies are for completeness only and make up only very minor contributions to
the LCIA results for this dataset (less than 0.1 percent).
Even after 60'000 years only a part of the deposited tailings are leached. With the modelled transfer
coefficients roughly 8% of the toxic potential in tailings is emitted, and 92% are retained in the
impoundment.
Comparison with residual material landfill
The transfer coefficients of the tailings landfill can be compared with the transfer coefficients of a
residual material landfill 7 from (Doka 2003), as shown in Fig. 2.10.
100.000%
0.000% 0.001% 0.010% 0.100% 1.000% 10.000% 100.000%
P
10.000%
S N
B
Mo
W Oxianions
Ca
1.000% Other solubles
Mg Sr Na
Other elements
Be Se
Cd Co Sc
Mn Ti Cl Br I
F 0.100%
ZnSn Ni K
Tl
Cu
V Cr
Fe Sb As 0.010%
Ag Al
Hg
Pb
Ba Si 0.001%
0.000%
short-term transfer coefficients residual material landfill (100a)
Fig. 2.10 Comparison of short-term transfer coefficients for residual material landfill (x axis) and tailings deposit (y
axis)
6
In such regions ecological planning horizons might be defined by very rare events like climate changes after meteorite
impact (cf. Cretaceous-Tertiary impact winter 65 million years ago) or by very slow processes (glaciation after continental
drift). For example, about 400 million years ago most regions that are tropical today were close to the south pole. However,
currently tropical regions drift towards the equator and not towards the poles.
7
The Swiss residual material landfill is for inert, inorganic waste materials sometimes solidified with cement.
Tab. 2.8 Average sulfidic tailings composition from Tab. 2.9 Modelled short- and long-term transfer
literature survey coefficients for the sulfidic tailings deposit
given as "kg emitted per kg deposited"
Element Short-term Long-term

transfer transfer
Element Content Data Geometric coefficients coefficients
in tailings points Standard (100a) (60'000a)
mg/kg (n) Deviation Sulfur 12.7% 100%
(geomean) Nitrogen 12.7% 100%
S 14600 57 1670% Phosphor 39.4% 100%
N 23.1 3 11600% Boron 8.77% 100%
P 493 85 384%
Chlorine 0.195% 69%
B 310 1 n.a.
Bromium 0.195% 69%
Cl 2 1 n.a.
Fluorine 0.195% 69%
Br 0.229 3 208%
Iodine 0.195% 69%
F 826 30 376%
Silver 0.00457% 2.74%
I n.a. 0 n.a.
Arsenic 0.00666% 3.92%
Ag 14.4 54 688%
As 284 87 919% Barium 0.000991% 0.595%
Ba 460 107 290% Cadmium 0.218% 100%
Cd 7.16 84 617% Cobalt 0.185% 100%
Co 18.6 62 404% Chromium 0.00898% 5.24%
Cr 42.6 114 234% Copper 0.0335% 20.1%
Cu 277 142 540% Mercury 0.00317% 1.9%
Hg 0.882 62 757% Manganese 0.159% 95.5%
Mn 875 109 426% Molybdenum 2.3% 100%
Mo 7.02 61 371% Nickel 0.0653% 39.2%
Ni 22.8 96 487% Lead 0.00195% 1.17%
Pb 468 139 857% Antimony 0.0077% 4.51%
Sb 79.2 44 812% Selenium 0.255% 78.4%
Se 6.81 13 661% Tin 0.0923% 55.4%
Sn 11.6 35 402% Vanadium 0.00937% 5.47%
V 57 69 255%
Zinc 0.0795% 47.7%
Zn 791 140 762%
Beryllium 0.283% 100%
Be 1.45 18 355%
Scandium 0.131% 41.4%
Sc 6.13 76 166%
Strontium 0.524% 100%
Sr 112 92 302%
Titanium 0.134% 80.3%
Ti 6.29 52 11500%
Thallium 0.0426% 25.6%
Tl 2.6 10 736%
W 10.9 19 414% Tungsten 1.28% 89.2%
Si 154000 24 306% Silicon 0.000913% 0.548%
Fe 45400 70 323% Iron 0.00955% 5.73%
Ca 11900 63 423% Calcium 1.27% 100%
Al 27800 60 262% Aluminium 0.00546% 3.28%
K 9600 56 229% Potassium 0.0941% 43.2%
Mg 7400 55 455% Magnesium 0.568% 100%
Na 1460 42 507% Sodium 0.676% 98.3%

It is apparent that oxianions have reduced relative mobility in the tailings deposit (red area in Fig.
2.10). Oxianions are more mobile at high pH values. Since residual landfills have high pH values, but
tailings deposit often have low pH values, the modelled behaviour is realistic. On the other hand there
is a cluster of elements with higher mobility in tailings deposits (blue area in Fig. 2.10), that contains
elements that are easier soluble at low pH values. While the uncertainties in these coarse models are
high, it is an encouraging fact that they succeed in depicting chemically reasonable mobility
behaviour, without any information on such mobility characteristics being entered into the models
beforehand. The results of the model are reasonable and not overly dependent on the variable nature of
the fundamental data. Otherwise such clustering of elemental types could not be distinctly observed. It
is a sign that the sample size of the applied literature data is large enough to represent typical tailings.
Tailings impoundment land use
Land use is based on a generic final height of a tailings impoundment of 50 meters. Thus, per kilogram
tailings with a density of 2200 kg/m3 a land area of 9.1·10-6 m2 is necessary. The land is assumed to be
converted from unknown origin to dump site. After an operational phase of 30 years, spontaneous
restoration of a consolidated state is assumed to be reached after one hundred years. The total
occupation time is therefore 130 years and the land occupation per kilogram tailings 1.18·10-3 m2a.
Outlook
Representation of sulfidic tailings disposal in the ecoinvent database is still very coarse: there is only
one single dataset for any sulfidic tailings deposit. Much remains to be done, which was not possible
due to time and money constraints: in the future the tailings disposal inventory should be differentiated
according to the mined metal, the tailings site climate and other important conditions.
Data quality
Uncertainty in this dataset is high, but not only due to a high aggregation level resulting from a
worldwide average. The subject matter is inherently uncertain due to the stochiastic processes in the
earth's crust, leading to very diverse ore bodies, and thus to very variable data relating to tailings even
for one single mining site. Hence, the level of uncertainty is high and likely to remain high, even if in
the future more differentiated tailings models for single sites or regions are created. The sample sizes
for almost all relevant (toxic) elements are in the range of 80 – 140 data points. Presuming the
literature data is representative, these sample sizes are large enough to yield significant LCIA results
for the observed data variability.
The uncertainty range for the emission data is derived according to the procedures applied in Doka
(2003). In a worst case, all tailings material will be weathered or eroded within 60'000 years. This
worst case is taken to represent the upper confidence value of a lognormal distribution. Together with
the inventoried mean emission, the geometric standard deviation is defined.
Rehabilitation and tailings management as it is done in certain parts, can alter the ecological impacts
greatly. Due to the lack of reliable data, no deliberate rehabilitation is assumed here.

Tab. 2.10 Inventoried exchanges for the disposal of sulfidic tailings.
disposal, sulfidic
SubCategory
tailings, off-site
Category
Unit
Name
Location GLO
InfrastructureProces
0
s
Unit kg
Sulfate water ground- kg 0.00559
Nitrate water ground- kg 0.000013
Phosphate water ground- kg 0.000595
Boron water ground- kg 0.0000272
Chloride water ground- kg 3.91E-09
Bromine water ground- kg 4.47E-10
Fluoride water ground- kg 0.00000161
Silver, ion water ground- kg 6.59E-10
Arsenic, ion water ground- kg 1.89E-08
Barium water ground- kg 4.56E-09
Cadmium, ion water ground- kg 1.56E-08
Cobalt water ground- kg 3.44E-08
Chromium VI water ground- kg 3.82E-09
Copper, ion water ground- kg 9.29E-08
Mercury water ground- kg 2.79E-11
Manganese water ground- kg 0.00000139
Molybdenum water ground- kg 0.000000162
Nickel, ion water ground- kg 1.49E-08
Lead water ground- kg 9.14E-09
Antimony water ground- kg 6.1E-09
Selenium water ground- kg 1.74E-08
Tin, ion water ground- kg 1.07E-08
Vanadium, ion water ground- kg 5.35E-09
Zinc, ion water ground- kg 0.000000629
Beryllium water ground- kg 4.11E-09
Scandium water ground- kg 8.05E-09
Strontium water ground- kg 0.000000587
Titanium, ion water ground- kg 8.41E-09
Thallium water ground- kg 1.11E-09
Tungsten water ground- kg 0.00000014
Silicon water ground- kg 0.00000141
Iron, ion water ground- kg 0.00000434
Calcium, ion water ground- kg 0.000151
Aluminum water ground- kg 0.00000152
Potassium, ion water ground- kg 0.00000904
Magnesium water ground- kg 0.000042
Sodium, ion water ground- kg 0.00000985
Sulfate water ground-, long-term kg 0.0383
Nitrate water ground-, long-term kg 0.0000893
Phosphate water ground-, long-term kg 0.000916
Boron water ground-, long-term kg 0.000283
Chloride water ground-, long-term kg 0.00000138
Bromine water ground-, long-term kg 0.000000158
Fluoride water ground-, long-term kg 0.000569
Silver, ion water ground-, long-term kg 0.000000395
Arsenic, ion water ground-, long-term kg 0.0000111
Barium water ground-, long-term kg 0.00000273
Cadmium, ion water ground-, long-term kg 0.00000714
Cobalt water ground-, long-term kg 0.0000186
Chromium VI water ground-, long-term kg 0.00000223
Copper, ion water ground-, long-term kg 0.0000556
Mercury water ground-, long-term kg 1.67E-08
Manganese water ground-, long-term kg 0.000835
Molybdenum water ground-, long-term kg 0.00000686
Nickel, ion water ground-, long-term kg 0.00000893
Lead water ground-, long-term kg 0.00000547
Antimony water ground-, long-term kg 0.00000357
Selenium water ground-, long-term kg 0.00000532
Tin, ion water ground-, long-term kg 0.00000643
Vanadium, ion water ground-, long-term kg 0.00000311
Zinc, ion water ground-, long-term kg 0.000377
Beryllium water ground-, long-term kg 0.00000145
Scandium water ground-, long-term kg 0.00000253
Strontium water ground-, long-term kg 0.000111
Titanium, ion water ground-, long-term kg 0.00000504
Thallium water ground-, long-term kg 0.000000664
Tungsten water ground-, long-term kg 0.00000961
Silicon water ground-, long-term kg 0.000843
Iron, ion water ground-, long-term kg 0.0026
Calcium, ion water ground-, long-term kg 0.0117
Aluminum water ground-, long-term kg 0.000911
Potassium, ion water ground-, long-term kg 0.00414
Magnesium water ground-, long-term kg 0.00735
Sodium, ion water ground-, long-term kg 0.00142
Transformation, from unknown resource land m2 0.00000909
Transformation, to dump site resource land m2 0.00000909
Occupation, dump site resource land m2a 0.00118

2.3.4 Disposal, non-sulphidic tailings, off-site in ecoinvent

The effects modelled in this module are
- land transformation
- land use
Tailings without considerable content of sulphur do not undergo ARD reactions. The pollution
potential is therefore lower than in sulfidic tailings for the classic heavy metal pollutants. However
also non-sulfidic tailings are mechanically compromised (ground into a dust), mixed with water and
exposed to atmospheric oxygen. This can result in distinctly different behaviour compared to the
original setting in the ore vein. No model for non-sulfidic tailings was created here, so no emissions to
soil or water are considered, solely the land transformation and land use are modelled with similar
figures as in 2.3.3.
The flux of this module and its representation in ecoinvent are presented in Fig. 2.12.
Data quality
Land use and occupation are calculated based on figures from only few sites. Rehabilitation and
tailings management as it is done in certain parts, can alter the ecological impacts greatly. Due to the
lack of reliable data, no rehabilitation is assumed here.
2.3.5 Disposal of overburden in ecoinvent

The effects modelled in this module are
- land transformation
- land use
Overburden is assumed to be piled on spherical heaps with a diameter of 20 m and a volume of
16'755 m3. With an in-situ density of 1.8 t / m3 this is 0.0418 m2 and 5.43 m2a per tonne overburden
disposed. The land is assumed to be converted from unknown origin to dump site. Restoration of the
natural state is assumed to be reached after one hundred years naturally, yielding an occupation of 130
years with the mine operation duration included.
No rehabilitation by human activities is assumed. Rehabilitation of coarse material often is done by re-
vegetation. Per m2 a mass of roughly 0.5 kg seeds would be necessary, which is 0.021 kg per tonne,
transported over a distance of 100 km ( = 2.1 10-3 tkm / t).
Overburden, which is refilled, is not inventoried at all.
The flux of this module and its representation in ecoinvent are presented in Fig. 2.13.
Data quality
Land use and occupation are calculated assuming a spherical heap. This can lead to overestimate the
land occupation of bigger piles. Rehabilitation and tailings management as it is done in certain parts,
can alter the ecological impacts greatly. Due to the lack of reliable data, no rehabilitation is assumed
here.

Tab. 2.11 Ecoinvent meta information for disposal processes in PGM winning operations (to be completed).
disposal, non-
disposal, sulfidic tailings, off- sulfidic disposal, non-
Name
site overburden, off- sulfidic tailings,
site off-site
Location GLO GLO GLO
Unit kg kg kg
Waste-specific short- and long-term
emissions to ground water from
Included Processes
tailings impoundment leachate.
Land transformation and occupation.
Amount 1 1 1
Entsorgung, sulfidische Berge, Entsorgung, nicht-sulfidischer Abraum,
Local Name
neben Gelände deponiert neben Gelände deponiert
Synonyms
General Comment to The module is designed for unspecific use
reference function for disposal of sulphidic tailings in non-
ferrous metal extraction. The module
includes land use. No rehabilitation is
assumed
CAS Number
Start Date 2003 2003 2003
End Date 2003 2003 2003
Average waste composition from
literature references. Transfer
Other Period Text
coefficients from prospective,
tailings-specific model.
No geographical Land use of average
Geography text context, inventory global mining
based on calculations operation
Mechanically stable tailings
Technology text impoundment. Spontaneous
renaturation after closure. No rehabilitation is assumed
Production Volume nA nA nA
Tailings model based on observed
leachate concentrations in literature.
Extrapolated to 60'000 years
Sampling Procedure
heeding chemical characteristics.
Initial waste composition from
various literature sources. Literature Literature
Uncertainty of waste input
composition data derived from large
literature survey. Mean long-term
emissions are the emissions during
a generic ecological planning
horizon (60'000a). Maximal long-
term emissions are the complete
emissions of all landfilled material. none none

Infras
OutputGroup
InputGroup
Expl truct disposal, Standard
Locati SubCate Uncertainty
anati 401 Name
on
Category
gory
urePr Unit sulfidic tailings, Type
Deviation GeneralComment
on oces 95%
off-site
s
662 Location GLO

493 InfrastructureProcess 0
403 Unit kg
4 Sulfate water ground- kg 0.005589578 1 3259% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Nitrate water ground- kg 1.30341E-05 1 3436% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Phosphate water ground- kg 0.000594832 1 1503% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Boron water ground- kg 2.72284E-05 1 698% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Chloride water ground- kg 3.90518E-09 1 1748% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Bromine water ground- kg 4.47031E-10 1 816% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Fluoride water ground- kg 1.61371E-06 1 2107% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Iodide water ground- kg 0 1 2345% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Silver, ion water ground- kg 6.58775E-10 1 7929% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Arsenic, ion water ground- kg 1.89113E-08 1 13021% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Barium water ground- kg 4.55655E-09 1 2213% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Cadmium, ion water ground- kg 1.56101E-08 1 5100% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Cobalt water ground- kg 3.43619E-08 1 2392% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Chromium VI water ground- kg 3.82071E-09 1 1353% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Copper, ion water ground- kg 9.29308E-08 1 4537% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Mercury water ground- kg 2.79266E-11 1 9593% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Manganese water ground- kg 1.39292E-06 1 2662% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Molybdenum water ground- kg 1.61573E-07 1 1677% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Nickel, ion water ground- kg 1.49016E-08 1 3605% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Lead water ground- kg 9.14304E-09 1 12295% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Antimony water ground- kg 6.09724E-09 1 10321% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Selenium water ground- kg 1.73658E-08 1 5736% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Tin, ion water ground- kg 1.07277E-08 1 2509% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Vanadium, ion water ground- kg 5.34745E-09 1 1509% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Zinc, ion water ground- kg 6.28537E-07 1 8028% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Beryllium water ground- kg 4.10953E-09 1 1842% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Scandium water ground- kg 8.05141E-09 1 649% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Strontium water ground- kg 5.87202E-07 1 1317% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Titanium, ion water ground- kg 8.41026E-09 1 100956% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Thallium water ground- kg 1.10784E-09 1 7804% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Tungsten water ground- kg 1.39689E-07 1 2153% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Silicon water ground- kg 1.4071E-06 1 2391% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Iron, ion water ground- kg 4.33674E-06 1 2126% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Calcium, ion water ground- kg 0.000150632 1 2238% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Aluminum water ground- kg 1.51915E-06 1 1648% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Potassium, ion water ground- kg 9.0406E-06 1 1018% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Magnesium water ground- kg 4.20117E-05 1 2716% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Sodium, ion water ground- kg 9.84907E-06 1 3288% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Sulfate water ground-, long-term kg 0.038302333 1 189% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Nitrate water ground-, long-term kg 8.93154E-05 1 189% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Phosphate water ground-, long-term kg 0.000915639 1 136% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Boron water ground-, long-term kg 0.000283128 1 207% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Chloride water ground-, long-term kg 1.3761E-06 1 472% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Bromine water ground-, long-term kg 1.57524E-07 1 472% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Fluoride water ground-, long-term kg 0.000568637 1 472% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Iodide water ground-, long-term kg 0 1 472% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Silver, ion water ground-, long-term kg 3.94534E-07 1 6318% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Arsenic, ion water ground-, long-term kg 1.11139E-05 1 4529% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Barium water ground-, long-term kg 2.73022E-06 1 26947% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Cadmium, ion water ground-, long-term kg 7.14215E-06 1 442% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Cobalt water ground-, long-term kg 1.85558E-05 1 455% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Chromium VI water ground-, long-term kg 2.22598E-06 1 3473% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Copper, ion water ground-, long-term kg 5.55941E-05 1 1103% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Mercury water ground-, long-term kg 1.6729E-08 1 8904% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Manganese water ground-, long-term kg 0.000834545 1 467% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Molybdenum water ground-, long-term kg 6.86218E-06 1 282% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Nickel, ion water ground-, long-term kg 8.925E-06 1 687% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Lead water ground-, long-term kg 5.47094E-06 1 14097% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Antimony water ground-, long-term kg 3.56598E-06 1 3982% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Selenium water ground-, long-term kg 5.32268E-06 1 439% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Tin, ion water ground-, long-term kg 6.43109E-06 1 566% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Vanadium, ion water ground-, long-term kg 3.11472E-06 1 3340% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Zinc, ion water ground-, long-term kg 0.000376615 1 612% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Beryllium water ground-, long-term kg 1.44861E-06 1 423% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Scandium water ground-, long-term kg 2.53012E-06 1 607% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Strontium water ground-, long-term kg 0.000111379 1 378% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Titanium, ion water ground-, long-term kg 5.03971E-06 1 488% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Thallium water ground-, long-term kg 6.64224E-07 1 919% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Tungsten water ground-, long-term kg 9.61238E-06 1 321% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Silicon water ground-, long-term kg 0.000842706 1 29169% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Iron, ion water ground-, long-term kg 0.002598786 1 3210% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Calcium, ion water ground-, long-term kg 0.01174368 1 319% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Aluminum water ground-, long-term kg 0.000910663 1 5346% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Potassium, ion water ground-, long-term kg 0.004139383 1 624% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Magnesium water ground-, long-term kg 0.007353726 1 373% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Sodium, ion water ground-, long-term kg 0.001421513 1 361% uncertainty is calculated from uncertainties in waste composition and transfer coefficients
4 Transformation, from unknown resource land m2 9.09091E-06 1 200% (3,3,1,1,1,5) & basic uncertainty of 2;
4 Transformation, to dump site resource land m2 9.09091E-06 1 200% (3,3,1,1,1,5) & basic uncertainty of 2;
4 Occupation, dump site resource land m2a 0.001181818 1 158% (3,3,1,1,1,5) & basic uncertainty of 1.5;
0 disposal, sulfidic tailings, off-site GLO waste man residual ma 0 kg 1
Fig. 2.11 Flows for “disposal, sulphidic tailings, offsite” and its representation in the ecoinvent
database.

mean value
Mean value
Source for
ecoinvent
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
name in
General
Output
Modul
Name
Input
Infra-
Type
Sub-
95%
Unit
sulfidic overburden,
Land trans- Transformation, from
 resource land 5.10E-05 m2 (IPPC, 2002) 1 2.06463 (1,3,1,3,3,4,8)
formation, ground disposal, non- unknown
Land trans- Transformation, to

 resource land 5.10E-05 m2 (IPPC, 2002) 1 2.06463 (1,3,1,3,3,4,8)
ormation, tailing off-site dump site
Land use for Occupation, dump
 average lifetime of 130 a assumed resource land 6.63E-03 m2a (IPPC, 2002) 1 1.58045 (1,3,1,3,3,4,7)
tailings site
disposal, non- residual

waste disposal, non-sulfidic
 sulfidic overburden, material No GLO 1.00E+00 kg
management overburden, off-site
off-site landfill
Fig. 2.12 Flows for “disposal, non-sulphidic tailings, offsite” and its representation in the ecoinvent database. The two tailings disposal modules are identical, except that from non
sulphuric disposal no emissions are inventoried.
mean value
Mean value
Source for
ecoinvent
Comment
Deviation
Category
Standard
structure
Remarks
category
Location
Process
name in
General
Output
Modul
Name
Input
Infra-
Type
Sub-
95%
Unit
Land trans- Transformation, from

sulfidic tailings,
resource land 5.10E-05 m2 (IPPC, 2002) 1 2.15521 (4,4,1,3,3,5,8)

disposal, non-
formation, ground unknown
Land trans-
off-site
Transformation, to
formation,  resource land 5.10E-05 m2 (IPPC, 2002) 1 2.15521 (4,4,1,3,3,5,8)
dump site
overburden
Land use for Occupation, dump
 average lifetime of 130 a assumed resource land 6.63E-03 m2a (IPPC, 2002) 1 1.6872 (4,4,1,3,3,5,7)
overburden site
disposal, non- residual
waste disposal, non-sulfidic
 sulfidic tailings, off- material No GLO 1.00E+00 kg
management tailings, off-site
site landfill
Fig. 2.13 Flows for “disposal, non-sulphidic overburden, offsite” and its representation in the ecoinvent database.

3. Literatur
3 Literatur
Althaus et al. (2004) Althaus H.-J., Blaser S., Classen M. and Jungbluth N. (2004) Life Cycle
Inventories of Metals. Final report ecoinvent 2000 No. 10. EMPA Dübendorf,
Swiss Centre for Life Cycle Inventories, Dübendorf, CH, Online-Version under:
www.ecoinvent.ch.
ATSDR 1997 ATSDR Agency for Toxic Substances and Disease Registry (1997) Public
Health Assessment Triumph Mine Tailings . Prepared by U.S. Department of
Health and Human Services , Public Health Service , Agency for Toxic
Substances and Disease Registry , Division of Health Assessment and
Consultation, Atlanta, Georgia. CERCLIS NO. IDD984666024. Download vom
13. Dezember 2006 von
http://www.atsdr.cdc.gov/HAC/PHA/triumph/tri_p2.html
ATSDR 2006 Agency for Toxic Substances and Disease Registry (2006) Public Health
Assessment For Bauer Dump & Tailings Blackhawk Resin Company Tooele
County, Utah July 25, 2006. Prepared by Environmental Epidemiology Program,
Utah Department of Health, USA. Download vom 3. Febraur 2007 von
http://www.atsdr.cdc.gov/HAC/PHA/Bauer%20DumpTailingsBlackhawkResin
Company/BauerDumpTailingsPHA072506.pdf
Products and Byproducts. MMSD - Mining, Minerals and Sustainable
Development, Online-Version under: http://www.iied.org/mmsd/mmsd_pdfs/
ayres_lca_main.pdf.
Boulet & Larocque 1998 Boulet M.P., Larocque ACL. (1998) A comparative mineralogical and
geochemical study of sulfide mine tailings at two sites in New Mexico, USA.
Envir. Geol. Volume 33, Numbers 2-3 / February, 1998. Download vom 20.1.07
von http://www.springerlink.com/content/c0uukhbem0ef43fd/
Doka (2003) Doka G. (2003) Life Cycle Inventories of Waste Treatment Services. Final
report ecoinvent 2000 No. 13. EMPA St. Gallen, Swiss Centre for Life Cycle
Doka 2008 Doka G. (2008) Life Cycle Inventory data of mining waste: Emissions from
sulfidic tailings disposal. Doka Life Cycle Assessments, Zurich, Switzerland,
April 2008. Retrieved at 03.04.2008 from
http://www.doka.ch/SulfidicTailingsDisposalDoka.pdf
Dorronsoro et al. 2002 Dorronsoro C., Martin F., Ortiz, I. Garcia I., Simon M., Fernandez E., Aguilar J.,
Fernandez J. (2002) Migration of Trace Elements from Pyrite Tailings in
Carbonate Soils. J. Environ. Qual. 31:829–835. Download vom 15. Dezember
2006 von http://jeq.scijournals.org/cgi/reprint/31/3/829.pdf
Downing & Gravel 2006 Downing B.W., Gravel J., (2006) Trace Element Geochemistry in Acid rock
drainage. Download vom 8 Dezember 2006 von
http://technology.infomine.com/enviromine/ard/Introduction/Trace.htm
EPA 1994 Technical Resource Document Extraction And Beneficiation Of Ores And
Minerals Volume 1 Lead-Zinc. EPA 530-R-94-011, U.S. Environmental
Protection Agency, Office of Solid Waste, Special Waste Branch, Washington,
DC. June 1994, Download vom 19. Juli 2003 von
http://www.epa.gov/epaoswer/other/mining/techdocs/leadzinc.pdf
Garcia-Meza et al. 2006 Garcia-Meza J.V., Carrillo-Chavez A., Morton-Bermea O. (2006) Sequential
extractions on mine tailingssamples after and before bioassays: implications on
the speciation of metals during microbial re-colonization. Environ Geol (2006)

3. Literatur
49: 437–448 . Download vom 21. Dezember 2006 von

http://www.springerlink.com/content/f145t7q511570073/fulltext.pdf
Hadjibiros et al. 2005 Hadjibiros K., Mantziaras I.D., Sakellariadis D.G., Giannakidou C. and Katsiri
A. (2005) Pollution Risk Assessment From European Mining Sites And
Preliminary Results From Tailings Dams In Greece. Department of Civil
Engineering, National Technical University of Athens. Proceedings of
"Protection And Restoration Of The Environment VIII", 3 - 7 July 2006 Chania ,
Greece. Download vom 22. Dezember 2006 von
http://www.itia.ntua.gr/~kimon/Chania.ppt , see also
http://www.ath.aegean.gr/srcosmos/showpub.aspx?aa=6555
Hernandez et al. 2007 Hernandez C.M., Banza A.N., Gock E. (2007) Recovery of metals from Cuban
nickel tailings by leaching with organic acids followed by precipitation and
magnetic separation. J Hazard Mater. 2007 Jan 2;139(1):25-30. Download vom
13. Dezember 2006 von
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=pubmed&do
pt=Abstract&list_uids=17084523
Geologisches Jahrbuch, Vol. Sonderhefte SH 7. Bundesanstalt für
Geowissenschaften und Rohstoffe, Hannover. ISBN 3-510-95829-2.
IAEA 2003 Database of Natural Matrix Reference Materials (2003). International Atomic
Energy Agency (IAEA). Download vom 15. Dezember 2006 von http://www-
naweb.iaea.org/nahu/nmrm/nmrm2003/default.htm
European Commission. Retrieved from http://www.jrc.es/pub/english.
cgi/0/733169
IPPC (2002) IPPC (2002) Integrated Pollution Prevention and Control (IPPC); Draft
Reference Document on Best Available Techniques for Management of Tailings
and Waste-Rock in Mining Activities. European Commission. Retrieved at
IPPC 2004 Integrated Pollution Prevention and Control (IPPC) Reference Document on
Best Available Techniques for Management of Tailings and Waste-rock in
Mining Activities. European Comission Joint Research Centre. Retreived
8.1.2007 from
http://www.jrc.es/pub/english.cgi/d1056340/25%20Reference%20Document%2
0on%20Best%20Available%20Techniques%20for%20Management%20of%20T
ailings%20and%20Waste-rock%20in%20Mining%20Activities%20-
%2021.2%20Mb
Koerth 2002 Koerth J. (2002) Expanded Engineering Evaluation/Cost Analysis (EEE/CA)
For The McLaren Tailings Site Cooke City, Montana. Montana Department of
Environmental Quality, Mine Waste Cleanup Bureau, May 2002. Download
vom 14. Februar 2007 von
http://deq.mt.gov/AbandonedMines/linkdocs/Mclaren/APPENDIX-A.pdf
Laliberte & Tremblay 2002 Laliberte D., Tremblay G. (2002) Metal, PCB, Dioxin and Furan Concentrations
in Fish and Sediments from Four Lakes in Northern Québec in 2001. Québec.
Ministère de l’Environnement. Direction du suivi de l’état de l’environnement.
Envirodoq no ENV/2002/0203. Report no. QE-129. Download vom 24.
Dezember 2006 von
http://www.mddep.gouv.qc.ca/eau/eco_aqua/chibougamau/rapport-en.pdf , s.

3. Literatur
auch http://www.beesum-communications.com/nation/archive/12-
20/OJ/poisoned.pdf"
Lazar 2002 Lazar F.F. (2002) Geochemisty of the environment in the areas of mining works
from Aries Valley (Apuseni Mountans, Romania). Abriged version of PhD
Thesis. Department of Geology, Babes-Bolyai University, Romania. Download
vom 20. Januar 2007 von http://bioge.ubbcluj.ro/~forray/pub/abridged.pdf
Martens et al. (2002) Martens P. N., Ruhrberg M. and Mistry M. (2002) Flächeninanspruchnahme des
Kupfererzberbaus. In: Erzmetall, 55(5/6), pp. 287-293.
Matheis et al. 1998 Matheis, G., Schreck, P. , Jahn, S.1 & Lorenz, R. (1998) Schwermetallaustrag
aus Haldenmaterial des Kupferschieferbergbaues: in-situ- Laugungsexperimente
in Helbra, Mansfelder Land. Institut für Angewandte Geowissenschaften,
Technische Universität Berlin. Download vom 15. Februar 2007 von
http://www.lagerstaetten.tu-berlin.de/forschung/Matheis/matheis.html
McLemore et al. 2006 McLemore VT., Donahue K, Phillips E, Dunbar N, Smith M, Tachie-Menson S,
Viterbo V, Lueth V W, and Campbell AR (2006) Petrographic, Mineralogical
And Chemical Characterization Of Goathill North Mine Rock Pile, Questa
Molybdenum Mine, Questa, New Mexico. Paper presented at the 2006 Billings
Land Reclamation Symposium, June, 2006, Billings, Mt. Published by ASMR,
3134 Montavesta Rd., Lexington, KY 40502. Download vom 23. Dezember
2006 von
http://geoinfo.nmt.edu/staff/mclemore/documents/Mclemore_NM_billings06_fin
al.doc
MIMI 2004 Höglund L.O., Herbert R. (Ed) (2004) MiMi - Performance Assessment - Main
Report. Mitigation of the environmental impact from mining waste (MiMi)
Lulea University of Technology, Sweden. Download vom 17. Januar 2007 von
http://www.misttra.org/download/18.c791f4103209a06ec80005948/MiMi-
PAMainReport.pdf
Muwanga 1997 Muwanga A. (1997) Environmental impact of mine waste disposal at Kilembe
copper mine, Uganda. Braunschweiger Geowiss. Arb. 21. Download vom 12.
Februar 2006 von http://www.iug.tu-
bs.de/users/walterpohl/html/uganda_waste_disposal/index.html
Nash 2000 Nash J.T. (2000) Hydrogeochemical Data For Historic Mining Areas, Humboldt
Watershed And Adjacent Areas, Northern Nevada. U.S. Geological Survey,
Denver CO, Open-File Report 00-459. Download vom 12. Dezember 2006 von
http://pubs.usgs.gov/of/2000/ofr-00-0459/HDOINTF.DOC , sowie Daten aus
http://pubs.usgs.gov/of/2000/ofr-00-0459/xls/HGX67C.XLS und
http://pubs.usgs.gov/of/2000/ofr-00-0459/xls/HW90MS.XLS
OK Tedi 2002 Tailing Geochemistry. Unreferenced Report from OK Tedi site. Download vom
29.2.07 von
http://www.oktedi.com/reports/reports/72/4112TailingGeochem.pdf?PHPSESSI
D=30a4eba40ced8c774ec66eb92235ba3f
Robinson et al. 2004 Robinson B., Bus A., Diebels B., Froehlich E., Grayson R. (2004) Tailings
Disposal Options for the Kensington Mine at Berners Bay Near Juneau, Alaska.
NOSB paper, 2004 Alaska Ocean Sciences Bowl. Download vom 6. Februar
2006 von http://www.uaf.edu/seagrant/nosb/papers/2004/midas-tailings.html
Thoms 2006 Thoms B. (2006) Fact Sheet Former Opp Mine Cleanup. Department of
Evironmental Quality, State of Oregon, Download vom 13. Dezember 2006 von
http://www.deq.state.or.us/wmc/pubs/factsheets/cu/FormerOppMineCleanup.pdf

3. Literatur
Tyler et al. 2004 Tyler G., Dubuisson C., Ruiz M.J., Carrasco R.. Sánchez-Rodas D., Pérez R.,
Sarmiento A.M., Nieto J.M. (2004) Optimization of Major and Trace Elements
Determination in Acid Mine Drainage Water Samples by USN-ICP-OES.
Download vom 12. Dezember 2006 von
http://cetac.com/pdfs/U5000_MineDrainagePoster.pdf
Van Zyl 2002 van Zyl D., Sassoon M., Digby C., Fleury A-M. , Kyeyune S. (2002) Mining for
the Future – Appendix J: Grasberg Riverine Disposal Case Study. Mining,
Minerals and Sustainable Development Project Report 68c, International
Institute for Environemnt and Development (IIED), World Business Council for
Sustainable Development, April 2002. Download vom 28.2.07 von
http://www.unr.edu/mines/mlc/presentations_pub/Pub_LVW/68c_mftf-j.pdf
Wels et al. 2000 Wels, C., Shaw, S., and Royle, M.(2000): "A Case History Of Intrinsic
Remediation Of Reactive Tailings Seepage For Questa Mine". Paper presented
at the ICARD 2000 Conference, Proceedings Volume 1, pp. 441-458, May 2000,
Denver, Colorado. Download vom 8. Dezember 2006 von
http://www.robertsongeoconsultants.com/papers/ICARDpaper.pdf
Ye et al. 2002 Ye Z. H., Shu W. S., Zhang Z. Q., Lan C. Y. and Wong M. H. (2002) Evaluation
of major constraints to revegetation of lead/zinc mine tailings using bioassay
techniques. Chemosphere,Volume 47, Issue 10 , June 2002, Pages 1103-1111.
Download vom 21. Dezember 2006 von
http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6V74-
455VKKN-
9&_coverDate=06%2F30%2F2002&_alid=512622669&_rdoc=1&_fmt=&_orig
=search&_qd=1&_cdi=5832&_sort=d&view=c&_acct=C000050221&_version
=1&_urlVersion=0&_userid=10&md5=809bb7fb052e71aa89d868d9b27c92c9
Sulovsky & Zeman 2000 Sulovsky P., Zeman J. (2000) Anthropogenic impacts on weathering processes:
man-induced atmosphere - water - rock interactions on global- to micro-scale
IGCP Project No. 405. Download vom 23. Februar 2007 von
http://www.sci.muni.cz/~sulovsky/stateart.htm
Cotter & Brigden 2006 Cotter J., Brigden K. (2006) Acid Mine Drainage: the case of the Lafayette
mine, Rapu Rapu (Philippines). GRL Technical Note 09/2006 October 2006,
University of Exeter, UK. Download vom 23. Februar 2007 von
http://www.greenpeace.to/publications_pdf/acid-mine-drainage.pdf
Price & Errington 1997 Price W.A., Errington J.C. (1998) Draft Guidelines For Metal Leaching and
Acid Rock Drainage at Minesites in British Columbia. Ministry of Employment
and Investment, Energy and Minerals Division, British Columbia, Canada,
October 1997.EMC 1998 Environmental Mining Council of BC (1998) Acid
Mine Drainage: Mining & Water Pollution Issues in BC. Environmental Mining
Council of BC and BC Wild, Victoria, British Columbia, Canada. Download
vom 21. Juli 2004 von http://www.miningwatch.org/emcbc/publications/amd.pdf
Gardner & Sampat 1998 Gardner G., Sampat P. (1998) Mind Over Matter: Recasting the Role of
Materials in Our Lives. Worldwatch Paper 144, December 1998. Download of
April 10, 2007 from http://www.worldwatch.org/system/files/EWP144.pdf
Hammarstrom et al. 2002 Hammarstrom J. M., Eppinger R.G., Van Gosen B.S., Briggs P.H., and Meier
A.L. (2002) Case study of the environmental signature of a recently abandoned,
carbonate-hosted replacement deposit: The Clayton Mine, Idaho. Geological
Survey Open-File Report 02-10. Download of April 10, 2007 from
http://pubs.usgs.gov/of/2002/of02-010/of02-010.pdf
Burtynsky 1996 Burtynsky E. (1996) Nickel Tailings No. 36, Sudbury, Ontario. Edward
Burtynsky Photography, Toronto, Ontario, Canada, Download vom 29. Januar

3. Literatur
2004 von
http://www.edwardburtynsky.com/WORKS/Breaking_Ground/Tailings/Tailings
_36.html
FAO 2000 FAO (2000) Global Agro-Ecological Zones system 2000 – Plate 01 Average
annual precipitation data. Food and Agriculture Organization of the United
Nations (FAO) Rome, Italy. International Institute for Applied Systems Analysis
(IIASA) Laxenburg, Austria. Download vom 13. Januar 2004 von
http://www.iiasa.ac.at/Research/LUC/GAEZ/plt/pa.htm?map=01
Spiess 2002 Spiess E. (2002) Schweizer Weltatlas – Nachgeführte und erweiterte Auflage.
Lehrmittelverlag des Kantons Zürich. 2002.

Part XV
Metal processing
Data v2.1 (2009)
Authors: Silvio Blaser, EMPA Dübendorf

Niels Jungbluth, ESU-services
Review: Roland Hischier, EMPA, St. Gallen

Mischa Classen, EMPA Dübendorf
Rolf Frischknecht, ESU-services
Summary
Summary and Comments on version 2.1

This part contains information and LCI of different metal forming and coating processes and with welding. The
modules include only the processes themselves and not the metals being formed, coated or welded.
The part corresponds to the original report in Althaus et al. (2004). The chapters “Cumulative Results and Inter-
pretation” of the original report have been deleted for the current update.
Version 2.1 remains unchanged compared to version 2.0 except for the meta information to the process “zinc
coating, pieces, adjustment per um”.
ecoinvent v2.1 report No. 10, Part XIV -i-

Table of Contents
Table of Contents
SUMMARY AND COMMENTS ON VERSION 2.0 .....................................................................I
1 INTRODUCTION ....................................................................................................... 1
2 STEEL FORMING ..................................................................................................... 2

2.1 General remarks .................................................................................................................... 2
2.2 Hot rolling of steel and chromium steel ................................................................................ 2
2.3 Sheet rolling of steel............................................................................................................ 15
2.4 Sheet rolling of chromium steel .......................................................................................... 32
2.5 Wire drawing of steel and chromium steel.......................................................................... 42
2.6 Pipe drawing of steel and chromium steel........................................................................... 54
2.7 Section bar rolling of steel................................................................................................... 55
2.8 Infrastructure of the rolling mill.......................................................................................... 58
3 ALUMINIUM FORMING ............................................................................................ 61

3.1 Sheet rolling of aluminium.................................................................................................. 61
3.2 Section bar extrusion of aluminium .................................................................................... 66
4 COPPER AND COPPER ALLOY FORMING................................................................... 71

4.1 Sheet rolling of copper ........................................................................................................ 71
4.2 Wire drawing of copper....................................................................................................... 76
4.3 Contouring of brass and bronze........................................................................................... 80
5 ZINC COATING OF STEEL ........................................................................................ 86

5.1 Zinc coating of steel coils.................................................................................................... 86
5.2 Zinc coating of pieces.......................................................................................................... 96
5.3 Zinc coating of pieces, correction of layer thickness per m............................................ 106
6 WELDING ............................................................................................................109
6.1 Definition and fundamentals of welding ........................................................................... 109
6.2 Manual Gas Welding of Steel ........................................................................................... 110
6.3 Arc welding of steel .......................................................................................................... 114
6.4 Arc welding of aluminium ................................................................................................ 120
7 ENAMELLING .......................................................................................................123
7.1 Introduction ....................................................................................................................... 123
7.2 Resources of Enamel ......................................................................................................... 123
7.3 Characterisation of the Product ......................................................................................... 123
7.4 Application of Enamels ..................................................................................................... 123
7.5 System Characterisation .................................................................................................... 124
7.6 Life Cycle Inventory of Enamelling.................................................................................. 125
7.7 Data Quality Considerations ............................................................................................. 128
7.8 Conclusions ....................................................................................................................... 128
ecoinvent v2.1 report No. 10, Part XIV - iii -

Table of Contents
8 SELECTIVE COATINGS ON METAL SHEETS FOR SOLAR ABSORBERS ........................130

8.1 Introduction ....................................................................................................................... 130
8.2 Resources used for Selective Coatings.............................................................................. 131
8.3 Characterisation of Metal Coatings ................................................................................... 131
8.4 Application of Selective Coatings..................................................................................... 132
8.5 System Characterisation .................................................................................................... 132
8.6 Life Cycle Inventories for Selective Coatings................................................................... 132
8.7 Data Quality Considerations ............................................................................................. 141
8.8 Cumulative Results and Interpretation .............................................................................. 142
8.9 Conclusions ....................................................................................................................... 143
9 SOFT SOLDERS (LEAD-FREE) AND SOLDERING........................................................145

9.1 Introduction ....................................................................................................................... 145
9.2 Reserves and resources of product .................................................................................... 145
9.3 Characterisation and use of soft solder.............................................................................. 145
9.4 System characterisation ..................................................................................................... 145
9.5 Life cycle inventories ........................................................................................................ 147
9.6 Data quality considerations ............................................................................................... 152
9.7 Conclusions ....................................................................................................................... 152
10 POWDER COATING ...............................................................................................154

10.1 Definition and fundamentals of powder coating ............................................................... 154
10.2 Coating powder ................................................................................................................. 154
10.3 Powder coating of steel ..................................................................................................... 157
10.4 Powder coating of aluminium ........................................................................................... 160
11 ANODISING OF ALUMINIUM ....................................................................................163

11.1 Definition and fundamentals of anodising ........................................................................ 163
11.2 Anodising in ecoinvent...................................................................................................... 163
12 REFERENCES ......................................................................................................168
ecoinvent v2.1 report No. 10, Part XIV - iv -

1. Introduction
1 Introduction
This part deals with the forming and coating processes and with welding of different metals.
Metals, can be subjected to plastic deformation in the solid state by the application of external forces,
i.e. it can be formed into new permanent shapes. Here, the term "forming" means the intentional and
controlled change of the geometry of a work piece, usually from a simple geometry to a complex one
with regard to the shape, size, accuracy and tolerances, appearance and properties. The mass, composi-
tion, and state of the material remain unchanged. If, however, the plastic limit is exceeded so that the
change in geometry is not controlled, forming with defects occurs, and the desired product is not ob-
tained. The phenomenon of plasticity results from the crystalline structure of metals. Their engineering
and industrial use depends on forming technology (Kümmerling (1997)).
Coating of metals is done for different purposes. In the version ecoinvent data v 1.01 zinc coating and
enamelling of steel and selective coating for solar collectors is addressed.
Many different welding techniques are industrially applied. Here two different techniques for welding
steel and one for welding aluminium are assessed.
All the processing data have in common that they relate only to the actual processes. Thus e.g. coating
of steel does not include the steel sheet to be coated. The same is true for forming and welding.

2. Steel forming
2 Steel forming
Authors: Silvio Blaser, Hans.Jörg Althaus, EMPA Dübendorf
Review: Roland Hischier
Acknowledgement
Thanks goes to Dr. Christoph Zeltner, from Stahl Gerlafingen AG in Gerlafingen, Switzerland, for the
information about the consumed energy of the hot and cold rolling process and the wire drawing proc-
ess; for the help in understanding the forming processes and the guided tour trough the steel forming
plant in Gerlafingen.
2.1 General remarks

The output of raw steel production are semi-finished products categorised in cast ingots, slabs, billets
and beam blanks (for further details on their properties see (IPPC (2001a), page 11)). At this point of
the production steps, the steel is relatively weak due to large and uneven metal crystals or grains. Roll-
ing the semi-finished products causes the coarse grain structure to recrystallise into a much finer grain
structure, giving greater toughness, shock resistance and tensile strength CRC (1998). Therefore in this
study, as a first processing step, the raw steel output (see Part II, chapter 4.9 of this report) is hot rolled
to different flat or long products.
After this first step, further forming processes like cold rolling, wire drawing, tubes drawing or section
bar rolling can be applied. All those processes are described in this chapter/section.
2.2 Hot rolling of steel and chromium steel

The term "hot rolling" or "hot forming" is used when heat is applied to the workpiece before forming
it between electrically powered rollers. In general, the temperature for hot forming is higher than the
recrystallization temperature (temperature ranging from 1050 to 1300 °C). If the temperature is too
high, a coarse-grained structure can be formed, which is undesirable (Kümmerling (1997)) and (IPPC
(2001a)).
In contrast to further processing steps like for example cold rolling, for the hot rolling process steel
and stainless steel are treated in approximately the same way. Therefore in this study, only one module
and process description for those two materials was made.
The advantages of hot forming include (Kümmerling (1997)):
- Great formability of the workpiece,
- Small forces required, as the yield stress of the workpiece is low,
- Large degree of deformation possible in one forming step, with consequent reduction in process-
ing time,
- Beneficial effect on the structure, and hence on the properties of the workpiece,
- Little or no work hardening (if this is not desired).

The most significant disadvantages of hot forming are:
- Very high cost of heating the workpiece compared with the energy required for forming,
- Capital investment for the heating furnaces,

2. Steel forming
- Very severe tool wear due to the very hard and brittle scale that forms on the surface of the work-
piece,
- Reduced working life of the tools due to the thermal load and increased wear.
Typical in-going slabs are 400 to 2500 mm wide and 40 to 500 mm thick and can be rolled to a thick-
ness of 1.2 to 20 mm by the hot rolling process. Products obtained from hot rolling are usually classi-
fied in two basic types according to their shape: flat and long products. In (IPPC (2001a)) production
volume of hot rolled products is given: "Total EU production in 1996 of hot rolled (HR) products was
127.8 million tonnes of which flats accounted for 79.2 million tonnes (ca. 62%). Germany is the larg-
est manufacturer of the flats, with 22.6 million tonnes followed by France with 10.7 million tonnes,
Belgium 9.9 million tonnes, Italy 9.7 million tonnes and UK 8.6 million tonnes. The vast majority of
HR flat products is wide strip. The remaining 38 % of HR products are long products with about 48.5
million tonnes in 1996. The two major manufacturing countries are Italy with about 11.5 million ton-
nes and Germany with 10.3 million tonnes; followed by UK (7 million tonnes) and Spain (6.8 million
tonnes). The largest part of the long products sector in tonnage terms is the production of wire rod
which stands for roughly a third of the total production followed by reinforcing bars and merchant
bars with an approximate share of one quarter of the production each." Fig. 2.1 shows an overview of
products manufactured for various purposes and applications.

2. Steel forming
Fig. 2.1 Overview of Hot Rolled Products (IPPC (2001a)).
Hot rolling usually comprise the following process steps (IPPC (2001a)):
- Conditioning of the input (scarfing, grinding),
- Heating to rolling temperature,
- Descaling,
- Rolling (roughing including width reduction, rolling to final dimension and properties),
- Finishing (trimming, slitting, cutting).

A short description of every single process step is given below in this chapter.
The infrastructure of a hot rolling mill comprises buildings where production processes take place and
for storage, surroundings and machinery for the single process steps (e.g.: furnace, rolls, cutting de-
vices, etc.) Infrastructure is described in detail in chapter 1.1.

2. Steel forming
2.2.1 Conditioning of the input (scarfing, grinding)

In IPPC (2001a) the surface conditioning process steps are described as follow:
"Slabs, blooms and billets can have surface defects, like cracks, folds or seams. Surface preparation of
rolling stock is necessary to ensure flawless surfaces of the rolled product. Surface defects of semifin-
ished products, can also be removed by the following processes.
Scarfing
Scarfing removes surface defects of carbon steel grades by an oxy-fuel flame. The flame is used to
rapidly melt and oxidise the steel surface while a separate supply of high pressure oxygen to the cutter
propels the slag product from the surface. (...)
The scale from machine scarfing is washed from the steel surface by high pressure water and led to a
water pit where the scale is removed by crane. The water is treated and recirculated. (...)
Dust emissions from scarfing operation are collected and, usually, abated by electrostatic precipitators
(wet or dry).
The consumption of fuel (natural gas, liquefied petroleum or plant gases) and oxygen for scarfing de-
pends on the dimensions of the rolling stock. Variations of 20 % in consumption and emissions are
possible when the e.g. thickness of a slab is increased from 200 mm to 250 mm.
Grinding
For stainless steels and special steel grades, thermal removal of the surface defects is not possible.
Therefore defects have to be eliminated by grinding. (...) Grinding machines are equipped with a roller
table on which the work piece is supported and on which it can be moved. The grinding wheel is
mounted on a carriage that permits it to be moved back and forth over the work piece. The unit is posi-
tioned in an enclosure, which collects the dust generated during grinding and ducts it to a bag filter be-
fore discharge to atmosphere."
2.2.2 Heating to rolling temperature

Reheating is done in different types of furnaces such as pit furnaces, pusher type furnaces, walking
beam furnaces, rotary hearth furnaces (which are the most usual type of continuous reheating furnaces)
or others. For details of these furnace types consult (IPPC (2001a)) and (Brandt & Weiler (1997)).
These furnaces are fired directly, usually by oil, natural gas or plant gases such as COG 1 and BFG 2 .
Depending on the heating media, different waste gas emissions will occur (mainly SO2 and NOx). In-
tegrated steel plants try to valorise as much as possible the gases arising within the works for internal
use. Hence blast furnace gas, coke oven gas, BOF gas, natural gas and a lot of different combinations
of these gases (so-called mixed gases) are used in continuous furnaces IPPC (2001a). No data was
available about the detailed consumptions of those gases, therefore in this study the whole energy con-
sumption for the reheating process is inventoried as natural gas. Emissions resulting from burning the
natural gas in the furnace, are not inventoried separately, because they are already included in the heat
module used "natural gas, burned in industrial furnace low-NOx >100kW".
2.2.3 Descaling
Scale, which is formed during heating, must be removed prior to rolling in order to avoid a contamina-
tion of the stock surface by scale impressed by the rolls (so-called 'rolled in matter').
1
COG: Coke oven gas
2
BFG: Blast furnace gas

2. Steel forming
A common method of descaling is breaking and spraying off the scale by means of high-pressure wa-
ter. Water with pressures of 120 to 250 bar (exceptionally 600 bar) is applied via flat jet nozzles onto
the material surface (IPPC (2001a)).
The main wastes of descaling are iron oxides, oil free and oily scale and waste water.
2.2.4 Rolling (roughing including width reduction, rolling to final dimension

and properties)
Hot rolling mills are classified by the type of product that they produce and by their design features.
For a more detailed insight in the different mill layouts and manufacturing sequences, see (IPPC
(2001a)), where information for blooming/slabbing mills, hot strip mills, plate mills, bar and rod mills,
structural/section mills, tube mills are reported. In this study no differentiation for the various mill
types are made.
Edging is usually applied in the manufacture of strip and plate for width reduction. Roughing depicts
the first bulk reduction done on the incoming hot roll stock for the production of strip, rod and sec-
tions. Generally, the roughing train consists of one or more horizontal stands for thickness reduction,
including edgers (vertical stands) situated in front of the stand for width adjustment (IPPC (2001a)).
Energy is used in form of electricity and the water consumption depends on whether the water flow
design is an open, a semiclosed or a closed circuit. Dust and fugitive oil are emitted to the air.
2.2.5 Finishing (trimming, slitting, cutting)

In general, the finishing mill consists of a number of rolling stands positioned after one another. The
stands have staggered roll gaps so that the thickness reduction to the requested final thickness is done
in one pass of the transfer bar. Crop shears arranged in front of the finishing stands serve to cut the
ends of the rough strip. Hot strips with length up to 2 km and even more are wound up on so named
down (sometimes up-) coilers at the end of the rolling mill by speeds up to 20 m/s and more. The fin-
ishing train of rod mills usually consists of rolling blocks with up to 10 sets of alternating horizontal
and vertical roller pairs (IPPC (2001a)).
2.2.6 Environmental issues of hot rolling

The main environmental issues of hot rolling are emissions to air, especially NOx and SOx; the energy
consumption of furnaces; (fugitive) dust emissions from product handling, rolling or mechanical sur-
face treatment; oil- and solid-containing effluents and oil-containing wastes (IPPC (2001a)).
Contaminated cooling and process water arising from processes of hot and cold forming of steel is
treated in internal waste water treatment plants. The reference contained no data about the composition
of the untreated wastewater but reported details about the pollutants composition of the purified water,
when leaving the treatment plant (Tab. 2.1).
The purifying process generates 5 kg sludge per cubic meter of treated water, which is assumed to be
disposed in a residual material landfill type with cement solidification. The composition of the arising
sludge is not known and is therefore calculated / estimated by G. Doka (see also Doka (2003)). As no
special heavy metal precipitating agents are used, the unknown sludge composition is derived propor-
tionally to the pollutant profile in the purified wastewater (Tab. 2.1). The water content of the sludge
is assumed to be 20%.
The values for the amount and composition of the treated wastewater are taken as proxy for the other
steel forming processes.

2. Steel forming
Tab. 2.1 Composition of purified waste water and of the arising sludge
Amount in puri- Amount in waste

fied waste water water treatment
sludge
kg/m3 kg/kg
TSS 2.65E-02 -
COD 6.88E-03 -
unspecified hydrocarbons 3.15E-04 splitted in H and C
Hydrogen - 1.10E-03
Carbon - 9.92E-03
Copper 4.84E-05 1.69E-03
Zinc 3.74E-05 1.31E-03
Cadmium 2.50E-05 8.75E-04
Aluminium 2.37E-04 8.28E-03
Lead 5.00E-05 1.75E-03
Chromium 9.00E-05 3.15E-03
Chromium 6+ 5.00E-06 -
Manganese 1.02E-04 3.57E-03
Iron 7.75E-04 2.71E-02
Nickel 1.41E-04 4.95E-03
Mercury 5.00E-06 1.75E-04
Chlorine 2.24E-04 7.82E-03
Oxygen n.a. 7.28E-01
2.2.7 Hot rolling in ecoinvent

Mean values are generally calculated as the geometric mean of the range given in IPPC (2001a).
Where the range is given from zero to a maximal value, and the geometric mean can't therefore be cal-
culated, the arithmetic mean is taken. The background information to the process "hot rolling, steel" is
given in Tab. 2.2, flows are shown in Fig. 2.2.
Data quality
Data is taken almost exclusively from IPPC (2001a), some estimations are also made. The values can
vary considerably depending on the degree of deformation.
Uncertainty is calculated assuming a lognormal distribution and the high value given in IPPC (2001a)
representing the 95 % geometric standard deviation. Where no range is given in IPPC (2001a), the
data quality is estimated by mean of the pedigree matrix, according to (Frischknecht et al. (2003)).

2. Steel forming
Tab. 2.2 Ecoinvent meta information for "hot rolling, steel"
Name hot rolling, steel

Location RER
Unit kg
Includes the process steps scarfing, grinding heating, de-scaling, rolling and fin-
Included Processes ishing. Semi-closed water circuit with water treatment plant is also included. Does
not include the material being rolled
Amount 1
Local Name Warm walzen, Stahl
Synonyms Hot forming
To achieve greater toughness, shock resistance and tensile strength, the raw
steel production outputs cast ingots, slabs, billets and beam blanks are hot rolled
ence function
to long or flat products or semi-finished products.
Start Date 1997
End Date 2002
Other Period Text
Geography text Data-set is representative for European Union
Average technique for EU. Assumption: heating 100% with natural gas. Furnaces
Technology text
of about 10 MW are approximated by furnace "> 100 kW"
Production Volume 127.8 million tonnes in 1996
Extrapolations none
The fact that no data were available about the percentage of companies repre-
sented by the study cited is considered in the uncertainties.

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Variations of 20% in consumption and emissions
propane/ butane, at
Propane  are possible w hen the e.g. thickness of a slab is oil f uels No RER 5.30E-04 k g IPCC 2001 1 1.22 (1,5,1,1,1,3,1)
refinery
increased f rom 200 to 250 mm.
Oxygen  chemicals inorganics No RER oxygen, liquid, at plant 7.15E-03 k g IPCC 2001 1 1.22 (1,5,1,1,1,3,1)
Dust emitted in cleaned w aste gas.
Corresponding geometric mean value of the
concentration is 0.024 g/m3. The air f low
 Dust <2.5 theref ore is 0.373 m3/kg dust. It is assumed that air unspecif ied Particulates, < 2.5 um 8.94E-06 k g IPCC 2001 1 8.94
particle size of the emitted dust is below 2.5
micrometers, because the source of the dust is a
combustion process.
Chromium dust Values in source f or oxides, metal content
hot rolling, steel, scarfing
 air unspecif ied Chromium 2.00E-08 k g IPCC 2001 1 4.47

to air calculated
Nickel dust to Values in source f or oxides, metal content
 air unspecif ied Nickel 2.00E-08 k g IPCC 2001 1 2.24
air calculated
Manganese Values in source f or oxides, metal content calculated f rom
 air unspecif ied Manganese 8.94E-08 k g IPCC 2001 1 2
dust to air calculated max value
Values in source f or oxides, metal content assuming
 Lead dust to air air unspecif ied Lead 1.77E-08 k g IPCC 2001 1 1.16
calculated lognormal
Copper dust to Values in source f or oxides, metal content distribution
 air unspecif ied Copper 7.18E-08 k g IPCC 2001 1 1.07
air calculated
calculated f rom mass balance (Dust - Sum(Cr-,Ni-
 Iron dust to air air unspecif ied Iron 6.06E-06 k g IPCC 2001 1 8.9
,Mn-,Pb-,Cu-Oxides))
From 4.4% to 58.7 % of dust f rom air cleaning
devices are landf illed. Given data represent f ew
manuf acturers, theref ore 20 % of total dust f rom disposal, dust,
residual
Filter Dust to air cleaning device is assumed to be landf illed. w aste unalloyed EA F steel,
 material No CH 4.42E-04 k g IPCC 2001 1 1.47
landf ill The rest is assumed to be recycled and thus not management 15.4% w ater, to
landf ill
inventoried. The proxy chosen to represent the residual material landf ill
dust stands f or dust disposal f rom steelmaking,
thus being quite accurate.
Assuming the air f low calculated by mean of the
 NOx air unspecif ied Nitrogen oxides 1.31E-05 k g IPCC 2001 1 2.05 (1,5,1,1,1,3,23)
dust emissions.
Assuming the air f low calculated by mean of the Carbon monoxide,
 CO air unspecif ied 2.24E-05 k g IPCC 2001 1 1.22 (1,5,1,1,1,3,24)
dust emissions. f ossil
 Scale >99% recycled --> not inventoried n.a. IPCC 2001
Fig. 2.2 Flows for "hot rolling, steel" and their representation in the ecoinvent database

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
It w as assumed, that the same air f low as f or
scarf ing is arising. A s the source of the dust is Particulates, > 2.5 um,
 Dust 2.5-10 air unspecif ied 5.11E-06 k g IPCC 2001 1 1.83
mechanical abrasion, it w as estimated, that half and < 10um
hot rolling, steel, grinding of the dust is betw een 2.5 and 10 micrometers
It w as assumed, that the same air f low as f or calculated f rom
scarf ing is arising. A s the source of the dust is max value
 Dust >10 air unspecif ied Particulates, > 10 um 5.11E-06 k g IPCC 2001 1 1.83
mechanical abrasion, it w as estimated, that half assuming
of the dust is bigger than 10 micrometers lognormal
Chromium (in distribution
 air unspecif ied Chromium 3.54E-07 k g IPCC 2001 1 5.77
Dust)
 Nickel (in Dust) air unspecif ied Nickel 2.24E-07 k g IPCC 2001 1 1.83
Manganese (in
 air unspecif ied Manganese 1.29E-07 k g IPCC 2001 1 1.58
Dust)
No data available about amount of sw arf output
 Sw arf f rom grinding. Sw arf is assumed to be n.a. IPCC 2001
completely recycled --> not inventoried
Scale (dry
 >99% recycled --> not inventoried 2.65E-03 k g IPCC 2001 1 13.2
material)
In continuous f urnaces blast f urnace gas, coke
oven gas, BOF gas, natural gas and dif f erent natural gas, burned in
heating
Natural Gas  combinations of these gases (mixed gases) are natural gas No RER industrial f urnace 1.56E+00 M J IPCC 2001 1 1.41
hot rolling, steel, furnace
systems
used. In this study the w hole consumption is >100kW
inventoried as natural gas.
Data f or the f urnace is given in (IPCC 2001) but
calculated f rom
not used in this study, because the f low s are
 Dust 3.16E-06 k g IPCC 2001 1 3.16 max value
already inventoried in the utilised EI-Module f or
assuming
natural gas in industrial f urnace >100kW.
lognormal
NOx f rom low -
 2.74E-04 k g IPCC 2001 1 1.83 distribution
NOx burner
 SO2 1.34E-05 k g IPCC 2001 1 44.7
 CO 6.52E-05 k g IPCC 2001 1 13
 Hydrocarbon 2.50E-06 k g IPCC 2001 1 2
Only 10.3% of total scale is landf illed and thus disposal, average
residual
inventoried. More detailed values f or single types w aste incineration residue,
 Furnace Scale material No CH 2.23E-04 k g IPCC 2001 1 21.7
of Mills are available in (IPCC 2001, p.65). management 0% w ater, to residual
landf ill
Deposition of f lue gas cleaning device as proxy material landf ill
Fig. 2.2 Flows for "hot rolling, steel" and their representation in the ecoinvent database (cont.)

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Only 3.6% of total scale is landfilled and thus is residual
hot rolling, w aste f urnace w astes, 0%
descaling
 Oil Free Scale inventoried. Main component of scale is assumed material No CH 4.83E-04 k g IPCC 2001 1 4.47
management w ater, to residual
steel, to be soot (elemental carbon). landf ill
material landf ill
disposal, used mineral

Only 5.6% of total scale is landfilled and thus is hazardous calculated f rom
w aste oil, 10% w ater, to
 Oily Scale inventoried. Oily w aste are aproximated used w aste No CH 2.13E-04 k g IPCC 2001 1 9.49 max value
management hazardous w aste
mineral oil incineration assuming
incineration
lognormal
Value calculated basing on the value given f or
distribution
Oil, Grease, the oil and grease emissions to air (as
 chemicals organics No RER lubricating oil, at plant 4.04E-03 k g IPCC 2001 1 2.83
Hydraulic Oil hydrocarbons) and the f act that they correspond
to 7% of the input.
Particulates Particulates originating f rom mill stands;
 (oxides, dust) inventoried as smaller than 2.5 micromrters air unspecif ied Particulates, < 2.5 um 8.94E-06 k g IPCC 2001 1 4.47
hot rolling, steel, hot rolling
<2.5 because of high temperatures.

Particulates
Particulates originating f rom coil handling stands; Particulates, > 2.5 um,
 (oxides, dust) air unspecif ied 2.24E-05 k g IPCC 2001 1 2.05 (1,5,1,1,1,3,26)
low er temperature produce coarser particles and < 10um
2.5-10
Particulates
Particulates originating f rom coil handling stands;
 (oxides, dust) air unspecif ied Particulates, > 10 um 2.24E-05 k g IPCC 2001 1 1.56 (1,5,1,1,1,3,25)
low er temperature produce coarser particles
>10
disposal, dust,
residual
w aste unalloyed EA F steel,
 Filter dust Collected f rom rolling stands material No CH 8.72E-04 k g IPCC 2001 1 8.72
management 15.4% w ater, to
landf ill
residual material landf ill calculated f rom
max value
NMV OC, non-methane
No values are given to make a distinction assuming
Fugitive Oil volatile organic
 betw een aromatic- und aliphatic hydrocarbons. air unspecif ied 2.83E-04 k g IPCC 2001 1 2.83 lognormal
Emissions compounds,
Inventoried as NMV OC. distribution
unspecif ied origin
Metallic By- Particulates originating f rom coil handling stands;

 1.02E-01 k g IPCC 2001 1 1.46
Products low er temperature produce coarser particles

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Inf rastructure f or w aste w ater treatment plant is
comprised in the inf rastructure module f or the
w hole rolling mill plant. V alue f or energy use is
Waste process Water, unspecif ied
 comprised in the broad range of uncertainty the resource in w ater 5.50E-03 m 3 IPCC 2001 1 2
w ater natural origin
energy data f or the rolling mill have. Water is
containig suspended solids, oil, etc. but no exact
hot rolling, steel, Waste water treatment plant
data about composition w ere available.

Flocculating
Assumption: same amount per cubic meter as f or construction lime, hydrated, loose,
agents  binder No CH 4.97E-05 k g IPCC 2001 1 1.2
w aste w ater f rom cold steel rolling materials at plant
(Ca(OH)2;95%)
Assumption: same amount per cubic meter as f or
w aste w ater f rom cold steel rolling. Personal chemicals inorganic, at
Polyelectrolytes  chemicals inorganics No GLO 1.45E-08 k g IPCC 2001 1 2
communication by G. Doka --> inorganic plant
chemicals as proxy
Discharged Treated eff luent is assumed to be discharged in
 5.50E-03 m 3 IPCC 2001 1 2
Water (DW) surf ace w ater --> not inventoried calculated f rom
TSS (Total max value
Suspended solids,
 Suspended w ater unspecif ied 1.46E-04 k g IPCC 2001 1 7.3 assuming
unspecif ied
Solids; in DW) lognormal
Chemical distribution
Oxygen COD, Chemical Oxygen
 w ater unspecif ied 3.78E-05 k g IPCC 2001 1 1.72
Demand (COD, Demand
in DW)
Biological
Oxygen BOD5, Biological Calculated from
 w ater unspecif ied 3.80E-05 k g 1 3.44
Demand (BOD, Oxygen Demand COD
in DW)
Dissolved
DOC, Dissolved Calculated from
 organic carbon w ater unspecif ied 1.20E-05 k g 1 3.44
Organic Carbon COD
(DOC, in DW)
Total organic
TOC, Total Organic Calculated from
 carbon (TOC, w ater unspecif ied 1.20E-05 k g 1 3.44
Carbon COD
in DW)
Hydrocarbon Hydrocarbons,
 w ater unspecif ied 1.73E-06 k g IPCC 2001 1 1.73
(in DW) unspecif ied

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
 Copper (in DW) w ater unspecif ied Copper, ion 2.66E-07 k g IPCC 2001 1 5.37
hot rolling, steel, Waste water treatment  Zinc (in DW)

Cadmium (in
w ater unspecif ied Zinc, ion 2.06E-07 k g IPCC 2001 1 9.35
 w ater unspecif ied Cadmium, ion 1.38E-07 k g IPCC 2001 1 2

DW)
A luminum (in
 w ater unspecif ied A luminum 1.30E-07 k g IPCC 2001 1 5.92
DW)
 Lead (in DW) w ater unspecif ied Lead 2.75E-07 k g IPCC 2001 1 2
calculated f rom
Chromium (in
 w ater unspecif ied Chromium, ion 4.95E-07 k g IPCC 2001 1 2 max value
DW)
assuming
Chromium6+
w ater unspecif ied Chromium V I IPCC 2001 1 2 lognormal
plant
 2.75E-08 k g
(in DW)
distribution
Manganese (in
 w ater unspecif ied Manganese 5.61E-07 k g IPCC 2001 1 2.55
DW)
 Iron (in DW) w ater unspecif ied Iron, ion 4.26E-06 k g IPCC 2001 1 2.58
 Nickel (in DW) w ater unspecif ied Nickel, ion 7.78E-07 k g IPCC 2001 1 14.1
Mercury (in
 w ater unspecif ied Mercury 2.75E-08 k g IPCC 2001 1 2
DW)
Free Chlorine
 w ater unspecif ied Chloride 1.23E-06 k g IPCC 2001 1 2.24
(in DW)
Assumption: same amount and composition per disposal, sludge f rom
Water residual
cubic meter as f or w aste w ater f rom cold steel w aste steel rolling, 20%
 Treatment material No CH 1.63E-02 k g IPCC 2001 1 1.22 (2,5,1,1,1,3,6)
rolling. Only a part of the sludge goes to landf ill -- management w ater, to residual
Sludge landf ill
> 59.4% invetoried material landf ill

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Only amount that is going to leave the system as
scrap to recycling. A ssumption basing on the steel, converter,
Primary steel  metals extraction No RER 5.00E-02 k g IPCC 2001 1 1.31 (4,5,1,1,1,3,3)
values f or cold sheet rolling taking into acount a unalloyed, at plant
hot rolling, steel, Overall better f ormability do to the high temperature
In the ref erence data w as available only f or the

rolling process in the narrow sense (70-140 electricity, medium
production
Electricity  kWh/t product), but additional electricity electricity No UCTE voltage, production 1.40E-01 k Wh IPCC 2001 1 1.14 (3,3,1,1,1,3,2)
mix
(estimated 40 kWh) is used f or the process UCTE, at grid
steps grinding, descaling and packing.
 Heat w aste air unspecif ied Heat, w aste 5.04E-01 M J IPCC 2001 1 1.09 (2,1,1,1,1,3,1)
Rolling shop  metals processing Y es RER rolling mill 1.62E-09 unit IPCC 2001 1 3.04 (3,4,1,1,1,4,9)
Includes the transports of all materials not
produced at the plant and the transport to
disposal of the w aste. 32t-lorry w as taken as a
proxy f or all this transports, theref ore transports transport (4,5,n.a.,n.a.,n.a
Lorry 32t  road No RER transport, lorry 32t 1.71E-02 tk m Estimation 1 2.09
to the municipal incineration are overestimated by systems .,n.a.,5)
one third (not relevant f or the overall process).
Distance to hazardous w aste disposal is
inventoried w ith 100 km
cardboard & packaging, corrugated

Paper; paper &
 corrugated No RER board, mixed f ibre, 4.72E-12 k g EA A 2000 1 1.38 (4,5,1,1,3,3,4)
hot rolling, steel, Packaging
cardboard cardboard
board single w all, at plant
packaging film, LDPE,

Plastics  plastics processing No RER 1.89E-11 k g EA A 2000 1 1.38 (4,5,1,1,3,3,4)
at plant
steel, converter,
Steel  metals extraction No RER 2.36E-11 k g EA A 2000 1 1.38 (4,5,1,1,3,3,4)
unalloyed, at plant
Steel coils  metals processing No RER sheet rolling, steel 2.36E-11 k g EA A 2000 1 1.38 (4,5,1,1,3,3,4)
saw n timber,
w ooden
Wood  extraction No RER sof tw ood, raw , air 1.62E-17 m 3 EA A 2000 1 1.38 (4,5,1,1,3,3,4)
materials
dried, u=20%, at plant
disposal, municipal
Packing material is disposed as mixed w aste w aste municipal solid w aste, 22.9%
 Waste No CH 7.08E-11 k g Estimation 1 1.31 (2,5,1,1,3,3,6)
w ith municipal w aste as proxy management incineration w ater, to municipal
incineration
Scrap, of f -
 recycled --> not inventoried 5.00E-02 k g Estimation 1
cuts, etc.
 Hot rolling metals processing No RER hot rolling, steel 1.00E+00 k g

2. Steel forming
2.3 Sheet rolling of steel

For many applications, the products of hot rolling are unsatisfactory, e.g., with respect to cross section,
surface quality, dimensional accuracy, and general finish, so that cold rolling is necessary. Cold rolled
products are mainly strips and sheets with high quality surface finish and precise metallurgical proper-
ties for use in high specification products.
In cold rolling, the properties of hot rolled strip products, e.g. thickness (typical final thickness from
0.1 to 3 mm), mechanical and technological characteristics, are changed by compression between roll-
ers without previous heating of the input. However, its temperature can increase during the forming
process, as the energy expended in forming is almost completely converted into heat. Frictional energy
at the contact surfaces of the workpiece, if established, is also converted to heat. In rapid adiabatic
processes the temperature can increase by >100 °C (IPPC (2001a); Wiegels (1997)).
Contrarily to hot rolling, the process route of cold rolling is different for steel (carbon-or (low-)alloy
steel) and for chromium steel (high alloy- or stainless steel). Therefore in this study those processes
are described in two different chapters (see chapter 1.1 for chromium steel sheet).
The advantages of cold forming include (Wiegels (1997)):
For the workpiece:
- Good dimensional accuracy and close tolerances
- Good surface properties
- Very small dimensions (sheet thickness ca. 10 μm, wire diameter ca. 10 μm)
- Work hardening (if desired)

For the tool:
- Reduced wear (effective lubricants can be used, the thermal load is low, and scaling does not oc-
cur)
Generally:
- No energy is required for heating the workpiece
Cold rolled sheets and plates production (wide strip) was 39.6 millions tonnes in 1996. The main pro-
ducing countries were Germany with 10.6 million tonnes followed by France with 6.3 million tonnes,
Italy 4.3 million tonnes, UK 4.0 million tonnes and Belgium 3.9 million tonnes. Cold rolled narrow
strip, obtained from cold rolling narrow hot strip or from slitting and cold rolling hot rolled sheet,
amounted to about 8.3 millions tonnes in 1994 (2.7 million tonnes of cold rolled and 5.5 million ton-
nes of slit strip). In 1994, Germany produced about 35 % of the slit strip, with 1.9 million tonnes, fol-
lowed by Italy and France each with a production of 0.9 million tonnes. (IPPC (2001a)).
The cold rolling mills for steel usually comprise the following process steps (IPPC (2001a)):
- Continuous pickling line,
- Cold rolling,
- Annealing,
- Temper mills,
- Inspection and finishing lines,
- Packaging lines for coils or sheets,

2. Steel forming
-
Roll shop.
A short description of every single process step is given below.
The infrastructure of cold rolling mill for carbon steel is described in chapter 1.1. It comprises build-
ings where production processes take place and for storage, surroundings and machinery fore the sin-
gle process steps (e.g.: furnace, rolls, cutting devices, etc.).
2.3.1 Pickling
The entire surface of the hot rolled coil is covered with a thin layer of scale containing oxides, which
must be removed prior to cold rolling. Normal steels are usually pickled in sulfuric acid at 95–100 °C,
or hydrochloric acid at 60–70 °C in three or four fully enclosed pickling baths. Before the actual pick-
ling process, there is usually a stretcher–bend–leveling device which also causes mechanical descal-
ing. This preliminary descaling is very important, as there must be a certain percentage of free iron
surface in the pickling bath, so that pickling can start. Once the strip or sheet is pickled, it must be
thoroughly rinsed with de-mineralised (or equal quality) water and subsequently dried. Oiling is done
either with rolling oil or anticorrosive oil. The used acid can be regenerated, in the case of sulfuric acid
by precipitating iron sulfate as its heptahydrate in a crystallizer. Hydrochloric acid can be regenerated
by decomposing the iron chloride to hydrochloric acid and iron oxide, by a spray process at 450 °C, or
in a fluidized bed at 850 °C (IPPC (2001a); Wiegels (1997))
2.3.2 Cold rolling

The following description is taken from (IPPC (2001a); Wiegels (1997)). After pickling, the descaled
hot-rolled strip is cold rolled passing through sets of rolls on reversing mills or on continuous in-line
mills. Many types of stands are used in the production of cold-rolled strip, e.g., two-high, four-high, or
larger.
Low carbon steels which are less hard, are generally rolled in multi-stand tandem mills due to a higher
capacity of these mills. They consist of a series of four or six-high stands. The strip enters the first
stand and undergoes an initial thickness reduction, further reductions are achieved in each subsequent
stand until the final gauge is attained. For low carbon strip generally an emulsion of 0.5 to 4 %
(10 - 20 % for direct application) of oil in water is necessary for lubrication, cooling of the strip, work
and back-up rolls, and for removal of iron-particles. In the rolling process the wide strip is mainly de-
formed in a longitudinal direction, with transverse deformation being practically negligible. For opti-
mum strip surface cleanliness it is important to avoid contamination of the tandem emulsion by the
hydraulic oil, by grease or by the cooling water (used to cool the emulsion).
In reversing stands, one roll pass can be immediately followed by another by reversing the direction of
the rollers. Reversing stands with many rolls exert a high pressure on the strip because the diameters
of the working rolls are small.
Flat cold-rolled strip achieve thickness variations of only a few thousandths of a millimetre and a high-
quality. Cold rolling causes work-hardening of the steel. This is reduced by an annealing stage.
2.3.3 Annealing
The following chapter is taken from (IPPC (2001a))
The basic stages of the annealing process comprise:
- Heating to annealing temperature (above 650 °C),
- Holding at annealing temperature,

2. Steel forming
- Cooling.
This annealing process may be carried out in batch furnaces or continuous furnaces. The annealing cy-
cle has a significant influence on the mechanical properties and, consequently, on the formability of
the steel strip. The main parameter controlling the annealing cycle is the temperature profile.
Continuous annealing and batch annealing are considered to be complementary processes and not fully
interchangeable. Production related issues are integral to the selection of the annealing technique.
Batch Annealing
Prior to the annealing process the strip may be cleaned (degreased) to produce a cleaner surface. The
most frequently used cleaning agents are phosphates, alkaline silicates, caustic soda and soda ash. The
steel sheet is unwound from the coil and passed through cleaning tanks, which may be stirred to im-
prove the cleaning effect. Subsequently, the steel is rinsed with water and recoiled. The degreasing so-
lution can be regenerated and recycled.
The cold rolled coiled strip is stacked in a hood furnace for annealing. The combustion chamber
(space between heating and protecting hood) is heated by oil or gas burners. Heat passes through the
protective hood into where the steel coils are stacked. A circulating fan provides as uniform tempera-
ture distribution as possible. The atmosphere in conventional plants is usually HNX gas (a nitrogen-
hydrogen mixture with a hydrogen content close to the flammability limit). The atmosphere can also
be 100 % H2 as in parts of the working cycle of high convection furnaces.
The heat passes into the coils through their outer edges, so that these areas are always hotter than the
inner windings, especially during heating up. The heat treatment causes the organic residues of the
emulsion to partly burn off; partly a distillation process takes place. Products of the reactions taking
place are CO/CO2, H2, FeOx and CH4. The strip is heated to re-crystallisation temperature, and an-
nealed at about 700 °C resulting in complete re-crystallisation of the cold rolled steel. For cooling the
coils, the heating hood is removed. The cooling effect can be enhanced by spraying water on the pro-
tecting hood; by covering it with a cooling bell and blowing air on it or by using a cooling by-pass
system which cools the protective atmosphere under the protective hood. The time necessary for an-
nealing depends on the annealing temperature and the weight of the charge and can take 2 to 7 days.
Annealing in this type of furnace is very slow and also the cooling rate is not suitable for certain steel
grades. Therefore an annealing process in a 100 % hydrogen atmosphere was developed yielding in
shorter annealing times and more rapid cooling.
Continuous Annealing
For continuous annealing the coils are welded together on the entry side of the plant and subjected to
the following process steps:
- Alkaline/electrolytic cleaning of the strips.
- Heating and holding at the required annealing temperature.
- Cooling (slow jet cooling, high gas jet cooling (using cooling rates up to 150 °C/s, hot water
quench (HOWAQ), overaging, roll cooling, final cooling, misting jet cooling).
The continuous annealing is carried out by passing the steel strip through a multi-zone heating furnace
with heating chamber, annealing chamber, cooling zone, tempering zone and a second cooling zone.
The steel is heated to a range between 650 °C and 830 °C and then cooled by gas jets, gas-water
sprays, contact rolls or water quenching, depending on the desired metallurgical properties. These fur-
naces are usually fired by gas (direct or indirect) or electrically heated. The steel can be protected by
an inert gas atmosphere or a reducing gas atmosphere in parts of the furnace. Continuous annealing
gives a shorter processing time (about 10 minutes), less intermediate storage, uniform mechanical
properties and a superior surface cleanliness. Continuous annealing is more appropriate for the produc-
tion of higher strength steels. The layout of the furnace facilitates energy recovery from the exhaust
gas.

2. Steel forming
2.3.4 Tempering
After annealing the surface finish and mechanical properties of the steel are modified in line with the
customer requirements. This is done by temper rolling which consists of subjecting the strip to a light
rolling pass with a thickness reduction between 0.3 and 2 %. Before tempering the strip temperature
must be less than 50 °C. Temper rolling is carried out in the temper mill which comprises typically
one or two four-high stands, although two-high or six-high stands are also possible (IPPC (2001a)).
2.3.5 Inspecting and finishing

Finishing comprises slitting into coils of different width strip and transversal shearing to obtain sheet.
Additional steps such as straightening, sampling, oiling and marking are also completed.
The finishing process comprises the following operations (IPPC (2001a)):
- Dimensional control (width, thickness and length),
- Inspection for surface defects and their removal,
- Sampling to determine the mechanical and technological properties, the strip roughness structure
and the content of residue of particles from roll wear and carbon on the strip surface (statistical
sampling),
- Trimming the coils to exact width,
- Straightening the strips to optimum flatness,
- Oiling the strips with electrostatic oiling machines or with oiling machines equipped with sprays
or rolls oiling machines (anticorrosive oils or prelubes),
- Marking the finished products with coil number, production date, etc.
- Welding of different smaller coils to bigger coils.
2.3.6 Packaging of coils or sheets

The material, now in its final form, is usually packed ready for despatch to the customer. The packag-
ing is designed to ensure that no damage occurs either during storage or transport to the end user. Con-
trol measures also have to be taken to prevent damage from mechanical and climatic sources. In addi-
tion the lifting and transport equipment in the production plant should also be designed and operated
to minimize damage to the steel. After finishing, packing of the individual items ordered by the cus-
tomers is done. The packing materials used includes: steel bands, paper, plastic, wood, seaworthy
packing, special packing (IPPC (2001a)).
2.3.7 Maintenance of the rolls

The activities generally consist of the dismantling of the chocks, the grinding of the rolls, the texturing
of the work rolls and the reassembling of the rolls. To satisfy the required strip tolerances and surface
standards the rolls must be regularly reground. During this grinding operation the rolls are usually
cooled and lubricated with a grinding emulsion. The spent emulsion can be recirculated and filtered,
but partial replacement is required periodically. In addition, used oil and grinding sludge are produced
which require disposal (IPPC (2001a)).

2. Steel forming
2.3.8 Environmental issues of sheet rolling of steel

The main environmental issues of cold rolling are: acidic wastes and waste water; degreaser fume,
acidic and oil mist emissions to air; oil-containing wastes and waste water (see chapter 2.2.6, Tab.
2.1); dust, e.g. from descaling and decoiling; NOx from mixed acid pickling and combustion gases
from furnace firing. As for acid emissions to air from cold rolling, these may arise from pickling and
acid regeneration processes. Emissions differ, depending on the pickling process used – basically the
acid used. Only little data were available for dust emissions from steel handling and descaling opera-
tions (IPPC (2001a)).
2.3.9 Steel sheet rolling in ecoinvent

culated, the arithmetic mean is taken. The background information to the process "sheet rolling, steel"
is given in Tab. 2.3, flows are shown in Fig. 2.3.
As no information on the shares of the different technologies for pickling, milling and annealing is
available it is assumed that the respective shares are 50% each.
Water emissions with the treated wastewater are given in chapter 2.2.6, Tab. 2.1.
Data quality
Data is taken almost exclusively taken from IPPC (2001a), some estimations are also made. The val-
ues can vary considerably depending on the degree of deformation.
representing the 95% geometric standard deviation. Where no range is given in IPPC (2001a), the data
quality is estimated by mean of the pedigree matrix, according to (Frischknecht et al. (2003)).

2. Steel forming
Tab. 2.3 Ecoinvent meta information for "sheet rolling, steel"
Name sheet rolling, steel

Location RER
Unit kg
Includes the process steps continuous pickling line, cold rolling, annealing,
Included Processes tempering, inspecting and finishing, packing coils or sheets, roll mainte-
nance. Does not include the material being rolled
Amount 1
Local Name Blech walzen, Stahl
Synonyms cold rolling
This process is to be used only for un- and low-alloyed steel. For many ap-
plications, the products of hot rolling are unsatisfactory, e.g., with respect to
General Comment to reference cross section, surface quality, dimensional accuracy, and general finish, so
function that cold rolling is necessary. Cold rolled products are mainly strips and
sheets with high quality surface finish and precise metallurgical properties
for use in high specification products.
Start Date 1997
End Date 2002
Other Period Text
Average technique for EU, for un- and low-alloyed steel. A 50-50 mix is in-
Technology text ventoried where different technologies are applied for the same process step
(e.g. pickling)
Production Volume Little less than 50 million tonnes in 1996
Extrapolations none
The fact that no data were available about the percentage of companies rep-
resented by the study cited is considered in the uncertainties.

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Only amount that is going to leave the system as steel, converter,
primary steel  metals extraction No RER 8.55E-02 k g IPCC 2001 1 1.31 (4,5,1,1,1,3,3)
w aste to recycling. unalloyed, at plant
HCl (33%) w ith hydrochloric acid, 30%
 chemicals inorganics No RER 3.97E-04 k g IPCC 2001 1 1.13
regeneration in H2O, at plant
sheet rolling, steel, HCl-pickling (Assumption 50% of sheet
A nticorrosive
 Lubricating oil as proxy chemicals organics No RER lubricating oil, at plant 7.07E-05 k g IPCC 2001 1 1.41
oils
Steam f or bath heat, unspecif ic, in
 chemicals organics No RER 2.29E-02 M J IPCC 2001 1 1.53
heating chemical plant
electricity, medium
Electrical production calculated f rom
 electricity No UCTE voltage, production 4.81E-03 k Wh IPCC 2001 1 2.31
energy mix max value
UCTE, at grid
assuming
Caloric energy natural gas, burned in
heating lognormal
(natural gas or  natural gas No RER industrial f urnace 3.16E-02 M J IPCC 2001 1 1.58
systems distribution
LPG) >100kW
Industrial &
w ater, deionised, at
demineralised  w ater supply production No CH 2.54E-02 k g IPCC 2001 1 2.55
plant
w ater
Concentration multiplied w ith the amount of
rolling)

 Dust to air exhaust gas f rom pickling tanks. Particle size air unspecif ied 5.00E-07 k g IPCC 2001 1 2
and < 10um
f rom chemical process: assumption 2.5-10 um.
 HCl to air air unspecif ied Hydrogen chloride 1.29E-07 k g IPCC 2001 1 2.05 (1,5,1,1,1,3,23)
calculated f rom
The sludge can be aproximated by Fe(OH)3. The
max value
amount of Fe in Fe(OH)3 (=76.66%) is
 Sludge cake soil unspecif ied Iron 3.05E-05 k g IPCC 2001 1 5.28 assuming
inventoried. A bout tw o thirds are recycled -->
lognormal
35% landf illed is inventoried
distribution
No value is given in ref erence. A s the oil input is
about the same f or HCl and H2SO4, it is disposal, used mineral
hazardous
Oil contamined assumed, that also the output is about the same. w aste oil, 10% w ater, to
 w aste No CH 5.55E-06 k g IPCC 2001 1 1.22 (1,5,1,1,1,3,6)
w aste Only 22.8% w hich are landf illed are inventoried, management hazardous w aste
incineration
the rest is recycled. Oil containig w astes are incineration
approximated w ith used mineral oil.
Scale dust
Ref erence contains neither composition nor
(collected by
 amount of collected dust. No inf ormation is n.a. IPCC 2001
dedusting
available about the share that can be recycled.
system)
Strip head / tail
 recycled --> not inventoried 1.50E-02 k g IPCC 2001 1
scrap
Fig. 2.3 Flows for "sheet rolling, steel" and their representation in the ecoinvent database

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
structure
category
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
sulphuric acid, liquid, at
H2SO4 (96%)  chemicals inorganics No RER 4.18E-03 k g IPCC 2001 1 1.2
plant
A nticorrosive
oils
Steam f or bath heat, unspecif ic, in
sheet rolling, steel, H2SO4-pickling
heating chemical plant
(Assumption 50% of sheet rolling)
electricity, medium
Electrical production
energy mix calculated f rom
UCTE, at grid
max value
 H2 air unspecif ied Hydrogen 7.27E-10 k g IPCC 2001 1 1.41
assuming
 SO4 air unspecif ied Sulf ate 3.46E-08 k g IPCC 2001 1 1.41
lognormal
 SO2 air unspecif ied Sulf ur dioxide 3.16E-07 k g IPCC 2001 1 1.58
distribution
Only 22.8% w hich are landf illed are inventoried, hazardous
Oil contamined w aste oil, 10% w ater, to
 the rest is recycled. Oily w astes are w aste No CH 5.55E-06 k g IPCC 2001 1 1.41
w aste management hazardous w aste
approxiamted by used mineral oil incineration
incineration
 Sludge cake soil unspecif ied Iron 1.47E-05 k g IPCC 2001 1 1.83
Scale dust
Ref erence contains neither composition nor
(collected by
 amount of collected dust. No inf ormation is n.a. IPCC 2001
dedusting
available about the share that can be recycled.
system)
calculated f rom
max value
Strip head / tail
 recycled --> not inventoried 2.04E-02 k g IPCC 2001 1 1.1 assuming
scrap
lognormal
distribution
Fig. 2.3 Flows for "sheet rolling, steel" and their representation in the ecoinvent database (cont.)

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Industrial & calculated f rom
demineralised  w ater supply production No CH 1.18E-02 k g IPCC 2001 1 1.69 max value
plant
w ater assuming
Rolling oil  sheet rolling, steel, Tandem Mill (Assumption 50% of sheet rolling) chemicals organics No RER lubricating oil, at plant 3.87E-04 k g IPCC 2001 1 2.58 lognormal
Other oils
(hydraulic,  no quantitative inf ormation n.a. IPCC 2001
Morgoil...)
heat, unspecif ic, in
Steam  chemicals organics No RER 8.66E-03 M J IPCC 2001 1 1.73
chemical plant
calculated f rom
electricity, medium
Electrical production max value
energy mix assuming
UCTE, at grid
lognormal
heating distribution
(natural gas or  natural gas No RER industrial f urnace 3.00E-03 M J IPCC 2001 1 6
systems
LPG) >100kW
Dust2.5-10 to Size distribution: mechanical process, no burning Particulates, > 2.5 um,
 air unspecif ied 2.40E-05 k g IPCC 2001 1 2.05 (1,5,1,1,1,3,26)
air process involved. and < 10um
Size distribution: mechanical process, no burning
 Dust>10 to air air unspecif ied Particulates, > 10 um 2.40E-05 k g IPCC 2001 1 1.56 (1,5,1,1,1,3,25)
process involved.
NMV OC, non-methane
Hydrocarbons volatile organic
to air compounds,
unspecif ied origin
NMV OC, non-methane
volatile organic
 Oil to air air unspecif ied 9.85E-06 k g IPCC 2001 1 2.05 (1,5,1,1,1,3,23)
compounds,
unspecif ied origin
Oil contamined Only 22.8% w hich are landf illed are inventoried, hazardous
 w aste the rest is recycled. Oily w astes are w aste No CH 2.28E-08 k g IPCC 2001 1 1.22 (1,5,1,1,1,3,6)
(disposed) approxiamted by used mineral oil incineration
incineration
Emulsion
 (internally recycled --> not inventoried 4.06E+00 k g IPCC 2001 1 1.62
calculated f rom
recycled)
max value
disposal, basic oxygen assuming
residual
Most of it is recycled --> 12.4% landf illed is w aste f urnace w astes, 0% lognormal
 sludge cake material No CH 7.20E-05 k g IPCC 2001 1 1.29
inventoried management w ater, to residual distribution
landf ill
material landf ill
Scrap, of f -
 recycled --> not inventoried 8.50E-03 k g IPCC 2001 1
cuts, etc.

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Industrial &
plant
w ater
Rolling oil  chemicals organics No RER lubricating oil, at plant 5.24E-05 k g IPCC 2001 1 1.05
electricity, medium
sheet rolling, steel, Reversing Mill
UCTE, at grid
assuming
heating lognormal
LPG) >100kW
NMV OC, non-methane
volatile organic
 HC to air air unspecif ied 4.61E-06 k g IPCC 2001 1 1.1
compounds,
unspecif ied origin
NMV OC, non-methane
volatile organic
 Oil to air air unspecif ied 5.40E-06 k g IPCC 2001 1 1.22 (1,5,1,1,1,3,6)
compounds,
unspecif ied origin

residual
Sludge cake Most of it is recycled --> 12.4% landf illed is w aste f urnace w astes, 0% calculated f rom
 material No CH 1.21E-04 k g IPCC 2001 1 1.03
(disposed) inventoried management w ater, to residual max value
landf ill
material landf ill assuming
lognormal
Emulsion
distribution
 (internally recycled --> not inventoried 4.37E-03 m 3 IPCC 2001 1 1.03
recycled)
Scrap, of f -
 recycled --> not inventoried 1.00E-02 k g IPCC 2001
cuts, etc.

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
electricity, medium
sheet rolling, steel, Batch annealing


UCTE, at grid
assuming
heating lognormal
LPG) >100kW
Inert gas
 no quantitative inf ormation n.a. IPCC 2001
(H2,N2)
Oil containing Oily w astes are approxiamted by used mineral hazardous
 w aste oil. Only 22.8% landf illed are inventoried, the rest w aste No CH 2.00E-05 k g IPCC 2001 1 1.22 (1,5,1,1,1,3,6)
(disposed) is recycled. incineration
incineration
SO2 - Arithmetic mean value, because the low er value
 air unspecif ied Sulf ur dioxide 5.00E-07 k g IPCC 2001 1 2
reversing mill of the range is zero.
SO2 -
Arithmetic mean value, because the low er value calculated f rom
 continuous air unspecif ied Sulf ur dioxide 1.03E-05 k g IPCC 2001 1 2
of the range is zero. max value
mills
assuming
NOx -
 air unspecif ied Nitrogen oxides 4.50E-05 k g IPCC 2001 1 10 lognormal
reversing mill
distribution
NOx -
 continuous air unspecif ied Nitrogen oxides 5.59E-06 k g IPCC 2001 1 22.4
mills

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Chemicals f or
alkaline and
 no quantitative inf ormation IPCC 2001
sheet rolling, steel, Continuous annealing (Assumption 50% of sheet
n.a.
electrolytic pre-
cleaning
electricity, medium
UCTE, at grid
assuming
heating lognormal
LPG) >100kW
Steam  chemicals organics No RER 5.95E-02 M J IPCC 2001 1 1.22 (1,5,1,1,1,3,1)
chemical plant
Inert gas
(H2,N2)
Wet temper
rolling agent
(f or rough  no quantitative inf ormation n.a. IPCC 2001
pass, electrical
sheet)
rolling)
A nticorrosive
oils
SO2 - Arithmetic mean value, because the low er value
reversing mill of the range is zero.
SO2 -
Arithmetic mean value, because the low er value
 continuous air unspecif ied Sulf ur dioxide 1.03E-05 k g IPCC 2001 1 2
of the range is zero.
mills
NOx -
 air unspecif ied Nitrogen oxides 4.50E-05 k g IPCC 2001 1 10
reversing mill calculated f rom
NOx - max value
 continuous air unspecif ied Nitrogen oxides 5.59E-06 k g IPCC 2001 1 22.4 assuming
mills lognormal
Waste heat distribution
 (steam, can be recycled --> not inventoried 2.69E-02 M J IPCC 2001 1 1.45
recovered)
Containing oil. Oily w astes are approxiamted by hazardous
Sludge cake w aste oil, 10% w ater, to
 used mineral oil. Most of it is recycled --> 12.4% w aste No CH 5.70E-06 k g IPCC 2001 1 5.11
(disposed) management hazardous w aste
landf illed is inventoried incineration
incineration
 Scale dust no quantitative inf ormation n.a. IPCC 2001
Oil sludge f rom
pre-cleaning

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Industrial &
plant
w ater
electricity, medium
energy

sheet rolling, steel, electricity
mix
No UCTE voltage, production
UCTE, at grid
1.52E-02 k Wh IPCC 2001 1 2.74
calculated f rom
max value
Tempering f luid

Temprering 5% solutions of w et temper rolling agents, nitrite
chemicals organics No RER lubricating oil, at plant 5.74E-04 k g IPCC 2001 1 5.22 assuming
concentrate based. Lubricating oil as proxy
lognormal
Anticorrosive
 Lubricating oil as proxy chemicals organics No RER lubricating oil, at plant 4.47E-05 k g IPCC 2001 1 1.12 distribution
oils
Oil Oily w astes are approximated by used mineral hazardous
 contaminated oil. Only 22.8% landf illed are inventoried, the rest w aste No CH 1.32E-05 k g IPCC 2001 1 1.12
w ater is recycled. incineration
incineration
Tempering /
 recycled --> not inventoried n.a. IPCC 2001
cooling f luid
 Sludge cake recycled --> not inventoried 3.51E-04 k g IPCC 2001 1 1.41
 Leakage oil soil unspecif ied Oils, unspecif ied 3.85E-05 k g IPCC 2001 1 1.12
Anticorrosive
oils
A ssumed as irrelevant --> input not inventoried,
Singing ink  inventoried f or transport and in a w aste mix f or 6.32E-06 k g IPCC 2001 1 12.6
sheet rolling, steel, Finishing (cutting,
w aste disposal
packaging f ilm, LDPE,
Film (clear)  plastics processing No RER 8.49E-05 k g IPCC 2001 1 1.06
at plant
No description available in ref erence --> paper & packaging kraf t paper,
Paper (blue)  No RER 2.68E-05 k g IPCC 2001 1 2.98
inspection, packing)
assumption kraf t paper as proxy cardboard papers unbleached, at plant

No description available in ref erence --> paper & packaging kraf t paper, bleached,
VCI Paper  No RER 1.60E-06 k g IPCC 2001 1 1.06
assumption kraf t paper as proxy cardboard papers at plant calculated f rom
Ref erence contained no f urther description of max value
the product. Steel sheet w as choosen as proxy. steel, electric, un- and assuming
FLEX-Stripes  metals extraction No RER 2.36E-04 k g IPCC 2001 1 1.1
V olume w as calculated w ith an assumed sheet- low -alloyed, at plant lognormal
thikness of 0.55 mm. distribution
FLEX-Stripes  metals processing No RER sheet rolling, steel 2.36E-04 k g IPCC 2001 1 1.1
saw n timber,
w ooden
Wood  extraction No RER sof tw ood, raw , air 1.85E-06 m 3 IPCC 2001 1 1.05
materials
electricity, medium
energy mix
UCTE, at grid
Only 22.8% landf illed are inventoried, the rest is hazardous
 recycled. Oil containig w aste is approxiamted by w aste No CH 2.93E-05 k g IPCC 2001 1 1.07
used mineral oil incineration
incineration
 Scrap recycled --> not inventoried 3.16E-02 k g IPCC 2001 1 1.26

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Grinding
emulsion  Lubricating oil as proxy chemicals organics No RER lubricating oil, at plant 5.66E-06 k g IPCC 2001 1 14.1
(f resh) calculated f rom
Shot grain One of the most important abrasive materials is max value
 chemicals inorganics No RER silicon carbide, at plant 4.47E-05 k g IPCC 2001 1 1.12
(SBT) taken as proxy assuming
Soluble oil lognormal
concentrate distribution
Chromium salt sodium dichromate, at
sheet rolling, steel, roll maintenance
 Na-dichromate as proxy chemicals inorganics No RER 1.50E-04 k g IPCC 2001 1 1.07

(PRETEX) plant
Dust2.5-10 to Size distribution: mechanical process w ithout Particulates, > 2.5 um,
air burning process and < 10um
Size distribution: mechanical process w ithout
 Dust>10 air unspecif ied Particulates, > 10 um 2.09E-08 k g IPCC 2001 1 1.56 (1,5,1,1,1,3,25)
burning process
 Chromium air unspecif ied Chromium 4.18E-09 k g IPCC 2001 1 5.06 (1,5,1,1,1,3,22)
disposal, sludge f rom
residual
Grinding w aste steel rolling, 20%
 Water treatment sludge as proxy material No CH 1.41E-04 k g IPCC 2001 1 1.41
Sludge management w ater, to residual
landf ill
material landf ill
inventoried in inf rastructure module --> not
 Grinding w heel 9.49E-07 k g IPCC 2001 1 1.05
separately inventoried
Only 22.8% landf illed are inventoried, the rest is hazardous
Waste w aste oil, 10% w ater, to
 recycled. Containing oil. Oily w astes are w aste No CH 1.21E-05 k g IPCC 2001 1 1.51 calculated f rom
emulsion management hazardous w aste
approximated by used mineral oil incineration max value
incineration
assuming
Oil Only 22.8% landf illed are inventoried, the rest is hazardous lognormal
 contaminated recycled. Oily w astes are approximated by used w aste No CH 3.40E-05 k g IPCC 2001 1 1.15 distribution
w aste mineral oil incineration
incineration
SBT (Shot Blast Texturing): This is a mechanical
SBT-dust2.5- Particulates, > 2.5 um,
 application of the texture onto the roll using grit air unspecif ied 2.24E-05 k g IPCC 2001 1 1.12
10 and < 10um
blasting. Theref ore the size distribution
 SBT-dust>10 application of the texture onto the roll using grit air unspecif ied Particulates, > 10 um 2.24E-05 k g IPCC 2001 1 1.12

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Demineralized w ater, deionised, at
 w ater supply production No CH 7.73E-02 k g IPCC 2001 1 1.55
Water plant
sheet rolling, steel, hydr fluids

Grease, oil,
hydraulic oil
Rolling oil  Lubricating oil as proxy chemicals organics No RER lubricating oil, at plant 9.95E-04 k g IPCC 2001 1 2.21
Hydraulic &
lubrication oils
electricity, medium
UCTE, at grid
assuming
Steam  Steam used as energy source chemicals organics No RER 4.47E-02 M J IPCC 2001 1 1.12 lognormal
chemical plant
distribution
disposal, steel in car
Disposed Rejects made mainly of steel. Only 34% landf illed w aste municipal shredder residue, 0%
 No CH 7.91E-03 k g IPCC 2001 1 2
w aste: rejects are inventoried, the rest is recycled. management incineration w ater, to municipal
incineration
Disposed Only 22.8% landf illed are inventoried, the rest is hazardous
 w aste: Used oil recycled. Oily w astes are approximated by used w aste No CH 7.63E-04 k g IPCC 2001 1 1.05
+ w ater mineral oil incineration
incineration
Water, cooling,
sheet rolling, steel,
Cooling w ater  Sum of the dif f erent process steps resource in w ater unspecif ied natural 2.70E-02 m 3 assumption 1 1.22 (1,5,1,1,1,3,12)
water and water
origin
One part of the incomin cooling and process
treatment
Cooling and w ater leaves the system as not contamined

 2.12E-02 m 3 assumption 1
process w ater evaporated w ater or included in discarded
w aste --> not inventoried
Flocculating calculated f rom
agents  binder No CH 5.42E-05 k g assumption 1 1.2 max value
materials at plant
(Ca(OH)2;95%) assuming
Personal communication by G. Doka --> inorganic chemicals inorganic, at lognormal
Polyelectrolytes  chemicals inorganics No GLO 1.58E-08 k g assumption 1 2
chemicals as proxy plant distribution

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
One part of the incoming cooling and process
w ater leaves the system as treated ef f luent; it is
assumed to be discharged in surf ace w ater -->
not inventoried. Inf rastructure f or w aste w ater
Treated w aste
 treatment plant is comprised in the inf rastructure 6.00E-03 m 3 assumption 1
w ater
module f or the w hole rolling mill plant. V alue f or
sheet rolling, steel, water and water treatment
energy use is comprised in the broad range of

uncertainty the energy data f or the rolling mill
have.
Suspended solids,
 TSS w ater unspecif ied 1.59E-04 k g assumption 1 1.57 (2,5,1,1,1,3,32)
unspecif ied
 COD w ater unspecif ied 4.13E-05 k g assumption 1 1.57 (2,5,1,1,1,3,32)
Demand
BOD5, Biological
 BOD w ater unspecif ied 4.13E-05 k g assumption 1 2.79 (5,5,5,5,5,5,32)
Oxygen Demand
DOC, Dissolved
 DOC w ater unspecif ied 1.33E-05 k g assumption 1 2.79 (5,5,5,5,5,5,32)
Organic Carbon
TOC, Total Organic
 TOC w ater unspecif ied 1.33E-05 k g assumption 1 2.79 (5,5,5,5,5,5,32)
Carbon
Hydrocarbons,
 Hydrocarbons w ater unspecif ied 1.89E-06 k g assumption 1 3.05 (2,5,1,1,1,3,34)
unspecif ied
 Cu w ater unspecif ied Copper, ion 2.90E-07 kg assumption 1 1.86 (2,5,1,1,1,3,37)
 Zn w ater unspecif ied Zinc, ion 2.24E-07 kg assumption 1 1.86 (2,5,1,1,1,3,37)
 Cd w ater unspecif ied Cadmium, ion 1.50E-07 kg assumption 1 1.86 (2,5,1,1,1,3,37)
 Al w ater unspecif ied A luminum 1.42E-06 kg assumption 1 1.86 (2,5,1,1,1,3,37)
 Pb w ater unspecif ied Lead 3.00E-07 kg assumption 1 1.86 (2,5,1,1,1,3,37)
 Cr w ater unspecif ied Chromium, ion 5.40E-07 kg assumption 1 1.86 (2,5,1,1,1,3,37)
 Cr6+ w ater unspecif ied Chromium V I 3.00E-08 kg assumption 1 1.86 (2,5,1,1,1,3,37)
 Mn w ater unspecif ied Manganese 6.12E-07 kg assumption 1 1.86 (2,5,1,1,1,3,37)
 Fe w ater unspecif ied Iron, ion 4.65E-06 kg assumption 1 1.86 (2,5,1,1,1,3,37)
 Ni w ater unspecif ied Nickel, ion 8.49E-07 kg assumption 1 1.86 (2,5,1,1,1,3,37)
 Hg w ater unspecif ied Mercury 3.00E-08 kg assumption 1 1.86 (2,5,1,1,1,3,37)
 Free Chlorine w ater unspecif ied Chloride 1.34E-06 kg assumption 1 1.57 (2,5,1,1,1,3,33)
Water residual
Only a part of the sludge goes to landf ill --> w aste steel rolling, 20%
 treatment material No CH 2.00E-02 k g assumption 1 1.57 (2,5,1,1,1,3,43)
66.5% invetoried management w ater, to residual
sludge landf ill
material landf ill

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Rolling Shop
 metals processing Y es RER rolling mill 1.62E-09 unit Estimation 1 3.04 (3,4,1,1,1,4,9)
(CR)
sheet rolling, steel, Overall

disposal, municipal
As a proxy, packing materials are mixed and w aste municipal solid w aste, 22.9%
 Waste-mix No CH 3.58E-04 k g Estimation 1 1.22 (2,5,1,1,1,3,6)
disposed as municipal w aste management incineration w ater, to municipal
incineration
(n.a.,5,n.a.,n.a.,
 Waste Heat air unspecif ied Heat, w aste 8.04E-01 M J IPCC 2001 1 1.56
n.a.,n.a.,7)
Cold rolling
 metals processing No RER sheet rolling, steel 1.00E+00 k g
sheet

2. Steel forming
2.4 Sheet rolling of chromium steel

The process route for high alloy steel (stainless steel) requires an initial annealing step prior to pick-
ling. This is necessary because, after the hot rolling process, the strip cools slowly, and carbon tends to
precipitate at the grain boundaries, owing to its reduced solubility at lower temperature. Annealing
causes the carbon to go back into solution. Rapid cooling of the strip then prevents the carbon from
precipitation by diffusion. Further more, several additional annealing and/or pickling steps during roll-
ing may be necessary.
Due to the different material characteristics, the field of application of chromium steel is very different
form the one of steel. The production volume of chromium steel is much smaller (estimated 5 %) than
3
the one of carbon and low alloyed steel .
The cold rolling mills for chromium steel usually comprise the following process steps (based on IPPC
(2001a)):
- Annealing (I) and pickling (I) of chromium steel,
- Cold rolling,
- Annealing (II) and pickling (II) of chromium steel,
- Tempering,
- Inspecting and finishing,
- Packing coils or sheets,

-
Roll maintenance.
A short description of the process steps which differs from the steel sheet rolling (annealing and pick-
ling I & II and cold rolling) is given below. For the other process steps see chapter 2.3.
The infrastructure of cold rolling mill for stainless steel is described in chapter 1.1. It comprises build-
ings where production processes take place and for storage, surroundings and machinery for the single
process steps (e.g.: furnace, rolls, cutting devices, etc.).
2.4.1 Annealing (I) and pickling (I) of chromium steel

The process of first annealing and pickling of chromium steel is described in (IPPC (2001a)).
Due to the presence of an oxide scale and chromium depleted layer formed during the hot rolling the
steel must also be descaled / pickled prior to cold rolling. In addition, the greater hardness of the steel
compared with carbon steels necessitates an initial annealing (I). The annealing process has three
components, heating to annealing temperature, temperature equalisation and cooling. Both continuous
and batch annealing processes may be used.
Ferritic grades are generally annealed as tightly wound coils in a batch heat treatment facility. Heating
may be achieved via gas firing or electrical heating. An inert protective atmosphere (nitro-
gen/hydrogen) is required for metallurgical reasons. Ferritic steels are generally annealed at steel tem-
peratures of up to 800 °C.
For continuous annealing of austenitic steels the steel coil is unwound and passed through one or more
heat treatment furnaces. An oxidizing atmosphere is needed to form a scale, which is rich in oxygen to
allow a better pickling. The waste gases are exhausted via flues under natural or forced draft. Strip
temperatures typically up to 1100 °C are necessary for annealing austenitic steels. The steel strip then
3
Personal communication on 25. February 03 from Dr. Ch. Zeltner, Stahl Gerlafingen, Swiss Steel Ltd.

2. Steel forming
passes through the cooling section and may be cooled using gas jets, air, water sprays or water
quenching. Continuous annealing is generally combined with a descaling/pickling plant to form a con-
tinuous annealing and pickling line.
After annealing the steel is descaled to produce a steel surface suitable for cold rolling. Mechanical
descaling such as shot blasting or scale breaking may be used prior to chemical pickling to remove the
heavy scale burden. However, careful control of such mechanical processes is necessary to minimise
damage to the strip surface, which could affect the final product quality. Final scale removal including
the removal of the chromium depleted layer beneath the scale is generally done by pickling in a mix-
ture of nitric and hydrofluoric acids at temperatures of up to 70 °C.
The chemical reactions that take place during the pickling of stainless steels are complex and involve
the removal of the adherent scale layer by the combined action of the acids used. For further details
consult directly the source (IPPC (2001a)).
Pickling line design is usually the same as for pickling of low alloy steels. Except, that on exit from
the last pickling tank the steel strip is rinsed with water to remove traces of acid, dried, but no applica-
tion of oil is necessary because of the corrosion resistance of the steel.
2.4.2 Cold rolling

The cold rolling mill is described in (IPPC (2001a)) as follows:
After the initial annealing and pickling the steel is generally rolled to the required thickness on revers-
ing cluster mills for a number of passes until the desired dimensions are achieved or until work hard-
ening necessitates further annealing. Reversing stands with many rolls exert a high pressure on the
strip because the diameters of the working rolls are small. These stands are therefore mainly used for
steels with work-hardening properties, e.g., stainless steels and electric sheet steels.
The process of cold reduction generates heat, part of which is transferred to the rolling oil which is
sprayed onto the steel and rolls for cooling as well as lubrication. This heat is subsequently dissipated
through heat exchangers to cooling water systems.
Mineral oils are generally used as rolling oil and close control of the oil cleanliness is necessary for
optimum performance. This may be achieved using oil filtration circuits, which incorporate media
element filters or powder pre-coat filter systems. Although the use of mineral oils predominates, emul-
sion cooling systems similar to those used for carbon steels may be employed. In such cases additional
precautions must be taken to control oil cleanliness in order to avoid strip marking. Collection hoods
are fitted to the mill and operate under continuous extraction to remove oil mist that is generated.
2.4.3 Annealing (II) and pickling (II) of chromium steel

Annealing and pickling are described in (IPPC (2001a); Wiegels (1997)).
Prior to the final annealing process, degreasing of the strip may be required to remove any contami-
nants from the steel surface. This can be achieved using an alkali based cleaning systems, which may
be incorporated in the entry section of continuous annealing lines. The steel coil is unwound and
passed through a series of tanks containing the cleaning solutions. Agitation of the solution is usually
incorporated to enhance the cleaning. Afterwards the strip is rinsed with water. De-mineralised water
can be used for the final rinsing step.
The process route for final annealing and pickling is dictated by the surface finish required. Examples
of this are EN standard finish ‘2R’ which requires a bright annealing treatment and surface finish ‘2B’
which requires annealing and pickling.

2. Steel forming
Bright annealing is generally conducted in an electrically heated or gas fired furnace with a protective
inert atmosphere of nitrogen and/or hydrogen. Both batch annealing and continuous annealing may be
used. Degreasing of the steel surface is required to prevent surface discolouration.
Material with a ‘2B’ surface finish is generally processed on continuous annealing and pickling lines.
The furnace design and operation is generally similar to that for the annealing of hot band material de-
scribed above. The steel is unwound and passed through one or more furnaces, which are usually gas
fired. Annealing takes place in an oxygen rich atmosphere. This is necessary to ensure that the scale
generated in the furnace is of a composition that is readily removable by the downstream chemical
pickling processes. Stainless steels are to an increasing extent being annealed in bright annealing
plants, as this process enables the bright surface produced by cold rolling, and typical of special steels,
to be retained.
Unlike the descaling of hot band material described above, mechanical descaling techniques cannot be
employed to assist in the removal of this scale due to the surface damage which would result for the
final cold rolled strip. Therefore only chemical pickling is applied. This is done in the same manner as
described above for hot band pickling. Scale of high alloy steel, due to the presence of oxides of alloy-
ing elements, cannot always be removed in just a single acid descaling step. Additional treatment or
pre-treatment processes may be necessary. These may include scale removal by an initial electrolytic
descaling process (using e.g. sodium sulfate) installed prior to the mixed acid section.
2.4.4 Environmental issues of sheet rolling of chromium steel

Unlike sulfuric acid or hydrochloric acid pickling solutions, regeneration is not possible, so that the
spent acids must be rendered harmless and disposed off (Wiegels (1997)). For other environmental is-
sues, see chapter 2.3.8.
2.4.5 Chromium steel sheet rolling in ecoinvent

culated, the arithmetic mean is taken. The background information to the process "sheet rolling, chro-
mium steel" is given in Tab. 2.4, flows are shown in Fig. 2.4.
Water emissions with the treated wastewater are given in chapter 2.2.6, Tab. 2.1.
Data quality
Data is almost exclusively taken from IPPC (2001a), some estimations are also made. The values can
vary considerably depending on the degree of deformation.

2. Steel forming
Tab. 2.4 Ecoinvent meta information for "sheet rolling, chromium steel"
Name sheet rolling, chromium steel

Location RER
Unit kg
Includes the process steps annealing (I) and pickling (I), cold rolling, annealing (II)
Included Processes and pickling (II), tempering, inspecting and finishing, packing coils or sheets, roll
maintenance. Does not include the material beeing rolled.
Amount 1
Local Name Blech walzen, Chromstahl
Synonyms
This process is to be used only for high alloyed (chromium) steel. For many appli-
cations, the products of hot rolling are unsatisfactory, e.g., with respect to cross
General Comment to refer- section, surface quality, dimensional accuracy, and general finish, so that cold
ence function rolling is necessary. Cold rolled products are mainly strips and sheets with high
quality surface finish and precise metallurgical properties for use in high specifica-
tion products.
Start Date 1997
End Date 2002
Other Period Text
Technology text Average technique for EU, for high-alloyed or chromium steel.
Production Volume Less than 2.5 million tonnes per year (Zeltner --> 5%Stahl)
Extrapolations none

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Primary Only part of the steel that leaves the system as chromium steel 18/8, at
 metals extraction No RER 5.16E-02 k g IPCC 2001 1 1.31 (4,5,1,1,1,3,3)
chromium steel scrap to recycling is inventoried plant
Industrial &
plant
sheet rolling, chromium steel, Pickling and annealing
w ater
One of the most important abrasive materials is
Shot  chemicals inorganics No RER silicon carbide, at plant 2.12E-03 k g IPCC 2001 1 1.41 calculated f rom
taken as proxy
max value
Nitric acid 1.4 kg HNO3 (50%) is used as proxy f or 1 kg nitric acid, 50% in H2O,
 chemicals inorganics No RER 7.67E-03 k g IPCC 2001 1 1.83 assuming
(HNO3 70%) HNO3 (70%) at plant
lognormal
Hydrof luoric hydrogen f luoride, at
 chemicals inorganics No GLO 4.33E-03 k g IPCC 2001 1 1.73 distribution
acid (HF 70%) plant
Hydrogen
0.7 kg of hydrogen peroxide (50%) is used as hydrogen peroxide,
peroxide (H2O2  chemicals inorganics No RER 3.83E-03 k g IPCC 2001 1 1.83
proxy f or 1 kg H202 (35%) 50% in H2O, at plant
35%)
Electrical energy needed to compress the air is
Compressed air  assumed to be included in the overall process n.a. IPCC 2001
electricity demand. The air is not inventoried
electricity, medium
UCTE, at grid
max value
heating assuming
systems lognormal
LPG) >100kW
distribution
Steam (acid heat, unspecif ic, in
heating) chemical plant
Metal oxide
and/or scale
particles, shot
f ragments
Particle size distribution f rom mechanical
Particulates, > 2.5 um, calculated f rom
 Dust2.5-10 descaling is assumed to be 50-50 smaller and air unspecif ied 7.07E-06 k g IPCC 2001 1 1.41
and < 10um max value
bigger than 10um
assuming
 Dust>10 air unspecif ied Particulates, > 10 um 7.07E-06 kg IPCC 2001 1 1.41
lognormal
 HF air unspecif ied Hydrogen f luoride 8.25E-07 kg IPCC 2001 1 4.12
distribution
 NOx air unspecif ied Nitrogen oxides 1.10E-04 kg IPCC 2001 1 36.5
 SO2 air unspecif ied Sulf ur dioxide 1.00E-06 kg IPCC 2001 1 1.22 (1,5,1,1,1,3,15)
Fig. 2.4 Flows for "sheet rolling, chromium steel" and their representation in the ecoinvent database

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
calculated f rom
sheet rolling, chromium
max value
steel, Pickling and

 Sludge cake soil unspecif ied Iron 5.48E-03 k g IPCC 2001 1 2.74 assuming
lognormal
annealing distribution
Spent pickle
no quantitative inf ormation. It is assumed to be
 bath / mixed n.a. IPCC 2001
recycled --> not inventoried
acid
SO2 f rom
f urnace
NOx f rom
 air unspecif ied Nitrogen oxides 1.22E-05 k g IPCC 2001 1 122
f urnace
Oil (mineral)  Lubricating oil as proxy chemicals organics No RER lubricating oil, at plant 2.58E-03 k g IPCC 2001 1 2
Oil f iltration
sheet rolling, chromium steel, High
media
electricity, medium
energy mix
UCTE, at grid calculated f rom
alloy reversing mill
NMV OC, non-methane max value

volatile organic assuming
 Oil Mean value of tw o mean values air unspecif ied 7.61E-05 k g IPCC 2001 1 1.2
compounds, lognormal
unspecif ied origin distribution
Assumption 50% f ilterdust. Only 12.4% landf illed disposal, dust,

Oil contamined residual
are inventoried. The proxy chosen to represent w aste unalloyed EA F steel,
 w aste f ilter material No CH 1.39E-04 k g IPCC 2001 1 1.25
the dust stands f or dust disposal f rom management 15.4% w ater, to
media landf ill
steelmaking, thus being quite accurate. residual material landf ill

Oil contamined hazardous
Assumption 50% oil contamined w aste. Only the w aste oil, 10% w ater, to
 w aste f ilter w aste No CH 1.39E-04 k g IPCC 2001 1 1.25
12.4% being landf illed are inventoried management hazardous w aste
media incineration
incineration
Scrap, of f -
 Recycled --> not inventoried 2.00E-02 k g IPCC 2001 1
cuts, etc.
Fig. 2.4 Flows for "sheet rolling, chromium steel" and their representation in the ecoinvent database (cont.)

2. Steel forming
Genera l Flow Information Re presentation in ecoinvent Uncerta inty Informations
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Industrial &
demineralised
sheet rolling, chromium steel,

plant
w ater
electricity, medium
UCTE, at grid
max value
Tempering f luid Tempering
 Good recycling rate --> assumed as not relevant 5.74E-04 k g IPCC 2001 1 5.22 assuming
concentrate
lognormal
A nticorrosive
 Lubricating oil as proxy chemicals organics No RER lubricating oil, at plant 4.47E-05 k g IPCC 2001 1 1.12 distribution
oils
Oil Oil containig w astes are approximated by used hazardous
 contaminated mineral oil. Only the 22.8% being landf illed are w aste No CH 1.32E-05 k g IPCC 2001 1 1.12
w ater inventoried incineration
incineration
Tempering /
 Recycled --> not inventoried n.a. IPCC 2001
cooling f luid
 Sludge cake Recycled --> not inventoried 2.83E-03 k g IPCC 2001 1 1.41
 Leakage oil soil industrial Oils, unspecif ied 3.85E-05 k g IPCC 2001 1 1.12
A nticorrosive
oils
A ssumed as irrelevant --> input not inventoried,
sheet rolling, chromium steel, Finishing
Singing ink  inventoried f or transport and in a w aste mix f or 6.32E-06 k g IPCC 2001 1 12.6
w aste disposal
Film (clear) plastics processing No RER IPCC 2001 1 1.06
(cutting, inspection, packing)
 8.49E-05 k g
at plant
No description available in ref erence --> paper & packaging kraf t paper,
Paper (blue)  No RER 2.68E-05 k g IPCC 2001 1 2.98
assumption kraf t paper as proxy cardboard papers unbleached, at plant
No description available in ref erence --> paper & packaging kraf t paper, bleached,
V CI Paper  No RER 1.60E-06 k g IPCC 2001 1 1.06
assumption kraf t paper as proxy cardboard papers at plant calculated f rom
Ref erence contained no f urther description of max value
the product. Steel sheet w as choosen as proxy. steel, electric, un- and assuming
FLEX-Stripes  metals extraction No RER 2.36E-04 k g IPCC 2001 1 1.1
V olume w as calculated w ith an assumed sheet- low -alloyed, at plant lognormal
thikness of 0.55 mm. distribution
FLEX-Stripes  metals processing No RER sheet rolling, steel 2.36E-04 k g IPCC 2001 1 1.1
saw n timber,
w ooden
Wood  extraction No RER sof tw ood, raw , air 1.85E-06 m 3 IPCC 2001 1 1.05
materials
electricity, medium
energy mix
UCTE, at grid
Oil containig w astes are approximated by used hazardous
 mineral oil. Only the 22.8% being landf illed are w aste No CH 2.93E-05 k g IPCC 2001 1 1.07
inventoried incineration
incineration
 Scrap recycled --> not inventoried 3.16E-02 k g IPCC 2001 1 1.26

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Grinding
emulsion  Lubricating oil as proxy chemicals organics No RER lubricating oil, at plant 5.66E-06 k g IPCC 2001 1 14.1
(f resh)
calculated f rom
Shot grain One of the most important abrasive materials is
 chemicals inorganics No RER silicon carbide, at plant 4.47E-05 k g IPCC 2001 1 1.12 max value
(SBT) taken as proxy
sheet rolling, chromium steel, roll maintenance
assuming
Soluble oil
 Lubricating oil as proxy chemicals organics No RER lubricating oil, at plant 5.48E-06 k g IPCC 2001 1 1.1 lognormal
concentrate
distribution
w ood preservative,
Chromium salt
 Inorganic salt containing chromium as proxy paintings production No RER inorganic salt, 1.50E-04 k g IPCC 2001 1 1.07
(PRETEX)
containing Cr, at plant
Dust2.5-10 to Size distribution: mechanical process w ithout Particulates, > 2.5 um,
air burning process and < 10um
Size distribution: mechanical process w ithout
 Dust>10 air unspecif ied Particulates, > 10 um 2.09E-08 k g IPCC 2001 1 1.56 (1,5,1,1,1,3,25)
burning process
 Chromium air unspecif ied Chromium 4.18E-09 k g IPCC 2001 1 5.06 (1,5,1,1,1,3,22)
residual
Grinding w aste steel rolling, 20%
 Water treatment sludge as proxy material No CH 1.41E-04 k g IPCC 2001 1 1.41
Sludge management w ater, to residual
landf ill
material landf ill
 Grinding w heel inventoried in inf rastructure module 9.49E-07 k g IPCC 2001 1 1.05
Oil containig w astes are approximated by used hazardous
Waste w aste oil, 10% w ater, to
 mineral oil. Only the 22.8% being landf illed are w aste No CH 1.21E-05 k g IPCC 2001 1 1.51
emulsion management hazardous w aste calculated f rom
inventoried incineration
incineration max value
disposal, used mineral assuming
Oil Oil containig w astes are approximated by used hazardous
w aste oil, 10% w ater, to lognormal
 contaminated mineral oil. Only the 22.8% being landf illed are w aste No CH 3.40E-05 k g IPCC 2001 1 1.15
management hazardous w aste distribution
w aste inventoried incineration
incineration
SBT-dust2.5- Particulates, > 2.5 um,
 application of the texture onto the roll using grit air unspecif ied 2.24E-05 k g IPCC 2001 1 1.12
10 and < 10um
 SBT-dust>10 application of the texture onto the roll using grit air unspecif ied Particulates, > 10 um 2.24E-05 k g IPCC 2001 1 1.12

2. Steel forming
Genera l Flow Information Re presentation in ecoinvent Uncerta inty Informations
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
-
Demineralized w ater, deionised, at
Water plant
Hydraulic fluids and lubricants

sheet rolling, chromium steel,
Grease, oil,
hydraulic oil
Rolling oil  Lubricating oil as proxy chemicals organics No RER lubricating oil, at plant 9.95E-04 k g IPCC 2001 1 2.21
Hydraulic &
lubrication oils
Electrical
management production
electricity, medium
calculated f rom
UCTE, at grid
assuming
Steam  Steam used as energy source chemicals organics No RER 4.47E-02 M J IPCC 2001 1 1.12 lognormal
chemical plant
distribution
disposal, steel in car
Disposed Rejects made mainly of steel. Only the 34% being w aste municipal shredder residue, 0%
 No CH 7.91E-03 k g IPCC 2001 1 2
w aste: rejects landf illed are inventoried, the rest is recycled. management incineration w ater, to municipal
incineration
Disposed Oil containig w astes are approximated by used hazardous
 w aste: Used oil mineral oil. Only the 22.8% being landf illed are w aste No CH 7.63E-04 k g IPCC 2001 1 1.05
+ w ater inventoried incineration
incineration
Water, cooling,
Cooling w ater  Sum of the dif f erent process steps resource in w ater unspecif ied natural 5.32E-02 m 3 IPCC 2001 1 1.22 (1,5,1,1,1,3,12)
origin
sheet rolling, chromium steel, Overall

Cooling and w ater leaves the system as not contamined
 4.80E-02 m 3 IPCC 2001 1 1.22 (1,5,1,1,1,3,12)
process w ater evaporated w ater or included in discarded
Rolling Shop
(CR)
process

disposal, municipal
w aste municipal solid w aste, 22.9%
 Waste-mix No CH 3.58E-04 k g EA A 2000 1 1.22 (2,5,1,1,1,3,6)
management incineration w ater, to municipal
incineration
(0,5,n.a.,n.a.,n.a
 Heat Waste air unspecif ied Heat, w aste 1.64E+00 M J IPCC 2001 1 1.56
.,n.a.,7)
Cold rolling sheet rolling, chromium
 metals processing No RER 1.00E+00 k g 1 1.22 (2,5,1,1,1,3,4)
sheet steel

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Flocculating calculated f rom
agents  binder No CH 4.74E-05 k g IPCC 2001 1 1.2 max value
materials at plant
(Ca(OH)2;95%) assuming
Personal communication by G. Doka --> inorganic chemicals inorganic, at lognormal
Polyelectrolytes  chemicals inorganics No GLO 1.38E-08 k g IPCC 2001 1 2
chemicals as proxy plant distribution
sheet rolling, chromium steel, Waste water treatment plant

Treated w aste
 treatment plant is comprised in the inf rastructure 5.25E-03 m 3 IPCC 2001
w ater
have.
Suspended solids,
 TSS w ater unspecif ied 1.39E-04 k g IPCC 2001 1 1.57 (2,5,1,1,1,3,32)
unspecif ied
 COD w ater unspecif ied 3.61E-05 k g IPCC 2001 1 1.57 (2,5,1,1,1,3,32)
Demand
BOD5, Biological
 BOD w ater unspecif ied 1.65E-06 k g IPCC 2001 1 1.2 (5,5,5,5,5,5,32)
Oxygen Demand
DOC, Dissolved
 DOC w ater unspecif ied 2.54E-07 k g IPCC 2001 1 2 (5,5,5,5,5,5,32)
Organic Carbon
TOC, Total Organic
 TOC w ater unspecif ied 1.96E-07 k g IPCC 2001 1 2.79 (5,5,5,5,5,5,32)
Carbon
Hydrocarbons,
 Hydrocarbons w ater unspecif ied 1.65E-06 k g IPCC 2001 1 3.05 (2,5,1,1,1,3,34)
unspecif ied
 Cu w ater unspecif ied Copper, ion 2.54E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Zn w ater unspecif ied Zinc, ion 1.96E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Cd w ater unspecif ied Cadmium, ion 1.31E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Al w ater unspecif ied A luminum 1.24E-06 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Pb w ater unspecif ied Lead 2.63E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Cr w ater unspecif ied Chromium, ion 4.73E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Cr6+ w ater unspecif ied Chromium V I 2.63E-08 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Mn w ater unspecif ied Manganese 5.35E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Fe w ater unspecif ied Iron, ion 4.07E-06 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Ni w ater unspecif ied Nickel, ion 7.42E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Hg w ater unspecif ied Mercury 2.63E-08 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Free Chlorine w ater unspecif ied Chloride 1.17E-06 kg IPCC 2001 1 1.57 (2,5,1,1,1,3,33)
Water Assumption: same amount per cubic meter as f or residual
w aste steel rolling, 20%
 treatment w aste w ater f rom cold steel rolling. Only a part material No CH 1.75E-02 k g IPCC 2001 1 1.57 (2,5,1,1,1,3,43)
management w ater, to residual
sludge of the sludge goes to landf ill --> 66.5% invetoried landf ill
material landf ill

2. Steel forming
2.5 Wire drawing of steel and chromium steel

Wire drawing is a process in which wire rods/wires are reduced in size by drawing them through cone-
shaped openings of a smaller cross section, so called dies. The input usually is wire rod of diameters
raging from 5.5 to 16 mm obtained from hot rolling mills in form of coils.
The European Union has the world largest wire drawing industry, followed by Japan and North Amer-
ica. It produces about 6 million tonnes of wire per year. Including the various wire products, such as
barbed wire, grill, fencing, netting, nails etc, the production of the sector reaches more than 7 million
tonnes per year. The largest producer of steel wire is Germany with 32 % (about 1.09 million tonnes)
of EU wire production, followed by Italy (approx. 22 %, 1.2 million tonnes), UK, Benelux (mainly
Belgium), France and Spain.
A typical wire drawing process line comprises the following steps (IPPC (2001a)):
 Pre-treatment of the wire rod (mechanical descaling, pickling).
 Dry or wet drawing (usually several drafts with decreasing die sizes).
 In some cases heat treatment (continuous-/discontinuos annealing, patenting, oil hardening).
 Finishing.
The final diameter size of dry drawn wire is between one and two millimetres, wet drawn wire has an
evan smaller diameter. A short description of every single process step is given below.
Wire is manufactured in different grades of steel: low carbon steel with carbon content of up to
0.25 %, high carbon steel with a carbon content of over 0.25 %, stainless and other alloy steel. Non-
alloy steel wire can be uncoated or coated with zinc, copper, brass, tin, nickel, chrome, plastic or var-
nish. Wire is sent as coils to further processing, like coating and manufacturing of finished products
(e.g. cable, mesh, barbed wire, wire fencing, grill, springs, nails).
The infrastructure of steel and chromium steel wire drawing is described in 1.1. It comprises buildings
where production processes take place and for storage, surroundings and machinery fore the single
process steps (e.g.: furnace, rolls, cutting devices, etc.).
2.5.1 Pre-treatment of the wire rod

Pre-treatment of wire rod is described in (IPPC (2001a)).
"Air-cooling after rolling produces an iron oxide-layer (scale) on the wire rod surface. This layer is
very hard, brittle and not deformable and must therefore be removed before any further processing can
take place. Two techniques for descaling wire rods are applied: mechanical descaling and chemical
pickling.
Mechanical descaling of wire rod

In the most common mechanical descaling method applied: wire bending, the wire is bent to cause the
brittle scale to peel off. Other descaling techniques, like sanding, brushing or shot blasting, are used as
a finishing step after wire bending to remove loose scale or are used as stand-alone techniques. Wire
bending, generates a relatively coarse scale, which generates only negligible amounts of dust.
Chemical descaling (pickling) of wire rod

In chemical descaling the mill scale is removed by dissolving in acid. Acids used are:
- Sulphuric or hydrochloric acid for low carbon wire.
- Hydrochloric acid for high carbon wire.

2. Steel forming
Pickling is mostly done in a batch process. Each roll of wire rod is immersed in the acid bath. The acid
slowly dissolves the oxide layer by transforming it into iron chlorides or sulphates. When pickling in
HCl an H2-inhibitor is used to suppress the reaction: 2 HCl + Fe -> FeCl2 + H2 and reduce the un-
wanted loss of iron. After pickling the wire rod coils are rinsed in water. This is usually done in cas-
cade rinsing (e.g. rinsing 3 times) to yield maximum rinsing efficiency and minimum water consump-
tion.
Application of Soap (Lubricant) Carrier

In some cases, soap carrier is applied to enhance the adhesion of the lubricant to the wire. A wide
range of soap carriers is available. Selection is made on economical basis and on required properties in
downstream processes. Traditional soap carriers include lime, borax and Zn-phosphate. Modern soap
carriers are typically mixtures of soluble salts, e.g. sodium and potassium sulphate, chloride, borax,
phosphate or silicate. They are adapted to a particular soap and a particular drawing situation. Soap
carrier is applied before drawing by dipping the wire in a watery soap carrier solution. This can be
done in a batch process, usually associated with chemical descaling, or continuous in association with
mechanical descaling.
Consumption and emissions associated with the application of soap carrier vary in dependence of the
type of soap carrier, thickness applied and the wire rod diameter. No quantitative data is available at
the moment.
If Zn-PO4 coating is used, a thick FePO4 sludge or cake is formed containing traces of zinc, soluble
phosphates and chlorate or nitrite coming from the Zn-PO4 bath. Also spent soap carrier bath may
arise as liquid waste. If drying is done in gas - or fuel fired ovens, the usual combustion products are
emitted."
2.5.2 Dry drawing of wire

Dry drawing is typically used to draw wire rod (> 5.5 mm) to a product diameter of 1 - 2 mm. The di-
ameter of wire is reduced by passing it through a series of dies with diminishing diameter. Before en-
tering the dies, the wire is passed through a dry lubricant. In most cases, soap-based lubricants are
used with the soap formulation depending on choice of fats from which the soap is produced, choice of
fillers or choice of additives. In exceptional cases (e.g. special steels, wire with special metal coatings)
other lubricants like pastes or oils can be used. The drawing operation heats both (wire and drawing
die) through friction of the wire. Cooling is done indirectly, by cooling the capstans in contact with the
wire with water (IPPC (2001a)).
2.5.3 Wet drawing of wire

Wet drawing is typically used to draw wire with an intermediate product diameter of 1 - 2 mm to final
diameter. The wire is also passed through a series of dies with diminishing diameter, but wire, dies and
capstans are immersed in a lubricant liquid that provides lubrication and cooling. Typically, soap or oil
emulsions (for some applications plain oil) are used. The heat resulting from drawing operation is
taken up by the lubricant, which is cooled indirectly with water (IPPC (2001a)).
2.5.4 Heat treatment of wire

Heat treatment of wire has different goals. Therefore different types of heat treatment are applied de-
pending on the type of steel (low carbon / high carbon / stainless) and the final use (desired ductility
and strength). Heat treatment also removes soap and lubricant residues thermally. The following heat
treatment types are applied (IPPC (2001a)):

2. Steel forming
- Batch annealing,
- Continuous (In-Line) annealing,
- Patenting,
- Oil hardening and tempering,
- Stress-relieving.
A considerable percentage of the output of the wire industry does not need any heat-treatment. The
heavy deformation of the metal crystal structure caused by drawing increases the hardness and the
strength of the wire in the axial direction (IPPC (2001a)).
Batch annealing
Drawing profoundly deforms the shape of the metal crystals of the wire. Annealing is one of the dif-
ferent methods to re-obtain a suited crystal shape. Batch annealing, typically used for low carbon steel
wire to obtain a very soft and ductile end product, is done in bell or pot furnaces, in chambers filled
with a either neutral or reducing protective gas (nitrogen, hydrogen, nitrogen/hydrogen mixtures and
partly oxidized natural gas) and typically heated by gas or fuel. The warming up from room to peak
temperature (roughly 700 °C) takes several hours; also cooling down again takes several hours. In or-
der to keep an over-pressure in the "pots" or "bells", a fraction of the protective gas is continuously
purged. In some cases, the wire is oiled immediately after annealing (IPPC (2001a)).
Continuous (In-Line) annealing

Continuous annealing or strand-annealing has a similar goal as batch annealing: restore a suited crystal
texture to the steel in the wire after drawing. But the desired crystal shape and metal properties are dif-
ferent from those that are desired when batch-annealing is applied. Continuous annealing is a typical
heat treatment method for low carbon products. Strand-annealing is a fast continuous process. The
wire is heated up to the recrystallisation temperature (500 - 700 °C) commonly by passing through a
molten lead bath, is kept at this temperature for a few seconds, and is cooled down again by quenching
in a water bath. This can be followed by in-line pickling with warm or cold HCl, othe acids or electro-
lytes, to remove oxides which also partly dissolves the potentially dragged lead. Pickling is followed
by a rinsing cascade (IPPC (2001a)).
For stainless steel, the heat-treatment is done under a protective gas atmosphere as if no protective at-
mosphere would be used, the stainless steel wire would be oxidized; removal of these oxides would
require pickling with special acids. The heating can be done by electricity (resistance heating, induc-
tive heating) or by combustion (IPPC (2001a)).
Patenting
Patenting is a heat-treatment method typically used for high-carbon and alloyed steel products to cre-
ate a special crystal structure, which allows easy further deformation. In contrast to annealing where
iron and iron/carbon compounds tend to be separated, patenting yields a structure where the carbon is
homogeneously distributed in the iron. Patenting is done by heating the wire to 850 - 1000 °C, then
cooling fast to 450 - 600 °C and keeping the wire for a while a this temperature, and finally quenching
in water (IPPC (2001a)).

2. Steel forming
Oil hardening and tempering

Oil hardening and tempering creates a special crystal structure in the steel with a high percentage of
martensite, resulting in increased hardness and wear resistance combined with good thoughness. First,
the wire is heated to 850 – 1000 °C, followed by rapid cooling. Oil-hardening is always followed by a
tempering or stress-relieving step, in order to remove the stresses caused by the extremely fast cooling.
This is done by heating the wire again to 300 –500 °C. Typically a normal oven is used, with electrical
heating or direct heating with a typical combustion gas, but also inductive heating can be used (IPPC
(2001a)).
Stress-relieving
The goal of stress-relieving is to remove internal stresses in the wire caused by a previous processing
steps, without changing the shape and structure of the steel crystal. The internal stresses can be caused
by deformation (mechanical stresses) or by fast cooling (thermal stresses). Stress-relieving as a stand-
alone heat-treatment step is a typical process step for the production of PC-strand (wire for prestressed
concrete). Stress-relieving can be done at various temperatures (200 - 500 °C), dependent on the de-
sired characteristics of the final product. Typically a normal oven is used, with electrical heating or di-
rect heating with a typical combustion gas, but also inductive heating can be used. After stress-
relieving, the wire is cooled relatively slowly in air or water (IPPC (2001a)).
2.5.5 Environmental issues of wire drawing of steel

The main environmental aspects of wire drawing are: air emissions from pickling, acidic wastes and
waste water; fugitive soap dust (dry drawing), spent lubricant and effluents (wet drawing), combustion
gas from furnaces and emissions and lead-containing wastes from lead baths (IPPC (2001a)).
2.5.6 Wire drawing of steel and chromium steel in ecoinvent

For mechanical wire rod preparation, only data about wire bending were available. No data were
available about the application of soap carrier, for batch annealing, for oil hardening and tempering,
annealing of stainless steel wire and stress relieving. For the heat treatment available data were inven-
toried only with 25 %, because 75 % of drawn steel underlies no further tempering.
The output is wire drawing for 1 kg of product. If coated wire is needed, one has to apply an additional
module, e.g. "zinc coating, coils" or "zinc coating, pieces".
culated, the arithmetic mean is taken. The background information to the process "wire drawing, steel"
is given in Tab. 2.5, flows are shown in Fig. 2.5.
Wastewater is estimated based on the wastewater from hot and cold rolling.
Data quality
Data is taken almost exclusively taken from IPPC (2001a), some estimations are also made. The val-
ues can vary considerably depending on the degree of deformation.

2. Steel forming
Tab. 2.5 Ecoinvent meta information for "wire drawing, steel"
Name wire drawing, steel

Location RER
Unit kg
Includes the process steps pre-treatment of the wire rod (mechanical descaling,
pickling), dry or wet drawing (usually several drafts with decreasing die sizes), in
Included Processes
some cases heat treatment (continuous-/discontinuous annealing, patenting, oil
hardening) and Finishing. Does not include coating and the material being rolled
Amount 1
Local Name Draht ziehen, Stahl
Synonyms
Wire drawing is a process in which wire rods/wires are reduced in diameter by
drawing them through cone-shaped openings of a smaller cross section, so called
dies. The input usually is wire rod of diameters raging from 5.5 to 16 mm obtained
ence function
from hot rolling mills in form of coils. The final diameter size of dry drawn wire is
between one and two millimetres, wet drawn wire has an even smaller diameter.
Start Date 1997
End Date 2002
Other Period Text
Average technique for EU. The processes of steel and stainless steel aren't fun-
Technology text
damentally different, thus this module covers both materials
Production Volume 6 to 7 million tonnes in 1996
Extrapolations none

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Only part of the steel that leaves the system as
steel, converter,
primary steel  scrap to recycling is inventoried. A ssumed to be metals extraction No RER 4.28E-02 k g Estimation 1 1.31 (4,5,1,1,1,3,3)
steel, Wire
unalloyed, at plant
drawing, 50% of scrap f rom sheet rolling
wire

Coarse scale, amount of dust is negligible. BOF residual
Scale (FeO and w aste f urnace w astes, 0%
 w aste that is used as proxy contains more CaO material No CH 3.16E-03 k g IPCC 2001 1 1.58
Fe3O4) management w ater, to residual
than is expected f or the scale. landf ill calculated f rom
material landf ill
max value
Value consider 50% of plants w ith, and 50% of hydrochloric acid, 30% assuming
HCL (32%)  chemicals inorganics No RER 1.21E-02 k g IPCC 2001 1 1.29
plants w ithout acid recovery in H2O, at plant lognormal
Value consider 50% of plants w ith, and 50% of sulphuric acid, liquid, at distribution
H2SO4 (98%)  chemicals inorganics No RER 8.84E-03 k g IPCC 2001 1 1.41
plants w ithout acid recovery plant
chemicals organic, at
H2 inhibitor  chemicals organics No GLO 3.16E-05 k g IPCC 2001 1 3.16
plant
wire drawing, steel, Chemical descaling
HCL pickling Assumption: the same percentage of the input is

 (acid vapour, evaporated as it is in the pickling process of cold air unspecif ied Hydrogen chloride 3.93E-06 k g IPCC 2001 1 1.62 (1,5,1,1,3,3,31)
droplets) to air sheet rolling
H2 from H2SO4
pickling
 (aerosols, air unspecif ied Hydrogen 1.54E-09 k g IPCC 2001 1 1.62 (1,5,1,1,3,3,31)
droplets of
acid)
SO4 f rom
H2SO4 pickling
 (aerosols, air unspecif ied Sulf ate 7.32E-08 k g IPCC 2001 1 1.31 (1,5,1,1,3,3,15)
droplets of
acid)
SO2 f rom
H2SO4 pickling
 (aerosols, air unspecif ied Sulf ur dioxide 6.68E-07 k g IPCC 2001 1 1.31 (1,5,1,1,3,3,15)
droplets of
acid)
Cl- from spent Value consider 50% of plants w ith, and 50% of
 w ater unspecif ied Chloride 9.98E-06 k g IPCC 2001 1 1.37 calculated f rom
HCL plants w ithout acid recovery
max value
SO4-- f rom Value consider 50% of plants w ith, and 50% of
 w ater unspecif ied Sulf ate 7.35E-06 k g IPCC 2001 1 1.5 assuming
spent H2SO4 plants w ithout acid recovery
lognormal
H+ from spent
 Negligible if correct H2-inhibitor is used 4.46E-07 k g IPCC 2001 1 1.41 distribution
acid
Fe++ f rom Value consider 50% of plants w ith, and 50% of
 w ater unspecif ied Iron, ion 9.34E-06 k g IPCC 2001 1 1.57 (2,5,1,1,1,3,31)
spent acid plants w ithout acid recovery
Fig. 2.5 Flows for "wire drawing, steel" and their representation in the ecoinvent database

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Dry draw ing
lubricant
A ssumption: lubricant is composed 50-50 of w ashing
(pref ormulated
alkali soap +

wire drawing, steel, Dry alkali soap and lubricating oil agents
tensides No RER soap, at plant 1.00E-03 k g IPCC 2001 1 2
calculated from
additives)
max value
Dry draw ing
assuming
drawing
lubricant  chemicals organics No RER lubricating oil, at plant 1.00E-03 k g IPCC 2001 1 2
lognormal
(lubricating oil)
distribution
Spent dry hazardous
Oil contamined w aste is approximated by used w aste oil, 10% w ater, to
 draw ing w aste No CH 2.00E-03 k g IPCC 2001 1 2
mineral oil management hazardous w aste
lubricant incineration
incineration
Soap dust
 n.a. IPCC 2001
emission
 Broken dies ne gligible IPCC 2001
Pref ormulated
w ashing
concentrated  tensides No RER soap, at plant 1.58E-03 k g IPCC 2001 1 3.16
agents
wire drawing, steel, Wet
soap emulsion
Pref ormulated
concentrated oil  chemicals organics No RER lubricating oil, at plant 1.58E-03 k g IPCC 2001 1 3.16
calculated from
emulsion
max value
drawing
Spent Wet
assuming
Draw ing
lognormal
Emulsion disposal, used mineral
Oil contamined w aste (containing COD (20-80 hazardous distribution
(including w aste oil, 10% w ater, to
 g/l) and Metals (0.2-1 g/l)) is approximated by w aste No CH 6.08E-02 k g IPCC 2001 1 3.54
sludge management hazardous w aste
used mineral oil. incineration
removed by incineration
f iltration/decant
ation)
 Broken dies ne gligible IPCC 2001
Fig. 2.5 Flows for "wire drawing, steel" and their representation in the ecoinvent database (cont.)

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
calculated f rom
max value
Only an estimated 25% of the output of the w ire lead, at regional
Lead  metals extraction No RER 9.68E-04 k g IPCC 2001 1 3.87 assuming
industry is annealed. storage
lognormal
distribution
energy consumption calculated as energy
Energy f or needed f or heating up the lead bath f rom ambient natural gas, burned in
heating
heating the lead  to melting temperature Overall ef f iciency of 25% natural gas No RER industrial f urnace 7.03E-05 M J IPCC 2001 1 1.38 (4,5,1,1,3,3,1)
systems
bath - gas assumed. A ssumption: 50% of plants using gas >100kW
and 50% of plants using f uel.
Energy f or light f uel oil, burned in
heating
wire drawing, steel, Continuous annealing
heating the lead  oil No RER industrial f urnace 7.03E-05 M J IPCC 2001 1 1.38 (4,5,1,1,3,3,1)
systems
bath - f uel 1MW, non-modulating
Only an estimated 25% of the output of the w ire
Lead w aste disposal, dust,
industry is annealed--> only 25% inventoried as residual
(lead oxides + w aste unalloyed EA F steel,
 dust from EAF process. The proxy chosen to material No CH 9.68E-04 k g IPCC 2001 1 3.87
spent bath management 15.4% w ater, to
represent the w aste probably overestimates landf ill calculated f rom
cover) residual material landf ill
some heavy metals (e.g. Cr). max value
SS in quench assuming
 w ater inventoried in the w ater treatment plant n.a. IPCC 2001 lognormal
overf low distribution
Pb in quench
 w ater inventoried in the w ater treatment plant 1.77E-06 k g IPCC 2001 1 3.16
overf low
Ref erence contained no data about the air f low
of w aste air. Thus the amount of exhaust gas
 Pb to air f rom HCL-pickling tanks f rom cold sheet rolling is air unspecif ied Lead 7.91E-09 k g IPCC 2001 1 5.11 (2,5,1,1,3,3,22)
assumd. Only an estimated 25% of the output of
the w ire industry is annealed.
f rom HCL-pickling tanks f rom cold sheet rolling is
 Dust to air assumd. Only an estimated 25% of the output of air unspecif ied 1.37E-07 k g IPCC 2001 1 1.62 (2,5,1,1,3,3,25)
and < 10um
the w ire industry is annealed. A ssumption: Dust
emissions f rom lead bath are betw een 2.5 and
10 um.
NMV OC, non-methane
f rom HCL-pickling tanks f rom cold sheet rolling is
volatile organic
 TOC to air assumd. Only an estimated 25% of the output of air unspecif ied 2.21E-07 k g IPCC 2001 1 2.1 (2,5,1,1,3,3,23)
compounds,
the w ire industry is annealed. NMV OC used as
unspecif ied origin
proxy w ith the assumption that C-content of
NMVOC is 80%.
Heating Emissions arising f rom f uel and gas consumption
 n.a. IPCC 2001
emissions are included in the invtoried burner modules

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
calculated from
max value
in-line HCL pickling - continuous

Only an estimated 25% of the output of the w ire hydrochloric acid, 30%
HCL (32% HCL)  chemicals inorganics No RER 7.91E-03 k g IPCC 2001 1 3.16 assuming
industry is annealed. in H2O, at plant
lognormal
distribution
Energy f or Of ten the bath is heated by residual heat in the
heating the HCl-  w ire, theref ore, as ref erence contained no data - n.a. IPCC 2001
bath (gas, f uel) -> not inventoried
annealing

 Total Cl- w ater unspecif ied Chloride 1.14E+00 k g IPCC 2001 1 1.35
industry is annealed.
calculated from
 Total H+ 3.32E-05 k g IPCC 2001 1 1.35 max value
assuming
 Fe w ater unspecif ied Iron, ion 4.84E-04 k g IPCC 2001 1 1.44 lognormal
distribution
HCl (emissions
 negligible 5.93E-09 k g IPCC 2001 1 2
f rom scrubber)
Lubricant burn-
 no data available n.a. IPCC 2001
of f
Particle size f rom chemical process: assumption
 Dust n.a. IPCC 2001
2.5-10 um. No data available --> not inventoried
calculated from
Patenting is done tipically f or high alloyed steel.
steel, Patenting
max value
wire drawing,
A mount of chromium steel w ire is about 10% of lead, at regional

Lead  metals extraction No RER 3.16E-04 k g IPCC 2001 1 3.16 assuming
total steel w ires. --> This step is inventoried f or storage
lognormal
only 10% of steel w ire.
distribution
Ref erence contained no data about the energy
consumption. A s proxy it w as calculated the natural gas, burned in
heating
Energy  energy needed f or heating up the steel w ire and natural gas No RER industrial furnace 4.13E-01 M J IPCC 2001 2 1.38 (4,5,1,1,3,3,1)
systems
the lead bath f rom ambient to the requested >100kW
temperature.

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Patenting is done tipically f or high alloyed steel.
Lead w aste A mount of chromium steel w ire is about 10% of disposal, dust,
residual calculated from
(lead oxides + total steel w ires --> only 10% inventoried as dust w aste unalloyed EA F steel,
 material No CH 3.16E-04 k g IPCC 2001 1 3.16 max value
spent bath f rom EA F process. The proxy chosen to management 15.4% w ater, to
landfill assuming
cover) represent the w aste probably overestimates residual material landf ill
lognormal
some heavy metals (e.g. Cr)
distribution
 SS inventoried in the w ater treatment plant n.a. IPCC 2001
 Pb inventoried in the w ater treatment plant 7.07E-07 k g IPCC 2001 1 3.16
of w aste air. As proxy, the amount of exhaust
gas f rom HCL-pickling tanks f rom cold sheet
wire drawing, steel, Patenting
 Pb to air rolling is taken. Patenting is done tipically f or high air unspecif ied Lead 1.41E-09 k g IPCC 2001 1 5.11 (2,5,1,1,3,3,22)
alloyed steel. A mount of chromium steel w ire is
about 10% of total steel w ires --> only 10%
inventoried
gas f rom HCL-pickling tanks f rom cold sheet
 Dust to air rolling is taken. A ssumption: Dust emissions f rom air unspecif ied Particulates, > 10 um 5.48E-08 k g IPCC 2001 1 1.62 (2,5,1,1,3,3,25)
lead bath are bigger than 10 um. Amount of
chromium steel w ire is about 10% of total steel
w ires --> only 10% inventoried
gas f rom annealing from cold sheet rolling is
taken. Patenting is done tipically for high alloyed
steel. Amount of chromium steel w ire is about
CO f rom Carbon monoxide,
 10% of total steel w ires --> only 10% air unspecif ied 5.48E-06 k g IPCC 2001 1 5.11 (2,5,1,1,3,3,17)
f urnace fossil
inventoried. As gas is burned w ith O2-shortage,
this CO-emission is added to the emissions
inventoried in the module for burning the natural
gas. A mount of chromium steel w ire is about
10% of total steel w ires --> only 10% inventoried
NOx f rom
 ne gligible IPCC 2001
f urnace
SO2 f rom
 ne gligible IPCC 2001
f urnace

2. Steel forming
General Flow Information Representation in ecoinvent Uncertainty Informa tions
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
structure
category
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Flocculating
A ssumption: w aste w ater treatment is construction lime, hydrated, loose,
agents  binder No CH 6.32E-05 k g IPCC 2001 1 1.31 (2,5,1,1,3,3,4)
comparable w ith the WWT of hot and cold rolling materials at plant
(Ca(OH)2;95%)
inorganic chemicals as proxy (personal
communication w ith G. Doka). A ssumption: chemicals inorganic, at
Polyelectrolytes  chemicals inorganics No GLO 1.84E-08 k g IPCC 2001 1 1.31 (2,5,1,1,3,3,4)
w aste w ater treatment is comparable w ith the plant
WWT of hot and cold rolling
wire drawing, steel, Waste water treatment plant

Treated w aste
 treatment plant is comprised in the inf rastructure 7.00E-03 m 3 IPCC 2001
w ater
have.
A ssumption: w aste w ater treatment is Suspended solids,

 TSS w ater unspecif ied 1.86E-04 k g IPCC 2001 1 1.57 (2,5,1,1,1,3,32)
comparable w ith the WWT of hot and cold rolling unspecif ied

 COD w ater unspecif ied 4.81E-05 k g IPCC 2001 1 1.57 (2,5,1,1,1,3,32)
Demand
BOD5, Biological
 BOD w ater unspecif ied 4.81E-05 k g IPCC 2001 1 2.79 (5,5,5,5,5,5,32)
Oxygen Demand
DOC, Dissolved
 DOC w ater unspecif ied 1.53E-05 k g IPCC 2001 1 2.79 (5,5,5,5,5,5,32)
Organic Carbon
TOC, Total Organic
 TOC w ater unspecif ied 1.53E-05 k g IPCC 2001 1 2.79 (5,5,5,5,5,5,32)
Carbon
Hydrocarbons,
 Hydrocarbons w ater unspecif ied 2.20E-06 k g IPCC 2001 1 3.05 (2,5,1,1,1,3,34)
unspecif ied
 Cu w ater unspecif ied Copper, ion 3.39E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Zn w ater unspecif ied Zinc, ion 2.62E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Cd w ater unspecif ied Cadmium, ion 1.75E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Al w ater unspecif ied A luminum 1.66E-06 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Pb w ater unspecif ied Lead 3.50E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Cr w ater unspecif ied Chromium, ion 6.30E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Cr6+ w ater unspecif ied Chromium V I 3.50E-08 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Mn w ater unspecif ied Manganese 7.14E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Fe w ater unspecif ied Iron, ion 5.42E-06 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Ni w ater unspecif ied Nickel, ion 9.90E-07 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Hg w ater unspecif ied Mercury 3.50E-08 kg IPCC 2001 1 1.86 (2,5,1,1,1,3,37)
 Free Chlorine w ater unspecif ied Chloride 1.57E-06 kg IPCC 2001 1 1.57 (2,5,1,1,1,3,33)
Water residual
Same composition as f or cold rolling of steel is w aste steel rolling, 20%
 treatment material No CH 3.50E-02 k g IPCC 2001 1 1.57 (2,5,1,1,1,3,43)
assumed management w ater, to residual
sludge landf ill
material landf ill

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Water used for dif f erent purposes: rinsing,
cooling, dillution, quenching w ater. Sum of the
Water 
wire drawing, steel, Overall dif ferent process steps, mainly consisting of
resource in w ater
Water, cooling,
unspecif ied natural 3.50E-02 m 3 IPCC 2001 1 1.38 (4,5,1,1,3,3,12)
cooling w ater, w ich amount has huge variations.
origin
A ssumption: w aste w ater treatment is
comparable w ith the WWT of hot and cold rolling
process

w ater leaves the system as not contamined
 w ater out 2.80E-02 m 3 IPCC 2001 1 1.38 (4,5,1,1,3,3,12)
evaporated w ater or included in discarded
electricity, medium Personal
Electrical No data available in IPPC 2001; therefore values production
 electricity No UCTE voltage, production 1.41E-01 k Wh communication 1 1.24 (3,5,1,3,1,3,2)
energy f rom Stahl Gerlaf ingen w ere taken mix
UCTE, at grid Dr. Ch. Zeltner
Rolling Shop
(CR)
transport (4,5,n.a.,n.a.,n.a
systems .,n.a.,5)

wire drawing, steel,
Paper; paper &

 corrugated No RER board, mixed f ibre, 4.72E-12 k g EAA 2000 1 1.38 (4,5,1,1,3,3,4)
cardboard cardboard
Packaging

Plastics  plastics processing No RER 1.89E-11 k g EAA 2000 1 1.38 (4,5,1,1,3,3,4)
at plant
steel, converter,
Steel  metals extraction No RER 2.36E-11 k g EAA 2000 1 1.38 (4,5,1,1,3,3,4)
unalloyed, at plant
Steel coils  metals processing No RER sheet rolling, steel 2.36E-11 k g EAA 2000 1 1.38 (4,5,1,1,3,3,4)
saw n timber,
w ooden
Wood  extraction No RER sof tw ood, raw , air 1.62E-17 m 3 EAA 2000 1 1.38 (4,5,1,1,3,3,4)
materials
disposal, municipal
wire drawing,
w aste municipal solid w aste, 22.9%

steel, Waste
 Waste No CH 7.08E-11 k g EAA 2000 1 1.22 (2,5,1,1,1,3,6)

treatment
management incineration w ater, to municipal

incineration
(n.a.,5,n.a.,n.a.,
 Heat w aste air unspecif ied Heat, w aste 5.09E-01 M J IPCC 2001 1 1.56
n.a.,n.a.,7)
Draw ing w ire,
 steel or metals processing No RER w ire draw ing, steel 1.00E+00 k g
chromium steel

2. Steel forming
2.6 Pipe drawing of steel and chromium steel

The production process and volume of drawn pipes (or seamless tubes) is described in (IPPC (2001a)).
Steel tubes cover three product categories, which differ considerably in manufacturing process, raw
materials, and investment requirements. These categories are classified as:
- Seamless tubes.
- Welded steel tubes of circular cross-section over 406.4 mm in outside diameter.
- Welded steel tubes of circular cross -section up to and including 406.4 mm in outside diameter to-
gether with welded steel tubes of non-circular cross-section of any perimeter.
In steel tube manufacture, the EU, which produced 11.8 million tonnes in 1996, (20.9 % of total world
production) is the largest producer followed by Japan and the U.S. The European steel tube industry
has a highly concentrated structure. Five countries – Germany (3.2 million tonnes), Italy (3.2 million
tonnes), France (1.4 million tonnes), United Kingdom (1.3 million tonnes) and Spain (0.9 million ton-
nes) - account for roughly 90 % of total EU production. About 26.4 % of the tubes produced are seam-
less tubes, of the remaining 73.6 % welded tubes about 18.9 % are of large diameter. The major part of
the total steel tube sector is small diameter tubes with a share of 54.7 %.
Nevertheless this type of pipes represent only about one quarter of total production, in this study the
focus was laid on the manufacturing of seamless pipes. The reason is that, if LCI data is needed about
welded tubes, they can be gained easily, combining different ecoinvent modules. In fact welded tubes
are produced from cold rolled or hot rolled steel strip; hot rolled wide strip; or thick plate. In principal
the production sequence comprises bending of the strip or plate to an open tube and closing the gap by
welding. Therefore ecoinvent modules of steel, of steel sheet production and welding are needed.
Seamless tube production consists basically of the following manufacturing steps:
- Heating of the input.
- Piercing (press-piercing or oblique rolling).
- Elongation/Stretching.
- Final Rolling.
- Heat Treatment.
The most important types of plants used today are mandrel mills, plug mills, push bench mills and
pilger mills. Raw material for the production of unalloyed and medium-alloy steel tubes are continu-
ously cast round billets or blooms, the raw material for high-chrome steel tubes are mainly round bars,
rolled from cast billets.
The input is heated to rolling temperature in rotary hearth furnaces fired with natural gas or oil. Then a
hollow billet (ingot) is produced by piercing. The most important hole-piercing process uses inclined
rollers. Longitudinal rolling, transverse rolling, or presses are used for the elongating stage. The most
important one is the continuous tube rolling process, as it is one of the most efficient and productive
processes.
Tubes for undemanding applications can be shipped immediately after rolling. For more exacting qual-
ity requirements, especially good toughness properties, separate heat treatment is usually carried out
after rolling. This can include normalising and tempering.
The infrastructure of steel and chromium steel pipes drawing is described in 1.1. It comprises build-
ings where production processes take place and for storage, surroundings and machinery fore the sin-
gle process steps (e.g.: furnace, rolls, cutting devices, etc.).

2. Steel forming
2.6.1 Pipe drawing of steel and chromium steel in ecoinvent

No data was available about the production process of seamless pipes. As an approximation it was de-
cided to compare the tube production process with one of the forming processes for steel and chro-
mium steel, described above, for which data was available. The wire drawing process was evaluated
being the one that describes the pipes drawing process the best. As the geometry of the pipe is more
complex as the geometry of the wire, pipe drawing is inventoried with 110% of wire drawing.
The background information to the process "drawing of pipes, steel" is given in Tab. 2.6, flows are
shown in Fig. 2.6.
Data quality
Uncertainty is estimated by mean of the pedigree matrix, according to (Frischknecht et al. (2003)).
2.7 Section bar rolling of steel

The production process of section bar rolling is not described in details in the studied literature and no
data about material and energy flows were available. An extract of the description found in (IPPC
(2001a)) is reported below.
As most other types of mills, section rolling mills are divided into roughing, intermediate and finishing
stands, preceded by a descaling machine installed between the reheating furnace and the roughing
mill. On leaving the finishing stand or train, the bars are divided to cooling bed length or to commer-
cial length by saws. After cooling on the cooling bed, the bars are straightened on mechanical or hy-
draulic straightening machines and cut to length by saws as required.
As conclusion from this description and the lack of data, it was decided to combine other ecoinvent
modules to reach an approximation of the relevant flows of the section bar rolling process.
The infrastructure of steel section bar rolling is described in 1.1. It comprises buildings where produc-
tion processes take place and for storage, surroundings and machinery fore the single process steps
(e.g.: furnace, rolls, cutting devices, etc.).
2.7.1 Section bar rolling of steel in ecoinvent

Due to the lack of data, the section bar rolling process is approximated by inventorying 10% of the
wire drawing process.
The background information to the process "section bar rolling, steel" is given in Tab. 2.6, flows are
shown in Fig. 2.7.
Data quality

2. Steel forming
Tab. 2.6 Ecoinvent meta information for "drawing of pipes, steel" and "section bar rolling, steel"
Name drawing of pipes, steel section bar rolling, steel

Location RER RER
Unit kg kg
The module describes the manufactur-
ing of seamless tubes. It is not repre-
sentative for welded tubes. It includes The module describes the rolling proc-
Included Processes
110% of the wire drawing process. ess of section bar. It includes 50% of
Does not include the material being the wire drawing process. Does not in-
rolled clude the material being rolled
Amount 1 1
Local Name Rohr ziehen, Stahl Profil walzen, Stahl
Synonyms drawing of tubes profile
As most other types of mills, section
rolling mills are divided into roughing,
intermediate and finishing stands, pre-
ceded by a descaling machine installed
General Comment to refer- Seamless tube production consists ba- between the reheating furnace and the
ence function sically of the following manufacturing roughing mill. After cooling on the cool-
steps: Heating of the input, Piercing ing bed, the bars are straightened on
(press-piercing or oblique rolling), Elon- mechanical or hydraulic straightening
gation/Stretching, Final Rolling and machines and cut to length by saws as
Heat Treatment. required
End Date 2002 2002
Other Period Text
Data-set is representative for European Data-set is representative for European
Geography text
Union Union
Average technique for EU. Because of
the lack of data about the production Average technique for EU. Because of
Technology text process, it was approximated with the the lack of data about the production
production process of steel wire draw- process, it was approximated with the
ing. production process of wire drawing.
Production Volume 3.3 million tonnes in 1996 No data available
Sampling Procedure Literature and estimation Literature and estimation
Extrapolations see technology see technology
The extrapolation made for the descrip- The extrapolation made for the descrip-
tion of the technology process and the tion of the technology process and the
fact that no data were available about fact that no data were available about
the percentage of companies repre- the percentage of companies repre-
sented by the study cited are consid- sented by the study cited are consid-
ered in the uncertainties. ered in the uncertainties.

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
pipes, steel
Tube  drawing of metals processing No RER w ire draw ing, steel 1.10E+00 k g Estimation 1 1.38 (4,5,1,1,3,3,4)
 Tube draw ing metals processing No RER draw ing of pipes, steel 1.00E+00 k g
Fig. 2.6 Flows for "drawing of pipes, steel" and their representation in the ecoinvent database
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Section bar
rolling, steel
section bar
 metals processing No RER w ire draw ing, steel 5.00E-01 k g Estimation 1 1.38 (4,5,1,1,3,3,4)
extruding
Section bar section bar rolling,

 metals processing No RER 1.00E+00 k g
extruding steel
Fig. 2.7 Flows for "section bar rolling, steel" and their representation in the ecoinvent database

2. Steel forming
2.8 Infrastructure of the rolling mill

2.8.1 Land use and buildings of the rolling mill
Data on land use of rolling mills are based on the map of the steel producing and forming plant of
Stahl Gerlafingen, an enterprise of the Swiss Steel Ltd.
The total area of the plant was read from a map with the scale of 1:6000. The buildings were estimated
covering 25 % of the total area of the plant, 75 % are therefore surroundings, which are further distin-
guished in 5 % green area and 95 % built up or asphalted surface.
The area of the buildings belonging to the steel manufacturing part of the plant could be read from the
map 4 , they were estimated to be occupying 20 % of the total build-up surface of the plant. Therefore
the total area, the surroundings and the built-up area belonging to the rolling mill amount to 80 % of
the total area of the plant. Only the surface of the administration buildings was divided into equal parts
(4000 m2) for the steel manufacturing and the steel rolling process. The total area of the rolling mill is
310'000 square meters, all other values were calculated basing on this value. The buildings were esti-
mated to be 10 m high and to have a lifetime of 50 years.
2.8.2 Machinery and disposal of the machinery of the rolling mill

Absolutely no data about the amount of machinery was available, therefore it was necessary to make a
rough estimation of the volume, respectively the volume of the main components of the machinery. It
is assumed that the machinery is basically composed of steel on a foundation of concrete. The volume
of the machinery was estimated to be 0.5 % of the buildings volume of the rolling process. The con-
crete fundament is assumed to have the same mass as the machines. Density of the steel is 7850 kg/m3,
that of the concrete 2300 kg/m3. The lifetime of the machinery was estimated to be 25 years.
The yearly output of the visited plant amounts to 616000 tonnes of formed steel.
The module for concrete disposal from building includes recycling and disposal and it is assumed that
100 % of the steel is recycled.
2.8.3 Infrastructure of the rolling mill in ecoinvent

The background information to the process "rolling mill" is given in Tab. 2.7, flows are shown in Fig.
2.8.
Data quality
4
Personal communication on 25. February 03 from Dr. Ch. Zeltner, Stahl Gerlafingen, Swiss Steel Ltd.

2. Steel forming
Tab. 2.7 Ecoinvent meta information for the "rolling mill"
Name rolling mill

Location RER
Unit unit
Data Set Version 1
Includes land occupation data for and land transformation data to buildings, as-
phalted and green surroundings; land transformation data from unknown; the
Included Processes
buildings; company internal roads, parking lots etc.; the main material of which
the machinery is composed and their disposal.
Amount 1
Local Name Walzwerk
Synonyms
The values for plant area were taken from a map received during the visit to a
electro-steel manufacturing and forming plant in Gerlafingen, Switzerland. The
output was 616000 tonnes of formed steel in 2002. The Lifetime is assumed to
ence function
be 50 years for infrastructure like buildings and routes, and 25 years for ma-
chinery and furnaces
Start Date 2002
End Date 2002
Other Period Text
The visited plant is in the range of size typical for European union (about a quar-
Geography text
ter of the maximal size). Data-set is representative for European Union
The visited plant produces hot rolled merchant bars, wire rod, concrete reinforc-
Technology text
ing bars, mats and cold rolled topar-S, topar-Rc and topar-Rs
Production Volume No data available
Sampling Procedure plant map and estimation
Extrapolations see technology and geography
The extrapolation made for the description of the technology process, the geo-
graphical extrapolations, the estimations for machinery weight and the fact that
no data were available about the percentage of companies represented by the
study cited are considered in the uncertainties.

2. Steel forming
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Land
Occupation, industrial Vist to one site,
occupation-  resource land 3.88E+03 m 2a 1 1.62 (4,5,1,3,1,n.a.,7)
area, built up estimation
buildings
Land
The value f or land use of the buildings and the Occupation, industrial Vist to one site,
occupation-  resource land 1.10E+04 m 2a 1 1.62 (4,5,1,3,1,n.a.,7)
green surroundings w as subtracted area estimation
surroundings
Land
occupation- Occupation, industrial Vist to one site,
 resource land 5.81E+02 m 2a 1 1.62 (4,5,1,3,1,n.a.,7)
surroundings- area, vegetation estimation
green
Land
Transf ormation, f rom Vist to one site,
transformation  resource land 3.10E+02 m 2 1 2.1 (4,5,1,3,1,n.a.,8)
unknow n estimation
f rom
Land
Transf ormation, to Vist to one site,
industrial area, built up estimation
to buildings
Land
Transf ormation, to Vist to one site,
rolling mill
industrial area estimation

to surroundings
Land
Transf ormation, to
transformation Vist to one site,
 resource land industrial area, 1.16E+01 m 2 1 2.1 (4,5,1,3,1,n.a.,8)
to surroundings- estimation
vegetation
green
Building-rolling construction building, hall, steel Vist to one site,
 buildings Y es CH 7.35E+01 m 2 1 3.09 (4,5,1,3,1,n.a.,9)
process processes construction estimation
Building- construction Vist to one site,
 buildings Y es RER building, multi-storey 4.00E+01 m 3 1 3.09 (4,5,1,3,1,n.a.,9)
administration processes estimation
Fundament of construction concrete, normal, at Vist to one site,
 concrete No CH 2.51E+01 m 3 1 1.3 (4,5,1,3,1,n.a.,4)
the machinery materials plant estimation
steel, electric, un- and Vist to one site,
Machinery  metals extraction No RER 5.77E+04 k g 1 1.3 (4,5,1,3,1,n.a.,4)
low -alloyed, at plant estimation
transport Vist to one site, (4,5,n.a.,n.a.,n.a
Lorry 32t  road No RER transport, lorry 32t 3.75E+04 tk m 1 2.09
systems estimation .,n.a.,5)
disposal, building,
w aste building concrete, not Vist to one site,
 Concrete No CH 5.77E+04 k g 1 1.3 (4,5,1,3,1,n.a.,6)
management demolition reinf orced, to f inal estimation
disposal
Vist to one site,
 Steel recycled --> not inventoried 5.77E+04 k g 1 1.3 (4,5,1,3,1,n.a.,6)
estimation
 Rolling mill metals processing Y es RER rolling mill 1.00E+00 unit
Fig. 2.8 Flows for "rolling mill" and their representation in the ecoinvent database

3. Aluminium forming
3 Aluminium forming
As for steel, the outputs of the aluminium production are slabs with relatively coarse shape, which can
be further formed to sheets, foils or section bars.
3.1 Sheet rolling of aluminium

The proceedings of rolling aluminium from an initial rough form to a well defined sheet is similar to
the processes taking place in steel forming, therefore it is not described in every detail.
An overview of the different process steps taking place, when Aluminium is rolled to sheet form, is
given in Fig. 3.1.
The basic input material for sheet fabrication is primary or secondary aluminium ingot of 500 to
700 mm thickness and a weight up to 25 tonnes. After heating the aluminium slab to around 500 °C, it
is hot rolled, tipically to 2-8 mm thickness, followed by cold rolling to the final gauge, 0.2 to 6 mm
(EAA (2000)).
As the process steps are similar to those of steel forming, also the infrastructure needed to roll alumin-
ium slabs into sheet form, is assumed to be similar. It is balanced the same amount of rolling mill per
volume unit of rolled material, as it is for the forming processes of steel.
3.1.1 Environmental issues of sheet rolling of aluminium

In the data referred in the considered literature (EAA (2000)), the main air and water emissions, are
arising from the combustion of energy carriers. Emissions to land are mainly hazardous and oily
wastes.
3.1.2 Sheet rolling of aluminium in ecoinvent

The inventory data reported were obtained from an industry-wide survey carried out in 1998. For the
sheet rolling industry segment, the survey coverage was 71% related to the total production output
(EAA (2000)).
The basic input material, primary aluminium is not balanced. The scrap losses of aluminium are
mainly internally recycled, this recycling process is not balanced. A small amount of scrap material is
not recycled, but is lost in waste. To obtain an even mass balance for1 kilogram of output product, it is
necessary to balance the 12 grams lost in waste as input of primary aluminium (Fig. 3.1).
The background information to the process "sheet rolling, aluminium" is given in Tab. 3.1, flows are
shown in Fig. 3.1
Data quality
A standard precision of 5% percent of the value is given in EAA (2000)(), to obtain a more detailed
view of the data quality, the uncertainty is estimated by completing the pedigree matrix, according to
(Frischknecht et al. (2003)).

Aluminium, ingot
1.012 kg
Sawing, scalping
Working
materials
Rolling (hot and cold)
Energy carriers
Solution heat treatment
Packing materials Finishing and packaging

0.012 kg
Emissions to
air, water and
land
1.000 kg System boundaries
Aluminium rolling,
sheet
Process step Materials & Energy
Fig. 3.1 Aluminium sheet rolling process; values are aluminium flows (modified from EAA (2000))

Tab. 3.1 Ecoinvent meta information for "sheet rolling, aluminium"
Name sheet rolling, aluminium

Location RER
Unit kg
Data Set Version 1
All the process steps, which can be attributed to semi-fabrication (sawing, scalp-
ing, hot rolling, cold rolling, solution heat treatment, finishing and packaging), are
included. Does not include the material being rolled, only the amount of scrap lost
Included Processes
in waste is balanced as primary aluminium input. Includes the transport of the ma-
terials to the plant, but does not include the transport of the product to the cus-
tomer.
Amount 1
Local Name Blech walzen, Aluminium
Synonyms profile rolling//section bar rolling
Aluminium ingots of 500-700 mm thickness and up to 25 tonnes are rolled, first
hot, then cold, to a final sheet thickness ranging from 0.2 to 6 mm. The module
ence function
can be applied also for section bar rolling of aluminium.
Start Date 2000
End Date 2002
Other Period Text Data refers to a survey made in 1998
Geography text Dataset is representative for European Union
Average technique for European Union. The infrastructure was assumed to be the
Technology text
same as for rolling steel.
Sampling Procedure Literature, estimation for transport and infrastructure
Extrapolations see technology and sampling procedure
Precision of the data set is referred to be of 5 % by the reference, nevertheless
uncertainty was estimated by filling out the uncertainty scoring

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Rolling aluminium casuses scrap losses w ich
Primary cannot be recovered, therefore 0.012 kg of aluminium, primary, at
 metals extraction No RER 1.20E-02 k g EAA 2000 1 1.07 (1,2,1,1,1,3,3)
A luminium primary aluminium per per kg of material rolled is plant
inventoried
Rolling Oil  chemicals organics No RER lubricating oil, at plant 3.80E-03 k g EAA 2000 1 1.07 (1,2,1,1,1,3,11)
Water, cooling,
Water  resource in w ater unspecif ied natural 6.00E-03 m 3 EAA 2000 1 1.07 (1,2,1,1,1,3,12)
origin
light f uel oil, burned in
heating
Fuel Oil  oil No RER industrial furnace 1.45E-02 M J EAA 2000 1 1.07 (1,2,1,1,1,3,1)
sheet rolling, aluminium, rolling
systems
1MW, non-modulating
heating
Gas  natural gas No RER industrial furnace low - 2.11E+00 M J EAA 2000 1 1.07 (1,2,1,1,1,3,1)
systems
NOx >100kW
electricity, medium
production
Electricity  electricity No UCTE voltage, production 5.47E-01 k Wh EAA 2000 1 1.07 (1,2,1,1,1,3,2)
mix
UCTE, at grid
It is assumed, that the same volume of material is
Rolling mill  rolled per year. Density of steel is 7850 kg/m3 metals processing Y es RER rolling mill 4.72E-09 unit EAA 2000 1 3.36 (4,5,1,3,4,5,9)
and f or aluminium 2700 kg/m3
Lorry 32 t  road No RER transport, lorry 32t 5.07E-03 tk m EAA 2000 1 2.09
systems .,n.a.,5)
NMV OC, non-methane
volatile organic
 VOC to air air unspecif ied 4.40E-04 k g EAA 2000 1 2 (1,2,1,1,1,3,23)
compounds,
unspecif ied origin
 Water 6.00E-03 m 3 EAA 2000 1 1.45 (4,5,2,1,3,5,4)
 COD in Water w ater unspecif ied 7.90E-05 k g EAA 2000 1 1.51 (1,2,1,1,1,3,32)
Demand
BOD5, Biological
 BOD w ater unspecif ied 7.90E-05 k g EAA 2000 1 2.79 (5,5,5,5,5,5,32)
Oxygen Demand
DOC, Dissolved
 DOC w ater unspecif ied 2.60E-05 k g EAA 2000 1 2.79 (5,5,5,5,5,5,32)
Organic Carbon
TOC, Total Organic
 TOC w ater unspecif ied 2.60E-05 k g EAA 2000 1 2.79 (5,5,5,5,5,5,32)
Carbon
Fig. 3.2 Flows for "sheet rolling, aluminium" and their representation in the ecoinvent database

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
disposal, hazardous
Hazardous w aste undergroun
sheet rolling, aluminium,

 contains 15% aluminium No DE w aste, 0% w ater, to 4.80E-03 k g EAA 2000 1 1.07 (1,2,1,1,1,3,6)
w aste management d deposit
underground deposit
hazardous
Oil is reprocessed/burnt; here it is assumed to be w aste oil, 10% w ater, to
 Oil w aste No CH 2.30E-03 k g EAA 2000 1 1.07 (1,2,1,1,1,3,6)
incinerated; contains 5% aluminium management hazardous w aste
incineration
rolling
incineration
disposal, municipal
Solid w aste, Municipal w aste as proxy; contains 33% w aste municipal solid w aste, 22.9%
 No CH 6.50E-03 k g EAA 2000 1 1.07 (1,2,1,1,1,3,6)
unspecif ied aluminum management incineration w ater, to municipal
incineration
disposal, aluminium,
aluminium losses during the rolling process that w aste municipal
 Scrap No CH 0% w ater, to municipal 9.02E-03 k g EAA 2000 1 1.09 (2,2,1,1,1,3,6)
are not included in the dif f erent w astes management incineration
incineration
 Heat w aste air unspecif ied Heat, w aste 1.97E+00 M J EAA 2000 1 1.07 (1,2,1,1,1,3,13)
aluminium, packaging
Paper; paper &

 corrugated No RER board, mixed f ibre, 1.00E-04 k g EAA 2000 1 1.07 (1,2,1,1,1,3,4)
cardboard-alsh cardboard
sheet rolling,

Plastics-alsh  plastics processing No RER 4.00E-04 k g EAA 2000 1 1.07 (1,2,1,1,1,3,4)
at plant
steel, converter,
Steel-alsh  metals extraction No RER 5.00E-04 k g EAA 2000 1 1.07 (1,2,1,1,1,3,4)
unalloyed, at plant
Steel-sh-alsh  metals processing No RER sheet rolling, steel 5.00E-04 k g EAA 2000 1 1.07 (1,2,1,1,1,3,4)
saw n timber,
w ooden
Wood-alsh  extraction No RER sof tw ood, raw , air 1.85E-07 m 3 EAA 2000 1 1.07 (1,2,1,1,1,3,4)
materials
Sheet rolling,
 metals processing No RER sheet rolling, aluminium 1.00E+00 k g
Aluminium
Fig. 3.2 Flows for "sheet rolling, aluminium" and their representation in the ecoinvent database (cont.)

3.2 Section bar extrusion of aluminium

"Extrusion is the process of forcing heated aluminium through a die under extreme force to produce
the desired shape. Extrusions can be made in literally thousands of complex shapes. Billets are pre-
heated and forced by a ram through one or more dies to achieve the desired cross-section. The product
is long in relation to its cross-sectional dimensions and has a cross section other than that of rod and
bar and pipe and tube" 5 . The starting material for aluminium extrusion production is an extrusion billet
(or extrusion ingot), coming from either primary or secondary aluminium production.
An overview of the different process steps taking place, when Aluminium is extruded to a profile, is
given in Fig. 3.3.
The same amount of rolling mill infrastructure balanced per volume unit of treated material for the
steel forming processes, is inventoried for the section bar extrusion of aluminium.
3.2.1 Environmental issues of section bar extruding of aluminium

In the data referred in the considered literature (EAA (2000)), the main air and water emissions, are
arising from the combustion of energy carriers. Emissions to land are mainly hazardous and oily
wastes.
3.2.2 Section bar extruding of aluminium in ecoinvent

The inventory data reported were obtained from an industry-wide survey carried out in 1998. For the
section bar extruding industry segment, the survey coverage was 30 % related to the total production
output (EAA (2000)).
The basic input material, primary aluminium is not balanced. The scrap losses of aluminium are
mainly internally recycled, this recycling process is not balanced. A small amount of scrap material is
not recycled, but is lost in waste. To obtain an even mass balance for1 kilogram of output product, it is
necessary to balance the 13 grams lost in waste as input of primary aluminium (Fig. 3.3).
The background information to the process "section bar extrusion, aluminium" is given in Tab. 3.2,
flows are shown in Fig. 3.2.
Data quality
A standard precision of 5% percent of the value is given in EAA (2000)(), to obtain a more detailed
view of the data quality, the uncertainty is estimated by completing the pedigree matrix, according to
(Frischknecht et al. (2003)).
5
http://www.energysolutionscenter.org/HeatTreat/MetalsAdvisor/aluminum/process_descriptions/extrusion.htm, accessed
24.03.03

Aluminium, ingot
1.013 kg
Sawing, scalping
Preheating
Working
materials
Extrusion
Energy carriers
Stretching, sawing
Ageing
Packing materials Packaging

0.013 kg
Emissions to
air, water and
land
Aluminium rolling,
profile
Fig. 3.3 Aluminium profile extrusion process; values are aluminium flows (modified from EAA (2000))

Tab. 3.2 Ecoinvent meta information for "section bar extrusion, aluminium"
Name section bar extrusion, aluminium

Location RER
Unit kg
Data Set Version 1
All the process steps, which can be attributed to semi-fabrication (sawing, scalp-
ing, preheating, extrusion, stretching, sawing, ageing and packaging), are in-
cluded. Does not include the material being extruded; only the amount of scrap
Included Processes
lost in waste is balanced as primary aluminium input. Includes the transport of the
materials to the plant, but does not include the transport of the product to the cus-
tomer.
Amount 1
Local Name Profil extrudieren, Aluminium
Synonyms profile extrusion
Heated aluminium is forced through one or more dies under extreme force to pro-
duce the desired shape. A great number of different shapes, and therefore possi-
ence function
ble applications, can be achieved by extrusion of aluminium.
Start Date 2000
End Date 2002
Other Period Text Data refers to a survey made in 1998
Average technique for European Union. The infrastructure was assumed to be the
Technology text
same as for rolling steel.
Sampling Procedure Literature, estimation for transport and infrastructure
Extrapolations see technology and sampling procedure
Precision of the data set is referred to be of 5 % by the reference, nevertheless
uncertainty was estimated by filling out the uncertainty scoring

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Primary aluminium, primary, at
 metals extraction No RER 1.30E-02 k g EAA 2000 1 1.09 (1,3,1,1,1,3,3)
A luminium plant
Caustic soda  chemicals inorganics No RER soda, pow der, at plant 2.80E-02 k g EAA 2000 1 1.09 (1,3,1,1,1,3,4)
Water, cooling,
Water  resource in w ater unspecif ied natural 1.70E-02 m 3 EAA 2000 1 1.09 (1,3,1,1,1,3,12)
origin
heating
Fuel Oil  oil No RER industrial furnace 2.77E-02 M J EAA 2000 1 1.09 (1,3,1,1,1,3,1)
section bar extrusion, aluminium, exxtrusion
systems
1MW, non-modulating
heating
Gas  natural gas No RER industrial furnace low - 3.00E+00 M J EAA 2000 1 1.09 (1,3,1,1,1,3,1)
systems
NOx >100kW
electricity, medium
production
Electricity  electricity No UCTE voltage, production 1.14E+00 k Wh EAA 2000 1 1.09 (1,3,1,1,1,3,2)
mix
UCTE, at grid
Rolling mill  rolled per year. Density of steel is 7850 kg/m3 metals processing Y es RER rolling mill 4.72E-09 unit EAA 2000 1 3.36 (4,5,1,3,4,5,9)
and f or aluminium 2700 kg/m3
Lorry 32t  road No RER transport, lorry 32t 2.25E-02 tk m EAA 2000 1 2.09
systems .,n.a.,5)
disposal, hazardous
 contains 15% aluminium No DE w aste, 0% w ater, to 1.60E-03 k g EAA 2000 1 1.09 (1,3,1,1,1,3,6)
underground deposit
hazardous
 Oil w aste No CH 1.70E-03 k g EAA 2000 1 1.09 (1,3,1,1,1,3,6)
incinerated. contains 5% aluminium management hazardous w aste
incineration
incineration
disposal, filter dust A l
residual
Spent 70% of f ilter dust w ith 0% w ater, as proxy f or w aste electrolysis, 0% w ater,
 material No CH 2.03E-02 k g EAA 2000 1 1.09 (1,3,1,1,1,3,6)
bath/sludge sludge w ith 30% w ater. Contains 15% aluminium management to residual material
landfill
landfill
disposal, municipal
Solid w aste, Municipal w aste as proxy. contains 33% w aste municipal solid w aste, 22.9%
 No CH 2.23E-02 k g EAA 2000 1 1.09 (1,3,1,1,1,3,6)
unspecif ied aluminium management incineration w ater, to municipal
incineration
disposal, aluminium,
Losses during the rolling process that are not w aste municipal
 Scrap No CH 0% w ater, to municipal 9.66E-04 k g EAA 2000 1 1.09 (2,2,1,1,1,3,6)
included in the dif f erent w astes management incineration
incineration
 Heat w aste air unspecif ied Heat, w aste 4.11E+00 M J EAA 2000 1 1.09 (1,3,1,1,1,3,13)
Fig. 3.4 Flows for "section bar extrusion, aluminium" and their representation in the ecoinvent database

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Paper, paper &
aluminium, packaging
corrugated No RER board, mixed f ibre, EAA 2000 1 1.09 (1,3,1,1,1,3,4)
section bar extrusion,
 3.00E-03 k g
cardboard-alpr cardboard

Plastic-alpr  plastics processing No RER 2.10E-03 k g EAA 2000 1 1.09 (1,3,1,1,1,3,4)
at plant
steel, converter,
Steel-alpr  metals extraction No RER 9.00E-04 k g EAA 2000 1 1.09 (1,3,1,1,1,3,4)
unalloyed, at plant
Steel-sh-alpr  metals processing No RER sheet rolling, steel 9.00E-04 k g EAA 2000 1 1.09 (1,3,1,1,1,3,4)
saw n timber,
w ooden
Wood-alpr  extraction No RER sof tw ood, raw , air 5.19E-05 m 3 EAA 2000 1 1.09 (1,3,1,1,1,3,4)
materials
Prof ile
section bar extrusion,
 extruding, metals processing No RER 1.00E+00 k g
aluminium
Aluminium
Fig. 3.4 Flows for "section bar extrusion, aluminium" and their representation in the ecoinvent database (cont.)

4. Copper and copper alloy forming
4 Copper and copper alloy forming

Copper and copper alloys are melted continuously or in batches and cast for the production of shapes
suited for the further fabrication step. The cast products are precursors for a variety of materials such
as sheets, strips, sections, bars, rods, wires and tubes (IPPC (2001b)).
4.1 Sheet rolling of copper

Slabs/cakes of copper or copper alloy are the start material for the fabrication of sheets and strips. The
material is preheated in gas or oil fired furnaces, hot and cold rolled and then sent to the finishing op-
eration, which includes re-rolling, cutting to length and width. Surface milling, annealing, pickling,
washing and drying are required as intermediate steps to produce high quality strips and sheets.
Hot rolling is usually done with a dual rolling mill equipped with benches up to 200 m and a final coil-
ing device. The cooling water for the rolls has small amounts of lubricant added for improving the at-
tachment to the steel rolls. The vapour generated is vented and the vent gas is de-misted prior to re-
lease to atmosphere.
Further cold rolling operations are then performed. Cold rolling results in a hardened metal. In most
cases the coil is annealed prior to cold rolling. Annealing is done under reducing conditions to avoid
oxidation. A protection gas “exogas” or nitrogen/hydrogen mixtures are used. Exogas is produced at
site from natural gas in a special reactor, which is indirectly fired. N2 and H2 are purchased and stored
at site in special tanks. The N2/H2 protection gas mixtures are produced from the storage tanks by mix-
ing the components in the ratio required. For annealing before cold rolling bell type furnaces are used
with electric heating or indirectly fired by natural gas or fuel oil. Tower type furnaces are applied for
intermediate annealing of pre-rolled coils.
The sheet thickness is further reduced by stepwise cold rolling operations on different reversing mills.
For cold rolling different types of mills called Duo, Quarto, Sexto, Sendzimir (12 rolls) are used. Roll-
ing mill designs as single-stand which are combined to an in-line multi-stand rolling mill are also ap-
plied. The combination of the different types applied depends on the thickness of the sheet on the coils
and on the finishing status the coil has already been reached which.
During cold rolling an emulsion or oil is used for roll protection. Therefore, the roll stands are vented
and the ventilation gases are cleaned by mechanical filters, wet electrostatic precipitators or scrubbing.
The emulsion and the oil is cleaned from metal and cracked oil particles are removed by paper or tex-
tile band filters.
This description given in (IPPC (2001b)) shows, that the principle for sheet rolling of copper is almost
the same of the one for sheet rolling of steel, chromium steel or aluminium. An approximative over-
view of the process is given in
As no data were available, the infrastructure it is assumed to be the same as for steel rolling. The same
amount of infrastructure per volume unit is balanced.
4.1.1 Environmental issues of sheet rolling of copper

The main environmental issues of the refined copper industry are air and water pollution. The facilities
generally have their own wastewater treatment facilities and wastewater recycling is usually practised.
Most potential wastes are reused. Some local aspects, like noise, are relevant to the industry. Due to
the hazardous nature of some solid and liquid waste streams, there is also a significant risk for soil

contamination unless they are stored and handled correctly. Fugitive emissions can be highly signifi-
cant (IPPC (2001b)).
Copper Slab
or Cake
1.010 kg
Working
materials
Preheating
Energy
carriers
Hot Rolling
(Cooling)
Water
Annealing Cold Rolling 0.010 kg

Emissions to
air, water
and land
Packing Finishing
materials
Sheet or
Strip rolling
Fig. 4.1 Copper sheet rolling process; values are copper flows

4.1.2 Sheet rolling of copper in ecoinvent

As no data were available in literature for the process of copper sheet rolling, the values for the differ-
ent in- and outputs are estimated by the values of aluminium sheet production and the hot rolling proc-
ess of steel.
The background information to the process "sheet rolling, copper" is given in Tab. 4.1, flows are
shown in Fig. 4.2.
Data quality
Data quality is estimated by filling in the pedigree matrix, according to (Frischknecht et al. (2003)),
taking into account the different more or less rough estimations and assumptions by approximating the
copper sheet rolling with the values of the aluminium sheet rolling process.
Tab. 4.1 Ecoinvent meta information for "sheet rolling, copper"
Name sheet rolling, copper

Location RER
Unit kg
Data Set Version 1
Includes the process-steps: preheating, hot rolling, cold rolling, annealing and fin-
ishing. Does not include the material beeing rolled, only the amount of scrap lost
Included Processes in waste is balanced as primary copper input . Includes the transport of the mate-
rials to the plant, but does not include the transport of the product to the cus-
tomer.
Amount 1
Local Name Blech walzen, Kupfer
Synonyms
The process of copper sheet rolling is similar to the process of aluminium sheet
rolling. Also the final outputs are similar in shape, even if the material properties
ence function
and thus the applications are very different
Start Date 2000
End Date 2002
Other Period Text Data on which assumptions are basing are from 1998
Technique describes the avarage of European Union. Data are estimated basing
Technology text on the data of aluminium sheet rolling, set in relation by mean of the volume of
material worked.
Sampling Procedure Estimation basing on Aluminium data
Uncertainty Adjustments sented by the study cited and the asumptions made, are considered in the uncer-
tainties.

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
It is assumed, that about the same volume of
copper, at regional
Primary Copper  primary material is lost due to scratch losses as metals extraction No RER 3.96E-02 k g Estimation 1 1.45 (4,5,2,1,3,5,3)
storage
it is for aluminium sheet production
Basing on values f or hot rolling of steel and
Rolling oil  chemicals organics No RER lubricating oil, at plant 4.00E-03 k g Estimation 1 1.45 (4,5,2,1,3,5,11)
aluminium sheet rolling
Basing on values f or hot rolling of steel and Water, cooling,
Water  aluminium sheet rolling considering the resource in w ater unspecif ied natural 6.00E-03 m 3 Estimation 1 1.45 (4,5,2,1,3,5,12)
temperatures reached in the process origin
Basing on values f or hot rolling of steel and
aluminium sheet rolling. A s no data w ere
available about the f raction of oil and gas heating
Fuel Oil  oil No RER industrial furnace 8.00E-01 M J Estimation 1 1.45 (4,5,2,1,3,5,1)
burners used, the amount of thermal energy is systems
1MW, non-modulating
distributed by one half f or f uel oil and natural gas
sheet rolling, copper, rolling
each.
heating
Gas  natural gas No RER industrial furnace low - 8.00E-01 M J Estimation 1 1.45 (4,5,2,1,3,5,1)
systems
NOx >100kW
electricity, medium
Basing mainly on the values of aluminium sheet production
Electricity  electricity No UCTE voltage, production 1.81E+00 k Wh Estimation 1 1.45 (4,5,2,1,3,5,2)
rolling mix
UCTE, at grid
Rolling Mill-Cu  rolled per year. Density of steel is 7850 kg/m3 metals processing Y es RER rolling mill 1.43E-09 unit Estimation 1 3.18 (4,5,2,1,3,5,9)
and f or copper 8920 kg/m3
The packing and w orking material is assumed to
Lorry 32 t  be transported over 300 km in Europe bef ore road No RER transport, lorry 32t 1.41E-02 tk m Estimation 1 2.09
systems .,n.a.,5)
using it in the rolling mill
NMV OC, non-methane
10% of rolling oil assumed (based on values of volatile organic
 VOC to air air unspecif ied 4.00E-04 k g Estimation 1 2.19 (4,5,2,1,3,5,23)
steel and aluminum rolling) compounds,
unspecif ied origin
 Water Same value as f or aluminium rolling assumed 6.00E-03 m 3 Estimation 1 1.45 (4,5,2,1,3,5,4)
 COD Same value as f or aluminium rolling assumed w ater unspecif ied 7.90E-05 k g Estimation 1 1.73 (4,5,2,1,3,5,32)
Demand
BOD5, Biological
 BOD Same value as f or aluminium rolling assumed w ater unspecif ied 7.90E-05 k g Estimation 1 2.79 (5,5,5,5,5,5,32)
Oxygen Demand
DOC, Dissolved
 DOC Same value as f or aluminium rolling assumed w ater unspecif ied 2.06E-05 k g Estimation 1 2.79 (5,5,5,5,5,5,32)
Organic Carbon
TOC, Total Organic
 TOC Same value as f or aluminium rolling assumed w ater unspecif ied 2.06E-05 k g Estimation 1 2.79 (5,5,5,5,5,5,32)
Carbon
Fig. 4.2 Flows for "sheet rolling, copper" and their representation in the ecoinvent database

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
disposal, hazardous
 Contains 15% copper No DE w aste, 0% w ater, to 1.55E-02 k g Estimation 1 1.45 (4,5,2,1,3,5,6)
sheet rolling, copper,

underground deposit
hazardous
 Oil w aste No CH 2.40E-03 k g Estimation 1 1.45 (4,5,2,1,3,5,6)
incinerated; Contains 5% copper management hazardous w aste
rolling
incineration
incineration
disposal, municipal
Solid w aste, w aste municipal solid w aste, 22.9%
 Municipal w aste as proxy; Contains 33% copper No CH 2.32E-02 k g Estimation 1 1.45 (4,5,2,1,3,5,6)
unspecif ied management incineration w ater, to municipal
incineration
Losses during the rolling process that are not
 Scrap 2.96E-02 k g Estimation 1 1.09 (2,2,1,1,1,3,6)
included in the dif f erent w astes. Recycled
 Heat w aste air unspecif ied Heat, w aste 6.51E+00 M J Estimation 1 1.45 (4,5,2,1,3,5,13)
The same amount of packing material as f or one
sheet rolling, copper,
Paper; cubic meter of aluminium sheet is inventoried. paper &

 corrugated No RER board, mixed f ibre, 3.30E-04 k g Estimation 1 1.45 (4,5,2,1,3,5,4)
cardboard Densities are 2700 kg/m3 f or aluminium and 8920 cardboard
kg/m3 f or copper.
packaging

Plastics  plastics processing No RER 1.32E-03 k g Estimation 1 1.45 (4,5,2,1,3,5,4)
at plant
steel, converter,
Steel  metals extraction No RER 1.65E-03 k g Estimation 1 1.45 (4,5,2,1,3,5,4)
unalloyed, at plant
Steel-sheet  metals processing No RER sheet rolling, steel 1.65E-03 k g Estimation 1 1.45 (4,5,2,1,3,5,4)
saw n timber,
w ooden
Wood  extraction No RER sof tw ood, raw , air 6.12E-07 m 3 Estimation 1 1.45 (4,5,2,1,3,5,4)
materials
 Copper Sheet metals processing No RER sheet rolling, copper 1.00E+00 k g
Fig. 4.2 Flows for "sheet rolling, copper" and their representation in the ecoinvent database (cont.)

4.2 Wire drawing of copper

Wire production consists in a first step of wire rod production, which can be compared and approxi-
mated by the process of sheet rolling with a smaller degree of deformation. In a second working step,
the wire rod is drawn through a series of dies and reduced to its final cross section. The description of
this process can well be approximated by the description of steel wire drawing, described in chapter
2.5.
As no data were available, the infrastructure it is assumed to be the same (per volume unit) as for steel
wire drawing.
4.2.1 Environmental issues of wire drawing of copper

The environmental issues of copper wire drawing, are te same as for the copper sheet production (see
chapter 4.1.1).
4.2.2 Wire drawing of copper in ecoinvent

No data were available in literature about the production process of copper wire drawing. As the pro-
duction process can be divided in two main steps, the assumptions and approximations for the bal-
anced values are made separately for these two steps.
The first step, the rolling of the wire rod, is similar to the rolling of the copper sheet. The second is
similar in the technique applied, but not in the values resulting, to the drawing of steel wire, because of
the greater formability of copper. Values balanced were estimated basing on aluminium and steel
forming processes. The background information to the process "wire drawing, copper" is given in Tab.
4.2, flows are shown in Fig. 4.3.
Data quality
Data quality is estimated by filling in the pedigree matrix, according to (Frischknecht et al. (2003)),
taking into account the different more or less rough estimations and assumptions by approximating the
copper wire drawing with the adapted values of the aluminium sheet rolling process.

Tab. 4.2 Ecoinvent meta information for "wire drawing, copper"
Name wire drawing, copper

Location RER
Unit kg
Data Set Version 1
Includes the production of wire rod and the further drawing of this to wire. Does
not include the material being rolled or drawn; only the amount of scrap lost in
Included Processes
waste is balanced as primary copper input. Includes the transport of the materials
to the plant, but does not include the transport of the product to the customer.
Amount 1
Local Name Draht ziehen, Kupfer
Synonyms
Wire rod production is comparable to sheet rolling leading to another final shape.
Further drawing leads to wires with cross sections ranging from 1.6 to 3.5 mm and
ence function
higher.
Start Date 2000
End Date 2002
Other Period Text Data on which assumptions are basing are from 1998
Technique describes the average of European Union. Data are estimated basing
Technology text on the data of aluminium sheet rolling, set in relation by mean of the volume of
material worked.
Sampling Procedure Estimation basing on Aluminium data
Uncertainty Adjustments sented by the study cited and the assumptions made, are considered in the un-
certainties.

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
It is assumed, that f or each step, the w ire rod
rolling and the w ire draw ing, about 50% of
copper, at regional
Primary Copper  primary material is lost due to scratch losses as metals extraction No RER 3.96E-02 k g Estimation 1 0.08 (4,5,2,1,3,5,3)
storage
it is for copper sheet production. Thus in the
same amount is inventoried.
The w ire rod production step requires less rolling
oil than sheet rolling (factor 0.8 assumed), due to
Rolling oil  the smaller degree of def ormation. For the w ire chemicals organics No RER lubricating oil, at plant 7.20E-03 k g Estimation 1 0.01 (4,5,2,1,3,5,11)
draw ing step the same amount as f or sheet
rolling is assumed.
The rod production step needs the same amount
wire drawing, copper, rolling
of cooling w ater as sheet rolling, as the material Water, cooling,

Water  is heated to the same temperature. The w ire resource in w ater unspecif ied natural 1.08E-02 m 3 Estimation 1 0.01 (4,5,2,1,3,5,12)
draw ing step requires less w ater (f actor 0.8 origin
assumed) because of f riction heat.
Basing on the amount of material to be heated.
A s no data w ere available about the fraction of light f uel oil, burned in
heating
Fuel Oil  oil and gas burners used, the amount of thermal oil No RER industrial furnace 8.00E-01 M J Estimation 1 0.83 (4,5,2,1,3,5,1)
systems
energy is distributed by one half for fuel oil and 1MW, non-modulating
natural gas each.
heating
Gas  natural gas No RER industrial furnace low - 8.00E-01 M J Estimation 1 0.83 (4,5,2,1,3,5,1)
systems
NOx >100kW
Wire rod rolling requires less, w ire draw ing more electricity, medium
production
Electricity  electrical energy f or the mechanical def ormation electricity No UCTE voltage, production 3.61E+00 k Wh Estimation 1 3.61 (4,5,2,1,3,5,2)
mix
step, than sheet rolling UCTE, at grid
Rolling Mill-Cu  rolled per year. Density of steel is 7850 kg/m3 metals processing Y es RER rolling mill 1.43E-09 unit Estimation 1 0 (4,5,2,1,3,5,9)
and f or copper 8920 kg/m3
The packing and w orking material is assumed to
Lorry 32 t  be transported over 300 km in Europe bef ore road No RER transport, lorry 32t 1.50E-02 tk m Estimation 1 0.03
systems .,n.a.,5)
using it in the rolling mill
NMV OC, non-methane
10% of rolling oil assumed (based on values of volatile organic
 VOC to air air unspecif ied 7.20E-04 k g Estimation 1 0 (4,5,2,1,3,5,23)
steel and aluminum rolling) compounds,
unspecif ied origin
Fig. 4.3 Flows for "wire drawing, copper" and their representation in the ecoinvent database

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
 Water Same value as f or aluminium rolling assumed 1.08E-02 m 3 Estimation 1 0.01 (4,5,2,1,3,5,4)
 COD Same value as f or aluminium rolling assumed w ater unspecif ied 7.90E-05 k g Estimation 1 0 (4,5,2,1,3,5,32)
Demand
BOD5, Biological
wire drawing, copper, rolling  BOD Same value as f or aluminium rolling assumed w ater unspecif ied 7.90E-05 k g Estimation 1 2.79 (5,5,5,5,5,5,32)
Oxygen Demand
DOC, Dissolved
 DOC Same value as f or aluminium rolling assumed w ater unspecif ied 2.06E-05 k g Estimation 1 2.79 (5,5,5,5,5,5,32)
Organic Carbon
TOC, Total Organic
 TOC Same value as f or aluminium rolling assumed w ater unspecif ied 2.06E-05 k g Estimation 1 2.79 (5,5,5,5,5,5,32)
Carbon
disposal, hazardous
 Contains 15% copper No DE w aste, 0% w ater, to 1.55E-02 k g Estimation 1 0.03 (4,5,2,1,3,5,6)
underground deposit
hazardous
 Oil w aste No CH 4.32E-03 k g Estimation 1 0 (4,5,2,1,3,5,6)
incinerated; Contains 5% copper management hazardous w aste
incineration
incineration
disposal, municipal
Solid w aste, w aste municipal solid w aste, 22.9%
 Municipal w aste as proxy; Contains 33% copper No CH 2.32E-02 k g Estimation 1 0.04 (4,5,2,1,3,5,6)
unspecif ied management incineration w ater, to municipal
incineration
Losses during the rolling process that are not
 Scrap 2.95E-02 k g Estimation 1 0.06 (2,2,1,1,1,3,6)
included in the dif f erent w astes. Recycled
 Heat w aste air unspecif ied Heat, w aste 1.30E+01 M J Estimation 1 13 (4,5,2,1,3,5,13)
The same amount of packing material as f or one
Paper; cubic meter of aluminium sheet product is paper &
copper, packaging
 corrugated No RER board, mixed f ibre, 3.30E-04 k g Estimation 1 0 (4,5,2,1,3,5,4)

cardboard inventoried. Densities are 2700 kg/m3 f or cardboard
wire drawing,
aluminium and 8920 kg/m3 f or copper.

Plastics  plastics processing No RER 1.32E-03 k g Estimation 1 0 (4,5,2,1,3,5,4)
at plant
steel, converter,
Steel  metals extraction No RER 1.65E-03 k g Estimation 1 0 (4,5,2,1,3,5,4)
unalloyed, at plant
Steel-sheet  metals processing No RER sheet rolling, steel 1.65E-03 k g Estimation 1 0 (4,5,2,1,3,5,4)
saw n timber,
w ooden
Wood  extraction No RER sof tw ood, raw , air 6.12E-07 m 3 Estimation 1 0 (4,5,2,1,3,5,4)
materials
 Copper w ire metals processing No RER w ire draw ing, copper 1.00E+00 k g
Fig. 4.3 Flows for "wire drawing, copper" and their representation in the ecoinvent database (cont.)

4.3 Contouring of brass and bronze

4.3.1 Fundamentals of forming of copper alloys
“Billets of copper or copper alloy are preheated and extruded and drawn into tubes with a variety of
diameters and thickness in a number of “passes”. The drawing machines use a variety of oils and soaps
as lubricant that contaminates the tubes and off cuts. The products are normally annealed and de-
greased before transport and the off cuts are de-oiled in a furnace or other degreasing process before
being returned to the furnace for melting.
Copper tubes may be also produced utilising an extrusion press with piercer in which billet sections
are extruded to tube shell pieces; the tube shell pieces are then rolled in a breakdown roller (Pilger
mill) and finally drawn to size in drawing blocks. The oil utilised (in small quantities) for the drawing
operations is neutralised in in-line degreasing/pickling systems often connected with the annealing
section.
Products are annealed in a variety of furnaces under reducing conditions using as protection gas “ex-
ogas” or hydrogen/nitrogen mixtures.
Wire, tube, strip and some other materials are pickled to produce a bright finish before packing or, in
the case of copper wire, to remove oxide from the surface prior to wire drawing. Sulphuric acid solu-
tions are used and sometimes a mixture of dilute sulphuric and nitric acids are used. In the latter case
nitric fumes can be emitted if higher strength acids are used.
Surface pickling of sheets and strips is usually performed with sulphuric acid (8 - 10%) but for some
special alloys, mixtures of sulphuric and nitric acids are used in continuous lines or automatic systems.
Gas cleaning is used to remove nitric acid fumes. The acid is changed from time to time in order to en-
sure that surface quality is not affected.
Spent acid is sent for treatment and recovery of metal in internal or external treatment plants. The
pickling systems are vented for worker protection. The products are rinsed and the rinse water is sent
for treatment, sludges are recycled if possible. For surface degreasing of rolled products, detergents
containing water are used. The spent water is cleaned by ultra filtration.” (IPPC (2001b))
4.3.2 Environmental issues of forming of copper alloys

Beside the energy consumption of the process, the emissions of metals and hydrocarbons to water
from degreasing and pickling are an environmental issue. IPPC (2001b) gives the values in Tab. 4.3
for the annual loads discharged to water from a copper semis production plant. Hydrocarbon emissions
are not measured. Emissions of “protection gas” can also be a problem:
“Typical CO production in a shaft furnace used for wire-rod or semis production is 2000 to11000
grams per tonne of copper. In some installations after-burning is used to remove hydrocarbons from
the gases when scrap coated with organic matter is processed. CO is also destroyed at the same time
and the emissions are reported to be ~ 45 grams per tonne of copper.” (IPPC (2001b))
4.3.3 Contouring of brass and bronze in ecoinvent

The processes are theoretically modelled according to the information in IPPC (2001b). Most of the
values for the flows in these processes are estimated from similar processes since very few quantitative
data is available.
The background information to the process "anodising, aluminium sheet" is given in Tab. 4.4, flows
are shown in Fig. 4.4 and Fig. 4.5.

Data quality
The uncertainty is estimated by completing the pedigree matrix, according to (Frischknecht et al.
(2003)). A lognormal distribution is assumed. The different more or less rough estimations and as-
sumptions are taken into account. Overall data quality is poor.
Tab. 4.3 Annual loads discharged to water from a copper semis production plant (IPPC (2001b))
Emission kg/a kg/m3 effluent

Cu 11 3.14E-04
Ni 3 8.57E-05
Zn 25 7.14E-04
Pb 1 2.86E-05
Cr 1 2.86E-05
As 0.01 2.86E-07
Cd 0.01 2.86E-07
Hg 0.01 2.86E-07
Sn 1 2.86E-05
Effluent discharge: 35000 m3/a.
Tab. 4.4 Ecoinvent meta information for contouring of brass and bronze
Name contour, brass contour, bronze

Location RER RER
Unit kg kg
Heating, rolling, degreasing / pickling Heating, rolling, degreasing / pickling
Included Processes
and waste water treatment. and waste water treatment.
Amount 1 1
Local Name Profilieren, Messing Profilieren, Bronze
Synonyms rolling brass rolling bronze
Most of the data used stem from proc- Most of the data used stem from proc-
esses similar to contouring brass. esses similar to contouring bronze.
ence function
Overall data quality is poor. Overall data quality is poor.
End Date 2002 2002
Other Period Text
Geography text
Technique describes the average of Technique describes the average of
Technology text European Union. Data are estimated European Union. Data are estimated
basing on data of similar processes. basing on data of similar processes.
Production Volume Unknown Unknown
Sampling Procedure Literature Literature
Extrapolations None None
Uncertainty Adjustments None None

General flow information Representation in ecoinvent Uncertainty

Infra So urc e
Pro c ess Cate Sub Lo c a Mo dul name in Me an StDv General
Input Output Remark s struc Unit me an Ty pe
Name go ry c atego ry tio n e c o invent value 95% Co mme nt
ture value
brass  met als ext ract io n No CH brass, at p lant 3 .96 E-02 kg Est imat e 1 2.3 9 (4,5,5,5,5,5,4 )
Same as f o r co pp er sheet ro lling
ro lling o il  chemicals organics No RER lubricat ing oil, at plant 4 .00 E-03 kg Est imat e 1 2.3 9 (4,5,5,5,5,5,4 )
f or deg reasing / p ickling. A mo unt as
sulphuric acid  HCl f o r d egreasing / pickling of st eel chemicals inorganics No RER sulphuric acid, liq uid, at plant 3.4 0E-0 1 kg Est imat e 1 2.3 9 (4,5,5,5,5,5,4 )
sheet f o r p owd er co at ing
wat er
wat er  prod uct io n No RER t ap wat er, at user 6.0 0E+00 kg Est imat e 1 2.3 9 (4,5,5,5,5,5,4 )
supp ly
contour, brass, part 1
elect ricit y elect ricit y, med ium volt age,

 elect ricit y prod uct io n mix No UCTE 1.8 1E+00 kW h Est imat e 1 2 .56 (5,5,5,5,5,5,2)
general prod uct io n UCTE, at g rid
heat , nat ural g as, at ind ust rial
heat f or curing  nat ural gas heat ing syst ems No RER 1.60 E+0 0 MJ Est imat e 1 2.3 9 (4,5,5,5,5,5,1)
f urnace low-NOx >10 0kW
t ranspo rt (4,5,n.a.,n.a.,n.a
Transpo rt  ro ad No RER t ranspo rt , lorry 3 2t 3 .84 E-02 t km Est imat e 1 2.0 9
syst ems .,n.a.,5)
Transpo rt  t rain No RER t ranspo rt , f reight , rail 2.3 0E-0 1 t km Est imat e 1 2.0 9
syst ems .,n.a.,5)
Inf rast ruct ure  met als processing Y es RER ro lling mill 1.4 3E-0 9 unit Est imat e 1 4 .25 (5,5,5,5,5,5,9)
 W ast e heat air unspecif ied Heat , wast e 6 .52E+00 MJ calculat io n 1 2 .56 (5,5,5,5,5,5,13 )
NM V OC, non-met hane vo lat ile
 CO t o air air unspecif ied organic co mpo unds, 4.50 E-05 kg IPPC 20 01 1 2 .61 (4,5,5,5,5,5,3 1)
unspecif ied o rig in
 wast e wat er 5.50E-0 3 m3 Est imat e
 COD wat er unspecif ied 7.90 E-05 kg Est imat e 1 2 .79 (5,5,5,5,5,5,32 )
Demand
Same value as f o r ho t rolling of st eel B OD5, B iolo gical Oxyg en
 B OD wat er unspecif ied 7.90 E-05 kg Est imat e 1 2 .79 (5,5,5,5,5,5,32 )
assumed Demand
DOC, Dissolved Org anic
 DOC wat er unspecif ied 2 .06 E-05 kg Est imat e 1 2 .79 (5,5,5,5,5,5,32 )
Carb on
 TOC wat er unspecif ied TOC, Tot al Org anic Carb on 2 .06 E-05 kg Est imat e 1 2 .79 (5,5,5,5,5,5,32 )
Fig. 4.4 Flows for "contour, brass" and their representation in the ecoinvent database


Infra So urc e
ture value
 Cu wat er unspecif ied Cop per, ion 1.73 E-06 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Ni wat er unspecif ied Nickel, ion 4 .71E-0 7 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
contour, brass, part 2  Zn wat er unspecif ied Zinc, io n 3 .93 E-06 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Pb wat er unspecif ied Lead 1.57E-07 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Cr Relat ion Cr / Cr V I as f o r wast e wat er unspecif ied Chro mium, ion 1.51E-07 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 wat er f ro m pickling st eel

Cr V I wat er unspecif ied Chro mium V I 6 .29 E-09 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 As wat er unspecif ied A rsenic, io n 1.57E-09 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Cd wat er unspecif ied Cadmium, io n 1.57E-09 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Hg wat er unspecif ied M ercury 1.57E-09 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Sn wat er unspecif ied Tin, io n 1.57E-07 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)

W ast e wat er disp osal, slud ge f ro m st eel
Same value as f o r ho t rolling of st eel wast e resid ual mat erial
 t reat ment No CH ro lling, 2 0% wat er, t o resid ual 1.6 3E-0 2 kg IPPC 20 01 1 2 .56 (5,5,5,5,5,5,6)
assumed management landf ill
sludg es mat erial landf ill
W ast e f ro m Zn rich f ilt er dust f rom EA F st eel disp osal, d ust , unallo yed EA F
wast e resid ual mat erial
 f ilt ers and used as proxy; A mo unt est imat ed as No CH st eel, 15.4% wat er, t o resid ual 4 .42 E-04 kg Est imat e 1 2 .56 (5,5,5,5,5,5,6)
sludg es f or hot ro lling o f st eel. mat erial landf ill
cont ouring
 met als processing No RER cont our, brass 1.00 E+0 0 kg
brass
Fig. 4.4 Flows for "contour, brass" and their representation in the ecoinvent database (cont.)


Infra So urc e
ture value
bronze  met als ext ract io n No CH bronze, at plant 3 .96 E-02 kg calculat io n 1 2.3 9 (4,5,5,5,5,5,4 )
ro lling o il  chemicals organics No RER lubricat ing oil, at plant 4 .00 E-03 kg Est imat e 1 2.3 9 (4,5,5,5,5,5,4 )
f or deg reasing / p ickling. A mo unt as
sulphuric acid  HCl f o r d egreasing / pickling of st eel chemicals inorganics No RER sulphuric acid, liq uid, at plant 3.4 0E-0 1 kg Est imat e 1 2.3 9 (4,5,5,5,5,5,4 )
sheet f o r p owd er co at ing
wat er
wat er  prod uct io n No RER t ap wat er, at user 6.0 0E+00 kg Est imat e 1 2.3 9 (4,5,5,5,5,5,4 )
supp ly
contour, bronze, part 1
elect ricit y elect ricit y, med ium volt age,

 elect ricit y prod uct io n mix No UCTE 1.8 1E+00 kW h Est imat e 1 2 .56 (5,5,5,5,5,5,2)
general prod uct io n UCTE, at g rid
heat , nat ural g as, at ind ust rial
heat f or curing  nat ural gas heat ing syst ems No RER 1.60 E+0 0 MJ Est imat e 1 2.3 9 (4,5,5,5,5,5,1)
f urnace low-NOx >10 0kW
Transpo rt  ro ad No RER t ranspo rt , lorry 3 2t 3 .84 E-02 t km Est imat e 1 2.0 9
syst ems .,n.a.,5)
Transpo rt  t rain No RER t ranspo rt , f reight , rail 2.3 0E-0 1 t km Est imat e 1 2.0 9
syst ems .,n.a.,5)
Inf rast ruct ure  met als processing Y es RER ro lling mill 1.4 3E-0 9 unit Est imat e 1 4 .25 (5,5,5,5,5,5,9)
 W ast e heat air unspecif ied Heat , wast e 6 .52E+00 MJ calculat io n 1 2 .56 (5,5,5,5,5,5,13 )
NM V OC, non-met hane vo lat ile
 CO t o air air unspecif ied organic co mpo unds, 4.50 E-05 kg IPPC 20 01 1 2 .61 (4,5,5,5,5,5,3 1)
unspecif ied o rig in
 wast e wat er 5.50E-0 3 m3 Est imat e
 COD wat er unspecif ied 7.90 E-05 kg Est imat e 1 2 .79 (5,5,5,5,5,5,32 )
Demand
Same value as f o r ho t rolling of st eel B OD5, B iolo gical Oxyg en
 B OD wat er unspecif ied 7.90 E-05 kg Est imat e 1 2 .79 (5,5,5,5,5,5,32 )
assumed Demand
DOC, Dissolved Org anic
 DOC wat er unspecif ied 2 .06 E-05 kg Est imat e 1 2 .79 (5,5,5,5,5,5,32 )
Carb on
 TOC wat er unspecif ied TOC, Tot al Org anic Carb on 2 .06 E-05 kg Est imat e 1 2 .79 (5,5,5,5,5,5,32 )
Fig. 4.5 Flows for “contour, bronze" and their representation in the ecoinvent database


Infra So urc e
ture value
 Cu wat er unspecif ied Cop per, ion 0.0 0E+00 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Ni wat er unspecif ied Nickel, ion 0.0 0E+00 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
contour, bronze, part 2  Zn wat er unspecif ied Zinc, io n 0.0 0E+00 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Pb wat er unspecif ied Lead 0.0 0E+00 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Cr Relat ion Cr / Cr V I as f o r wast e wat er unspecif ied Chro mium, ion 0.0 0E+00 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Cr V I wat er f ro m pickling st eel wat er unspecif ied Chro mium V I 0.0 0E+00 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 As wat er unspecif ied A rsenic, io n 0.0 0E+00 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Cd wat er unspecif ied Cadmium, io n 0.0 0E+00 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Hg wat er unspecif ied M ercury 0.0 0E+00 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)
 Sn wat er unspecif ied Tin, io n 0.0 0E+00 kg IPPC 20 01 1 6 .45 (5,5,5,5,5,5,35)

W ast e wat er disp osal, slud ge f ro m st eel
Same value as f o r ho t rolling of st eel wast e resid ual mat erial
 t reat ment No CH ro lling, 2 0% wat er, t o resid ual 1.6 3E-0 2 kg IPPC 20 01 1 2 .56 (5,5,5,5,5,5,6)
assumed management landf ill
sludg es mat erial landf ill
W ast e f ro m Zn rich f ilt er dust f rom EA F st eel disp osal, d ust , unallo yed EA F
wast e resid ual mat erial
 f ilt ers and used as proxy; A mo unt est imat ed as No CH st eel, 15.4% wat er, t o resid ual 4 .42 E-04 kg Est imat e 1 2 .56 (5,5,5,5,5,5,6)
sludg es f or hot ro lling o f st eel. mat erial landf ill
cont ouring
 met als processing No RER cont our, bronze 1.00 E+0 0 kg
brass
Fig. 4.5 Flows for “contour, bronze" and their representation in the ecoinvent database (cont.)

5. Zinc coating of steel
5 Zinc coating of steel

The hot-dip galvanising of steel involves the application of a zinc coating onto the surface of steel by a
process wherein the steel is immersed into molten zinc to apply the coating. The zinc reacts with the
steel to form the galvanised coatings. The time the steel is immersed in the zinc along with post-
galvanising treatment controls the coating thickness, appearance and other characteristics. Hot dip gal-
vanised coatings are applied to steel to improve the anti-corrosion performance of the steel to ensure
that it lasts as long as possible with a minimum of maintenance or to provide a decorative appearance.
Since zinc has to be heated to a temperature of approximately 455 °C or higher for the galvanising
process to be implemented correctly, the process is referred to as a “hot dip” process.
There are two common practices for applying zinc in hot-dip galvanising. One involves the application
of zinc onto steel sheet as it is passed as a continuous ribbon of steel through a bath of molten zinc at
high speeds. Hence, the term “continuous” hot-dip galvanising.
The other process involves the application of a zinc coating to the surface of steel parts after the parts
have been fabricated. This process is not a continuous process in that the parts are immersed as a dis-
crete “batch” into the zinc bath. Hence, the names “batch”, “after fabrication” and “general” galvanis-
ing are applied to this process (extracted from Pankert (1997) and internet 6,7 ).
5.1 Zinc coating of steel coils

The production process is described in (IPPC (2001a)) and two URL-addresses6,8 .
As mentioned above, the continuous galvanising process is used to describe the application of a zinc
coating onto the surface of steel sheet. In the hot dip coating process, the steel is continuously passed
through the molten metal, at speeds up to 200 meters per minute. The sheet thickness might be as thin
as 0.25 mm or less to as thick as 6.3 mm. In the molten-metal bath, an alloying reaction between the
two metals takes place, leading to a good bond between coating and substrate. As the alloy layer is
hard and brittle, steel is immersed for a very brief time, in order to keep the alloy layer very thin to ac-
commodate subsequent forming into the final shape. As the strip emerges from the molten bath, it
drags out excess metal, much like when an object is pulled rapidly from a container of water. Through
the use of an air-wiping process, a controlled amount (thickness) of coating usually expressed as
weight of coating per unit area is allowed to remain on the strip surface.
Metals suitable for the use in hot dip coating are those, which have a melting point low enough to
avoid any thermal changes in the steel product, like for example aluminium, lead, tin, and zinc. In this
study only the case of zinc was analysed.
In general continuous coating lines for sheet comprise the following steps:
- An entry-end welder to join the trailing edge of one coil to the leading edge of the succeeding coil,
- Surface cleaning by means of chemical and/or thermal treatment,
- Heat treatment; annealing furnace to heat the steel to high temperatures to impart the desired me-
chanical properties (strength and formability) to the steel sheet,
6
http://www.galvinfo.com/GalvInfoNote%20PDF's/GalvInfo9.pdf, accessed 03.03.03
7
http://www.corp.indgalv.com.au/technical/pdf/zfile02.pdf, accessed 03.03.03
8
http://www.galvinfo.com/GalvInfoNote%20PDF's/ginote2.pdf, accessed 03.03.03

- Immersion in a bath of molten metal (zinc for this study),
- Finishing treatment (cooling section, temper mill, tension leveller, surface treatment section and a
recoiler).
Continuous wire galvanising plants involve the following steps: pickling, fluxing, galvanising, finish-
ing. Wire is hot dip coated with zinc (galvanised) or tin. Zinc coating is mainly applied as corrosion
protection. Tin gives the wire a shine appearance and also provides a solder adhesive layer. As the
flows for the two processes are similar, in this study only the continuous galvanising lines for sheets
and trips were inventoried. A more detailed description of every single step is given below.
The infrastructure of continuous hot dip zinc coating comprises buildings where galvanising processes
take place and for storage, surroundings and machinery fore the single process steps. The assumed
lifetime for the infrastructure is 50 years. As no information is available, the infrastructure is approxi-
mated by the infrastructure for rolling.
The production of continuous hot dip coating lines in EU 15 was 15 Mt in 1997. The vast majority of
coated products, as shown in Fig. 5.1, were galvanised steel.
Fig. 5.1 Production share for different hot dip coatings IPPC (2001a)
5.1.1 Entry section

As one coil is being processed through the coating bath, another is being welded to the trailing edge of
the first coil. Thus, the process is truly “continuous”, the coating line may indeed operate continuously
for days without interruption. No data is available about the welding process therefore it is neglected.
Hot dip galvanising of cold rolled strip is more common than galvanising hot rolled strip. The process-
ing is similar, except that for hot rolled products additional descaling (pickling, to remove hot mill
scale) is necessary. Pickling of unannealed cold rolled coil is done for activation. Both processes are
done in hydrochloric acid, followed by rinsing. The technology is the same as the pickling process de-
scribed in the cold rolling section 2.3.1.
It is essential that the surface of the steel coil is free of impurities such as grease, oil or abraded iron
fines to ensure the entire surface is exposed to the coating product and strong adherence is achieved.
Although degreasing is not necessary when the strip is subsequently heat treated, it is often applied in
any case. The cleaning process involves several or all of the following steps:
- Alkaline degreasing via immersion or by spraying; maybe coupled with brushing.
- Alkaline degreasing by an electrolytic system; maybe coupled with brushing.

- Rinsing with water with intermediate brushing between rinsing tanks.
- Drying.
Description is taken from (IPPC (2001a)) and the internet 9 .
5.1.2 Heat treatment

From the cleaning section, the strip passes into the heating (annealing) furnace to soften the full-hard
strip, to impart the desired strength and formability to the steel and to allow the steel to reach the re-
quired temperature before dipping. In the annealing furnace, the strip is maintained under a reducing
gas atmosphere to remove any vestiges of oxides on the steel surface. The gas atmosphere is composed
of hydrogen and nitrogen. This oxide-reduction step, iron oxide being converted to iron by reaction
with the hydrogen, is very important to obtain complete wetting of the steel surface during the short
time that it is immersed into the coating bath.
The following types of furnaces are used:
- Sendzimir Furnace; this type of furnace is no longer used in modern coating lines, but some older
plants might still operate this horizontal continuous furnace,
- Direct flame furnace (D.F.F; non oxidizing type); this type of furnace cleans the coil surface and
does not require any prior degreasing of the metal, but air emissions are generated as the remain-
ing oil from the surface is burnt,
- Radiant Tube Furnace (R.T.F.); this furnace is a variation with indirect heating. The combustion
gases circulate through radiant tubes and do not get in contact with the strip. The annealing fur-
nace is combined with an alkaline degreasing equipment.
Fuels used are desulphured coke oven gas and (inventoried in this study) natural gas. Description is
taken from (IPPC (2001a)) and the internet 10 .
5.1.3 Hot dipping (galvanising)

The exit end of the furnace is connected directly to the molten coating bath metal to prevent the heated
steel strip from re-oxidising (in the air) prior to reacting with the molten coating metal. The galvanis-
ing bath consists of one or more tanks, usually made of ceramic material. These pots, which can be ei-
ther fixed or mobile, contain the molten metal at a temperature of 440 – 490 °C, through which the
strip is passed. The bath contains zinc and any other required additive (i.e. antimony, lead or alumin-
ium). The bath contains a sufficient amount of molten zinc in order to prevent the wide fluctuations in
the operating temperature that may appear under maximum operating conditions. Most pots are nowa-
days heated by electrical induction systems. However, natural gas as fuel for heating the pots is an al-
ternative.
As the steel passes through the molten zinc bath the surface is coated, to some extent with layers of
different iron-zinc alloys, but due to the high speed of the steel strip (max. 180 - 200 m/min) and the
short exposure time, the coating (between 20 and 45 m thick (Frischknecht et al. (1996))) mainly
consist of zinc.
Temperature control of the bath is essential, as a high temperature will increase the rate of oxidation at
the surface of the molten bath, resulting in increased ash generation. A low bath temperature will in-
crease the viscosity of the molten coating metal, thus inhibiting the formation of a thin coating layer.
9
10

The bath temperature is therefore optimised to meet the individual quality requirements of the final
coated product.
In the coating bath, the strip passes around a submerged roll and then exits the bath in a vertical direc-
tion. At the exit point, a set of air or nitrogen knives (high pressure air jets) wipe off excess molten
metal leaving behind a closely controlled thickness of molten metal. The system is usually controlled
by an automatic gauge, which measures the thickness of the coating using X-ray measurement tech-
nology or a similar technique. Description is taken from (IPPC (2001a)) and the internet10.
5.1.4 Cooling, post treatment and finishing

The coating is then gradually cooled by air coolers to allow the metal to freeze on the steel surface.
Freezing of the coating, or solidification, has to be accomplished before the strip contacts another roll
to avoid transferring the coating onto the roll. Thus, coating-process lines usually have a high tower
above the coating bath.
Following the coating, the steel strips are usually subjected to post-treatment in order to prevent sur-
face damages and flaws, like white rust caused by water condensation in case of insufficient air access.
Voluminous corrosion products as zinc hydroxides (white rust) affect further processing, such as e.g.
painting. The formation of condensate during transportation and storage is to be generally prevented.
The resistance against white rust is considerably increased by the application of post treatment, like:
- Oiling; applying a wet film coat with thickness between 0.25 - 3 g/m2/side,
- Passivation; a treatment with solvents containing chromic acids, applied by spraying or applicator-
rolls, with coat thickness of a few nanometer only,
- Phosphating; applying a phosphate coat of approx. 1 - 1.8 g/m².

To give the steel a special surface appearance, smoothness etc. as required by the customers or to meet
width tolerances, the following finishing operations may be applied:
- Mini or No Spangle Treatment; the strip is either cooled more rapidly or steam sprayed with a
chemical additive (usually phosphate based) or sprayed with zinc powder,
- Mat Mill; the strip is skin-passed to obtain a mat finish surface.
- For certain applications, an edge cutting must be done in order to obtain the width tolerance
needed and to avoid little dog bone defects.
The continuous strip is sheared at the weld that was made at the entry end of the line to remove the
weld as well as to maintain coil-to-coil identity. Description is taken from (IPPC (2001a)) and the in-
ternet 11 .
5.1.5 Environmental issues of continuous zinc coating process of steel

Main environmental issues concerning this sub-sector are acidic air emissions, wastes and wastewater;
air emissions and energy consumption of furnaces, Zinc-containing residues, oil- and chrome-
containing wastewater (IPPC (2001a)).
As wastewater is treated in the factory, post treatment wastewater composition and ancillary products
for the water treatment as given in IPPC (2001a) are used to calculate the expenditures and the water
emissions of the water treatment process.
11

5.1.6 Continuous hot dip zinc coating in ecoinvent

In (IPPC (2001a)) most values are given in ranges, the mean values are generally calculated as the
geometric mean of the given range. When the lower value of the range is zero, then the arithmetic
mean of the range is taken (what correspond to half of the maximal value). Values taken from
(Frischknecht et al. (1996)) are already mean values and do not have any information about minima
and maxima or distribution.
The background information to the process "zinc coating, coils" is given in Tab. 5.1, flows are shown
in Fig. 5.2.
Data quality
For values represented by the geometric mean, uncertainty is calculated assuming a lognormal distri-
bution and the high value given in (IPPC (2001a)) representing the 95% geometric standard deviation.
For the arithmetic means of ranges given in (IPPC (2001a)) and the values from (Frischknecht et al.
(1996)), the uncertainty is estimated by completing the pedigree matrix, according to (Frischknecht et
al. (2003)).

Tab. 5.1 Ecoinvent meta information for "zinc coating, coils"
Name zinc coating, coils

Location RER
Unit m2
Includes the process steps surface cleaning (by means of chemical and/or ther-
mal treatment), heat treatment, immersion in a bath of molten zinc and finishing
Included Processes
treatment. Also includes zinc input and transportation to coiling plant. Does not
include the steel strip or sheet being coated.
Amount 1
Local Name Bandverzinkung
Synonyms hot dip coating//galvanizing//sendzimir
This module is applicable, if continuous zinc coating (galvanising) of hot or cold
rolled strip- or wire-coils has to be inventoried. Data given per tonne of coated
product are transformed by dividing them by the mean surface area of 64 m2/t
(corresponding to ca. 4.2 mm thickness). The zinc coating layer is between 20 to
45 m thick. The ecoinvent module "zinc coating, pieces, adjustment per um" can
be used for adjusting for thinner or thicker zinc layers of coated coils, but it 1.53
ence function
m of it have to be inventoried per um layer thickness adjustment. This is be-
cause the zinc consumption per um is higher for coil coating than it is for piece
coating. For the chemicals and the zinc, 100 km of 32t lorry transport were inven-
toried. The softened industrial water is assumed to be produced at the coating
plant. Continuous wire coating can be approximated with this module.
Start Date 1996
End Date 2002
Other Period Text Reference from 1996 refers partly to data from 1993.
Technology text Average technique for EU.
Production Volume 12 million tonnes in 1997
Some values are from single or very few plants. For other values no data were
Uncertainty Adjustments available about the percentage of companies represented by the study cited. This
was considered in the uncertainties.

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
hydrochloric acid, 30% Frischknecht et (1,n.A.,3,1,1,n.A
HCl-in  chemicals inorganics No RER 3.40E-01 k g 1 1.11
in H2O, at plant al. 1996 .,4)
A ssumption: same amount per m2 product
A nticorrosive
 surf ace as f or HCl-pickling in cold rolling sheet. chemicals organics No RER lubricating oil, at plant IPPC 2001 1 1.41 see text
zinc coating, coils, Pickling
1.77E-02 k g
oils
Lubricating oil as proxy
Frischknecht et (2,n.A.,2,1,1,n.A
 HCl to air air unspecif ied Hydrogen chloride 5.05E-02 k g 1 2
al. 1996 .,23)
calculated from
surf ace as f or HCl-pickling in cold rolling sheet.
max value
 Sludge cake soil unspecif ied Iron 9.94E-03 k g IPPC 2001 1 5.28 assuming
lognormal
inventoried. About tw o thirds are recycled -->
distribution
surf ace as f or HCl-pickling in cold rolling sheet. hazardous
 Only 22.8% w hich are landf illed are inventoried, w aste No CH 1.39E-06 k g IPPC 2001 1 1.56 (1,5,1,1,1,3,41)
the rest is recycled. Oil containig w astes are incineration
incineration
approximated w ith used mineral oil.
A ssumption: same amount per kg product as for
Strip head / tail
 HCl-pickling in cold rolling sheet. Recycled --> not 4.69E-01 k g IPPC 2001 1
scrap
inventoried
calculated from
sodium hydroxide, max value
zinc coating, coils, Degreasing
Detergents  NaOH as proxy chemicals inorganics No RER 50% in H2O, 3.52E-03 k g IPPC 2001 1 2 assuming
production mix, at plant lognormal
distribution
NaOH,(PO4)2-
 Na as proxy (amount ref ers to Na in NaOH) air unspecif ied Sodium 4.49E-07 k g IPPC 2001 1 2.05 (1,5,1,1,1,3,23)
air
NaOH,(PO4)2-
 Na as proxy (amount ref ers to Na in NaOH) w ater unspecif ied Sodium, ion 4.04E-03 k g IPPC 2001 1 1.56 (1,5,1,1,1,3,33)
w ater
calculated from
A ssumption: detergents not contained in the oily hazardous max value
 Oil sludge. Oil containing w astes are approximated w aste No CH 7.03E-04 k g IPPC 2001 1 2 assuming
by used mineral oil disposal incineration lognormal
incineration
distribution
hazardous
Oil containing w astes are approximated by used w aste oil, 10% w ater, to
 Oily sludge w aste No CH 2.81E-03 k g IPPC 2001 1 1.56 (1,5,1,1,1,3,41)
mineral oil disposal management hazardous w aste
incineration
incineration
 Fe2O3, FeSO4 Ref erence contained no data n.a. IPPC 2001
Fig. 5.2 Flows for "zinc coating, coils" and their representation in the ecoinvent database

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
hydrogen, liquid, at
Hydrogen  chemicals inorganics No RER 1.92E-03 k g IPPC 2001 1 1.83
plant
zinc coating,
coils, Heat
treatment
Nitrogen  chemicals inorganics No RER nitrogen, liquid, at plant 6.30E-01 k g IPPC 2001 1 2.48
ammonia, liquid, at Frischknecht et

A mmonia  chemicals inorganics No RER 3.06E-03 k g 1 3.58
regional storehouse al. 1996 calculated from
CO- included in the inventoried module for natural gas max value
 8.30E-05 k g IPPC 2001 1 4.39
RTFf urnace burning (overall process) assuming
NOx- included in the inventoried module for natural gas lognormal
 1.19E-03 k g IPPC 2001 1 1.38
RTFf urnace burning (overall process) distribution
zinc for coating, at
Zinc metals extraction No RER IPPC 2001 1 1.74
coating
 6.79E-01 k g
, coils,
regional storage
zinc
 Plate scrap recycled --> not inventoried 1.15E-01 k g IPPC 2001 1 4.9
Zn containing
 recycled --> not inventoried 3.52E-02 k g IPPC 2001 1 2
material
Ref erence contained no data. A ssumption: Same
amount per m2 product surf ace as f or cold
coating,
Oil-in  chemicals organics No RER lubricating oil, at plant 3.54E-02 k g IPPC 2001 1 1.22 (1,5,1,1,1,3,4)
rolling chromium steel sheet. Lubricating oil as
coils,
zinc
proxy
Now adays, the oily protection is mainly applied
 Oil-out by electrostatic oily machines, w hich produce ne gligible IPPC 2001
negligible emission levels
Corrosion
inhibitors, partly  Ref erence contained no data n.a. IPPC 2001
zinc coating, coils, Waste water
Zn-based
Biocides:
sodium
 Ref erence contained no data n.a. IPPC 2001
hypochlorite,
treatment plant
sodium bromide
Flocculating
agents: Ref erence contained no data. A ssumption: Same
aluminium sulphate,
A luminium  amount per kg product as f or steel f orming chemicals inorganics No RER 6.07E-05 k g IPPC 2001 1 1.22 (2,5,1,1,1,3,4)
pow der, at plant
sulphate, process. A luminium sulphate as proxy.
polymers
w ater leaves the system as treated ef fluent; it is
assumed to be discharged in surface w ater -->
not inventoried. Infrastructure f or w aste w ater
 Waste w ater 6.72E-03 m 3 IPPC 2001
treatment plant is comprised in the infrastructure
module. V alue f or energy use is comprised in the
range of uncertainty of inventoried overall
energy.
Fig. 5.2 Flows for "zinc coating, coils" and their representation in the ecoinvent database (cont.)

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Suspended solids, calculated from
 TSS w ater unspecif ied 5.33E-06 k g IPPC 2001 1 8.53
unspecif ied max value
 Iron w ater unspecif ied Iron, ion 1.15E-05 kg IPPC 2001 1 8.17 assuming
zinc coating, coils, Waste water treatment plant  Nickel w ater unspecif ied Nickel, ion 1.98E-07 kg IPPC 2001 1 3.16 lognormal
 Zinc w ater unspecif ied Zinc, ion 1.56E-06 kg IPPC 2001 1 20 distribution
 Lead w ater unspecif ied Lead 1.31E-07 kg IPPC 2001 1 5.06 (1,5,1,1,1,3,35)
 Chromium w ater unspecif ied Chromium, ion 1.44E-06 kg IPPC 2001 1 2 calculated from
 Cr6+ w ater unspecif ied Chromium V I 6.72E-08 kg IPPC 2001 1 2 max value
COD, Chemical Oxygen assuming
 COD w ater unspecif ied 4.28E-04 k g IPPC 2001 1 5.48
Demand lognormal
BOD5, Biological calculated f rom
 BOD w ater unspecif ied 4.28E-04 k g 1 11 Tw ice the
Oxygen Demand COD
uncertainty of
DOC, Dissolved calculated f rom
 DOC w ater unspecif ied 1.99E-05 k g 1 13.4 COD
Organic Carbon TOC
calculated from
max value
TOC, Total Organic
 TOC w ater unspecif ied 1.99E-05 k g IPPC 2001 1 6.69 assuming
Carbon
lognormal
distribution
Hydrocarbons,
 Hydrocarbons w ater unspecif ied 1.16E-06 k g IPPC 2001 1 3.05 (1,5,1,1,1,3,34)
unspecif ied
 Oil and f at w ater unspecif ied Oils, unspecif ied 6.94E-06 k g IPPC 2001 1 6.08
calculated from
 Phosphorus w ater unspecif ied Phosphorus 1.17E-07 k g IPPC 2001 1 9.35
max value
Water Waste w ater treatment sludge f rom steel f orming disposal, sludge f rom
residual assuming
treatment processes is inventoried as proxy. 1.2 times the w aste steel rolling, 20%
 material No CH 7.26E-03 k g IPPC 2001 1 3.87 lognormal
sludge (dry amount is inventoried to accuont f or the w ater management w ater, to residual
landfill distribution
material) content of the proxy material landf ill
Waste w ater treatment sludge f rom steel f orming disposal, sludge f rom
Chemical residual
processes is inventoried as proxy. 1.2 times the w aste steel rolling, 20%
 treatment material No CH 2.25E-03 k g IPPC 2001 1 5.06 (1,5,1,1,1,3,35)
amount is inventoried to accuont f or the w ater management w ater, to residual
sludges landfill
content of the proxy material landf ill
Frischknecht et
 Chloride w ater unspecif ied Chloride 3.13E-05 k g 1 1.56 (1,5,1,1,1,3,33)
al. 1996

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Demineralised w ater, deionised, at
 w ater supply production No CH 4.03E+00 k g IPPC 2001 1 1.55
Water plant
Process and Water, unspecified
 resource in w ater Y es CH 1.18E-02 m 3 IPPC 2001 1 1.22 (1,5,1,1,1,3,12)
cooling w ater natural origin
Process and cooling w ater evaporated
 Water (uncontamined --> not inventoried) or already 9.16E-03 m 3 IPPC 2001 1 1.22 (1,5,1,1,1,3,12)
included in the w aste disposal
heating
Natural gas  natural gas No RER industrial furnace 1.84E+01 M J IPPC 2001 1 1.47
systems calculated from
>100kW
max value
electricity, medium
production assuming
Electricity  electricity No UCTE voltage, production 3.41E-01 k Wh IPPC 2001 1 1.78
mix lognormal
zinc coating, coils, Overall process
UCTE, at grid
distribution
Hot w ater  chemicals organics No RER 4.64E-01 M J IPPC 2001 1 1.48
chemical plant
Frischknecht et (1,n.A.,3,1,1,n.A
Soda  chemicals inorganics No RER soda, pow der, at plant 1.67E-03 k g 1 1.11
al. 1996 .,12)
Transport, transport Frischknecht et (4,5,n.a.,n.a.,n.a
 road No RER transport, lorry 32t 1.52E-01 tk m 1 2.09
Lorry 32t systems al. 1996 .,n.a.,5)
 Dross recycled n.a IPPC 2001
Particle size f rom mechanical process: Frischknecht et
 Dust >10 air unspecif ied Particulates, > 10 um 5.70E-03 k g 1 1.57 (1,5,2,1,1,3,25)
assumption 50 % >10 um. al. 1996
Particulates, > 2.5 um, Frischknecht et
 Dust 2.5-10 2.5-10 um. In addition 50 % of the mechanical air unspecif ied 1.71E-02 k g 1 2.05 (1,5,2,1,1,3,26)
and < 10um al. 1996
abrasion
Frischknecht et
 NH3 air unspecif ied Ammonia 1.09E-01 k g 1 2.05 (1,5,2,1,1,3,23)
al. 1996
Frischknecht et
 As air unspecif ied Arsenic 3.71E-09 k g 1 1.57 (1,5,2,1,1,3,31)
al. 1996
Frischknecht et
 Cd air unspecif ied Cadmium 2.56E-08 k g 1 1.57 (1,5,2,1,1,3,31)
al. 1996
Frischknecht et
 Pb air unspecif ied Lead 6.63E-08 k g 1 1.57 (1,5,2,1,1,3,31)
al. 1996
Frischknecht et
 Zn air unspecif ied Zinc 1.31E-05 k g 1 1.57 (1,5,2,1,1,3,31)
al. 1996
Frischknecht et
 Fe air unspecif ied Iron 2.73E-05 k g 1 1.57 (1,5,2,1,1,3,31)
al. 1996
10% of rolling mill f or steel f orming per kg output
Inf rastructure  metals processing Y es RER rolling mill 2.54E-12 unit estimation 1 3.07 (4,4,1,3,1,4,9)
as rough approximation
 Heat w aste air unspecif ied Heat, w aste 1.23E+00 M J IPPC 2001 1 1.22 (1,5,1,1,1,3,13)
Zinc coating of
 metals processing No RER zinc coating, coils 1.00E+00 m 2
coil

5.2 Zinc coating of pieces

The production process is described in (IPPC (2001a)) and on the internet 12 .
In a sense, the general or “batch” process is the same as the continuous process in that the objective is
to apply a continuous coating of corrosion-resistant zinc onto the surface of steel. However, the prac-
tices to achieve this end result are very different.
As stated above, the second galvanising process involves the application of zinc onto a “fabricated”
shape. That means, the steel is shaped into the final product, and then dipped into molten zinc to apply
a zinc coating. The items to be coated in batch galvanising plants are steel fabrications, such as nails,
screws and other very small items; lattice grates, construction parts, structural components, light poles
and many more. Tubes are sometimes also galvanised in conventional batch coating plants. These
items are coated either one at a time or, in the case of small parts, as a number of parts contained in a
“basket”. Hence, the terms “batch” or “after fabrication” are applied to this process.
Galvanising, usually comprises the following process steps:
- Degreasing – physically cleaning and degreasing steel parts,
- Pickling – removing scale, rust, and oxide from steel parts,
- Fluxing – applying a flux coating to reduce surface oxide and allow for the growth of a zinc-iron
intermetallic layer necessary for proper galvanising,
- Galvanising – dipping parts in the kettle containing molten zinc and lead.
- Finishing – excessiv zinc removal, cooling, inspection.

A galvanising plant essentially consists of a series of treatment or process baths. The steel is moved
between tanks and dipped into the baths by overhead cranes. A more detailed description of every sin-
gle step is given below.
The infrastructure of batch hot dip zinc coating, comprises buildings where galvanising processes take
place and for storage, surroundings and machinery fore the single process steps. The mean lifetime for
the infrastructure is 50 years. As no information is available, the infrastructure is approximated by the
infrastructure for rolling.
In 1997 the tonnage of galvanised steel was about 5 million. The largest share was produced in Ger-
many with 1.4 million tonnes and 185 galvanising plants (in 1997). Second largest producer was Italy
with 0.8 million tonnes (74 plants), followed by UK and Ireland with 0.7 million tonnes (88 plants)
and France 0.7 million tonnes (69 plants). The zinc consumption of the EU galvanising industry (ex-
cluding Greece and Luxembourg) in 1997 was 381188 t. Galvanised steel is used in construction,
transport, agriculture, power transmission and other applications where good corrosion protection and
long life are essential.
5.2.1 Degreasing of steel

During the degreasing step organic contaminants such as coolants and lubricants are removed from the
fabricated steel in a hot alkali solution. These surface contaminants need to be removed prior to pick-
ling so that the surface can be “wetted” by the pickling solution. The steel is dipped in baths contain-
ing surfactants, which remove oil and grease from the metal surface by emulsifying. The resulting un-
stable emulsions float on the surface of the bath and can be removed by gravitational separators,
skimmers, micro- or ultrafiltration etc.
12
http://www.oit.doe.gov/nice3/factsheets/ferrotech.pdf, accessed 03.03.03

Concentration, bath temperature and immersion time of the work-pieces determine the efficiency of
the degreasing baths. The normal temperature range for indirectly heated degreasing baths is 30 –
70 °C, although in some cases hot degreasing is applied at a temperature of about 85 °C. The baths
consists of a sodium-hydroxide solution (1 - 10 %) plus other alkaline reagents, such as soda, sodium
silicate, condensed alkaline phosphates and borax, and specific surfactants, emulsifying agents and
dispersion agents. After degreasing, rinsing is necessary to prevent carry-over of degreasing agents
which would shorten the life of pickling baths and reduce the re-usability of the bath.
An alternative method is acidic degreasing; but it is not considered in this study.
Description is taken from (IPPC (2001a)) and the internet 13 .
5.2.2 Pickling of steel

Pickling involves the immersion of parts into an acid solution (typically either heated sulfuric acid or
ambient temperature hydrochloric acid) to remove surface scale or rust (both oxides of iron), casting
skin, roll skin or grid. Therefore a galvanising plant usually comprises a series of pickling baths with
different acid concentrations ranging from 2 - 16 %, normally 12 – 16 % when freshly prepared. To
prevent excessive pickling of steel items, especially in pickling high tensile steels, and to protect the
steel pickling vats, pickling inhibitors (e.g. hexamethylenetetramine) are added to the bath.
During operation the iron content of the pickling bath increases, while the amount of free acid de-
creases, making it necessary to top up the bath occasionally by adding fresh acid. Iron-(II)-chloride
has a limited solubility in HCl. When this maximum is reached, pickling becomes impossible, but usu-
ally the pickling bath has to be replaced even sooner at lower FeCl2 concentrations. Discarding con-
centrations of 170 g FeCl2/l (=75 g Fe2+ /l) and 100 to 120 g Fe/l have been reported. Raising the tem-
perature of the pickling bath can enable its use at relatively high concentrations of FeCl2, e.g. 175 to
200 g/l with the bath at 35 °C, with some increase in emissions.
Pickling in plants with open pre-treatment is usually done at ambient air temperature; plants with an
enclosed pre-treatment sometimes operate with higher acid temperatures. Gaseous emissions of the
pickling agent can arise from the pickling baths, depending on concentration and temperature of the
bath, and from the pickled items. Hydrogen vesicles formed during the pickling process can also con-
tain acid droplets.
Sometimes it is necessary to clean the suspension devices of zinc coatings, to remove faulty coatings
from steel fabrications or to de-zinc fabrications whose coatings have to be renewed. This is com-
monly done by dipping in diluted pickle acid. This process step is called stripping. When pickling and
stripping are carried out in the same treatment vat, pickle liquors are created which contain iron- and
zinc chloride.
Rinsing is a very important step in the galvanising process as it prolongs the life of subsequent treat-
ment baths, reduces the generation of waste and increases the re-usability of by-products. After de-
greasing and pickling the fabricated steel is therefore rinsed/dipped in water baths, which are some-
times heated. Water from rinsing can be used to prepare fresh pickling or degreasing baths as a way of
recycling water and minimising aqueous blowdown. Description is taken from (IPPC (2001a)) and the
internet 14
13
14

5.2.3 Fluxing of steel

The purpose of fluxing is to enable liquid zinc to wet the surface of the steel, a necessary prerequisite
for the galvanising reaction, and with ammonium chloride-containing fluxes, to provide additional
pickling (cleaning of the surface) during hot dipping. At temperatures of above 200 °C the ammonium
chloride in the flux decomposes into NH3 and HCl, which results in an additional pickling effect.
Fluxing can be either “dry” or “wet".
In dry fluxing, the steel is immersed in a fluxing bath, usually an aqueous solution of zinc chloride and
ammonium chloride, typically maintained at 40 – 80 °C. Cold fluxing is possible but reduces the po-
tential for air-drying of the work after removal from the flux bath.
Typical bath characteristics are: ZnCl2 150 - 300 g/l
NH4Cl 150 - 300 g/l
Density: 1.15 – 1.30 g/ml
Dissolved iron: < 2 g/l
pH of flux baths may be in the range of 1 to 5.0 (normally adjusted to approximately 4.5).
Ammonium chloride in a typical good flux often accounts for 40 – 60 % of the total flux salt. It pro-
vides quick drying and better removal of iron oxides from the surfaces of the items, but also causes
more fume, ash and dross formation during the coating process. Zinc chloride prevents oxidation of
the work-piece surfaces. This is particularly important when the drying time is long.
The iron content of the flux bath is extremely important for process control, economy and environ-
ment. To reduce the environmental impact of ammonium chloride during dipping, some galvanisers
have changed to 'smoke-reduced' fluxes in which ammonium chloride has partly or completely been
substituted by potassium chloride.
After withdrawing the work-pieces from the flux bath, some of the water from the adhering fluxing
fluid evaporates. Further drying is sometimes achieved in dedicated drying. Drying of the work helps
reduce splashing and ejection of metal from the zinc bath as the work is dipped, a benefit which is in-
creased if the work retains heat after leaving the dryer, i.e. if preheat is applied.
A small number of galvanising works, especially those with exceptionally demanding fluxing re-
quirements for intricate parts, operate an alternative process, called wet fluxing. In this process the
fluxing agents flow as a layer of molten salt on the surface of the galvanising bath. Steel parts to be
galvanised are passed through the flux layer into the zinc bath. Then the molten salt layer is drawn
back from the surface by means of a rake to allow the steel parts to be withdrawn from the galvanising
bath without further contact with the flux. Description is taken from (IPPC (2001a)) and the internet 15
5.2.4 Galvanising of steel

The fluxed steel parts are slowly lowered into a bath of molten zinc. The steel reacts with the zinc to
form a coating consisting of a series of zinc-iron alloy layers topped by a layer of pure zinc when the
parts are withdrawn from the bath. The batch process requires that the part be immersed for much
longer times than for the continuous process, typically measured in minutes (up to 30 minutes for the
heaviest structural parts), not seconds. There are two reasons for needing longer immersion times. One
is to allow the part to reach the bath temperature, because immersion of a relatively cold thick-walled
large pipe, for example, means that as the steel is first immersed, it freezes a skin of zinc onto its sur-
face. For the coating to bond metallurgically to the steel, the pipe has to be heated to “remelt” the zinc.
Then, additional time is required to develop the iron/zinc alloy bond zone. For most steels, a longer
15

immersion time provides a thicker alloy layer (in Frischknecht et al. (1996), values of 65 m for both-
sided and of 130 m for one-sided coating are given), a thicker alloy bond layer is often desired to
provide a longer life to the final product, that is, a longer life before the onset of rust. Like the zinc it-
self, the alloy layer is galvanically protective to the steel part; thus, a thicker alloy layer means longer
life. Yes, the alloy layer is hard and brittle, but since the part is already fabricated, there is no need for
additional forming.
The molten zinc has a temperature of 440 – 475 °C. Kettle dimensions vary greatly; typical dimen-
sions are 7 m long by 1.4 m wide by 2.6 m deep. The kettle is normally heated externally, commonly
by gas or oil-fired burners.
The zinc bath usually also contains very small amounts of other metals, which either are impurities of
the zinc input or are added as alloying elements. A typical bath composition is: Zinc 98.9 wt-%, lead
1.0 wt-%, iron 0.03 wt-%, aluminium 0.002 wt-%, cadmium 0.02 wt-%, traces of other metals (e.g. tin,
copper). Aluminium and lead are added because of their influence on the thickness and the appearance
of the coating (lead causes the formation of the typical, attractive large spangled surface, which
through the years was “the way to identify galvanised coatings”). Lead helps to wet the steel before
galvanising and the zinc to flow from the surface after galvanising. In some instances today, bismuth
is being substituted for lead to achieve free drainage of the excess zinc. Lead can also be used to pro-
tect the kettle. The thickness of the kettle's steel walls is measured on a regular basis to prevent break-
ing. A small number of plants carry out 'high temperature galvanising' using refractory lined vessels,
which allow the zinc bath to be operated at higher temperatures, usually about 530 °C. Description is
taken from (IPPC (2001a)) and the internet 16
5.2.5 Finishing
The steel parts are withdrawn from the zinc bath; excessive zinc is removed by wiping or, in some
cases, by rattling. They are then cooled and inspected. Small surface imperfections are repaired and
the parts are removed from the jigs and made ready for dispatch. After hot dip coating some steel
products are quenched in water to give them special properties. As protection against white rust, the
products may be covered with oil emulsions or may be chromated. Other details of variations of this
process step for pipes and small components are also reported in (IPPC (2001a)).
5.2.6 Environmental issues of the batch process for zinc coating of steel
The main environmental issues for batch galvanising are emissions to air (HCl from pickling, and dust
and gaseous compounds from the kettle); spent process solutions (degreasing solutions, pickling baths
and flux baths), oily wastes (e.g. from cleaning of degreasing baths) and zinc-containing residues (fil-
ter dust, zinc ash, hard zinc), (IPPC (2001a)).
As wastewater is treated in the factory, post treatment wastewater composition and ancillary products
for the water treatment as given in IPPC (2001a) are used to calculate the expenditures and the water
emissions of the water treatment process.
5.2.7 Hot dip zinc coating (batch process) in ecoinvent

In (IPPC (2001a)) most values are given in ranges, the mean values are generally calculated as the
geometric mean of the given range. When the lower value of the range is zero, then the arithmetic
mean of the range is taken (what correspond to half of the maximal value). Values taken from
16

(Frischknecht et al. (1996)) are already mean values and do not have any information about minima
and maxima or distribution.
The background information to the process "zinc coating, pieces" is given in Tab. 5.2, flows are shown
in Fig. 5.3
Data quality
For values represented by the geometric mean, uncertainty is calculated assuming a lognormal distri-
bution and the high value given in (IPPC (2001a)) representing the 95% geometric standard deviation.
For the arithmetic means of ranges given in (IPPC (2001a)) and the values from (Frischknecht et al.
(1996)), the uncertainty is estimated by completing the pedigree matrix, according to (Frischknecht et
al. (2003)).
Tab. 5.2 Ecoinvent meta information for "zinc coating, pieces"
Name zinc coating, pieces

Location RER
Unit m2
Includes the process steps degreasing, pickling, fluxing, galvanising (melt zinc
Included Processes
coating) and finishing. Does not include the fabricates being coated.
Amount 1
Local Name Stückverzinkung
Synonyms hot dip coating//galvanising
This module is applicable, if batch zinc coating of semi-finished steel products,
that's shape doesn't change anymore, has to be balanced. Data given per tonne
of coated product had to be transformed dividing it by the mean surface area of
60 m2/t. For the chemicals and the zinc, 100 km of 32t lorry transport were bal-
General Comment to refer- anced. The softened industrial water is assumed to be produced at the coating
ence function plant. The coating layer is 65um thick, if the surface is coated on both sides and
130 m thick, if coating is done on one side only. Due to the fact, that data are
given per square meter and not per kg, the energy inventoried for heavy pieces is
to low and for light pieces to high. If the data would have been given per kg, the
error would have been the contrary.
Start Date 1996
End Date 2002
Production Volume 5 million tonnes in 1997

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
sodium hydroxide,
Degreasing solution of 10 % needed, --> 1/5 of 50% solution
chemicals inorganics No RER 50% in H2O, 6.33E-03 k g IPPC 2001 1 2
agent is inventoried
zinc coating, pieces, Degreasing
Water, unspecif ied
Water_  resource in w ater 1.67E-04 m 3 IPPC 2001 1 2
natural origin
Assumption: each energy carrier is used f rom heating heat, light fuel oil, at
Oil_  oil No RER 3.35E-01 M J IPPC 2001 1 2 calculated f rom
the same number of plants systems industrial f urnace 1MW
max value
natural gas, burned in assuming
heating
Natural gas  natural gas No RER industrial f urnace 3.35E-01 M J IPPC 2001 1 2 lognormal
systems
>100kW distribution
electricity, medium
production
Electricity  electricity No UCTE voltage, production 9.29E-02 k Wh IPPC 2001 1 2
mix
UCTE, at grid
heat, unspecific, in
w aste heat  chemicals organics No RER 3.35E-01 M J IPPC 2001 1 2
chemical plant
Liquor and Including discarded degreasing bath. Recycled --
 4.50E-02 k g IPPC 2001 1 2
sludge > not inventoried
hazardous
 Oily sludge_ disposed w aste No CH 2.67E-03 k g IPPC 2001 1 1.56 (1,5,1,1,1,3,41)
incineration
incineration
Hydrochloric hydrochloric acid, 30%
 chemicals inorganics No RER 3.20E-01 k g IPPC 2001 1 2.09
acid (30%) in H2O, at plant
chemicals organic, at
zinc coating, pieces, Pickling
Inhibitor  hexamethylenetetramine chemicals organics No GLO 1.67E-03 k g IPPC 2001 1 2

plant
Water, unspecif ied
Water_  resource in w ater 2.92E-04 m 3 IPPC 2001 1 2
natural origin
electricity, medium
production calculated f rom
Electricity  electricity No UCTE voltage, production 1.04E-01 k Wh IPPC 2001 1 4
mix max value
UCTE, at grid
assuming
Assumption: same amount of w aste gas per kg
Hydrogen lognormal
 of product as f or HCl-pickling in cold sheet rolling air unspecif ied Hydrogen chloride 1.18E-06 k g IPPC 2001 1 7.07
chloride distribution
of steel
Assumption: same amount of w aste gas per kg
of product as f or HCl-pickling in cold sheet rolling Particulates, > 2.5 um,
 Dust air unspecif ied 5.00E-05 k g IPPC 2001 1 2
of steel. Assumption: particle f rom chemical and < 10um
process are betw een 2.5 and 10 um
Waste acid and
 assumed as recycled --> not inventoried 3.33E-01 l IPPC 2001 1 2
sludge
Discarded Ref erence contained no data about composition
 n.a. IPPC 2001
pickle liquor and amount of discarded pickling liquor.
Fig. 5.3 Flows for "zinc coating, pieces" and their representation in the ecoinvent database

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
A ssumption: same amount of w aste gas per kg
Hydrogen
 of product as f or HCl-pickling in cold sheet rolling air unspecif ied Hydrogen chloride 1.18E-06 k g IPPC 2001 1 7.07
pieces, Pickling
chloride
zinc coating, of steel
A ssumption: same amount of w aste gas per kg
calculated from
max value
of product as f or HCl-pickling in cold sheet rolling Particulates, > 2.5 um, assuming
 Dust air unspecif ied 5.00E-05 k g IPPC 2001 1 2
of steel. Assumption: particle f rom chemical and < 10um lognormal
process are betw een 2.5 and 10 um distribution
Waste acid and
 assumed as recycled --> not inventoried 3.33E-01 l IPPC 2001 1 2
sludge
Discarded Ref erence contained no data about composition
 n.a. IPPC 2001
pickle liquor and amount of discarded pickling liquor.
Hydrochloric hydrochloric acid, 30%
 chemicals inorganics No RER 5.00E-02 k g IPPC 2001 1 2
coating,
acid (30%) in H2O, at plant

pieces,
zinc
Water, unspecified
Water  resource in w ater 5.83E-05 m 3 IPPC 2001 1 2
natural origin calculated from
Spent strip max value
 assumed as recycled --> not inventoried 2.24E-02 k g IPPC 2001 1 1.12
liquor assuming
lognormal
Flux agent chemicals inorganic, at
 Inorganic chemicals as proxy chemicals inorganics No GLO 2.50E-02 k g IPPC 2001 1 2 distribution
zinc coating, pieces,
(ZnCl2 x NH4Cl) plant
Water, unspecified
Water  resource in w ater 1.67E-04 m 3 IPPC 2001 1 2
natural origin
Fluxing
Energy f or heating up the flux bath is in some

Energy  n.a. IPPC 2001
case supplied by internally recovered heat
calculated from
max value
 Spent f lux assumed as recycled --> not inventoried 4.08E-02 k g IPPC 2001 1 2.45 assuming
lognormal
distribution
Iron hydroxyde
 n.a. IPPC 2001
sludge
Fig. 5.3 Flows for "zinc coating, pieces" and their representation in the ecoinvent database (cont.)

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Water, unspecified
Water  Inventoried f or the overall process resource in w ater 1.67E-04 m 3 IPPC 2001 1 2
natural origin
V alues f rom the tw o references are almost the zinc for coating, at
Zinc  metals extraction No RER 1.05E+00 k g IPPC 2001 1 3.16
same regional storage
Recovered zinc  recycling --> not inventoried 1.25E-01 k g IPPC 2001 1 2
A ssumption: 49% of the plants is using f uel as heating heat, light f uel oil, at
Oil  oil No RER 1.25E+01 M J IPPC 2001 1 2.36
energy carrier for heating up the zinc bath systems industrial furnace 1MW
calculated from
A ssumption: 49% of the plants is using natural natural gas, burned in
heating max value
Natural gas  gas as energy carrier f or heating up the zinc natural gas No RER industrial furnace 1.25E+01 M J IPPC 2001 1 2.36
systems assuming
bath >100kW
lognormal
electricity, medium
A ssumption: 2% of the plants is using electricity production distribution
Electricity  electricity No UCTE voltage, production 1.41E-01 k Wh IPPC 2001 1 2.36
zinc coating, pieces, Rinsing
as energy carrier f or heating up the zinc bath mix

UCTE, at grid
Extracted gas
 7.07E+01 m 3 IPPC 2001 1 2.83
f rom kettle
 Dust>10 air unspecif ied Particulates, > 10 um 2.58E-03 k g IPPC 2001 1 3.87
 Zinc air unspecif ied Zinc 4.47E-04 k g IPPC 2001 1 3.16
Hydrogen
 air unspecif ied Hydrogen chloride 1.00E-04 k g IPPC 2001 1 1.41
chloride
 Lead ne gligible IPPC 2001
Floating on the surf ace of the galvanizing bath,

ash is skimmed off before removing the dipped
w ork-pieces. Zinc content is 40 to 90 % and
 Ash 1.67E-01 k g IPPC 2001 1 2.5
makes it valuable for recycling either directly in
the plant by secondary zinc industry --> not
inventoried
Returned to the secondary zinc industry f or
calculated from
recovery of the zinc content or to the zinc
 Dross 2.04E-01 k g IPPC 2001 1 2.45 max value
chemicals industry f or the manuf acture of zinc
assuming
oxide --> not inventoried
lognormal
Combustion
distribution
 gas: (NOx, Inventoried in the fuel combustion modules 2.12E+01 m 3 IPPC 2001 1 2.55
CO/CO2, SO2)
disposal, dust,
residual
w aste unalloyed EA F steel,
 Filter dust Dust f rom low -alloied EAF-steel as proxy material No CH 2.45E-01 k g IPPC 2001 1 2.45
management 15.4% w ater, to
landfill
residual material landf ill

Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
Ref erence contained no data. A ssumption: Same
Flocculating aluminium sulphate,
 amount per kg product as f or coil coating. chemicals inorganics No RER 1.51E-05 k g IPPC 2001 1 1.22 (2,5,1,1,1,3,4)
agents pow der, at plant
A luminium sulphate as proxy.
w ater leaves the system as treated ef fluent; it is
zinc coating, pieces, Waste water treatment plant
assumed to be discharged in surface w ater -->
not inventoried. Infrastructure f or w aste w ater
treatment plant is comprised in the infrastructure
 Waste w ater 1.67E-03 m 3 IPPC 2001 1 1.22 (2,5,1,1,1,3,4)
module. V alue f or energy use is comprised in the
range of uncertainty of inventoried overall
energy. A ssumption: ratio betw een treated and
untreated w ater is the same as f or coil coating
process.
A ssumption: same amount per m3 treated w ater, Suspended solids,
 TSS w ater unspecif ied 3.74E-06 k g IPPC 2001 1 1.57 (2,5,1,1,1,3,32)
as f or coil coating process unspecif ied
 Iron w ater unspecif ied Iron, ion 4.10E-07 kg IPPC 2001 1 5.06 (2,5,1,1,1,3,35)
 Nickel w ater unspecif ied Nickel, ion 1.67E-08 kg IPPC 2001 1 5.06 (2,5,1,1,1,3,35)
 Zinc w ater unspecif ied Zinc, ion 2.63E-07 kg IPPC 2001 1 5.06 (2,5,1,1,1,3,35)
 Lead w ater unspecif ied Lead 2.51E-08 kg IPPC 2001 1 5.06 (2,5,1,1,1,3,35)
 Chromium w ater unspecif ied Chromium, ion 1.10E-07 kg IPPC 2001 1 5.06 (2,5,1,1,1,3,35)
 Cr6+ w ater unspecif ied Chromium V I 1.67E-08 kg IPPC 2001 1 5.06 (2,5,1,1,1,3,35)
 COD w ater unspecif ied 2.20E-04 k g IPPC 2001 1 1.57 (2,5,1,1,1,3,32)
Demand
BOD5, Biological
 BOD w ater unspecif ied 2.20E-04 k g IPPC 2001 1 2.5 (5,5,5,5,5,5,32)
Oxygen Demand
DOC, Dissolved
 DOC w ater unspecif ied 4.95E-06 k g IPPC 2001 1 2.45 (5,5,5,5,5,5,32)
Organic Carbon
TOC, Total Organic
 TOC w ater unspecif ied 4.95E-06 k g IPPC 2001 1 1.57 (2,5,1,1,1,3,32)
Carbon
Hydrocarbons,
 Hydrocarbons w ater unspecif ied 1.98E-06 k g IPPC 2001 1 3.05 (2,5,1,1,1,3,34)
unspecif ied
 Oil and f at w ater unspecif ied Oils, unspecif ied 1.73E-06 k g IPPC 2001 1 1.57 (2,5,1,1,1,3,32)
 Phosphorus w ater unspecif ied Phosphorus 1.50E-06 k g IPPC 2001 1 1.57 (2,5,1,1,1,3,33)
Water Waste w ater treatment sludge f rom steel f orming disposal, sludge f rom
residual
treatment processes is inventoried as proxy. Assumption: w aste steel rolling, 20%
 material No CH 1.51E-03 k g IPPC 2001 1 1.22 (2,5,1,1,1,3,6)
sludge (dry same amount per m3 treated w ater, as f or coil management w ater, to residual
landfill
material) coating process. material landf ill

Source for
ecoinvent
Comment
Deviation
Standard
Location
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
transport Frischknecht et (4,5,n.a.,n.a.,n.a
Lorry 32t  road No RER transport, lorry 32t 1.34E-01 tk m 1 2.09
systems al. 1996 .,n.a.,5)
Frischknecht et
Soda  zinc coating, pieces, chemicals inorganics No RER soda, pow der, at plant 1.67E-03 k g
al. 1996
1 2.05 (1,5,2,1,1,3,5)
A mmonia 
Overall process chemicals inorganics No RER
ammonia, liquid, at
5.10E-04 k g
Frischknecht et
1 1.22 (1,5,2,1,1,3,4)
regional storehouse al. 1996
Water w ater, deionised, at Frischknecht et
 w ater supply production No CH 3.33E+00 k g 1 1.23 (2,5,2,1,1,3,12)
(sof tened) plant al. 1996
Process and cooling w ater evaporated
(uncontamined --> not inventoried) or already
Frischknecht et
 Water included in the w aste disposal. Assumption: ratio 2.51E-03 m 3 1 1.23 (2,5,2,1,1,3,12)
al. 1996
betw een treated and untreated w ater is the
same as f or coil coating process.
10% of rolling mill f or steel f orming per kg output
Inf rastructure  metals processing Y es RER rolling mill 2.71E-12 unit estimation 1 3.15 (4,5,1,3,3,4,9)
as rough approximation
 Heat w aste air unspecif ied Heat, w aste 1.26E-01 M J IPPC 2001 1 1.22 (1,5,1,1,1,3,13)
Particle size f rom mechanical process: Frischknecht et
 Dust>10 air unspecif ied Particulates, > 10 um 2.58E-05 k g 1 1.57 (1,5,2,1,1,3,25)
assumption 50 % >10 um. al. 1996
Particulates, > 2.5 um, Frischknecht et
 Dust 2.5-10 2.5-10 um. In addition 50 % of the mechanical air unspecif ied 7.73E-05 k g 1 2.05 (1,5,2,1,1,3,26)
and < 10um al. 1996
zinc coating, pieces, Overall process
abrasion
Frischknecht et
 HCl air unspecif ied Hydrogen chloride 2.04E-04 k g 1 2.05 (1,5,2,1,1,3,23)
al. 1996
Frischknecht et
 NH3 air unspecif ied Ammonia 7.27E-04 k g 1 2.05 (1,5,2,1,1,3,23)
al. 1996
Frischknecht et
al. 1996
Frischknecht et
al. 1996
Frischknecht et
al. 1996
Frischknecht et
al. 1996
Frischknecht et
 Fe air unspecif ied Iron 2.33E-05 k g 1 1.57 (1,5,2,1,1,3,31)
al. 1996
Frischknecht et
 Chloride w ater unspecif ied Chloride 3.13E-05 k g 1 1.57 (1,5,2,1,1,3,31)
al. 1996
Frischknecht et
 Ion Zinc w ater unspecif ied Zinc, ion 3.13E-05 k g 1 1.57 (1,5,2,1,1,3,31)
al. 1996
Frischknecht et
 Ion Iron w ater unspecif ied Iron, ion 9.38E-05 k g 1 1.57 (1,5,2,1,1,3,31)
al. 1996
coating of
 metals processing No RER zinc coating, pieces 1.00E+00 m 2
pieces

5.3 Zinc coating of pieces, correction of layer thickness per m

If zinc layer thickness other than the double-sided 65 m or the one-sided 130 m are requested, the
amount of zinc needed and the arising heavy metals emissions (evaporating from zinc) have to be
adapted proportionally to the m-change of layer thickness. The remaining flows depend basically on
the coated piece surface area and have therefore not to be changed.
5.3.1 Zinc coating of pieces, correction of layer thickness per m in

ecoinvent
Per m of additional or subtracted layer thickness, the module has to be inventoried one time. This
module can be applied also for changing the layer thickness when continuous galvanising of coils is
done. In this case the module has to be inventoried 1.53 times per m of changed thickness layer, be-
cause the zinc consumption per m is higher. Data are taken from (Frischknecht et al. (1996)) where
no information about minima and maxima or distribution are given.
The background information to the process "zinc coating, pieces, adjustment per um" is given in Tab.
5.3, flows are shown in Fig. 5.4.
Data quality
(2003)). A lognormal distribution is assumed.

Tab. 5.3 Ecoinvent meta information for "zinc coating, pieces, adjustment per um"
Name zinc coating, pieces, adjustment per um

Location RER
Unit m2
Data Set Version 1
Includes the process steps galvanising (melt zinc coating) and finishing. Does not
Included Processes
include the fabricates being coated.
Amount 1
Local Name Stückverzinkung, Korrektur pro um
Synonyms hot dip coating//galvanising
This module is applicable if the coating layer is thicker than the one given for "zinc
coating, pieces", or for "zinc coating, coils". Applying it for the latter, the module
has to be multiplied by the factor 1.53 for correct use. The Dataset is not applica-
ence function
ble for thinner coating layers. Data given per square meter of product surface cor-
rected by one m.
Start Date 1996
End Date 2002
Extrapolations none

Source for
ecoinvent
Comment
Deviation
Standard
Location
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
transport Frischknecht et (4,5,n.a.,n.a.,n.a
Lorry  road No RER transport, lorry 32t 1.50E-03 tk m 1 2.09
zinc coating, pieces, systems al. 1996 .,n.a.,5)
adjustment per um
zinc for coating, at Frischknecht et
Zinc  metals extraction No RER 1.54E-02 k g 1 1.22 (1,5,2,1,1,3,12)
regional storage al. 1996
Frischknecht et
al. 1996
Frischknecht et
al. 1996
Frischknecht et
al. 1996
Frischknecht et
al. 1996
zinc coating, pieces,
 Coated product metals processing No RER 1.00E+00 m 2
adjustment per um
Fig. 5.4 Flows for "zinc coating, pieces, adjustment per m" and their representation in the ecoinvent database

6. Welding
6 Welding
Author: Hans.Jörg Althaus, EMPA Dübendorf
6.1 Definition and fundamentals of welding

Welding is defined (DIN 1910, Part 1: 07.81 and ISO 857: 1990) as the bonding of several parts made
of the same or different materials to form a single unit, where, in the weld region, either the tempera-
ture is raised to the melting point, the parts are pressed together, or both. Weld filler materials are also
used as necessary. The welded joint can be separated only by destroying it. This definition applies to
the welding of both metals and thermoplastic materials (Brandt & Weiler (1997)).
Weldability (see DIN 8528, Part 1: 06.73): A component fabricated by welding is weldable if, after
fabrication, it fulfils all the functions required of it at its point of use for the duration of its expected
lifetime. It is not weldable if it does not fulfil these functions as a result of any defects arising from a
welding process. One of the most important preconditions for a good welded joint is that impurities on
the surface of the workpiece such as thick scale, rust, oil, and paint should be removed as completely
as possible. It is certainly possible to weld over a protective coating, but this can affect the integrity of
the weld (Brandt & Weiler (1997)).
The methods of welding metals can be divided into two groups:
In fusion welding of metals, the junction area is melted to form a pool of molten metal, which consists
of parent metal, either alone or with added filler metal (minus metal lost in the process). The zone sur-
rounding the molten pool is to a greater or lesser extent affected by changes to its structure and proper-
ties, depending on the degree and duration of heating and the rate of cooling. This depends on the heat
flux, which in turn is very dependent on the welding method.
In compression welding of metals, the workpieces are pressed together at the welding region, and, in
most processes, are also heated. The heat is applied by various means, e.g., electrically (resistance or
induction heating), or by flames (Brandt & Weiler (1997)).
This study concentrates on the fusion welding, differing between gas welding of steel, arc welding of
steel and arc welding of aluminium. Fig. 6.1 shows the different processes and sub-processes of the
fusion welding.

6. Welding
Fig. 6.1 Fusion welding processes (Brandt & Weiler (1997)).
6.1.1 Environmental issues of the welding process

Environmental issues in the welding process are mainly discussed in literature from a point of view of
health and safety for the welder. Incorrect operation of a welding process can lead to hazards to per-
sonnel, especially from electric current, atmospheric pollutants (smoke, vapours, and gases), radiation,
and noise. Precaution against fire and explosion is also very important. Fire can be produced directly
by flames, or can begin slowly due to unnoticed molten material impelled over long distances, often
only becoming serious after some hours. Fuel gases (e.g. acetylene, propane, etc.) and oxygen, consti-
tute the main source of danger for explosions.
The concentrations of atmospheric pollutants such as airborne particulates (fume), vapours, and gases
must not exceed the legal limits, i.e., the MAK ("Maximale Arbeitsplatzkonzentrationen"; maximal
concentration on the workplace) values. Those limits are only valuable to guarantee human health
when workers are exposed to the harming gases during the workprocess. They are not setting limits
and they don't tell anything about the environmental pollution itself, because they can be maintained
simply by diluting the concentration by mixing the polluted air with fresh air.
The airborne particulates contain lead, chromium, cadmium, nickel, mercury, copper, and zinc. They
are formed as metal vapour during the processes of welding and soldering, and usually become oxi-
dised and suspended in the air as fine dust. The harmful gases formed can include nitrogen oxides
(NO, NO2) from flames, ozone and sometimes phosgene from the intense light emitted by electric
arcs, and carbon monoxide when welding with a carbon dioxide gas shield.
6.2 Manual Gas Welding of Steel

In manual gas welding (or autogenic welding; descriptions taken from Brandt & Weiler (1997)) the
heat required is produced by burning a mixture of acetylene and oxygen. The flame impinges directly
on the welding area, and the filler metal is fed by hand.
In manual gas welding, the rate of heat flow to the workpiece is low compared with other welding
methods, and the process is comparatively slow. For this reason, and because of the high thermal con-
ductivity of metal, the structure of the neighbouring parent metal is considerably affected. In spite of
its disadvantages, ca. 10% of all welding is by the autogenic method.

6. Welding
The autogenic torch is simple, lightweight, and cheap. It is much used for repairs and for simple jobs
in the workshop. The gas supply consists simply of oxygen and acetylene bottles with tubing to con-
vey these gases to the welding area. The main uses of autogenic torches are for welding pipelines in
chemical plant construction, general industrial plants, shipbuilding, and heating plants, where its value
depends more on the manual skill of the welder than on high melting rates.
Acetylene is used exclusively as the fuel gas in gas welding, as of all the hydrocarbons it has the high-
est heat of formation per unit mass, and gives the highest combustion temperature (Fig. 6.2), combus-
tion rate, and hence heat flux and heat transfer from the flame to the workpiece. This is especially use-
ful for the rapid heating of small surfaces. To prevent scaling and decarburization of the workpiece,
the acetylene flame is adjusted to a neutral state, i.e., with no excess of either oxygen or fuel gas (vol-
ume ratio of oxygen to acetylene ca. 1.05). The rest of the oxygen required by the stoichiometry is
provided by the surrounding air.
Fig. 6.2 Temperature profile along the flame axis when burning acetylene and propane Brandt & Weiler (1997).
As autogenic welding is done manually with no or very little infrastructure- and material demand for
the machinery, those inputs are neglected in this study.
6.2.1 The manual gas welding process of steel in ecoinvent

For a lack of information on acetylene and oxygen consumption and on emissions the following as-
sumptions are made:
The amount of acetylene used is estimated assuming that the energy needed for gas welding is five
times the energy needed for automated arc welding (MAG). The heating value of acetylene is 51.2
MJ/kg (Bach et al. (1999)).
The volume ratio of oxygen / acetylene = 1.05. Thus the mol ration of oxygen / acetylene also = 1.05.
With the molar weight of O2 = 32 g/mol and of C2H2 = 26 g/mol the mass ration of oxygen / acetylene
is calculated to be 32 / (26*1.05) = 1.17.
The metal, the nitrogen oxide and the ozone emissions are taken from the MAG welding process.
The CO and CO2 emissions are calculated to reflect the formation by the acetylene burning. Because
of the neutral to slightly reducing conditions chosen for welding, the CO / CO2 ratio is rather high. It is
assumed to be 1:100 (mol/mol). Thus per kg (= 38.46 mol) of acetylene burned 3.35 kg CO2 (= 76.16
mol) and 0.021 kg CO (= 0.7616 mol) are formed.
Furthermore it is assumed that 1% of the acetylene is emitted to air by lighting the torch and changing
the bottles.

6. Welding
The transports of the steel feed and gas is inventoried according to the quality guidelines (see
Frischknecht et al. (2003))
The background information to the process "welding, gas, steel" is given in Tab. 6.1, flows are shown
in Fig. 6.3.
Data quality
Tab. 6.1 Ecoinvent meta information for "welding, gas, steel"
Name welding, gas, steel

Location RER
Unit m
Includes the acetylene welding of non alloyed steel and the transports of the filler
Included Processes
rod and the protective gas to the place of use.
Amount 1
Local Name Schweissen, Autogen, Stahl
Synonyms
General Comment to refer- Based partly on rather rough estimates and on input data of poor quality (acety-
ence function lene). Not to be used if welding is of importance in the system considered.
Start Date 1997
End Date 2002
Other Period Text
Geography text Data-set is representative for Europe
Technology text Acetylene welding with a wire consumption of 0.0536 kg/m
Extrapolations none

6. Welding
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
steel, low -alloyed, at
Steel  metals extraction No RER 5.36E-02 k g calculated 1 1.38 (4,5,1,1,1,5,4)
plant
A rgon  chemicals inorganics No RER argon, liquid, at plant 2.37E-02 k g calculated 1 1.38 (4,5,1,1,1,5,4)
carbon dioxide liquid,
Carbon dioxide  chemicals inorganics No RER 4.11E-03 k g calculated 1 1.38 (4,5,1,1,1,5,4)
at plant
Oxygen  chemicals inorganics No RER oxygen, liquid, at plant 9.15E-04 k g calculated 1 1.38 (4,5,1,1,1,5,4)
calculated f rom
electricity, medium personal
production
Electricity  electricity No UCTE voltage, production 2.75E-02 k Wh communication, 1 1.32 (3,5,1,3,1,5,2)
mix
UCTE, at grid Tenconi SA ,
Airolo
steel 600 km, gases 100 km. Gases tw ice the transport (4,5,n.A .,n.A .,n.
transport rail train No RER transport, f reight, rail 6.66E-02 tk m estimated 1 2.09
welding, arc, steel

w eight f or the bottles systems A .,n.A .,5)
steel and gases 50 km. Gases tw ice the w eight transport (4,5,n.A .,n.A .,n.
transport lorry  road No RER transport, lorry 32t 5.55E-03 tk m estimated 1 2.09
f or the bottles systems A .,n.A .,5)
construction electronics for control
Inf rastructure  pow er transistor buildings No RER 1.00E-05 k g estimated 1 3.59 (5,5,5,3,3,5,9)
processes units
 Cr (II, III) air unspecif ied Chromium 2.81E-05 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
 Cr (V I) air unspecif ied Chromium V I 3.00E-08 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
 Cu air unspecif ied Copper 2.10E-07 kg Künniger 1998 1 1.92 (1,5,5,3,3,5,31)
 Fe air unspecif ied Iron 8.74E-05 kg Künniger 1998 1 1.92 (1,5,5,3,3,5,31)
 Mn air unspecif ied Manganese 1.13E-05 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
 Mo air unspecif ied Molybdenum 2.16E-06 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
 Ni air unspecif ied Nickel 1.13E-05 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
 Zn air unspecif ied Zinc 4.20E-07 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
sum of oxides of the metal emissions and the
 partikel <2.5 air unspecif ied Particulates, < 2.5 um 1.41E-04 k g calculated 1 3.41 (4,5,5,3,3,5,20)
f luoride above
Carbon monoxide,
 CO air unspecif ied 6.67E-05 k g Künniger 1998 1 5.42 (1,5,5,3,3,5,17)
fossil
 NOx air unspecif ied Nitrogen oxides 2.00E-06 k g Künniger 1998 1 1.92 (1,5,5,3,3,5,16)
 O3 air unspecif ied Ozone 9.06E-06 k g Künniger 1998 1 1.92 (1,5,5,3,3,5,31)
 Waste Heat air unspecif ied Heat, w aste 9.90E-02 M J calculated 1 1.32 (3,5,1,3,1,5,13)
 MA G w elding metals processing No RER w elding, arc, steel 1.00E+00 m
Fig. 6.3 Flows for "welding, gas, steel" and their representation in the ecoinvent database

6. Welding
6.3 Arc welding of steel

The following description of the welding processes is taken from Brandt & Weiler (1997). A good
overview of the arc welding process in German language is given in Nies (1992).
In arc welding the heat source is the electric arc between the welding electrode and the workpiece. The
arc may be exposed, but is then nearly always surrounded by a gas shield (see below), or it may be
covered by a layer of powder (submerged arc welding). In both cases, access of air (oxygen, nitrogen,
and water vapour) is restricted.
Protective gases are either inert (almost exclusively argon), reactive (carbon dioxide), or mixed gases
with a high proportion of inert gas, e.g., argon, and 1–9 vol.% oxygen or 18 vol.% carbon dioxide.
These not only have a protective effect, but in some cases a metallurgical effect also.
Welding powders used in the submerged arc process (DIN 8557, Part 1: 04.81 and EN 756: 1992) sta-
bilise the arc and reduce the rate of heat loss and also have a protective effect. Metallurgical effects in-
clude deoxidisation, prevention or replacement of burn-off losses, and sometimes addition of alloying
elements. The powders consist mainly of oxides of silicon, manganese, calcium, barium, magnesium,
titanium, sodium, potassium, and iron, and also some fluorides, e.g., of calcium and so-
dium/aluminium. The many types of powder can be classified according to their principal components
(manganese silicate type, calcium silicate type, basic aluminate type, etc.).
The electrical energy is fed to the arc from controllable and variable sources of electric current, e.g.,
transformers, rectifiers, or converters. Arc welding processes operate at a higher heat flow rate then
gas welding, so that higher welding and melting rates can be achieved and the heat-affected zones are
less extensive.
There are several different types of arc welding processes (see also Fig. 6.1).
Manual Arc Welding

In manual arc welding with coated electrodes, the arc is formed between the melting end of the elec-
trode and the workpiece, causing the coating to melt and partially vaporise. The weld pool and the
molten droplets during transfer to the workpiece are protected from the surrounding air by the gases
produced from the coating.
The coating also performs a number of other functions:
- It facilitates metallurgical reactions that clean the weld pool
- It stabilises the arc
- It compensates for the loss of alloying elements from the cored wire and the parent metal
- It can introduce seed nuclei into the weld pool and hence grain refine the weld metal
- It can control the basicity or acidity of the melt
- It can alloy with the material of the welded joint and hence modify its chemical composition
Cored wire is usually used to achieve a composition that matches that of the alloy being welded as
closely as possible. The special requirements of manual electrodes are met with the aid of the coating
used. The coatings on manual welding electrodes have compositions similar to those of submerged arc
welding powders, and usually contain silicon, manganese, calcium, titanium, magnesium, aluminum,
potassium, and sodium, as, well as gas formers such as cellulose for certain applications.
Two important types of steel electrode are:
- The rutile (titanium dioxide) type with a high proportion of titanium dioxide in the coating, with
which simple manual welding tasks can be quickly carried out

6. Welding
- The basic lime (calcium carbonate) type with a high proportion of calcium carbonate in the coat-
ing, which can produce welds with the best quality characteristics, especially toughness
High-performance electrodes have coatings containing a high proportion of metal in powder form.
These give a higher melting rate than normal electrodes for the same electricity consumption.
Welding with Cored Wire

Cored wire is a tubular metal electrode filled with material similar to that used for the coating on nor-
mal electrodes. Its particular advantage is that it can be fed from a long coil, and can be welded con-
tinuously. This gives a faster average welding speed than with manual electrode welding. During
welding, the cored wire slides through a burner close to the welding area. Shielding gas can also be
injected. The use of cored wires containing gas-producing substances can enable welding to be carried
out without a gas shield under difficult conditions, e.g., in wind.
Tungsten Inert Gas (TIG) Welding

In these processes, the arc is formed between the tip of an infusible tungsten electrode and the work-
piece. The function of the electrode is only to conduct electricity and produce the arc. The filler metal
is fed to the weld pool from one side, either by hand in rod form or by an automatic system as wire.
Argon is usually used as the protective gas, occasionally helium or argon–helium mixture. The tung-
sten electrode is alloyed with up to 4% thorium dioxide to reduce the work of electron emission at the
tip of the electrode and to increase the maximum current.
Welds of the highest quality can be produced by TIG welding. It is best to use direct current with the
workpiece as the positive pole when welding high-alloy (chromium–nickel) steels, nickel-based al-
loys, titanium, zirconium, tantalum, and their alloys. Alternating current is used when welding metals
that form high-melting oxides, e.g., aluminium and its alloys, as continuous reversal of the current di-
rection continuously removes the oxide film from the weld pool.
Pulsed current arc welding is a variation of the TIG process. This is very useful for welding thin sheet
metal, as it minimises the extent of the heat-affected zone and hence the distortion of the sheet. The
weld pool is so small that dripping of molten metal does not occur, even when welding in difficult
situations. For high-quality TIG welding, transistorised pulsed current welding robots are available
that can be very precisely controlled by a variety of computer programs.
Plasma Welding
With the plasma arc, the gas rate is usually much higher than with the TIG arc. Also, the temperature
of the arc is increased by a factor of 2–4 and the energy density by a factor of ca. 10. This is achieved
by greatly constricting a TIG arc together with the protective gas stream surrounding it with the aid of
a water-cooled copper nozzle, forcing the entire gas stream to pass through the arc.
In plasma welding, the plasma jet is always surrounded by a protective gas layer. As the plasma jet
usually penetrates into the sheet, the result is a key-lock effect similar to that produced by autogenic
rightward welding. There is an upper limit to the weldable sheet thickness, but the welding speed is
significantly higher than in other arc welding processes. The amount of heat applied per unit length of
weld is small.
A very wide range of power outputs can be used in plasma welding, and the process can therefore be
used for sheet thicknesses in the approximate range 0.01–10 mm. So-called microplasma welding is a
low-power process using an "untransferred" arc, i.e., between the tungsten electrode and the nozzle
wall. In the medium to higher power range, the plasma welding arc is "transferred," i.e., it is between
the tungsten electrode and the workpiece. The usual plasma gases are argon and hydrogen.

6. Welding
Plasma welding is also used for resurfacing metal, which can be plated over large areas to a thickness
of ca. 2–8 mm. It is mainly used in the construction of chemical equipment. There are two variations:
plasma welding of powder coatings and plasma welding of wire.
In plasma welding of powder coatings, the powder is fed from a holding vessel to the plasma stream,
where it melts. In the hot wire process, two wires are fed in parallel to the welding zone. Electric cur-
rent from an auxiliary source is passed through the wires, heating them (because of their ohmic resis-
tance) to a temperature just below their melting point close to the welding zone. The plasma burner
oscillates and can therefore apply metal to a wide fusion zone.
Metal Inert Gas (MIG) and Metal Active Gas (MAG) Welding
In MIG and MAG welding, high-power processes, the arc is produced between a wire electrode, fed
from a coil, and the workpiece. This method of supplying the electric current gives a high electrode
current density (ca. 100 A/mm2) and hence a high melting rate.
The wire electrode is fed to the electric arc by a motor whose speed can be varied according to the
wire diameter and electric current. The current to the welding wire is fed through a tubular contact in
the "welding pistol" close to the arc. Only direct current is used in the MIG and MAG process, the
electrode usually being the positive pole. The metal is then transferred as small droplets and penetra-
tion is deep. If the electrode is negative, larger drops of metal are transferred and there is less penetra-
tion.
The difference between MIG and MAG lies in the gas used:
 The metal inert gas (MIG) process uses an inactive, i.e., noble gas shield
 In the metal active gas (MAG) process the protective gas is active, e.g., carbon dioxide or mixtures
or argon with carbon dioxide and/or oxygen is used
The MIG process, with argon (or sometimes helium) as the protective gas, can be used to weld work-
pieces of thickness >3 mm made of non-ferrous metals and alloy steels. With pulsed direct current of
variable magnitude and pulse frequency, the transfer of droplets can be controlled. In this pulsed arc or
pulsed MIG process, each pulse of current causes a droplet of metal that has been melted in the back-
ground current phase to be thrown off. The droplet size and frequency can be influenced by the type of
pulse used. The pulsed MIG process is very suitable for welding in difficult situations.
The MAG process is mainly used for welding plain carbon and low-alloy steels. As the protective gas
has an oxidising effect, the filler metal must contain a high proportion of deoxidising components
(e.g., manganese, silicon, aluminium). The metallurgical reaction that takes place gives increased nu-
cleation and hence grain-refined weld metal.
Sources of Electric Current

Power transistors are increasingly used for controlling the electric current. They can be used as digital
switches or analog regulators. As their power losses are higher, analog regulators are used mainly for
lower-power equipment. As switches (high frequency), they can be in the primary or secondary cir-
cuit, i.e., the transistors can be between the mains supply and the transformer or the transformer can be
between the mains and the transistors. With the first arrangement the equipment is lightweight, but
very sensitive to voltage peaks. Equipment of the second type is heavier.
Submerged Arc Welding

In submerged arc welding, the arc produced between the automatically fed filler wire and the work-
piece is covered by powder, and is located in a cavity formed within powder that has become molten.
The powder flows in the direction of welding, preceding the welding zone of the metal sheet. Only a

6. Welding
small amount of powder is consumed by the welding process, and the unused powder is sucked away
and reused. The powder blanket essentially performs the same function as the coating on manual elec-
trodes. The welding speed is high, and in equipment with several filler wires the melting capacity is
also high. The process can be used only in situations where the sheet metal is not steeply sloped.
Narrow Gap Welding

Heavy section workpieces can be joined economically and with minimum effect on the parent metal
by the so-called narrow gap welding process. The best of the various methods for filling a vertical nar-
row gap with weld metal is the MIG/MAG process. In this gas-shielded process, the filler wire and
protective gas are cooled in a sword-shaped sheathed tube and fed to the welding point in the narrow
gap. There are two systems:
- Two wires, which have been subjected to plastic deformation to give a predetermined internal
stress, are guided into the two corners of the welding gap, giving a defect-free root position
- In the second system, a wire preformed to a waved shape so that it alternately fills each corner of
the region to be welded inside the narrow gap performs this function.
Narrow gap welding can also be carried out by the submerged technique. While gas-shielded narrow
gap welding can be used for gap widths of ca. 10 mm, the gap must be wider for submerged welding.
6.3.1 Automated arc welding of steel in ecoinvent

The module provides values per meter weld seam by MAG-welding with a mixture of 83% Ar, 13%
CO2 and 4% O2 as protective gas. The gas demand is 16 l/min. The low alloyed steel wire is 1.2 mm in
diameter and 6 m/min is used. The welding speed is 1 m/min 17 . Thus the consumption of wire per me-
ter of weld seam is 0.0536 kg.
The power consumption is calculated based on the value given in Künniger & Richter (1997) and the
assumption that the energy input corresponds linearly to the amount of wire used. The emissions (ex-
cept for the particulate emission that is calculated as the sum of all the metal emission) are given in
Künniger & Richter (1998).
The transports of the steel feed and gas is inventoried according to the quality guidelines (see
Frischknecht et al. (2003))
For the source of power (infrastructure) the unspecific electronic device is used as proxy. A value of
1E-5 kg/m is assumed (this value corresponds to the life time production of 10 km of weld seam per
kg power transistor).
The background information to the process "welding, arc, steel" is given in Tab. 6.2, flows are shown
in Fig. 6.4.
Data quality
17
personal communication, G. Willhelm, EMPA Dübendorf, 08.05.03

6. Welding
Tab. 6.2 Ecoinvent meta information for "welding, arc, steel"
Name welding, arc, steel

Location RER
Unit m
Includes the MAG-welding of non alloyed steel and the transports of the filler rod
Included Processes
and the protective gas to the place of use.
Amount 1
Local Name Schweissen, Lichtbogen, Stahl
Synonyms MAG//MIG
Different arc welding technologies can have very different emissions. Only to be
used for the technology describes or as a rough proxy if welding is not important
ence function
in the system considered.
Start Date 1997
End Date 2002
Other Period Text
MAG-welding of unalloyed steel with 83% Ar, 13% CO2 and 4% O2 protective gas
Technology text
and a wire consumption of 0.0536 kg/m.
Extrapolations none

6. Welding
Source for
ecoinvent
Comment
Deviation
Standard
Location
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
steel, low -alloyed, at
Steel  metals extraction No RER 5.36E-02 k g calculated 1 1.38 (4,5,1,1,1,5,4)
plant
acetylene, at regional estimated /
acetylen  chemicals organics No CH 9.76E-03 k g 1 1.68 (4,5,1,1,4,5,1)
storehouse calculated
oxygen  chemicals inorganics No RER oxygen, liquid, at plant 1.14E-02 k g calculated 1 1.68 (4,5,1,1,4,5,4)
steel 600 km, acetylene 600 km, oxygen 100 km. transport (4,5,n.A .,n.A .,n.
transport rail  train No RER transport, f reight, rail 4.62E-02 tk m estimated 1 2.09
Gases tw ice the w eight f or the bottles systems A .,n.A .,5)
steel, acetylene and oxygen 50 km. Gases tw ice transport (4,5,n.A .,n.A .,n.
transport lorry  road No RER transport, lorry 32t 4.80E-03 tk m estimated 1 2.09
the w eight f or the bottles systems A .,n.A .,5)
welding, gas, steel
Inf rastructure  neglected 0.00E+00

 Cr (II, III) air unspecif ied Chromium 2.81E-05 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
 Cr (V I) air unspecif ied Chromium V I 3.00E-08 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
 Cu air unspecif ied Copper 2.10E-07 kg Künniger 1998 1 1.92 (1,5,5,3,3,5,31)
 Fe air unspecif ied Iron 8.74E-05 kg Künniger 1998 1 1.92 (1,5,5,3,3,5,31)
 Mn air unspecif ied Manganese 1.13E-05 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
 Mo air unspecif ied Molybdenum 2.16E-06 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
 Ni air unspecif ied Nickel 1.13E-05 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
 Zn air unspecif ied Zinc 4.20E-07 kg Künniger 1998 1 5.42 (1,5,5,3,3,5,22)
sum of oxides of the metal emissions and the
 partikel <2.5 air unspecif ied Particulates, < 2.5 um 1.41E-04 k g calculated 1 3.41 (4,5,5,3,3,5,20)
f luoride above
 CO2 air unspecif ied Carbon dioxide, f ossil 3.24E-02 k g calculated 1 1.73 (4,5,5,3,3,5,14)
Carbon monoxide,
 CO air unspecif ied 2.03E-04 k g calculated 1 5.47 (4,5,5,3,3,5,17)
fossil
 NOx air unspecif ied Nitrogen oxides 2.00E-06 kg Künniger 1998 1 1.92 (1,5,5,3,3,5,16)
 O3 air unspecif ied Ozone 9.06E-06 kg Künniger 1998 1 1.92 (1,5,5,3,3,5,31)
 acetylen to air air unspecif ied Ethyne 9.76E-05 kg estimated 1 3.22 (5,5,5,5,5,5,23)
 w aste heat air unspecif ied Heat, w aste 5.00E-01 MJ calculated 1 1.68 (4,5,1,1,4,5,13)
 Gas w elding metals processing No RER w elding, gas, steel 1.00E+00 m
Fig. 6.4 Flows for "welding, arc, steel" and their representation in the ecoinvent database

6. Welding
6.4 Arc welding of aluminium

This chapter is mainly based on Brandt & Weiler (1997):
The main problem in welding of aluminium is dealing with the film of aluminium oxide scale which
adheres extremely strongly and has a very high melting point. As the thermal conductivity of alumin-
ium metal is also very high, there must be an ample heat supply so the weld pool does not solidify too
rapidly or contain nonmetallic particles or gas-filled pores.
The simplest process, welding with manual electrodes, is widely used and is completely satisfactory
for simple joints. Qualitatively good welds can be produced by the TIG and MIG processes. The pre-
ferred protective gas is helium. The alternating current removes the layer of aluminium oxide. If the
weld is carefully prepared and the auxiliary materials have been correctly chosen (surface cleanliness),
mechanically sound low-porosity welded joints are obtained. Recent investigations have indicated that
high-power processes such as submerged arc welding are also suitable.
Owing to the low melting point of aluminium, the low viscosity of the molten metal, and the small dif-
ference in density between the metal and its oxide, manual welding of aluminium and its alloys needs
highly skilled welders.
6.4.1 Arc welding of aluminium in ecoinvent

Because no quantitative information on aluminium arc welding could be obtained, the following as-
sumptions are made:
The volume of filler rod used is estimated to be the same as for MAG welding of steel, i.e. 6.79E-6 m3.
Thus 0.0183 kg/m weld seam of filler rod is used. The filler rod is assumed to be composed of 90% Al
and 10% Si.
Energy consumption is calculated from the current and the voltage commonly used for aluminium
MIG welding 18 . With 150 A and 15V a power of 2250 W are obtained. Assuming a welding speed of
0.5 m/min the energy consumption is calculated to 75 Wh/m weld seam.
Helium is assumed to be used as protective gas. A flow of 10 l/min is assumed. Thus 20 l/m weld seam
helium is consumed.
The total particulate matter emissions are assumed to be as high as for MAG welding of steel. The
composition however is different. Zn, Al and Si are thought to be dominant. We assume 60% Al, 30%
Zn and 10% Si.
The background information to the process "welding, arc, aluminium" is given in Tab. 6.3, flows are
shown in Fig. 6.5.
Data quality
(2003)). A lognormal distribution is assumed. The data are based on rough estimations and of poor
overall quality.
18
http:/www.brazing.com/techguide/procedures/aluminium.asp, accessed 17.03.03

6. Welding
Tab. 6.3 Ecoinvent meta information for "welding, arc, aluminium"
Name welding, arc, aluminium

Location RER
Unit m
Data Set Version 1
Includes the arc welding of aluminium and the transports of the filler rod and the
Included Processes
protective gas to the place of use.
Amount 1
Local Name Schweissen, Lichtbogen, Aluminium
Synonyms MAG
General Comment to refer- Based on rather rough estimates. Not to be used if welding is of importance in the
ence function system considered.
Start Date 1997
End Date 2002
Other Period Text
Technology text MIG Welding of aluminium with helium as protective gas.
Extrapolations none

6. Welding
Source for
ecoinvent
Comment
Deviation
Location
Standard
Category
category
structure
Remarks
Output
General
Process
name in
Modul
Name
value
value
Input
Infra-
mean
Sub-
Mean
Type
Unit
A luminium  metals extraction No RER 1.65E-02 k g estimate 1 1.93 (5,5,5,3,3,5,3)
mix, at plant
Silizium  metals extraction No RER 1.83E-03 k g estimate 1 1.93 (5,5,5,3,3,5,3)
mix, at plant
argon, crude, liquid, at
Helium  chemicals inorganics No RER 3.57E-03 k g estimate 1 1.93 (5,5,5,3,3,5,3)
plant
electricity, medium
production
UCTE voltage, production
welding, arc, aluminium
Electricity  electricity No 7.50E-02 k Wh estimate 1 1.93 (5,5,5,3,3,5,2)

mix
UCTE, at grid
transport (4,5,n.A .,n.A .,n.
transport rail  train No RER transport, f reight, rail 1.53E-02 tk m estimate 1 2.09
systems A .,n.A .,5)
transport (4,5,n.A .,n.A .,n.
transport lorry  road No RER transport, lorry 32t 1.27E-03 tk m estimate 1 2.09
systems A .,n.A .,5)
construction electronics for control
Inf rastructure  buildings No RER 1.00E-05 k g estimate 1 3.59 (5,5,5,3,3,5,9)
processes units
 He air unspecif ied Helium 3.57E-06 kg estimate 1 2.16 (5,5,5,3,3,5,31)
 Al air unspecif ied Aluminum 8.45E-05 kg estimate 1 2.16 (5,5,5,3,3,5,31)
 Zn air unspecif ied Zinc 4.23E-05 kg estimate 1 2.16 (5,5,5,3,3,5,31)
 Si air unspecif ied Silicon 1.41E-05 kg estimate 1 2.16 (5,5,5,3,3,5,31)
 Particulates air unspecif ied Particulates, < 2.5 um 1.41E-04 kg estimate 1 3.59 (5,5,5,3,3,5,27)
Carbon monoxide,
 CO air unspecif ied 6.67E-05 k g estimate 1 2.16 (5,5,5,3,3,5,31)
fossil
 NOx air unspecif ied Nitrogen oxides 2.00E-06 k g estimate 1 2.16 (5,5,5,3,3,5,31)
 O3 air unspecif ied Ozone 9.06E-06 k g estimate 1 2.16 (5,5,5,3,3,5,31)
 Waste Heat air unspecif ied Heat, w aste 2.70E-01 M J estimate 1 1.93 (5,5,5,3,3,5,13)
Welding
 metals processing No RER w elding, arc, aluminium 1.00E+00 m
Aluminium
Fig. 6.5 Flows for "welding, arc, aluminium" and their representation in the ecoinvent database

7. Enamelling
7 Enamelling
Author: Niels Jungbluth, ESU-services, Uster
Review: Mischa Classen, Hans-Jörg Althaus, Rolf Frischknecht
Acknowledgement
We thank Gabriela Lowitz for the information provided in a personal communication.
Summary
Enamel is a glass of a particular chemical composition and physical nature determined for the surface
protection of metal. The major interest for the enamelling process in this database is the application for
hot water boilers. The inventory has been elaborated for the European situation. The basic inventory is
split up into the major process stages degreasing, corroding, cleaning and firing as well as an inventory
for the materials used for enamelling. The life cycle inventory is based on the data investigated in a
study for the new single layer enamelling process. General assumptions are used for transport dis-
tances. The inventory of enamelling can be considered as a basis for the assessment of this process in
connection with technical applications.
7.1 Introduction
Enamel is a glass of a particular chemical composition and physical nature determined for the surface
protection of metal. This results in a composite material of glass and metal, which combines positive
qualities of both materials. The process of firing this glass on metal is called enamelling and thus the
object, which has had its surface protected and improved, is referred to as enamelled (Wendel-Email
(2002)). Some of the early vitreous enamelled relics date back to the 13th Century BC (IVE (2002)).
The word enamel comes from the High German word “smelzan” and later from the Old French “es-
mail”. In American English it is referred to as Porcelain Enamel. It should not be confused with paint,
which is sometimes called ‘enamel’. Paints cannot be enamel. They do not have the hardness, heat re-
sistance and colour stability that is only available with real vitreous enamel (IVE (2002)).
7.2 Resources of Enamel

First of all an enamel frit is melted from natural and synthetic materials in furnaces. Silica sand (40%-
49%), boric anhydride (15%-17.5%) and sodium monoxide (15%-17%) are the major ingredients of
enamel frits (Lowitz (1998)). After that, it is ground to become enamel slurry or powder enamel. De-
pending on the nature and size of the object to be enamelled, the metal pieces get a wet or a dry appli-
cation of enamel. This is done by dipping, spraying, pouring, flooding or powdering. Afterwards the
objects are fired at temperatures of between 550°C and 1,000°C.
7.3 Characterisation of the Product

The smooth glass-like surface is hard; it is scratch, chemical and fire resistant. It is easy to clean and
hygienic (IVE (2002)).
7.4 Application of Enamels

One can recognise enamel as the material used to produce the now highly collectable advertising signs
produced during the early 20th Century. The “Hovis” and “Virol” signs were part of the everyday
street scene. Cook stoves will almost certainly have a vitreous enamelled oven and the higher quality

7. Enamelling
cookers will use it on the outer parts. A cast iron or steel bath may also have been vitreous enamelled.
Less obvious are the storage silos on farms, usually blue or green (IVE (2002)).
Enamel is become more used as a protective coating for hot water tanks, like solar boilers and water
heaters, since the availability of appropriate powder application equipment for such products (Ditmer
(2002)).
Due to its mechanical, thermal, optical and chemical properties, enamel offers a good surface protec-
tion for any enamelable metal. Not only can metals be protected against corrosion by being enamelled,
but they can also be made more interesting by the use of colour (Wendel-Email (2002)). Vitreous
enamel can be applied to most metals. For jewellery and decorative items it is often applied to gold,
silver, copper and bronze. For the more common uses, it is applied to steel or cast iron. There are some
specialised uses on stainless steel and aluminium (IVE (2002)).
7.5 System Characterisation

Fig. 7.1 shows a typical layout of a modern powder enamelling plant for hot water tanks. This plant
has a production capacity of circa 100 -120 products / hour and consists of the following equipment :
 Shot blaster
 Powder enamel system
 Automatic transfer device
 Enamelling furnace
 Cooling tunnel
Fig. 7.1 Layout of a powder enamelling plant for hot water tanks (Ditmer (2002)).
The boilers are loaded onto a conveyor near the testing station after the welding department and then
they are transported to the shot blaster. They remain stationary in front of the sliding doors until this
machine is ready for operation. The doors will open automatically at the end of each cycle and the
boilers will move to the blasting positions where they are clamped into a centring device. During the

7. Enamelling
first stage the boilers will be centred and rotated, while a rigid multi nozzle lance enters the product to
blast the interior. Remaining shot medium will be washed away by means of compressed air during the
next stage. At the end of this cycle boilers will be moved to the automatic powder application system,
which also comprises of two stages (Ditmer (2002)).
At the first stage the boilers will be centred while the interior is sprayed with powder enamel by means
of extended automatic guns mounted on a PLC controlled reciprocator. At the second stage, the spray-
ing operation will be repeated in order to obtain the required enamel layer thickness. An automatic
transfer unit picks the boilers one-by-one from the pre-treatment / application conveyor and places
them carefully on the furnace conveyor (Ditmer (2002)). Fig. 7.2 shows a picture of an enamelling
plant for hot water tanks.
Fig. 7.2 Picture of an enamelling plant for hot water tanks (Ditmer (2002)).
7.6 Life Cycle Inventory of Enamelling

The environmental impacts of enamelling processes have been investigated by Lowitz (1998), right
side of Tab. 7.1). Three different processes were investigated in this Ph.d. thesis per m2 of enamelled
steel sheet:
 At the time of investigation the major process used two stages with a basic and an upper enamel
coating which is fired in two production stages. The enamel use is together 710g/m2.
 An improved process uses also two layers which are fired in one process. The enamel use is to-
gether 390g/m2.
 A new production process uses only one enamel layer and one firing process. This powder coating
technology is used in the meantime by many manufacturers. The enamel use is 600g/m2.
The life cycle inventory is based on the data provided by Lowitz (1998) for the new single layer
enamelling process. According to an actual statement of the author this is the most important process
in the mean time. They have been investigated from 1993-1998 in Germany. Further information
about the type of steel sheet were not available.
Tab. 7.1 shows the unit process raw data and the data investigated by Lowitz (1998). The major inter-
est for the enamelling process in this database is the application for hot water boilers. An important
share of enamelled boilers is imported to Switzerland from different countries. Thus the inventory has
been elaborated for the European situation.

7. Enamelling
The basic inventory is split up into the major process stages degreasing, corroding, nickel-plating,
cleaning and firing as well as an inventory for the materials used for enamelling. According to the in-
formation provided by an expert 19 is the process of nickel-plating not used for the enamelling of boil-
ers (which are the major product of interest). It has thus not been considered for the inventory.
Gas, oil and electricity are used as energy carriers by different manufacturers, but the share of these
energy carriers in today plants was not known. Lowitz (1998) assumed that all energy is delivered by
electricity which was at this time the dominating energy used. The energy for the pre-treatment proc-
esses was not split up to the different stages. For some special chemicals there are no background data
available. Their use has been assessed with the generic data for organic or inorganic chemicals. Trans-
ports of materials are calculated with the standard transport distances. Data for some process specific
emissions are also shown in the study of Lowitz (1998). They are considered for the unit process raw
data.
Data for the infrastructure and land use of such facilities were not available. They have been roughly
assumed with data for a metal coating plant. More Information about threshold limits for effluents and
air emissions in Germany can be found in dev (1999). But, it was not possible to relate these informa-
tion to the functional unit investigated in this inventory. Further background information for the enam-
elling process is given by Petzold (Petzold & Pöschmann (1992)).
19
Personal communication with E. Voß, Wendel-Email, 14.6.2002. Corroding or sandblasting are used as pre-treatment proc-
esses.

7. Enamelling
Tab. 7.1 Unit process raw data and data quality indicators for the enamelling process (Lowitz (1998)
Infrastructu
Uncertain
Deviation
Standard
Location
nickel-plating
enamel
95%
Name Unit enamelling GeneralComment degreasing corroding (not cleaning firing
materials
considered)
Location RER DE DE DE DE DE DE
InfrastructureProcess 0 0 0 0 0 0 0
Unit m2 m2 m2 m2 m2 m2 m2
output enamelling RER 0 m2 1.00E+0
(3,4,2,3,3,5); Other energy carriers
technosphere electricity, medium voltage, production UCTE, at grid UCTE 0 kWh 1.44E+1 1 1.35 possible. Some manufacturers use 2.57E+0 2.57E+0 2.57E+0 2.57E+0 1.17E+0 5.56E+0
also gas or oil.
aluminium oxide, at plant RER 0 kg 4.60E-2 1 1.26 (3,4,2,3,1,5); Literature 4.60E-2
Borax, anhydrous, powder, at plant RER 0 kg 2.89E-1 1 1.26 (3,4,2,3,1,5); Literature 2.89E-1
boric acid, anhydrous, powder, at plant RER 0 kg - 1 1.26 (3,4,2,3,1,5); Literature 1.00E-2
(3,4,2,3,3,5); Literature, special
chemicals inorganic, at plant GLO 0 kg 4.02E-2 1 1.35 chemicals (corroding agent, nickel 2.20E-2 3.00E-2 1.82E-2
sulphate)
(3,4,2,3,3,5); Literature, Organic
chemicals organic, at plant GLO 0 kg 1.80E-2 1 1.35 1.80E-2
adsorbers
limestone, milled, packed, at plant CH 0 kg 3.43E-2 1 1.26 (3,4,2,3,1,5); Literature 3.43E-2
potassium nitrate, as K2O, at regional storehouse RER 0 kg 2.70E-2 1 1.26 (3,4,2,3,1,5); Literature 2.70E-2
silica sand, at plant CH 0 kg 3.05E-1 1 1.26 (3,4,2,3,1,5); Literature 3.05E-1
(3,4,2,3,1,5); Literature, Cleaning
soap, at plant RER 0 kg 5.00E-3 1 1.26 5.00E-3
agents
sulphuric acid, liquid, at plant RER 0 kg 3.00E-2 1 1.26 (3,4,2,3,1,5); Literature 3.00E-2
metal coating plant RER 1 unit 3.33E-7 1 3.95 (5,5,1,3,5,na); Rough estimation
tap water, at user RER 0 kg 7.30E+1 1 1.26 (3,4,2,3,1,5); Literature 2.00E+0 1.10E+1 2.00E+0 6.00E+1 6.00E-4
(4,5,na,na,na,na); Standard

10 Metals v2.1

Uploaded by

Copyright:

Available Formats

10 Metals v2.1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

10 Metals v2.1

Uploaded by

Copyright:

Available Formats

Swiss Centre

for Life Cycle

Life Cycle Inventories of Metals

Mischa Classen, Hans-Jörg Althaus, Silvio Blaser,

ecoinvent v2.1 report No. 10

Dübendorf, March 2009

ecoinvent Board: Gérard Gaillard (Chair), Agroscope Reckenholz-

Institutes of the ecoinvent Centre:

Participating consultants: Doka Life Cycle Assessments, Zürich

Software Support: ifu Hamburg GmbH

Project leader: Roland Hischier, ecoinvent Centre, Empa,

ecoinvent v2.1 report No. 10 - ii -

Life Cycle Inventories of Metals

Editor: Hans-Jörg Althaus

Authors: Mischa Classen, Hans-Jörg Althaus, Silvio Blaser,

Reviewer: Roland Hischier (Empa), Roberto Dones (PSI),

Contact address: ecoinvent Centre

Responsibility: This report has been prepared on behalf of one or

Terms of Use: Data published in this report are subject to the

Liability: Information contained herein have been compiled

© Swiss Centre for Life Cycle Inventories / 2009

ecoinvent v2.1 report No. 10 - iv -

Data v1.1 (2004) Data v2.1 (2009)

ecoinvent report v2.1 No. 10 -v-

Author: Hans-Jörg Althaus, EMPA, Dübendorf

Review Roland Hischier, EMPA, St. Gallen

Comments on this version 2.1

ecoinvent v2.1 report No. 10 -i-

COMMENTS ON THIS VERSION 2.0 .....................................................................................I

TABLE OF CONTENTS .................................................................................................... III

2 RESERVES AND RESOURCES OF MATERIALS............................................................. 3

3 CHARACTERISATION OF THE MATERIALS ................................................................... 5

4 USE / APPLICATION OF MATERIAL ............................................................................ 7

5 SYSTEM CHARACTERISATION .................................................................................. 9

ecoinvent v2.1 report No. 10 - iii -

ecoinvent v2.1 report No. 10 -1-

2 Reserves and Resources of Materials

Energy re- natural nuclear hydro

ecoinvent v2.1 report No. 10 -3-

3 Characterisation of the materials

3.2 Aluminium hydroxide

3.3 Aluminium Oxide

ecoinvent v2.1 report No. 10 -5-

Tab. 3.3 Physical properties of Aluminium

Property Value Unit

ecoinvent v2.1 report No. 10 -6-

4 Use / Application of Material

Tab. 4.1 Uses of aluminium precursors other than aluminium production

bauxite aluminium hydroxide aluminium oxide

ecoinvent v2.1 report No. 10 -7-

5.1.1 Production process and infrastructure

5.1.4 Bauxite mining in ecoinvent

- bauxite, at mine Fig. 5.1

- mine, bauxite Fig. 5.2

- recultivation, bauxite mine Fig. 5.3

ecoinvent v2.1 report No. 10 -9-

Tab. 5.1 Meta information for the bauxite mining processes

ecoinvent v2.1 report No. 10 - 10 -

General Flow information Representation in ecoinvent Uncertainty