Geoenvironmental units (GEUs) in waste dumps

from Collahuasi
David Arcos
Amphos 21 Consulting S.L., Spain
Wolf von Igel
Amphos 21 Chile Ltda., Chile
Lissette G. Flores
Compaa Minera Doa Ins de Collahuasi, Chile
Alberto Acua
Compaa Minera Doa Ins de Collahuasi, Chile

ABSTRACT
Geoenvironmental units (GEUs) are rocks extracted from a mining operation with similar
characteristics in terms of lithology, mineralogy and alteration, the destination of which is the waste
dumps or other operational elements that will remain after mine closure. It is expected that
different GEUs behave differently during their surface exposure, resulting in different acid rock
drainage potential. Their definitions can be made even before the feasibility studies using the block
models and the features included in it. Also, if the mining plan is available, the amount of different
GEUs can be calculated, obtaining information concerning their representativity and distribution at
the end of the operation in the waste dump.
It is known that one of the main problems when assessing the potential acid rock drainage of a
waste dump is the representativity of the samples. Using GEUs, samples can be selected from the
exploratory boreholes, solving any representativity uncertainty. Therefore, samples from waste
dumps can be collected even before the start of the mining operation, and the existence of the waste
dump and acid rock drainage assessment can be performed at any stage of the mining operation.
This procedure has been applied to the waste dump of the Rosario deposit in Collahuasi. The
selection of samples from the different GEUs ensured that at least 85% of the materials to be
deposited in the waste dumps at the end of the operation are represented. A series of laboratory
analyses and tests has been conducted on selected samples, allowing for the establishment of their
potential to generate acid rock drainage. The behaviour of the whole waste dump in terms of acid
rock drainage can be evaluated on the basis of the distribution of the different GEUs within the
waste dump and the knowledge obtained from the experimental data for each unit.

INTRODUCTION
One of the most relevant environmental problems associated with mining operations is acid rock
drainage (ARD). The process of ARD generation is caused by the interaction of water with rock
materials under atmospheric conditions. There are several aspects that enhance the potential
formation of ARD, one of the most important being the exposure of underground rocks to the
surface, allowing for the oxidation of their pyrite content and increasing the reactive surface of the
rocks due to rock handling during the mining operation. Assessing the ARD potential as a function
of time is of paramount importance to allow for the implementation of prevention and/or
remediation techniques. However, the materials deposited in waste rock dumps can have different
properties, both from mineralogical and hydrological (including grain size distribution)
perspectives. These differences can be found even in a single waste rock dump and, therefore, the
potential ARD generation can be different depending on the type of rock material. For this reason,
it is important to assess the potential ARD generation of each type of material from the waste rock
dump as well as its contribution to the out flowing water.
When facing the assessment of the ARD from a waste rock dump, the difficulties are related to the
complex structure of the dump itself (Figure 1). It is necessary to determine how the different
materials are distributed in the dump and their representativity with respect to the total amount of
materials deposited in the dump. In addition, depending on whether the ARD assessment is
addressed before the operation (during the feasibility stage) or when the mine has already been
operative for several years, different problems have to be faced. In the first case, there is no waste
rock dump constructed yet, so it is impossible to obtain direct samples from it; whereas in the
second case, taking representative samples of the different materials deposited, even from a single
type of material, is almost impossible as some of the material types are already covered (especially
those extracted during the early stages of operation).

Figure 1 View of part of the waste rock dump from Collahuasi

Taking into account all the difficulties associated with the characterisation of the materials
deposited in a waste rock dump and the assessment of their potential ARD generation, there is a
need for establishing a methodology that can address all the problems described above. The first
approach consists of grouping all those materials that have similar properties and potentially have
similar behaviour from the geochemical point of view and, therefore, for their potential ARD
generation. These groups of materials have been called geoenvironmental units (GEUs), in contrast
with the geometallurgical units (GMUs), which are defined according to their metallurgical
behaviour and their destination in the mineral processing scheme of a mine. The comparison of
these two types of units is not senseless, as the definition procedure for both types of units share
many aspects in common. For instance, the definition of GEU can be done at any phase of a mining
project and samples can be taken directly from exploratory drill cores, ensuring a proper set of

samples representing all the materials that can be deposited, not only in a waste rock dump, but in
other operative elements that can remain after mine closure.
There are not many attempts to classify rock materials from mining operations in terms of ARD
prediction. Only very recently some methodologies have been described to classify rock materials
according to their ARD behaviour (Verburg et al., 2009 and Parbhakar Fox et al., 2011). However,
these methodologies are more focused on the experimental procedure for rock type characterisation
than on the use of operational data for mine planning. In the present case, it is suggested to use
information from mine operation to classify rock materials that will be deposited in waste dumps,
allowing determining the representativity of the different rock types and the sampling
methodology. Moreover, a further step could be the characterisation procedure for the different
units that can follow different approaches, as the one presented in Parbhakar Fox et al. (2011).

GEU DEFINITION METHODOLOGY

The GEU are defined on the basis of the mineralogical composition. Therefore, the descriptors used
in the block model, such as lithology, mineral zone and alteration type, are entirely useful for the
GEU definition. Coupling this definition of GEU in the block model with mine planning, where the
destinations of every block are assigned (i.e. waste dump, heap leaching or processing plant), the
representativity of every GEU in the waste rock dump can be defined. As an example, the definition
of the block destination from the Rosario open pit in Collahuasi has been used together with the
descriptors mineral zone, lithology and alteration type to define the GEUs and their
representativity in the Rosario waste rock dump (Figure 2). From the information updated in
November 2010, the total amount of rock already deposited in the Rosario waste dump is
488 million tonnes. Most of the materials correspond to the mineral zones leached (58.18%) and
primary pyritic (35.59%), accounting for up to 93.77% of the total mass. Taking the example of the
leached mineral zone, more than half of the materials deposited in the waste dump correspond to
the lithology dacite. From this leached dacite combination constrain, the most represented types of
alteration are argillic, moderate sericite and chlorite sericite. The GEU leached dacite argillic,
leached dacite moderate sericite, and leached dacite chlorite sericite represent 11.95%, 3.79% and
10.83%, respectively, of all the material deposited in the waste dump. This procedure has to be
applied to all combinations of mineral zone, lithology and alteration types until an acceptable
representativity of all the material deposited in the dump is achieved, i.e. > 85% for the Rosario
waste rock dump.

Figure 2 Example of GEU definition for the Rosario waste rock dump

In addition, as the main acid generation reaction is related to the oxidation of pyrite, the content
and grain size of this mineral has to be included in the GEU definition process. This is equivalent to
the consideration of copper and its reactivity with different acids for the definition of GMU in a
mining operation of a porphyry copper deposit. However, the measurement of the pyrite content
for each block is not straightforward, at least at a reasonable cost. Thus, a procedure to model the
pyrite content has to be considered. A reasonable way to carry this pyrite content model can be
based on the sulphur analysis (sulphur as sulphide) of the samples used for the block model
definition together with a classical mineralogical analysis of selected samples. A reasonable number
of samples for mineralogical characterisation must be collected representing most of the GEUs
defined following the previous procedure. The correlation between sulphur and mineralogical
analyses will allow the definition of a pyrite content model and the assignment of the pyrite content
and associated grain size to every block. The implementation of the pyrite content to GEU
definition can be made by means of defining ranges of pyrite content. These ranges can be based on
the results of acid base accounting (ABA) analyses, where samples with low, uncertain and high
acid generation potential define three groups of pyrite content. However, this group division has to
be made with care, as there are site specific conditions, such as the amount of additional minerals
that can contribute to acid generation or neutralising minerals.
At this stage, a large number of defined GEUs can result, which make it difficult to manage in a
mining operation. Therefore, it is necessary to reduce the number of units by considering some site
specific data, as there could be similarities in the mineralogy of different units defined a priori or in
their behaviour during geochemical tests (i.e. humidity cell tests). In the case of Collahuasi, up to
29 GEUs were defined to represent 90% of the material deposited in the Rosario waste dump, prior
to considering the pyrite content (Table 1).

Table 1 Defined GEU for the materials deposited in the Rosario waste dump of Collahuasi
Mineral zone

Lithology

Alteration

Mass
(tones)

Representativity
%

Leached

Andesite

Argillic

7,524,986

1.70

Accumulated
representativity
%
1.70

Leached

Andesite

Chlorite sericite

9,325,602

2.10

3.80

Leached

Andesite

Silicification

4,039,456

0.91

4.71

Leached

Dacite

Argillic

58,294,322

13.14

17.85

Leached

Dacite

Chlorite sericite

52,821,302

11.91

29.76

Leached

Dacite

No altered

7,493,899

1.69

31.45

Leached

Dacite

Sericitic

18,468,953

4.16

35.62

Leached

Dacite

Silicification

10,043,982

2.26

37.88

Leached

Other

Argillic

24,846,091

5.60

43.48

Leached

Other

Chlorite sericite

10,091,953

2.28

45.76

Leached

Argillic

12,688,406

2.86

48.62

Chlorite sericite

15,392,164

3.47

52.09

Silicification

6,149,060

1.39

53.47

Primary Pyritic

Sedimentary
units
Sedimentary
units
Sedimentary
units
Andesite

Chlorite sericite

5,924,666

1.34

54.81

Primary Pyritic

Andesite

Sericitic

3,767,016

0.85

55.66

Primary Pyritic

Andesite

Silicification

5,290,369

1.19

56.85

Primary Pyritic

Dacite

Argillic

15,150,445

3.42

60.27

Primary Pyritic

Dacite

Chlorite sericite

29,132,672

6.57

66.84

Primary Pyritic

Dacite

Sericitic

22,751,907

5.13

71.97

Primary Pyritic

Dacite

Silicification

17,012,540

3.84

75.80

Primary Pyritic

Sericitic

4,357,464

0.98

76.78

Primary Pyritic

Not
determined
Other

Argillic

7,040,162

1.59

78.37

Primary Pyritic

Other

Chlorite sericite

5,500,812

1.24

79.61

Primary Pyritic

Other

Quartz sericite

5,740,307

1.29

80.91

Primary Pyritic

Sedimentary
units
Sedimentary
units
Sedimentary
units
Sedimentary
units
Sedimentary
units

Argillic

6,518,964

1.47

82.38

Chlorite sericite

8,796,407

1.98

84.36

Quartz sericite

6,125,971

1.38

85.74

Sericitic

4,511,373

1.02

86.76

Silicification

17,852,535

4.03

90.78

Leached
Leached

Primary Pyritic
Primary Pyritic
Primary Pyritic
Primary Pyritic

GEU SAMPLING AND LOCATION IN THE WASTE ROCK DUMP

Once GEUs are defined, a sampling strategy can be designed. Samples can be obtained directly
from the exploratory boreholes used for the block model, ensuring the units of interest are properly
sampled. This procedure allows taking samples from the different GEUs even prior to starting the
mining operation and dumping waste rocks, so the methodology can be applied during the
feasibility phase of the mining operation. It is also worth noting that direct sampling from the drill
cores allow taking samples from GEUs that can be covered by more recent materials, if the process
is started some years after the initiation of the mining operation.
Another interesting aspect of defining these GEUs is that once they can be associated with mine
planning and, therefore, to the build up of the waste dump, they can be traced and located within
the waste dump. For instance, in the Rosario waste dump, the two main units corresponding to
leached and primary pyritic materials can be related to the time plan of extraction from the open pit
(Figure 3).

Figure 3 Time evolution of the percentage of the two main units deposited in the
Rosario waste rock dump of Collahuasi

If, in addition, there is geographic information concerning the waste dump growth, then the
distribution of units within the dump can be mapped and used as a tool for control in the case of
verification of ARD. This distribution within the waste dump can also be used for assessment
purposes, such as implementing this distribution in hydrogeochemical models, or even for
application and monitoring remediation techniques at a cost effective rate. In the case of the Rosario
waste rock dump, the distribution of the percentage of primary pyritic materials based on the
graphic from Figure 3 has been applied to the waste rock dump mapping.

CONCLUSION
The definition and use of GEUs in the assessment, prevention and remediation process of potential
ARD in a mine site has been demonstrated to be very useful and applicable to any stage of the
mining operation, from the feasibility to the closure phases. GEUs are defined by using information
6

from the block model and mine planning, although some additional information has to be recorded
in order to take into consideration some site specific aspects. The application of this procedure
ensures that all types of materials deposited in a waste rock dump are properly considered and
represented. The GEUs can be used for further sampling using exploratory drill cores facilitating
and lowering the operative cost of sampling an already built waste rock dump. The combination
with mine planning allows identifying the amount, representativity and exact location of every
GEU in the waste dump as a function of time. The information generated can be used for ARD
assessment purposes as well as for the decision making on their management and application of
prevention and/or remediation techniques at a cost effective rate.

REFERENCES
Parbhakar Fox, A.K., Edraki, M., Walters, S. & Bradshaw, D. (2011) Development of a textural index for the
prediction of acid rock drainage, Minerals Engineering, in press.
Verburg, R., Bezuidenhout, N., Chatwin, T. & Ferguson, K. (2009) The global acid rock drainage guide
(GARD Guide), Mine Water and Environment, vol. 28, pp. 305 310.