Weda Bay, Halmahera, Ni-Co
Weda Bay, Halmahera, Ni-Co
Weda Bay, Halmahera, Ni-Co
LATERITE PROJECT
By
M.G. Baillie
TABLE OF CONTENTS
1. INTRODUCTION 1
2. RESOURCE DEFINITION 4
3. METALLURGY 8
6. PRE-FEASIBILITY STUDY 15
1. INTRODUCTION
1.1. GENERAL
The Weda Bay nickel and cobalt deposits are located in the central part of Halmahera in
Eastern Indonesia. Figure 1 shows the location of Halmahera within Eastern Indonesia
and Figure 2 shows the concession area within Halmahera. The area is held as a
Seventh Generation Contract of Work (COW) by PT Weda Bay Nickel (WBN). WBN is
jointly owned by Strand Minerals (Indonesia) Pty Ltd (Strand) (90%) and PT Aneka
Tambang (Antam) (10%). Strand is a wholly owned subsidiary of Weda Bay Minerals
Inc. (WDA), a company listed on the Toronto Stock Exchange in Canada and Antam is
an Indonesian mining company, owned 60% by the Government.
The extensive areas of laterite discovered at Weda Bay have developed over
serpentinised harzburgites and dunites forming part of a larger ophiolite basement
complex which extends from central Halmahera some 400 km eastward through
numerous smaller islands to Waigeo off the north coast of Irian Jaya. The Weda Bay
project region occurs at the western margin of this ocean floor sequence, and
weathering of these rocks has produced laterite profiles that are enriched in nickel and
cobalt.
1
Figure 2 Location of Project on Halmahera
Application for a COW over 120 500 Ha of Central Halmahera was made in April 1996
with exploration beginning soon after under an interim exploration license (SIPP). These
preliminary investigations quickly revealed the presence of potentially economic nickel
and cobalt mineralisation in several coastal laterites.
2
Several agreements were signed in early 2000 with OM Group, Inc. (OMG), a US based
company involved in the refining of nickel and cobalt metal and the manufacture and
marketing of specialist nickel and cobalt chemicals and products. These agreements
provided the funding for ongoing work and included a take or pay sales and purchase
agreement for the output of the initial phase of production. Since mid-2000, ongoing
work has further defined the resources, metallurgy and site logistics for input into a pre-
feasibility study (PFS) to form the basis for further detailed metallurgical testing and
engineering and ultimate development of the project.
3
2. RESOURCE DEFINITION
A systematic approach to exploration was adopted with the initial aim to demonstrate
that sufficient resources were available to support future development. This resource
development program began in late 1996 with the evaluation of the regional potential of
the COW and was carried out using a combination of the interpretation of remotely
sensed data (aerial photographs, Landsat TM and Radarsat imagery) to provide targets
for aerial reconnaissance mapping. This targeted the identification of the rounded
landforms which are generally indicative of economic laterite development and was
followed by a helicopter survey to enable visual confirmation by identifying the distinctive
red-orange colouration of the laterites. These methods proved to be very successful.
The task was made considerably easier in the cleared and settled coastal regions and
areas of logging activity due to the presence of an extensive network of disused forestry
tracks. These not only exposed the laterites from beneath the sometimes thick jungle
canopy, but also provided 4WD access to areas of laterite development.
A series of airborne surveys were conducted between late 1996 and mid 1998 that
identified five major lateritic provinces and regions within the Weda Bay COW (Figure
3). These occur over large and widespread areas of the COW and include the following:
The Big Kahuna (BK) Province in the central-western region of the COW.
Resource definition has taken place in the Jira River and Big Kahuna deposits
which occur in the southern region of this province. A series of additional
interpreted laterite bodies extending to the north-east from this region have not
yet been subjected to detailed evaluation. In common with the Jira River and Big
Kahuna deposits, these appear to be large, dissected lateritic plateaus.
The Santa Monica region, is a large dissected plateau located in the south-
central region of the COW in the northern part of the Coastal Province.
Preliminary evaluation of Santa Monica began in 1999 and confirmed its the
potential. A portion of this deposit has recently been classified as a measured
resource.
4
Figure 3 Location of Laterite Occurrences
Initial resource definition was commenced in areas serviced by the network of forestry
roads in the Weda Bay region and the southern regions of the Big Kahuna province.
Evaluation was conducted by means of hand-excavated test pits and reconnaissance
diamond drilling along the major access tracks to determine general profile
characteristics and grade so that different areas could be ranked in terms of
prospectivity. The more prospective regions were systematically gridded, mapped and
diamond drilled on 100 m x 200 m patterns using man-portable drill rigs. This work was
completed in December 1997 and defined indicated and inferred resources of
approximately 90 Mdt.
5
specialized programs during 1998. These included detailed studies on the density
variations through the laterite profile by the excavation of a series of test pits and the
collection of undisturbed samples through drilling with hollow (Shelby) tubes for detailed
laboratory tests. Alternative resource evaluation techniques were also trialled during
this period, including Ground Penetrating Radar (GPR) which is a geophysical technique
used to continuously map variations in and through the laterite profile.
Exploration and resource evaluation activities since late 2001, have concentrated on
improving confidence in the estimation of known resources rather than increasing the
global resource base. The Santa Monica resource data, compiled by the Company over
the past three years, has been comprehensively reviewed by an independent geological
consultant, International Mining Consultants Pty Ltd (IMC). IMC have constructed a
computer based block model of the resource and completed an independent resource
estimate that includes the first measured resources at SM. The estimate for SM provided
by IMC at 1% nickel or 0.1% cobalt cut off is given in Table 1.
The measured resource represents part of the limonite horizon only. The indicated and
inferred resources are combined limonite and saprolite. The SM resource is expected
to form the first feed to the proposed plant.
The total measured indicated and inferred resource base within the Weda Bay Contract
of Work comprising the independent estimate of Santa Monica resources as described
above and the Companys polygonal resource estimates for all other areas at 1% nickel
or 0.1% cobalt cut off grades is shown in Table 2.
6
Table 2 Summary of Drilled Resources at Weda Bay
7
3. METALLURGY
The majority of the Coastal resources are limonitic or transitional in nature or consist of
saprolite and are unsuited to pyrometallurgical treatment processes.
Batch scale metallurgical tests in the second half of 1997 demonstrated that all laterite
types from Coastal regions were amenable to treatment by the high pressure acid leach
(PAL) process. High recoveries of both nickel (98%) and cobalt (94%) were obtained for
all laterite types at a temperature of 250oC. Leaching kinetics were fast with extraction
completed after around 30 minutes, indicating the high reactivity of Coastal ores. Acid
consumption increased from 250 kg/tonne for limonite to 750 kg/tonne for saprolite.
Coastal limonites and saprolites thickened satisfactorily, with solids densities in the 34-
40% w/w range, but transitional ores which comprise a significant proportion of the
Coastal laterites, exhibite poor settling and thickening behaviour, reaching only 25-30%
w/w solids density. Composite samples representing likely plant feed from Coastal
regions tended to behave like the transitional material.
The pulps of leached residues from these Coastal composites also thickened and settled
poorly and indicated that the application of the PAL process to these ores required
further development.
As the exploration program progressed, it became evident that the plateau deposits, in
particular Santa Monica, were much larger, and of more consistent composition than the
Coastal laterites. Furthermore, the Santa Monica deposit contained more saprolitic
laterite with the ratio of saprolite to limonite being close to 1:1, and only minor amounts
of transitional material. By 1999, metallurgical testwork had focused on Santa Monica
ore with Coastal and Big Kahuna resources treated as potential supplementary ore.
Following a series of preliminary tests, systematic feed thickening tests were carried out
using a selected flowsheet with different laterite types at ambient temperature, with an
initial pulp density of 3% w/w solids and ore ground to -212m.
Solids densities of 30-40% w/w for limonite with acceptable settling rates for all
samples.
Solids densities of 45-50% w/w for saprolite with fast settling rates but some
residual cloudiness in the liquor.
Solids densities of 27-40% w/w for transitional material with generally slow
settling rates.
8
Grinding finer than -212m had a significant negative impact on saprolite settling
behaviour.
These tests on individual laterite types were followed by tests on composite blends made
up by adding saprolite to limonite containing minor amounts of transitional material, to
increase the MgO content to around 9-10%. The selection of this MgO level was based
on the behaviour of the leached pulp (see section on leach residue thickening) and these
blends are referred to as PAL feed blends (PFBs). The feed thickening characteristics
of these blends indicated that a satisfactory pulp density of 36% w/w solids could be
obtained using high rate thickeners. Liquors from PFB thickening were generally clear.
Separate PAL testwork was carried out on limonites (3-6% MgO) and PFBs (9-10%
MgO). The following conditions were selected based on the results of initial evaluations:
A series of tests on samples from different parts of the deposit showed consistently high
extractions of nickel (96.5%) and cobalt (94.5%) for all samples tested. Some general
observations from the testwork were:
Impurity pickup (iron and aluminium) was minimal but increased with higher
acidity level.
Overall, the leaching behaviour of Santa Monica limonites and PFBs in the PAL test
program was very encouraging.
Additional tests were performed on blends of Santa Monica PFBs and Coastal
composites. These showed that leach kinetics improved to near the levels of Coastal
material alone when blending Santa Monica PFBs and Coastal composites in ratios of
1:1 and 3:1. The extractions of nickel and cobalt were unaffected.
9
Table 3 Data from selected Pressure Acid Leach tests conducted on Santa Monica
limonite and PFB samples
BasicTest Conditions
o
Temperature ( C) 255 255 255
Leach Time (mins) 60 60 60
% Solids 26 26 26
Acid Addition (t/t) 0.33 0.365 0.59
Grind Size - 212m - 212m -212 m
Final Metal
Extractions (%)
Nickel 96.6 97.8 96.8
Cobalt 94.1 93.9 94.8
Magnesium 84.0 68.0 94.1
Aluminium N/A 34 27.0
Leach residue thickening tests were carried out on limonites and PFBs at ambient
temperature with an initial pulp density of 4-5% w/w solids using a selected flocculent
and a simulated barren liquor wash.
Final thickened solids densities of 40-50% w/w were obtained with limonite samples and
37-47% w/w on PFBs. In practice, it appears that that 40% w/w solids densities can be
readily achievable for limonite but represents an upper limit for a PFB. The unit settling
areas increase with higher MgO content.
Leach residue settling tests were also carried out with blends of the Santa Monica PFB
and Coastal composites at blending ratios of both 1:1 and 3:1. With the 3:1 blend, the
thickening behaviour was similar to a 100% Santa Monica PFB, but at 1:1 there was a
significant negative impact. These tests confirm that there is scope to blend moderate
amounts of Coastal ores with Santa Monica ore in PAL plant feed.
10
4. ATMOSPHERIC LEACHING OF SAPROLITE
Studies in 1998 had indicated the potential of atmospheric leaching for high magnesium
content ore (saprolite). This was particularly relevant for Santa Monica ore which
contains significant quantities of saprolite, much of which cannot be incorporated into
PAL feed.
CCD CCD
Tests have confirmed that Santa Monica saprolite is highly reactive and is suitable for
neutralization of PAL pulps. With a slurry prepared by the pressure acid leaching of a
PFB, up to 80% of the nickel and 95% of the cobalt were recovered from saprolite used
to neutralize the excess acid at a PAL/SN neutralization feed ratio of 4:1. The final
discharge acidity after SN was typically 5 g/l H2SO4, confirming the high degree of
reactivity of the ground Santa Monica saprolite. There was a low degree of iron and
aluminum co-extraction.
Tests were also conducted on leach pulps after neutralization to determine the impact of
SN on leach residue thickening. These tests indicated that solids densities in the range
of 40-47% w/w could be obtained at reasonable settling rates. The data indicated that
11
coarser saprolite material fed to SN resulted in faster thickening kinetics although the
final solids densities were generally independent of the saprolite grind size.
However, flocculent only tests generally produced poor quality supernatants. This was
resolved by the use of commercial coagulants, which resulted in only slightly slower
thickening rates but clear supernatants.
Combined PAL/ATL/SN circuit trials were also undertaken on Santa Monica samples.
The key results of these trials were:
A separate ATL circuit can recover 95% of the nickel and 85% of the cobalt from
Santa Monica saprolite at 90C in leach times of 2 hours or less. High co-
extraction of iron with the nickel and cobalt was observed, and the acid demand
was typically one tonne of acid per tonne of solids feed.
Approximately 80% of the nickel and over 70% of the cobalt was recovered from
saprolite fed to a subsequent SN stage (PAL:ATL:SN feed ratios of 4:1:1) with no
significant impurity pickup.
Leach residue thickening tests on the subsequent pulps indicated that final solids
densities of 40% w/w solids or higher can be achieved at acceptable settling
rates.
12
5. THE PROJECT CONCEPT
The first two of these factors have been described in some detail in the previous
sections, together with their impact on the development of a project concept. The
remaining two factors impacting on the project concept are discussed below.
Of the projects currently operating with the PAL process only one, the Moa Bay project in
Cuba, was designed to produce an intermediate product. All the Australian projects
were designed to produce metal, although the Cawse operation has recently been
modified to produce an intermediate product.
For the Weda Bay project, an early decision was made to produce an intermediate
product. This decision was influenced by several factors.
Firstly, a PAL based operation involves complex chemical processes, many of which are
sequential and interdependent. If metal is produced, the number of steps and their
interdependency is greater, adding to the potential for difficulties during startup and
operation. There can be little doubt that this factor has been contributory to the
prolonged startup difficulties of the Australian PAL plants. The Weda Bay project is
located on a remote island in East Indonesia where such problems would be magnified.
Hence it was decided to limit the complexity of the initial phase of development by
deferring the metal refining stage.
Secondly, there is currently an excess of nickel refining capacity around the world and
many refineries are seeking a long term secure supply of intermediate product. These
refineries are already operating efficiently and can offer payment terms based on high
recovery factors, eliminating the ramp-up impact of a new refinery. Historically, many of
these refineries utilized nickel sulphide concentrates as feed, but a shortage of suitable
concentrates has been developing for some years, and a sulphide intermediate product
from a PAL process is an excellent substitute.
Finally, and in Weda Bays case, it has been possible to write a sales contract which also
guarantees revenue from the significant cobalt content.
13
5.2. LOGISTICS
One of the significant advantages of the Weda Bay project is its excellent logistics.
While the location is remote, it is easily accessible and other important logistic features
are equally favourable. These include:
The main resources (including Santa Monica) are close to the coast and are
generally accessible by road.
There is a relatively flat coastal strip suitable for the construction of a processing
plant and support facilities.
No people live in the resource areas or in the areas identified for the processing
plant site.
There is a plentiful permanent fresh water supply available from several rivers.
There is deep, sheltered bay suitable for a port located immediately adjacent the
proposed plant site.
Quality limestone deposits suitable for use in the process plant exist within the
COW.
There are plentiful sources of rock suitable for aggregate and road building.
A deep trench, suitable for Deep Sea Tailings Placement (DSTP) has been located
within 3 kilometers of the treatment plant.
The project concept which has been developed consists of a mine, commencing with the
large, consistent Santa Monica orebody and a processing facility to treat 4 million dry
tonnes for annum of limonitic ore of approximately 9% MgO using PAL and 1 million dry
tonnes per annum to a saprolite neutralization circuit. The possibility of supplementing
this concept with some direct atmospheric leaching of saprolite will be considered later.
The processing plant will be located immediately adjacent the port site, along with other
support facilities. The initial product will be a mixed nickel/cobalt sulphide for which a
long term sales contract has already been signed with a Finnish refinery. The output of
the plant is expected to be in the order of 60,000 tonnes of nickel per annum and 5000
tonnes of cobalt per annum.
14
6. PRE-FEASIBILITY STUDY
A number of scoping studies were completed during 2001 to define the size and nature
of facilities to be incorporated into a Pre Feasibility Study (PFS). These studies
indicated that the incorporation of saprolite neutralization into the PAL flowsheet
provided significant benefits by increasing output of intermediate product with only a
modest increase in capital. It was also shown that atmospheric leaching of additional
saprolite has the potential to further improve economics.The scoping studies also
showed a production rate of around 60,000 tonnes of nickel per annum is desirable to
provide robust returns.
A nominal production rate of 60,000 tonnes per annum of nickel and 5000 tonnes
per annum of cobalt in a sulphide intermediate product.
The mining strategy is still to be optimized and may be affected by additional resource
development.
A joint venture between Kvaerner E & C and Hatch was appointed to prepare a
preliminary design and cost estimate for the project. Based on a schedule of plant feed,
prepared by International Mining Consultants and the metallurgical testwork results
outlined in Section 4 of this report. Figure 5 is a conceptual layout of the major proposed
project facilities.
The capital cost estimate provided from the PFS, completed in April 2002, was US$902
million. Operating costs were estimated to be US$0.78 per pound of nickel in sulphide
intermediated before cobalt credits.
15
N
C o n s tr u c ti o n
1000 m Cam p
54 0 0 0 m N
R oa d
C o n s tr u c t io n
I n d u s t r ia l
A re a P rocess
P la n t
A re a
Layd own Tug
A re a M o o r in g s
W a te r
R e s e r v o ir
C o n ta in e r
S to r a g e
C o n s tr u c ti o n
O ffi c e s
52000 mN R oad S h ip p in g B e r th
Ta n k F a r m Cape
U lie
A ir p o r t B a rg e
la n d in g s W E D A B A Y
16