Content uploaded by Stacey Leichliter
Author content
All content in this area was uploaded by Stacey Leichliter on Sep 12, 2018
Content may be subject to copyright.
SME Annual Meeting
Feb. 24 - 27, 2013, Denver, CO
1 Copyright © 2013 by SME
Preprint 13-066
USING GEOMETALLURGICAL MODELS TO AID IN METALLURGICAL TESTWORK FOR PRE-FEASIBILITY PROJECTS, LA
COLOSA, COLOMBIA
S. Leichliter, AngloGold Ashanti, Colorado Springs, CO
R. Jahoda, AngloGold Ashanti Colombia, Cajamarca, Colombia
P. Montoya, AngloGold Ashanti Colombia, Cajamarca, Colombia
ABSTRACT
The La Colosa, Colombia porphyry gold deposit is currently in the
pre-feasibility stage. This large deposit has many types of gold
mineralization: native gold, electrum, and tellurides. These variations
in the gold mineralization lead to variability in the gold recovery. The
final process flowsheet may include gravity concentration with cyanide
leaching. Metallurgical testwork is utilized to identify the zones of ore
that may be problematic to recover. To help select the samples that
represent the fluctuating variability zones, geometallurgical models are
constructed using mineralogy and geology data. These zones are then
wireframed and compared to the geological and structural models for
spatial correlation. Samples for the variability testwork will be selected
from these different zones to identify and analyse the effect the
different gold mineralization types have on the recovery. By using
geometallurgy to aid in the sample selection for metallurgical testwork,
the variability in the ore body can be understood and constrained to aid
in process optimization.
INTRODUCTION
The La Colosa porphyry gold deposit is located approximately 7
kilometers northwest of the town of Cajamarca in the department of
Tolima in Colombia (Figure 1) (Lodder et al., 2010). It is wholly owned
by AngloGold Ashanti and is currently in the pre-feasibility stage of
permitting for mining.
La Colosa is on the eastern flank of the Central Cordillera. The
deposit is a large gold-only porphyry style deposit located in a cluster
of porphyritic intrusions. It is called a gold-only porphyry deposit due to
the trace amounts of copper and molybdenum present and the high
concentration of gold (Sillitoe, 2000; Sillitoe, 2007; and Lodder et al.,
2010). The gold mineralization presents as native gold, electrum, gold
tellurides, gold-silver tellurides, and gold associated with sulfides
(pyrite) (Lodder et al., 2010).
Geometallurgy is a key component to aid in sample selection for
the metallurgical variability testwork. The variability in the mineralogy
as well as its metallurgical response needs to be understood and
modeled to optimize recovery and reduce risk.
HISTORY
The La Colosa deposit was discovered in 2006 as a part of a
massive exploration campaign through Colombia by AngloGold
Ashanti in 2004 (Lodder et al., 2010). After the initial exploration
drilling in 2008, a resource of 467 tons of gold (Lodder et al., 2010)
was calculated. Currently, the site has completed over 110 diamond
drill holes for over 34,000 meters. The inferred resource estimate as of
2011 was 516 Mt at an average gold grade of 0.98 g/t (AngloGold
Ashanti, 2011). The La Colosa deposit is currently in the pre-feasibility
stage of permitting for mining.
GEOLOGY
La Colosa is located in Middle Cauca metallogenic belt of the
Central Cordillera in Colombia (Sillitoe, 2007). This area lies along a
structural zone at the junction of the Romeral-Cauca fault and
Palestina fault systems (Lodder et al., 2010). The Romeral-Cauca
fault is located to the west of the deposit with the Palestina fault to the
southeast (Ceidel et al., 2003; Sillitoe and Lodder et al., 2010). The La
Colosa normal fault runs along the eastern side of the deposit, striking
north to northwest and dipping northeast. There are numerous smaller
normal faults which strike north to northwest and east to northeast
(Figure 2) (Horner, 2011).
Weste rn
Cordillera
Eastern
Cordillera
Central
Cord illera
Ibague
La Colosa
Bogota
Colombia
Figure 1. Location of the La Colosa deposit and surrounding
cordilleras. Original map was from Sadalmelik, I., 2007.
The La Colosa deposit consists of multiple intrusions through the
schist and hornfels host rock (Lodder et al., 2010). The intrusions are
similar in texture and are determined by cross-cutting relationships.
These intrusions are the following rock types: early diorites,
intermineral diorites, and late dacites/quartz diorites (Figure 2) (Sillitoe,
2007 and Lodder et al., 2010).
The alteration assemblages associated with the gold
mineralization are potassic, potassic-calcic, and sodic-calcic (Sillitoe,
SME Annual Meeting
Feb. 24 - 27, 2013, Denver, CO
2 Copyright © 2013 by SME
2007 and Lodder et al., 2010). There are minor amounts of propylitic,
intermediate argillic, and quartz-sericite alteration. Sodic-calcic
alteration usually overprints the potassic or potassic-calcic alteration
(Sillitoe, 2007 and Lodder et al., 2010). The major vein types
associated with gold are quartz-sulfide (A type), quartz-sulfide with
suture (B type), sulfide veins (S type), and sulfide-quartz veins (D type)
(Sillitoe, 2007 and Lodder et al., 2010). Gold grades are associated
with the vein frequency (higher number of veins, higher gold values)
(Lodder et al., 2010).
Figure 2. Geologic and structural map of the La Colosa deposit
(Horner, 2011).
GOLD MINERALIZATION
The gold mineralization at La Colosa consists of native gold,
electrum, gold tellurides, gold-silver tellurides, and gold locked with
sulfides (pyrite) (Leichliter et al., 2011 and Leichliter, 2012). Higher
gold grades are also associated with a structure along the contact with
the hornfels (host rock) which strikes northwest. The tellurides appear
to concentrate near this same contact region, but are also found in all
the major rock types (Leichliter et al., 2011 and Leichliter, 2012).
Gold mineralization appears to be predominantly related to the
veining, although a portion of the gold is disseminated in the matrix of
the early diorites. The overall minerals directly associated with the
gold mineralization are K feldspar and pyrite (Leichliter et al., 2011 and
Leichliter, 2012). The average gold grain size is 3-4 microns (very fine
grained), although there has been coarse gold grains observed at 35-
50 microns (Leichliter et al., 2011 and Leichliter, 2012) (Figure 3).
The native gold grains have an average grains size of about 4
microns with dominant mineral associations with K feldspar and pyrite
(Leichliter et al., 2011 and Leichliter, 2012). Locked native gold grains
were found in plagioclase and pyrrhotite (Leichliter et al., 2011 and
Leichliter, 2012).
Figure 3. Examples of mineral gold grains under the SEM (Leichliter,
2012).
The gold telluride grains have an average grain size of 3 microns
and are associated with K feldspar and pyrite with a majority liberated
(Leichliter et al., 2011 and Leichliter, 2012). A small amount of the
gold telluride grains locked was in K feldspar and pyrite (Leichliter et
al., 2011 and Leichliter, 2012).
The gold-silver telluride grains have an average grain size of 2
microns and are found with apatite, albite, and pyrite. Most grains are
locked in the same minerals they are associated (Leichliter et al., 2011
and Leichliter, 2012).
GEOMETALLURGY
By understanding the gold mineralization and its deportment, the
possible recovery methods can be evaluated. This is where
geometallurgy is vital in understanding the variability in the gold
mineralization and its recovery response. Metallurgical variability
testwork allows for the understanding and quantifying of that recovery
response of the different ore types. Geometallurgy can assist in the
sample selection by predicting the variable recoveries due to
mineralogical and geological parameters.
Recovery
The recovery flowsheet for the La Colosa deposit is still being
evaluated. Gravity concentration, flotation concentration, and cyanide
leaching have been analyzed for efficiency. Gold and gold-silver
tellurides are not known to recover well with cyanide leaching
(Marsden and House, 2006). The grain size of the gold is fine, so
gravity may or may not be the best method. Since the two
SME Annual Meeting
Feb. 24 - 27, 2013, Denver, CO
3 Copyright © 2013 by SME
predominant minerals associated with the gold are K feldspar and
pyrite, flotation followed by cyanide leach might be an option.
Using what information and data available, models were
constructed to determine areas to sample for gold amendable to
flotation or gravity concentration and the zones with high possibility of
tellurides. A gold cut-off grade of 1.0 g/t was used for the models.
For the model to predict zones for possible flotable and gravity
gold, multi-element assays for gold and sulfur were analyzed and
grouped into domains for flotable gold (high gold and high sulfur) and
free gold (gravity) (high gold and low sulfur) (Figure 4).
Figure 4: Domains showing possible flotatable and free (gravity) gold.
Using these domains, wire frames were constructed to show
spatial relationships and locations. The contact between the host rock
(hornfels/schist) and the intrusives was also plotted to see the
relationship between the different major lithologies (Figure 5). The
light pink wire frames are intermineral diorites and the light blues are
early diorites.
Figure 5. Free (green-blue) and flotable (pink) modeled at 1.0 g/t gold
(Oblique view). Contact is purple.
Looking at the resultant wire frames and their geologic
relationships, it appears that a majority of the flotable gold (dark pink)
is located above the contact with hornfels/schist (purple). Also, the
free (gravity) gold wire frames (green-blue) appear to be mostly below
that contact.
To construct the telluride model, similar methods were utilized to
determine the telluride-rich zones of the deposit (Figure 6). The
domains were then again wire framed to show spatial relationships and
locations. The same contact and lithologies were used again to show
any relationships (Figure 7).
Looking at the telluride-rich wire frames (green) and their
relationship with the major lithologies, it appears that the telluride-rich
zone is along the contact, but is both above and below it.
Comminution
Geometallurgy is used to understand and constrain the variability
in the comminution response of the different rock types at the La
Colosa deposit. A combination of comminution, geometallurgical, and
mineralogical testwork was used to develop an initial comminution
model. This preliminary model provides targets in zones with
variability for future comminution testwork. This initial comminution
model was developed by P. Montoya (2011).
Figure 6. Domains showing possible telluride-rich gold.
Figure 7. High tellurium group (bright green) modeled at a gold grade
of > 1.0 g/t (Oblique view).
The comminution testwork consisted of JKRBT Lite (A*b) and
Bond Mill Work index (BMWi) tests (Montoya et al., 2011). The
geometallurgical testwork included Equotip and GeMCi (Montoya et al.,
2011). Mineralogical testwork gathered were multi-element assays,
QXRD data, and Mineral Liberation Analyzer (MLA) data (Montoya et
al., 2011). These datasets were combined with the data in the drill
hole database (i.e. density, rock type, alteration, minerals, and
veinlets) (Montoya et al., 2011).
Statistical analysis, using principal components analysis, class-
based analysis, predictive modeling, and process performance
domaining, was used to develop and estimate comminution indices
A*b and BMWi (Montoya et al., 2011).
Rock type usually is a primary control for constraining the
comminution response, but at La Colosa this does not apply (Figures 8
and 9) (Montoya et al., 2011). At La Colosa it was determined that
rock type, alteration, and gold grade do not control and constrain the
variability in the comminution response (Montoya et al., 2011).
Figure 8. BMWI response as related to rock type and gold grade
(Montoya et al., 2011).
SME Annual Meeting
Feb. 24 - 27, 2013, Denver, CO
4 Copyright © 2013 by SME
Figure 9. A*b response as related to rock type and gold grade
(Montoya et al., 2011).
The gangue mineralogy (from mineralogy testwork) was used in
the principal components analysis (PCA) to constrain the variability
within the comminution dataset (Montoya et al., 2011). The broad
mineralogical groups associated with the comminution response were:
feldspar, chalcopyrite, and pyrite (soft rocks) and Equotip, magnetite,
chlorite, albite, and density (hard rocks) (Montoya et al., 2011). A
discriminant diagram was produced from the PCA to define the
different classes observed in the multivariate scatterplot. Multiple
linear regressions proxy support models were constructed by class to
estimate values for A*b and BMWi (Montoya et al., 2011). The BMWi
values ranged from 12.8 to 23.62, and the A*b values ranged from
27.82 to 44.23 (Montoya et al., 2011).Using the regression equations
developed, A*b and BMWi was estimated for the drill hole database
and the results were domained and wireframed (Figures 10 and 11)
(Montoya et al., 2011).
Figure 10. Wireframes of the different BMWi domains for La Colosa
(Montoya et al., 2011).
CONCLUSION
Using geometallurgy to predict the recovery and comminution
behavior due the mineralogy, aids in sample selection so that a wide
variety of testwork can be completed to determine the most efficient
method to recover the gold. By constructing domains to represent
possible zones of flotable, free, and telluride-rich gold, future drill
targeting and sampling can help define the areas of varying
metallurgical response. Wireframes for the comminution response
variables, BMWi and A*b, can also provide targets for further
metallurgical comminution testwork. Delineating these variabilities will
help in optimizing the comminution and recovery processes and
reducing the risk associated with extracting the gold.
Figure 11. Wireframes of the different A*b domains for La Colosa
(Montoya et al., 2011).
REFERENCES
1. AngloGold Ashanti, 2011, Mineral Resource and Reserve
Report 2011, www.anglogold.co.za.
2. Ceidel, F., Shaw, R.P., and Caceres, C., 2003, “Tectonic
assembly of the Northern Andean block” , in Bartolini, C., R.T.,
and Blickwedw, J., eds., The Circum-Gulf of Mexico and
Carribean: Hydrocarbon habitats, basin formation, and plate
techtonics, American Association of Petroleum Geologists
Memoir, v. 79, p. 815-848.
3. Horner, J., 2011, “Structural Geologic Study (PFS) Final Report
for La Colosa Gold Mine”, iC consulenten group, AngloGold
Ashanti internal report. 58p.
4. Leichliter, S., Hunt, J., Berry, R., Keeney, L., Montoya, P.,
Chamberlain, V., Jahoda, R., and Drews, U., 2011, “Development
of a Predictive Geometallurgical Recovery Model for the La
Colosa, Porphyry Gold Deposit, Colombia”. GeoMet 2011
conference proceedings, AusIMM., p. 85-91.
5. Leichliter, S., 2012, “Gold Deportment and Geometallurgical
Recovery Model for the La Colosa, Porphyry Gold Deposit,
Colombia”, Thesis, University of Tasmania, Australia, 186 p.
6. Lodder, C., Padilla, R., Shaw, R., Garzon, T., Palacio, E., and
Jahoda, R., 2010, “Discovery History of the La Colosa Gold
Porphyry Deposit, Cajamarca, Colombia”, Society of Economic
Geologists Special Publication, v. 15, p. 19-28.
7. Marsden, J.O. and House, C.I., 2006. The Chemistry of Gold
Extraction, Second Edition, Society of Mining, Metallurgy, and
Exploration, Inc., Littleton, Colorado, 651 p.
8. Montoya, P., Keeney, L., Jahoda, R., Hunt, J., Berry, R., Drews,
U., Chamberlain, V., and Leichliter, S., 2011, “Geometallurgical
Modelling Techniques Applicable to Pre-Feasibility Projects: La
Colosa Study.” GeoMet 2011 conference proceedings, AusIMM.,
p. 103-111.
9. Sadalmelik, I., 2007, Topographic map of Colombia, Colombia
Topography.png,
http://commons.wikimedia.org/wiki/File:ColombiaTopography.png.
10. Sillitoe, R.H., 2000, “Gold-Rich Porphyry Deposits: Descriptive
and Genetic Models and Their Role in Exploration and
Discovery”, Society of Economic Geologists, SEG Reviews, v. 13,
p. 315-345.
11. Silltoe, R.H., 2007, Preliminary Geological Model for the Colosa
Porphyry Gold System: Bogota, Colombia, AngloGold Ashanti
internal report, 13 p.
