Figure 1 - uploaded by Stacey Leichliter
Content may be subject to copyright.
Shake leach assay testwork setup. (http://www.opticsplanet.com/eberbach-heavy-dutyshaker-two-speed-eberbach-6010.html)
Source publication
The goal of this project was to develop a predictive geometallurgical recovery model for the La Colosa porphyry gold deposit using the gold deportment, analytical data (multi-element assays), mineralogy, and recovery data. The aim of geometallurgy is to reduce risk and uncertainty by understanding the variability within the ore body, to increase th...
Contexts in source publication
Context 1
... shake leach assay is a quick, optimistic analysis to determine the amount of cyanide extractable gold. Testwork involves a crushed or pulverised rock, 200 mesh (74 µm), and sample placed into a vessel with a cyanide solution (NaCN) and placed on an agitator (Eberbach shakers) for 1 hour (Figure 1). The solution is then analysed for the concentration of gold extracted by the cyanide. ...
Context 2
... predominant alteration is potassic with biotite replacing amphibole and plagioclase. Intrusion breccia EBX1 has clasts of E1 in a medium to very fine-grained matrix ( Figure 11). This intrusive breccia is not a common rock compared to the later intrusive phases. ...
Context 3
... dominant alteration is potassic with similar textures to E1. Early diorite unit, E3, is a fine-grained diorite porphyry with a groundmass composed of quartz, plagioclase, K feldspar, and amphibole ( Figure 14). Phenocrysts are plagioclase and amphibole (< 3.5 mm) with an abundance of approximately 20% (Gil- Rodriguez, 2010). ...
Context 4
... The second intermineral unit, I2, is a diorite porphyry with a very fine-grained matrix of plagioclase, amphibole, and quartz ( Figure 18). The plagioclase, amphibole, and quartz phenocrysts (40%) are fine to medium grained (< 5 mm) (Gil-Rodriguez, 2010). ...
Context 5
... dacite or late diorite porphyry unit in Figure 16 is the largest body of this group. ...
Context 6
... forms a stock with a diameter of 1.6 km in the north to northeastern part of the deposit. It has a very fine-grained matrix of plagioclase, amphibole, biotite, and quartz ( Figure 19). Plagioclase, quartz, biotite, and amphibole phenocrysts (approximately 30%) are fine to coarse-grained (< 1 cm) (Gil-Rodriguez, 2010). ...
Context 7
... geological and alteration cross sections show the lithologies ( Figure 41 (Table 5) were also compared with the QXRD data for the weakly altered diorite and hornfels samples (Table 6). This comparison between the strongly altered Phase 1 and Phase 2 XMOD samples and the weakly altered diorite and hornfels ...
Context 8
... size distribution curves for all the gold-silver telluride samples showed the largest gold-silver telluride was 22 µm, and the smallest was at the detection limit of 0.87 µm ( Figure 51 and Table 14). A grain size distribution curve for the two grains found in the early diorite showed the largest grain was 9.6 µm with the smallest at 8.1 µm (Table 14). ...
Context 9
... to the few gold-silver telluride grains observed, it was difficult to understand the grain size distribution for the early diorite and hornfels. Figure 51: Grain size distribution of gold-silver tellurides (n = 63) in the Phase 1 and Phase 2 samples by lithology: early diorite (n=2), intermineral diorite (n=60), and hornfels (n=1). ...
Context 10
... recoveries for the drill hole data were then estimated using the same regression equations established for the classes from the diagram ( Table 23). The recovery data was domained and then wireframed to check the spatial continuity throughout the deposit ( Figure 71). ...
Context 11
... used to develop the XMOD recovery model. The recovery data was the same dataset used with the multi-element model for Model 2, so no outliers were identified or removed ( Figure 81). The same methods that were used in Model 1 and Model 2 multi- element were utilised to construct this model. ...
Context 12
... input variables for the PCA used in this were pyrite, albite, and actinolite (Table 30). The eigenvectors from the PCA analysis showed albite and gold having the largest contribution in PC1 (Table 31) There were two classes identified within the clusters, high arsenopyrite and low arsenopyrite (Figures 100 and 101). A discriminant diagram was constructed from these classes ( Figure 101). ...
Context 13
... eigenvectors from the PCA analysis showed albite and gold having the largest contribution in PC1 (Table 31) There were two classes identified within the clusters, high arsenopyrite and low arsenopyrite (Figures 100 and 101). A discriminant diagram was constructed from these classes ( Figure 101). A regression model was calculated for each class to estimate the recovery by bottle roll test (extractable gold) (Table 32). ...
Context 14
... regression models showed a high strength of fit (R 2 > 0.96) and an average relative error of ~11% when compared to the analytical error of the tests (fire assay and shake leach assay) (< 10%). When the measured shake leach recovery was plotted against the measured shake leach recovery (ppm) (Figure 102), it showed that the calculated gold Page | 137 recovery was between 70% and 90% while the measured shake leach recovery had a greater spread or more variability with values between 40% and 100%. Looking at the gold grades as a possible cause of the larger errors in the calculated shake leach recoveries (ppm), it appears that the higher gold grade, the greater the error ( Figure 103). ...
Context 15
... the measured shake leach recovery was plotted against the measured shake leach recovery (ppm) (Figure 102), it showed that the calculated gold Page | 137 recovery was between 70% and 90% while the measured shake leach recovery had a greater spread or more variability with values between 40% and 100%. Looking at the gold grades as a possible cause of the larger errors in the calculated shake leach recoveries (ppm), it appears that the higher gold grade, the greater the error ( Figure 103). This could be that the coarser or larger gold grains aren't completely leaching with the cyanide. ...
Context 16
... simple linear regression model was calculated using the fire assay and the measured shake leach assay data (Figure 104), so this model would be considered the simplest model to predict recoverable gold. The regression analysis equation to predict cyanide recoverable gold was: An F -test was performed on the first recovery model and a linear recovery model based on the prior testwork (Figures 105 and 106). ...
Context 17
... simple linear regression model was calculated using the fire assay and the measured shake leach assay data (Figure 104), so this model would be considered the simplest model to predict recoverable gold. The regression analysis equation to predict cyanide recoverable gold was: An F -test was performed on the first recovery model and a linear recovery model based on the prior testwork (Figures 105 and 106). An F-test allows for the comparison of two different models to determine which model is best fitted to the data. ...
Context 18
... difference in absolute errors is negligible. The estimated shake leach assays were then wireframed and compared with Model 1's wireframes ( Figure 107). There was noticeable difference in the wireframes, which could be due to the mineralogical aspects used in the geometallurgical model (Model 1). ...
Context 19
... the Phase 1 sample data was estimated using the bottle roll test recovery regression equations and domained, using the same parameters as in Model 1 (low <70%, average 70%-85%, and high >85%). These domains were wireframed (3 dimensional views) to visualize the variability in the estimated bottle roll gold recovery and to check spatial continuity and relationships with the geology (Figure 108). When comparing the location of the gold recovery domains with lithology maps, the wireframes overlapped the major lithologies (Figures 7 and 108). ...
Context 20
... domains were wireframed (3 dimensional views) to visualize the variability in the estimated bottle roll gold recovery and to check spatial continuity and relationships with the geology (Figure 108). When comparing the location of the gold recovery domains with lithology maps, the wireframes overlapped the major lithologies (Figures 7 and 108). The low recovery areas compared with the location of early diorite lithologies (Figures 7 and 108). ...
Context 21
... comparing the location of the gold recovery domains with lithology maps, the wireframes overlapped the major lithologies (Figures 7 and 108). The low recovery areas compared with the location of early diorite lithologies (Figures 7 and 108). The areas with high gold recovery by bottle rolls corresponded with areas underlain by the intermineral diorites (Figures 7 and 108). ...
Context 22
... low recovery areas compared with the location of early diorite lithologies (Figures 7 and 108). The areas with high gold recovery by bottle rolls corresponded with areas underlain by the intermineral diorites (Figures 7 and 108). The results of this study suggested that much of the La Colosa ore (with the exception of mineralisation hosted in the early diorites) may be amenable to gold extraction via cyanide leaching. ...
Context 23
... not yet available for Phase 2 samples, so this model could not be verified at this When comparing the wireframes of the Phase 2 estimated shake leach gold recoveries with those of Model 1 (Figure 108), there was some difference. Mainly, the high recovery domain was smaller than Model 1( Figure 111). ...
Context 24
... not yet available for Phase 2 samples, so this model could not be verified at this When comparing the wireframes of the Phase 2 estimated shake leach gold recoveries with those of Model 1 (Figure 108), there was some difference. Mainly, the high recovery domain was smaller than Model 1( Figure 111). The average and low recovery domains were consistent with Model 1. Differences in the wireframes could possibly be due to average relative error in the geometallurgical model. ...
Similar publications
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 lea...
The strategic mine plan is a crucial step for the success of mining companies, and for its development, it is necessary to use a number of inter-related variables that are usually estimated independently. These variables include operational data that is traditionally isolated in information islands between the different departments in the mine or t...
Citations
... Mineral Liberation Analyser, QEMSCAN, Mineralogic, TIMA-X), allow quantitative data on these characteristics to be acquired both quickly and relatively cheaply. Due to the wealth of information these techniques provide, they are commonly applied in geometallurgical studies Frenzel et al., 2018;Kern et al., 2018;Leichliter, 2013). They are also frequently used to numerically constrain the deportment of rare and precious metals (e.g. ...
Scanning electron microscope-based automated mineralogy studies are readily associated with quantitative results, providing one of the foundations of geometallurgical studies. Despite the importance of quantitative data for such studies, and efforts to reduce statistical errors, the reporting of uncertainties is rare. This contribution illustrates how bootstrap resampling can be used to provide robust estimates of statistical uncertainties for the modal mineralogy, metal deportment and all relevant textural attributes of a sample or a series of samples. Based on a case study of the Bolcana Au-Cu porphyry deposit in the South Apuseni Mountains, Romania, the impact of insufficient sampling statistics on quantitative mineralogical studies is illustrated. Quantitative analyses of the mineralogy and microfabric of milled ore samples from seven 40 m drill core intervals from the Bolcana Prospect were conducted using a Mineral Liberation Analyser (MLA), complemented by electron probe micro-analysis. Bootstrap resampling was then applied to assess how many grain mount surfaces should be analysed to achieve statistically robust results for both Cu and Au mineralogy, deportment and textural attributes. Despite variable mineralogy, grades and mineralisation styles, estimated statistical uncertainties on Cu deportment are consistently low. In contrast, uncertainties for Au deportment are so high that most reported values for important characteristics are statistically meaningless. This is mainly attributed to the pronounced nugget effect for Au mineralisation, exacerbated by the small sample size analysed by MLA. An unfeasible number of measurements would be necessary to provide robust figures for the deportment of minor/trace elements and minerals, along with other tangible mineralogical properties, such as mineral associations. The results of this case study demonstrate that statistical uncertainties need to be carefully incorporated when considering the results of automated mineralogical studies and their impact on geometallurgical models. This is particularly relevant for studies of precious metal ores.
... Mineral Liberation Analyser, QEMSCAN, Mineralogic, TIMA-X), allow quantitative data on these characteristics to be acquired both quickly and relatively cheaply. Due to the wealth of information these techniques provide, they are commonly applied in geometallurgical studies Frenzel et al., 2018;Kern et al., 2018;Leichliter, 2013). They are also frequently used to numerically constrain the deportment of rare and precious metals (e.g. ...
Scanning electron microscope-based automated mineralogy studies are readily associated with quantitative results, providing one of the foundations of geometallurgical studies. Despite the importance of quantitative data for such studies, and efforts to reduce statistical errors, the reporting of uncertainties is rare. This contribution illustrates how bootstrap resampling can be used to provide robust estimates of statistical uncertainties for the modal mineralogy, metal deportment and all relevant textural attributes of a sample or a series of samples. Based on a case study of the Bolcana Au-Cu porphyry deposit in the South Apuseni Mountains, Romania, the impact of insufficient sampling statistics on quantitative mineralogical studies is illustrated. Quantitative analyses of the mineralogy and microfabric of milled ore samples from seven 40 m drill core intervals from the Bolcana Prospect were conducted using a Mineral Liberation Analyser (MLA), complemented by electron probe micro-analysis. Bootstrap resampling was then applied to assess how many grain mount surfaces should be analysed to achieve statistically robust results for both Cu and Au mineralogy, deportment and textural attributes. Despite variable mineralogy, grades and mineralisation styles, estimated statistical uncertainties on Cu deportment are consistently low. In contrast, uncertainties for Au deportment are so high that most reported values for important characteristics are statistically meaningless. This is mainly attributed to the pronounced nugget effect for Au mineralisation, exacerbated by the small sample size analysed by MLA. An unfeasible number of measurements would be necessary to provide robust figures for the deportment of minor/trace elements and minerals, along with other tangible mineralogical properties, such as mineral associations. The results of this case study demonstrate that statistical uncertainties need to be carefully incorporated when considering the results of automated mineralogical studies and their impact on geometallurgical models. This is particularly relevant for studies of precious metal ores.
... No existe en la literatura gran información sobre aplicaciones de la geometalurgia a yacimientos de oro, muy probablemente debido a que un problema fundamental es el muestreo de este tipo de yacimientos (Pitard, 1989a y b;Goodall y Scales, 2007). Los trabajos más destacados con sistemas auríferos proceden de tesis desarrolladas en universidades australianas (Leichliter et al., 2011;Bonnici, 2012;Leichliter, 2013). El objeto del presente trabajo es el de dar a conocer los resultados más relevantes de un ensayo de modelo de recuperación geometalúrgica efectuado en la mina La Herradura en Sonora (Romero-Valle, 2017), una de las minas de oro más importantes de México, tomando en cuenta el análisis de la información obtenida a partir de estudios con microscopía de reflexión, concentración gravimétrica, Mineral Liberation Analyzer (MLA) y pruebas de cianuración dinámica. ...
La Herradura mine in Sonora is one of the most important gold districts in Mexico producing more than 5.5 million ounces of gold over 17 years. It is part of an orogenic gold deposits belt with a northwest-southeast direction for 300 km long and 50 kmwide. The mineralization consists of veins and quartz networks veinlets formed in a brittle-ductil geologic environment and it is hosted in Proterozoic quartz-feldspar gneisses. The ore bodies, defined by a 0.3 g/t Au cut-off grade, have tabular forms up to 1 km in length, 1 km in depth and 100 m in width. The visual control for mineralization is the abundance of quartz veins and veinlets, and a persistent sericitic hydrothermal alteration. Different techniques were used in this work with special focus on the Mineral Liberation Analizer (MLA) program to prove the possibility of predicting recoverable gold in the mine for the primary sulphide zone. Three geometallurgical zones (Zones A, B and C) were defined by ore composites and gravimetric concentrates from the same composites. Modal composition of the concentrates is quartz, feldspar and muscovite (sericite), and a metallic mineralogy of pyrite, sphalerite, galena, magnetite, gold and tellurides of gold and silver. Gold is identified as inclusions in pyrite or in gangue minerals like quartz, albite, orthoclase or ankerite, as well as coating pyrite crystals. The gold composition is electrum with 74 % Au and 26 % Ag; the presence of petzite (Ag3AuTe2) and stutzite (AgTe) were also identified. Recovery constants were calculated for each geometallurgical zone, which were introduced to the resource model of more than 14 million ounces of gold, indicating that error range in recoverable gold is less than 4 % for Zone A, 6 % for Zone C and 13 % for Zone B, in relation with gold recovery calculated with traditional methods. These results could be acceptable to applicate this methodology to La Herradura deposit. The most important error range in the Zone B is interpreted as due to a nugget effect, which isvery common in such mineral deposits. It is also concluded that the secondary milling process currently incorporated to the metallurgical plant is probably unnecessary, so its removal would result in a significant saving in energy and therefore in the economy of the mine.
... The gold mineralization at La Colosa consists of native gold, electrum, gold tellurides, gold-silver tellurides, and gold locked with sulfides (pyrite) Leichliter, 2012). Higher gold grades are also associated with a structure along the contact with the hornfels (host rock) which strikes northwest. ...
... 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, 2012). ...
... 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 liberatedLeichliter, 2012). A small amount of the gold telluride grains locked was in K feldspar and pyriteLeichliter, 2012).The gold-silver telluride grains have an average grain size of 2 microns and are found with apatite, albite, and pyrite. ...
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.
... Metallurgical and geometallurgical studies on composited and selective samples indicate that gold occurs predominantly as native gold, as electrum, and in minor quantities as gold tellurides and gold-silver tellurides (Leichliter, 2013). Gold occurs as isolated grains and as inclusions or fracture fillings in pyrite, pyrrhotite, and silicate minerals such as feldspar and quartz (Fig. 12). ...
The La Colosa porphyry Au deposit is located on the eastern flank of the Central Cordillera of Colombia, within the Middle Cauca metallogenic belt. The deposit contains more than 800 t (22.37 Moz) Au at grades up to 0.8 g/t and is hosted by a composite porphyry stock of dioritic to tonalitic composition, which was emplaced into Triassic-Cretaceous schists of the Cajamarca Complex in the late Miocene (~8 Ma). The country rocks underwent two ductile deformation events, including development of shear zones, folds, and penetrative foliation, prior to emplacement of the stock. Subsequent brittle deformation reactivated preexisting N- and NNE-trending structures and formed secondary faults due to a change from right- to left-lateral shear sense on regional faults. This switch in stress orientations is attributed to a new plate configuration in the mid-Miocene. The left-lateral movement along regional faults favored emplacement of intrusive centers in dilational pull-apart zones, including the La Colosa stock within the regional Palestina fault zone. The La Colosa porphyry stock was intruded in three stages, termed early, intermineral, and late, all within a relatively short time interval of ~1.1 m.y. The early and intermineral stages are diorite porphyries and related intrusion breccias, whereas the late stage consists of quartz diorite and tonalite porphyries. The intrusions caused contact metamorphism and hydrothermal alteration that partially obliterated the original texture and composition of the schistose country rocks. The early and intermineral stages are dominated by potassic alteration, with local chloritic alteration in the core of the intermineral stage, sodic-calcic alteration in the deeper parts of the stock, and propylitic alteration confined to the late stage. Three gold mineralization events are recognized at La Colosa. The first was of porphyry style, during which hypersaline fluids (40-50 wt % NaCl equiv) formed predominantly A- and S-type veinlets and caused multistage wall-rock silicification accompanied by potassic and sodic-calcic alteration. The early-stage intrusions contain the highest gold grades varying from 0.75 to 1 g/t Au, associated with pyrite and minor chalcopyrite, molybdenite, and magnetite in the porphyries and with pyrrhotite-pyrite-melnikovite in the country rocks. In the intermineral-stage intrusions the gold grades drop to 0.5 to 0.75 g/t Au, and pyrrhotite and pyrite are the major sulfides. Gold grades reach low values of <0.3 g/t Au in the late-stage porphyries. The second gold-precipitating event formed sheeted veinlets of drusy quartz and pyrite with centimeter-wide halos of albite-sericite-pyrite overprinting all other alteration types at the deposit. A ~200°C hydrothermal brine (21-28 wt % NaCl equiv) deposited gold at high grades (>1.5 g/t Au over >10 m drill core intervals) within N-striking normal faults that developed during and after emplacement of the porphyry stock. The third mineralization event was supergene, with Au enrichment confined to late porphyries and characterized by sulfide boxworks, resulting in gold grade increases from 0.3 to 1.2 g/t Au. © 2018 Gold Open Access: this paper is published under the terms of the CC-BY-NC license.
With the increase in complexity of mineral deposits, understanding the grade variability is of high significance in getting enough information about the deposit, preventing financial losses and identifying optimized plans during and after exploitation. Likewise, mineralogical variability, grade, alteration … can highly effect metal recovery. Modeling these parameters is seriously substantial in order to have an optimized exploitation of the mine and to minimize the risks and costs. This study used the dataset of 91 drill holes from Sari Gunay epithermal gold deposit located in north-west of Iran to model gold grade and the recovery resulted from leaching tests. Initially, the oxide and sulphide zones were distinguished using Indicator Kriging and Simulation. Then the gold grade and recovery in each zone were modeled with the application of sequential Gaussian simulation and the developed block model. The grade-recovery curve was then drawn to estimate the total tonnage amount of the deposit which was related to the sulphide zone in more than 70%. Then with the help of recovery simulation data, the was considered a preferable parameter in depicting the curve and estimating the reserve.
The quantitative and qualitative assessment of gold grains from samples of glacial till is a well-established method for exploring gold deposits hidden under glaciated cover. This method, which is widely used in the industry and has resulted in numerous successes in locating gold deposits in glaciated terrain, is still based on artisanal gravity separation techniques and visual identification. However, being artisanal, it is limited by inconsistent recoveries and difficulties associated with visually identifying the predominantly small gold grains. These limitations hinder its capacity to decipher subtle or complex signals. To improve detection limits through the recovery of small gold grains, a new approach has recently been introduced into the industry, which is commercially referred to as the “ARTGold” procedure. This procedure involves the use of an optimized miniature sluice box coupled with an automated scanning electron microscopy routine. The capabilities of this improved method were highlighted in this study by comparing till surveys conducted around the Borden gold deposit (Ontario, Canada) using the conventional and improved methods at both local and regional scales. Relative to that with the conventional approach, the improved method increased the recovery of gold grains from samples (regional and down-ice mineralization) by almost one order of magnitude. (regional and down-ice mineralization), dominantly in regard of the small size fractions. Increasing the counts in low-abundance regional samples allows for a better discrimination between background signals and significant dispersions. The described method offers an alternative for improving the characterization of gold dispersal in glaciated terrain and related gold deposit footprints.
The Biely Vrch deposit in the Western Carpathians is assigned to the shallow, sulfide-poor porphyry gold deposit type and has an exceptionally low Cu/Au ratio. According to 3-D geochemical models, there is a limited spatial correlation between Au and Cu due to the primary introduction of gold by a salt melt and Cu by low-density vapor. Despite a rough spatial correlation of gold grades with quartz stockwork intensity, gold is hosted mostly by altered rock, exclusively in native form. Three main gold mineral assemblages were recognized here. In the deepest parts of the system, the K- and Ca-Na silicate gold assemblage is associated with minerals of high-temperature alteration (plagioclase, K-feldspar, actinolite), with gold grades and fineness depending on depth and potassium content of the host rock: K-silicate alteration hosts the lowest fineness gold (~ 914), whereas Ca–Na silicate alteration has the highest (~ 983). The intermediate argillic gold assemblage is the most widespread, with gold hosted mainly by chlorite, illite, smectite, and interstratified illite–chlorite–smectite minerals. The gold fineness is mostly variable (875–990) and inherited from the former gold mineral assemblages. The latest advanced argillic gold assemblage has its gold mostly in kaolinite. The extremely high fineness (~ 994) results from gold remobilization by late-stage aqueous magmatic-hydrothermal fluids. Uncommon bonanza-grade appears where the earlier gold mineral assemblages were further enriched by this remobilized gold. Primary precipitation of gold occurred during ascent and cooling of salt melts at 450 to 309 °C, mostly during retrograde quartz solubility.
