- Email: [email protected]
sulfur-oxidizing mixotrophic bacteria
Improving Gold Recovery from Refractory Gold Ores Through Biooxidation using Iron-Sulfur-Oxidizing/Sulfur-Oxidizing Mixotrophic Bacteria M.Z. Mubarok, R. Winarko, S.K. Chaerun, I.N. Rizki, Z.T. Ichlas PII: DOI: Reference:
S0304-386X(16)30756-3 doi:10.1016/j.hydromet.2016.10.018 HYDROM 4454
To appear in:
Hydrometallurgy
Received date: Revised date: Accepted date:
2 April 2016 20 September 2016 17 October 2016
Please cite this article as: Mubarok, M.Z., Winarko, R., Chaerun, S.K., Rizki, I.N., Ichlas, Z.T., Improving Gold Recovery from Refractory Gold Ores Through Biooxidation using Iron-Sulfur-Oxidizing/Sulfur-Oxidizing Mixotrophic Bacteria, Hydrometallurgy (2016), doi:10.1016/j.hydromet.2016.10.018
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Improving Gold Recovery from Refractory Gold Ores Through Biooxidation using Iron-Sulfur-Oxidizing/Sulfur-Oxidizing Mixotrophic Bacteria
T
M.Z. Mubaroka*, R. Winarkoa, S.K. Chaeruna,b, I.N. Rizkic and Z. T. Ichlasa Department of Metallurgical Engineering, Faculty of Mining and Petroleum Engineering, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, Indonesia
b)
Geomicrobiology-Biomining & Biocorrosion Laboratory, Microbial Culture Collection Laboratory, Biosciences and Biotechnology Research Center, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, Indonesia
c)
GeoAssay Metallurgical Laboratory, Geoservices Company, Cikarang Barat 17530, Indonesia
NU
SC R
IP
a)
MA
*Corresponding author: [email protected], +62 (22) 2502239
AC
CE P
TE
D
Abstract The effect of biooxidation (BIOX) pre-treatment of refractory gold concentrates using mixotrophic and chemolithotrophic bacteria on the gold extraction during cyanidation of the concentrates at neutral pH was studied. A series of biooxidation and cyanidation of the biooxidized concentrates was carried out using three strains of bacteria; two mixotrophic bacteria of SKC1 and SKC2 and a chemolithotrophic bacteria Acidithiobacillus ferroxidans (AC). Two distinct types of refractory gold concentrates with high sulfur content (i.e. S>20%) and low sufur content (i.e. S<5%) were used. The experimental results showed that biooxidation with any of the three bacteria generally showed positive effect on the gold extraction. The highest gold extraction (91.4%) from high sulfur-concentrate was achieved by BIOX with SKC2 for 14 days. It was 18% higher than the extraction level of direct cyanidation of the untreated concentrate. Biooxidation with AC, on the other hand, resulted in only slight increase of gold extraction from both concentrates due to the high solution pH of >5.0 which is not a suitable living environment for the bacteria. Cyanide consumption for cyanidation of the biooxidized concentrates with both mixotrophic bacteria was significantly increased due to reactions of sulfur and iron species that were precipitated during the biooxidation step with cyanides to form thiocyanate and ferrocyanide ions. Keywords: biooxidation, mixotrophic bacteria, gold, refractory ore, preg-robbing, neutral pH Introduction Cyanidation is the most common method for gold extraction from its ore or concentrate. However, direct cyanidation is only effective for treating non-refractory ores. Direct cyanidation of refractory ores would result in inadequate gold recoveries due to various reasons. The inadequate recoveries can be associated with the presences of fine gold entrapped within sulfide minerals and/or carbonaceous materials in the ores (Aylmore and Jaffer, 2012, Yang et.al., 2015). The fine gold particles which are locked inside non-porous sulfide minerals cannot be extracted due to hindrance of the reaction between gold and cyanide ions, while the presence of naturally occurring carbonaceous matters (i.e. elemental
ACCEPTED MANUSCRIPT
T
carbons, organic carbons or hydrocarbons) can lead to adsorption of the dissolved aurocyanide complex ions (Miller et.al., 2005, Helm et.al., 2009, Yang et.al, 2013). This phenomenon is referred to as gold loss due to preg-robbing mechanism. Principally, in order to achieve high gold recovery from refractory ores, ore pre-treatment is required prior to cyanidation leaching.
NU
SC R
IP
The application of bacterial oxidation (biooxidation) for pre-treatment of refractory sulfide gold concentrates is now considered as a proven commercial technique. The advantages of this method compared to roasting or pressure oxidation are that this method is relatively easier to operate and control, less expensive as it does not require high energy costs or expensive autoclaves, and more environmentally-friendly (Brierley 2013, Brierley, 2008; Cui and Zhang, 2008). Biooxidation method uses the bacteria to oxidize sulfide minerals to liberate gold particles from the sulfide matrix that making them readily accessible to the cyanidation stage.
D
MA
The oxidation mechanisms can take place directly or indirectly. Direct mechanism involves physical contact between the bacteria and the sulfide minerals, such as pyrite (FeS2), pyrrhotite and other iron(II) sulfide (FeS), arsenopyrite (4FeAsS) and chalcopyrite (CuFeS2), which then react with dissolved oxygen to convert sulfide- sulfur to sulfate or elemental sulfur according to Equation (1) through (4): (1)
4FeS + 8O2 + 2H2SO4 bacteria 4Fe2+ + 6SO42- + 4H+
(2)
CE P
TE
2FeS2 + 7O2 + 2H2O bacteria 2Fe2+ + 4SO42- + 4H+
4FeAsS + 13O2 + 6H2O bacteria 4Fe2+ + 4SO42- + 12H++ 4AsO4
(3)
CuFeS2 + O2 + 2H2SO4 bacteria Fe2+ + Cu2+ + 2SO42- + 2S + 2H2O
(4)
AC
Indirect mechanism involves oxidation-reduction cycle of ferrous and ferric ions in mineralsolution interface during BIOX process according the following reaction: (Sand et.al., 2001). Fe2+ + ¼O2 + H+ bacteria Fe3+ + ½H2O
(5)
The ferric ion generated by Reaction (5) further plays a role in subsequent oxidation of metal(II) sufide (MS) into its divalent ions and elemental sulfur according to the following reaction: MS + 2Fe3+ M2+ + 2Fe2+ + S˚
(6)
Several microorganisms such as heterotrophic bacteria and fungi are capable in passivation of the surface of carbonaceous matters or decompose them (Yang et al., 2013). In the passivation mechanism, the surface of carbonaceous matters is coated with extracellular polymeric substance (EPS) produced by the microorganism (Wang et.al., 2000; Yang et.al, 2003, Fu et.al., 2010). Several types of fungi can secrete several enzymes that can break the structure of the carbonaceous matters (Amankwah et al., 2005). Brierley and Kulpa (1992) reported that heterotrophic bacteria belonged to the family of Pseudomonas are able to
ACCEPTED MANUSCRIPT passivate the surface of carbonaceous matters and have beneficial effect in increasing gold extraction efficiency from carbonaceous preg-robbing ore.
NU
SC R
IP
T
Researches on biooxidation of refractory gold ores containing both sulfide minerals and carbonaceous materials are generally conducted in two steps, in which the sulfide minerals are oxidized biologically at the first step and then the presence carbonaceous matters is resolved at the second step (Amankwah et al., 2005). The following bacteria are commonly used in biooxidation processes: Acidithiobacillus ferrooxidans, Acidithiobacillus thiooxidans and Leptospirillum ferrooxidans (Brierley and Brans, 1994; Brierley, 1995; Hackl, 1997; Rawlings, 1998). These bacteria are chemolitotrophs and are not resistant to organic substances. Unlike chemolithotrophic bacteria, mixotrophic bacteria have the ability to use both organic and inorganic substances as their carbon sources. These bacteria are also able to grow on wider pH range than chemolithotrophic bacteria. Several types of mixotrophic bacteria have the ability to oxidize iron and/or sulfur (i.e. sulfide-sulfur) for obtaining their energy.
CE P
TE
D
MA
In the present study, two mixotrophic bacteria SKC1 and SKC2 and a chemolithothropic bacterium Acidithiobacillus ferrooxidans (AC), were used for biooxidation of two types of refractory gold concentrates with distinct characteristics. The first type concentrate has high sulfur content, while the second one has low-sulfur content with preg-robbing tendency. The effect of biooxidation with these strains on gold extraction as well as on cyanide and lime consumption was studied. Control experiments without inoculation of the bacteria were also carried out to determine the effect of the dissolved ions in the medium on the cyanidation efficiency.
AC
Materials and Methods Samples of Gold Concentrates Refractory gold concentrates used in the investigations were obtained by flotation of highsulfidation and low sulfidation gold ores originally from Sumatra and Sulawesi Region, in Indonesia. The concentrate with high sulfur content (sample A) contained 20.55% of total sulfur, while low-sulfur concentrate (sample B) contained 5.05% of total sulfur. Each sample was milled with rod mill and sieved to have particle size distribution of P80 -74 µm. The result of semi-quantitative X-ray diffraction (XRD) of the ore concentrates is shown in Table 1. Chemical composition of the concentrate samples is presented in Table 2. Table 1. Semi-quantitative XRD analyses of the gold concentrate samples. Mineral Pyrite Feldspare Quartz Dolomite Calcite Gypsum Mica
Weight percentage (%) Sample A Sample B 30.5 5.1 31.0 8.9 20.2 37.6 6.0 4.6 15.7 1.1 3.4 4.4 21.3
ACCEPTED MANUSCRIPT
IP SC R
Sample B (%) 4.56 0.11 0.11 0.02 0.74 4.79 0.31 0.37 24.45 2.48 18.0 5.0 1.85
NU
Sample A (%) 15.32 0.10 0.02 0.10 0.55 0.82 0.09 0.36 19.68 7.16 46.0 20.55 0.01
MA
Element Fe As Pb Cu Al Ca K Mg Si Au (g/t) Ag (g/t) Total S Total C
T
Table 2. Elemental analyses of the gold concentrate samples.
AC
CE P
TE
D
Bacterial culture and growth media Iron- and sulfur-oxidizing mixothrophic bacteria (designated by SKC1) and sulfur-oxidizing mixotrophic bacteria (designated by SKC2) were used in the present study. SKC1 is the Gram-negative mixotrophic bacterium, which has the abilities of producing EPS (extracellular polymeric substances) and oxidizing iron and sulfur, but its ability to oxidize iron is greater than sulfur. It was isolated from a mine area of nickel laterite ore in South Sulawesi, Indonesia. SKC2 is the Gram-negative bacterium, which was isolated from the sediments of the Domas crater of Tangkuban Perahu Mountain in, Bandung, West Java, Indonesia. It is capable of oxidizing sulfur and growing well in Luria–Bertani (LB) and Febroth medium. Growth of both bacteria occurred at pH 2-9 with optimum pH of 3. Chemolithothrophic bacterium of Acidithiobacillus ferroxidans (designated by AC) was also used as comparison. Each bacterium was cultured in a 500-mL Schott bottle that was shaken on a rotary shaker at 180 rpm and room temperature for 5 days. The three bacteria were growth in different media. AC was cultured in a modified iron-broth medium containing 0.5 g/L MgSO4.7H2O, 3.0 g/L (NH4)2SO4, 0.1 g/L KCl, 0.5 g/L K2HPO4 and 0.01 g/L Ca(NO3)2 supplemented with 15.0 g/L of FeSO4.7H2O and 5.0 g/L of Na2S2O3.5H2O as the sources of energy of the bacteria. Bacterium of SKC-1 was also cultured in a modified iron-broth medium but with the addition of 0.5 g/L tryptone. Bacterium of SKC-2 was cultured in Luria-Bertani broth containing 10 g/L tryptone, 5 g/L yeast extract, 10 g/L NaCl, 5 g/L Na2S2O3.5H2O and 2 g/L FeSO4.7H2O. All materials, except for FeSO4.7H2O and Na2S2O3.5H2O, were dissolved in sterilized distilled water before addition into the medium which was already sterilized. The medium used in the BIOX step was different with the medium used during culturing. The amounts of FeSO4.7H2O and Na2S2O3.5H2O added into the medium for AC and SKC-1 during biooxidation experiments were reduced to 10 g/L and 1 g/L, respectively, as the concentrates themselves already contained sulfide minerals. For SKC-2, only 1 g/L of
ACCEPTED MANUSCRIPT FeSO4.7H2O was added into the medium. The initial pHs of all biooxidation media were adjusted to 3.0 with concentrated sulfuric acid. All inorganic reagents used were of Analytical Reagent (AR) grade.
MA
NU
SC R
IP
T
Biooxidation Procedure The biooxidation experiments were carried out using 500-mL Schott bottle as the reactor with working volume of 350 mL. Non-sterilized solid sample of 60 g and 270 mL of sterilized solution media were placed in the reactor and the pH of the slurry was measured prior to inoculation. Bacterial culture of 30 mL volume was then added into the reactor for inoculation and to achieve a slurry density of 20% (w/v). The slurry was shaken on a rotary shaker under shaking speed of 180 rpm at room temperature for 14 days. pH of the slurry was not adjusted or maintained at a certain value. Control tests were also carried out at the same conditions but without bacterial inoculation for comparison and for determining the effect of medium solution used on the oxidation of the sulfide minerals. During each test, the pH values, oxidation reduction potential (ORP) and dissolved iron concentration were measured every 24 hours. Sampling for dissolved iron analysis was carried out by taking a portion of slurry which was then centrifuged at 10,000 rpm for 10 minutes to separate the solid and aqueous phases.
AC
CE P
TE
D
Bottle Roll Cyanidation Test The biooxidized concentrate samples were leached with a cyanide solution at 40% solid percentage, free cyanide (CN-) concentration of 1000 mg/L, pH 10-10.5 and retention time 48 hours. Solution samples were taken at 2, 4, 8, 24 and 48 hours for measurements of CNconcentration and the concentrations of the dissolved gold and silver. The cyanide concentration was maintained at 1000 mg/L with the addition of solid NaCN. The pH was maintained by lime addition. After 48 hours, the residue was washed and dried for gold and silver analysis. Direct cyanidation tests were also carried out to the concentrates without pretreatment to estimate the efficiency of biooxidation. Preg-robbing Test using Barrick Goldstrike Mines Incorporated (BGMI) Method BGMI tests were performed to determine the preg-robbing tendency of the ore concentrates. A total carbon of concentrate B (1.85%) is an initial indication for preg-robbing tendency of this concentrate. The test was carried out by mixing 10 g of concentrate with 20 mL of standard gold solution containing 3 mg/L of gold. The mixture was then agitated with a mechanical stirrer for 15 minutes. The residue was then filtered and separated from the aqueous phase and the filtrate was analyzed by Atomic Absorption Spectrophotometry (AAS) to determine the gold concentration in the aqueous phase. The indication of preg-robbing was shown by the reduction of gold concentration in the aqueous phase after the test is finished (Goodall, et al. 2005). Analytical Method The elemental composition of the mineral was measured using Inductively Coupled Plasma (ICP MS) and X-ray Fluorescence (XRF) analyses. Semi-quantitative mineral analysis was performed by X-Ray Diffraction (XRD). Mineragraphy analysis was performed using optical microscopy. Prior to the optical microscopy analysis, the concentrate samples were
ACCEPTED MANUSCRIPT
SC R
IP
T
concentrated via dense media separation technique. The values of pH and ORP during BIOX tests were measured using pH/ORP meter equipped with a pH/ORP probe. Dissolved iron concentration in aqueous solution was measured using AAS. Total sulfur and carbon concentrations in the sample were analyzed using LECO®sulfur/carbon analyzer. Gold and silver concentrations in solid residues were measured via fire assay finished by ICP analyses. During cyanidation, cyanide concentration in the aqueous solution was measured by titration using AgNO3 as titrant and rhodamine as indicator while the dissolved gold and silver were measured using AAS.
D
MA
NU
Results and Discussion Sample Characterization Mineragraphy analysis results of thin polished section of concentrate samples are depicted in Figure 1. It is revealed that very fine gold grains (<10 m) were encapsulated in the sulfide minerals. Observation of sample A showed that some gold particles were present inside chalcopyrite and chalcocite minerals. Observation of sample B showed that some gold particles were associated with chalcocite. In addition, gold particles were also found in the form of electrum. Semi-quantative XRD analysis revealed that pyrite and quartz were predominant minerals in the sample A. In sample B, on the other hand, only quartz was predominant, while sulfide minerals content was low (5.1%).
Chalcopyrite
CE P
TE
(a)
AC
Chalcocite
Gold 10m (b)
Gold
10m
Electrum Chalcocite
Figure 1. Micrographs of (a) sample
A and (b) sample B.
ACCEPTED MANUSCRIPT
NU
SC R
IP
T
Biooxidation of the Concentrates Profiles of the slurry pH and Eh during biooxidation tests of Sample A (high sulfur content) are shown in Figure 2a and 2b, respectively. The pH of the biooxidation media, which were initially 3.0, increased to higher values immediately after addition of the concentrates due to the presence of acid-consuming minerals. Profiles of the slurry pH and Eh during biooxidation tests of Sample B (low sulfur content) are shown in Figure 3a and 3b, respectively. The content of carbonates in sample B in the forms of dolomite and calcite (~20%) was higher than that of sample A (~6%) and led to the higher slurry pH of the sample B. The reduction of slurry pH during biooxidation was significant as SKC1 were used in the BIOX of both sample A and Sample B. The slurry pH of BIOX sample A with AC decreased from 6.3 to 5.8 which occurred after day-8, while BIOX of sample B using SKC1 demonstrated pH reduction from 7.2 to 6.3 after day-6. The reduction of pH was caused by H+ generation from oxidation reactions of the sulfide minerals according to Reaction (1) and (2). 8
MA
(a) 7 6
D
4
TE
pH
5
3
A-SKC1
CE P
2
A-SKC2
1
A-AC
0
AC
0
1
2
3
4
5
6
7
8
9
10
8
9
10
Time (days)
600
(b)
500
Eh (mV)
400 300 200 100 0 A-SKC1
A-SKC2
A-AC
-100 0
1
2
3
4
5
6
7
Time (days) Figure 2. Profiles of (a) pH alteration during BIOX of Sample A using SKC-1, SKC-2 and AC bacteria (b) Eh alteration during BIOX of Sample A using SKC-1, SKC-2 and AC bacteria
ACCEPTED MANUSCRIPT 8.00
(a) 7.00 6.00
T
4.00
IP
pH
5.00
B-SKC1
2.00
B-SKC2 1.00
B-AC
0.00 0
1
2
3
4
5
SC R
3.00
6
7
8
9
10
NU
Time (days) 700
B-SKC1
500
D
400
A-AC
300
TE
Eh (mV) vs. SHE
B-SKC2
MA
(b)
600
200
CE P
100 0
-100
AC
0
1
2
3
4
5
6
7
8
9
10
Time (days)
Figure 3. Profiles of (a) pH alteration during BIOX of Sample B using SKC-1, SKC-2 and AC bacteria (b) Eh alteration during BIOX of Sample B using SKC-1, SKC-2 and AC bacteria
Profiles of dissolved iron concentrations as a function of time in each biooxidation test are illustrated in Figure 4. Redox potential and dissolved iron concentration in each biooxidation test showed varying trend as can be seen in Figure 3 and 4, respectively. The alteration trend of redox potential was different for each of the three bacteria. The redox potential in biooxidation test using AC was relatively constant for both sample A and B after day-2. Analysis of the dissolved iron concentration during BIOX of Samples A and B using AC bacteria detected no dissolved iron in the solution after day-2 and onward. Redox potential in the biooxidation tests with SKC-1 showed staggering trend, while those with SKC2 showed decreasing trend. In the BIOX test using SKC1, oxidization of iron performs more predominantly than the oxidation of sulfur. The high pH values resulted in unstable iron ions in the solution that leads to iron precipitation. However, SKC1 is able to generate high amounts of EPS (SK Chaerun, unpublished data 2014) that has an ability to form Fe(III)-EPS anionic complex and avoid iron precipitation. Consequently, the redox
ACCEPTED MANUSCRIPT
SC R
IP
T
potential values in A-SKC1 and B-SKC1 tests were showing staggering trend during the BIOX process. During day-0 to day-6, iron concentrations in A-SKC1 and B-SKC1 experiments showing decreasing trend because SKC1 was still adapting to the environment. After 14 days of biooxidation, iron concentration was increased to 17.5 mg/L and 42.2 mg/L for A-SKC1 and B-SKC1, respectively. At this condition, the BIOXs of samples A and B using SKC 1 bacteria tend to oxidize sulfur than iron. The iron concentrations during BIOX test with SKC2 were showing increasing trend although the pH of the solutions were higher than 5.0. This was also due to generation of EPS by SKC2 bacteria that were able to maintain iron in its ionic state.
TE
D
MA
NU
Cyanidation Test Results Profiles of cumulative gold extraction from concentrate samples A and B for various pretreatment conditions and for direct cyanidation are presented in Figure 5.a and Figure 5.b, respectively. The direct cyanidation results showed that gold extraction from sample A and B were 73.0% and 47.8%, respectively. These low gold extraction levels give strong indication that the concentrates used were of refractory types. BGMI tests showed that sample A was not showing preg-robbing characteristics. In contrast, sample B tends to be preg-robbing, as the gold concentration in the aqueous phase of sample B was reduced after the equilibration. Hence, the gold extraction from sample B was lower than that from sample A due to loss of dissolved aurocyanide complex adsorbed by carbonaceous matters present is sample B, in addition to entrapment of fine gold particles inside the sulfide minerals.
AC
CE P
The biooxidized concentrates were leached in a cyanide solution containing 1000 mg/L of CN- for 48 hours. Biooxidation with SKC2 was able to increase the gold extraction from concentrate sample A to 91.4% which was 18.4% higher than gold extraction from untreated concentrate. On the other hand, biooxidation of sample A with SKC1 slightly reduced gold extraction from 73% (without pre-treatment) to 67.8%. As has been earlier discussed, although both bacteria are able to oxidize the sulfide minerals, SKC1 preferentially oxidize iron than sulfur. During biooxidation of the sulfide mineral, sulfur species formed on the surface of mineral which cannot be oxidized properly would result in passivation of the mineral. Hence, even the gold particles have already been liberated; it may not be leached effectively. The preg-robbing test with BGMI Method revealed the preg-robbing tendency of the concentrate B from significant reduction of gold concentration in solution after 15 minute (i.e. from 3 mg/l downs to 2.14 mg/l or 29%). Similar phenomenon did not occur for the preg-robbing test using sample A. As has been indicated by the results of total carbon analysis, total carbon content of sample B is significantly higher than sample A (Table 1). Moreover, concentrate B has significant content (21.3%) of mica muscovite which is belongs to clay minerals that can also have preg-robbing properties.
ACCEPTED MANUSCRIPT
300
A-AC
250
B-AC
T
200
IP
150 100
SC R
Fe concentration (mg/L)
350
50 0 0
2
4
6
NU
Time (days) (a) 1600
A-SKC1
MA
1200 1000
D
800 600 400
CE P
200
B-SKC1
TE
Fe concntration (mg/L)
1400
0
2
4
6 8 Time (days)
10
12
14
10
12
14
(b)
60 Fe concntration (mg/L)
AC
0
A-SKC2 B-SKC2
50 40 30 20 10 0 0
2
4
6 8 Time (days)
(c) Figure 4. Dissolved iron concentration profiles versus biooxidation time of sample A and B using (a) pure culture AC (A-AC and B-AC), (b) iron- and sulfur- oxidizing mixotrophic bacteria of SKC1 (A-SKC1 and BSKC1) and (c) sulfur-oxidizing mixotrophic bacteria of SKC2 (A-SKC2 and B-SKC2).
ACCEPTED MANUSCRIPT 100
(a)
80 70
50
A-Direct cyanidation
40
A-AC
30
A-SKC1
20
A-SKC2
10
A-Control
0 12
24 Time (hours)
100
80 70
D
60
48
MA
(b)
90
36
NU
0
TE
50 40 30
CE P
%Cumulative Au extracted
IP
T
60
SC R
%Cumulative Au extracted
90
20 10
B-Direct cyanidation B-AC B-SKC1 B-SKC2 B-Control
0
AC
0
12
24
36
48
Time (hours)
Figure 5. Profiles of the cumulative gold extraction from cyanidation of (a) sample A and (b) sample B with biooxidation using pure culture AC (A-AC and B-AC), iron- and sulfur- oxidizing mixotrophic bacteria SKC1 (A-SKC1 and B-SKC1) and sulfur-oxidizing mixotrophic bacteria SKC2 (A-SKC2 and B-SKC2), control tests without bacterial inoculation (A-Control and B-Control) and direct cyanidation
As shown in Figure 5.a, cyanidation of the biooxidized sample A has lower extraction slopes than the direct cyanidation which indicated that the rate of gold dissolution of the biooxidation products was slower. This was associated with re-dissolution of the precipitated sulfur and iron species that were precipitated during the biooxidation step. The dissolution reactions for gold and silver was then competing with the reactions of thiocyanate and ferrocyanide formations, and hence lowering the dissolution rate of gold. Biooxidation of sample B with SKC1 improved gold extraction level from 47.8% (without pretreatment) to 62.8%. SKC1 bacterium has an ability to use both organic and inorganic carbon as their carbon sources that leads to reduction of preg-robbing effect of carbonaceous
ACCEPTED MANUSCRIPT
T
matters in Sample B. However, part of gold is still un-exposed and encapsulated within the sulfide minerals. Biooxidation of sample B with SKC2 slightly increases gold extraction of un-oxidized concentrate from 47.8% to 51.3%. This bacterium was also able to consume carbonaceous matters in sample B due to its mixotrophic nature but not as effective as SKC1. Moreover, SKC2 also oxidized sulfide minerals in sample B as it also did to sample A.
NU
SC R
IP
The cyanidation of the biooxidized concentrated with AC for sample A and B showed nearly similar results to that of direct cyanidation. The gold extraction percentage for sample A and B were 73.0% and 50.3%, respectively. The gold extraction for the control tests were 71.8% and 54.6% for sample A and sample B, respectively. These control tests were carried out using similar biooxidation medium of the BIOX tests using AC bacteria. Insignificant improvements of gold recoveries from sample A and sample B which were pre-treated with AC bacteria indicated inappropriate growth of AC bacteria during the course of the biooxidation process.
AC
CE P
TE
D
MA
Cyanide and Lime Consumption NaCN and lime consumptions for cyanidation of the biooxidized concentrates with mixotrophic bacteria was higher than that of direct cyanidation for both samples as illustrated in Figure 6. NaCN consumption for cyanidation of sample A was increased from 3.72 kg/ton to 9.99 kg/ton and 8.37 kg/ton when the concentrate samples were pre-treated with SKC1 and SKC2, respectively. This was also similar for sample B, in which NaCN consumption was increased from 3.01 kg/ton to 7.77 kg/ton and 6.66 kg/ton, for the BIOX pretreatments using SKC1 and SKC2, respectively. Higher NaCN consumption was due to the addition of sulfur and iron nutrient during the biooxidation process which was precipitated during BIOX and re-dissolved in the cyanidation step. The excess of sulfur reacts with cyanide ions to form thiocyanate anion complexes, while dissolved iron reacts with cyanides ions to form ferrocyanide complexes. Lime consumptions for cyanidation of the biooxidized concentrates with SKC2 were decreased from 3.99 kg/ton to 2.87 kg/ton for sample A and from 3.25 kg/ton to 1.7 kg/ton for sample B. Reducing of lime consumption for cyanidation of bioxidized concentrates by SKC1 and SKC 2 might be in correlation with hydroxyl (OH-) ions generation and ammonification (i.e. the production of NH3) by bacterial cells. Since the concentrates contain significant amount of silica, an increase in solution pH in experimental systems with silica may occur through bacterial production of CO2 with concomitant release of OH-, creating a highly alkaline microenvironment around active cells (Schultze Lam, Fortin, Davis and Beveridge, 1996). During ammonification by bacterial cells, hydrogen ion is consumed when amino acids from dead cells are metabolized, thereby forming NH3 (Lee and Beveridge, 2001). The increased pH remained stable due to the presence of silica which presumably supported the buffering capacity of the solution (Chaerun, Tazaki, Asada and Kogure, 2005).
ACCEPTED MANUSCRIPT 12
Sample A
NaCN
Lime
10
T
6
IP
kg/ton
8
2 0 A-Direct A-Control cyanidation
A-AC
9 8
Sample B
kg/ton
6 5 4
Lime
D
3
TE
2
0
NaCN
A-SKC2
MA
7
1
A-SKC1
NU
(a)
SC R
4
CE P
Direct B-Control cyanidation
B-AC
B-SKC1
B-SKC2
AC
(b) Figure 6. Cyanide and lime consumptions during direct cyanidation and cyanidation of the biooxidized concentrates (a) sample A and (b) sample B with pure culture AC (A-AC and B-AC), iron- and sulfur- oxidizing mixotrophic bacteria SKC1 (A-SKC1 and B-SKC1) and sulfur-oxidizing mixotrophic bacteria SKC2 (A-SKC2 and B-SKC2).
Conclusions The present study demonstrated that biooxidation of the refractory gold concentrates using iron- and sulfur-oxidizing mixotrophic bacterium of SKC1 or sulfur-oxidizing mixotrophic bacterium of SKC2 was able to increase gold extraction during cyanidation. The bacteria were able to treat both sulfide type of refractory gold concentrate and preg-robbing type at neutral pH. The SKC1 and SK2 bacteria are able to generate EPS that forms Fe(III)-EPS anionic complex which prevents passivation of sulfide minerals caused by iron and sulfur precipitation on the surface of the minerals during the BIOX process. However, cyanide consumption of the biooxidized products with mixotrophic bacteria was increased due to reactions of precipitated sulfur and iron during cyanidation with cyanide ions to form thiocyanate and ferrocyanide ions. Biooxidation mechanisms using SKC1 and SKC2 bacteria were dissimilar and still poorly understood. More comprehensive analysis of the formed EPS substances and morphology analysis of the biooxidation products need to be further studied.
ACCEPTED MANUSCRIPT
SC R
IP
T
Acknowledgements This research work was supported by the Geomicrobiology-Biomining & Biocorrosion Laboratory, Biosciences and Biotechnology Research Center of Institut Teknologi Bandung. We thank Mr. Wiku Padmonobo, Manager of Geo-metallurgy Department of PT Geoservice, for providing gold concentrate sample and analytical supports as well as permission to publish the research result.
AC
CE P
TE
D
MA
NU
References Aymore, M.G. and Jaffer, A., 2012. Evaluating process options for treating some refractory ores. ALTA 2012 International Gold Conference, May 31-June 1 2012, Burswood Convention Centre, Perth, Western Australia Amankwah, R.K., Yen, W.T., Ramsay, J.A., 2005. A two-stage bacterial pretreatment process for double refractory gold ore. Minerals Engineering 18(1): 103-108. Brierley, J., Kulpa, C.F., 1992. Microbial consortium treatment of refractory precious metal ores: USA, 5127942. Brierley, C.L., 1995. Bacterial oxidation. Engineering and Mining Journal, 196, 42–44. Brierley, C.L., Brans, R., 1994. Selection of Bactech’s thermophilic biooxidation process for Youanmi Mine. In: Biomine ’94, Australian Mineral Foundation, Glenside, South Australia, pp. 5.1– 5.7. Brierley, C.L., 2008. How will biomining be applied in future, Transactions of Nonferrous Metal Society of China 18, 1302-1310. Brierley, C.L., 2013. Mining Biotechnology: Research to Commercial Development and Beyond, Biomining: Theory, Microbes and Industrial Processes, 3-17. Celep, O., Alp, I., Deveci, H., 2011. Improved gold and silver extraction from a refractory antimony ore by pretreatment with alkaline sulphide leach. Hydrometallurgy 105, 234239. Chaerun, S.K., Tazaki,K., Asada, R. and Kogure, K, 2005. Interaction between clay minerals and hydrocarbon-utilizing indigenous microorganisms in high concentrations of heavy oil: implications for bioremediation, Clay Minerals Vol. 40, p.105-114. Chowdhury, F., Ojumu, T.V., 2014. Investigation of ferrous-iron biooxidation kinetics by Leptospirillum ferriphilum in a novel packed-column bioreactor: Effects of temperature and jarosite accumulation. Hydrometallurgy 141, 36-42. Cui, J., Zhang, L.,2008. Metallurgical recovery of metals fromelectronic waste: A review, Journal of Hazardous Materials 158, 228-256. Diao, M., Nguyen, T.A.H., Taran, E., Mahler, S., Nguyen, A.V., 2014. Differences in adhesion of A. thiooxidans and A. ferrooxidans on chalcopyrite as revealed by atomic force microscopy with bacterial probes. Frizan V., Giaveno A., Chiacchiarini P., Donati E., 2003. Bioleaching of Argentinean sulfide ores using pure and mixed cultures. 15th International Biohydrometallurgy Symposium (IBS 2003)
ACCEPTED MANUSCRIPT
AC
CE P
TE
D
MA
NU
SC R
IP
T
Fu, K., Lin, H., Luo, D., Jiang, W., Zeng, P., 2014. Comparison of bioleaching of copper sulphides by Acidithiobacillus ferrooxidans. African Journal of Biotechnology 13, 664672 Fu, Y., Yang, H., Cui, R., Fan, Y., Wang, C., 2010. Role of extracellular polymers of bacteria during arsenic-bearing gold concentrate bio-oxidation process. The Chinese Journal of Nonferrous Metals, 20(10): 2057-2062 (in Chinese) Goodall, W.R., Leatham, J.D., Scales, P.J., 2005. A new method for determination of pregrobbing in gold ores. Minerals Engineering 18, 1135-1141. Gu, S., Su, L., Chen, M., Sun X., Zhou, H., 2010. Bio-leaching effects of Leptospirillum ferriphilum on the surface chemical properties of pyrite. Mining Science and Technology 20, 287-291. Hackl, R.P., 1997. Commercial applications of bacterial-mineral interactions. In: McIntosh, J.M., Groat, L.A. (Eds.), Biological- Mineralogical Interactions. Mineralogical Association of Canada, pp. 143–167. Hansford, G.S., Chapman, J.T., 1992. Batch and continuous biooxidation kinetics of a refractory gold-bearing pyrite concentrate. Minerals Engineering, Vol. 5, No. 6, pp. 597-612. Harris, D.C., 1997. Quantitative Analytical Chemistry. W.H. Freeman and Company, New York, NY. Helm, M., Vaughan J., Staunton, W.P. and Avraamides, J., (2009): An investigation of the carbonaceous component of preg-robbing gold ores. Proceedings of World Gold Conference 2009, 139-144. Jonglertjunya, W., 2003. Bioleaching of chalcopyrite. Thesis. Department of Chemical Engineering School of Engineering, The University of Birmingham. Lee, J.U. and Beveridge, T.J., 2001. Interaction between iron and Pseudomonas aeruginosa biofilms attached to Sepharose surfaces, Chemical Geology Vol. 180, p. 67-80. Livesey-Goldblatt, E., Norman, P., Livesey-Goldblatt, D.R., 1983. Gold recovery from arsenopyrite/pyrite ore by bacterial leaching and cyanidation. In: Rossi, G. and Torma, A.E. (Eds.), Recent Progress in Biohydrometallurgy. Assoc. Mineraria Sarda, Iglesias, pp. 627-641. Marsden, J.O., House, C.I., 1992. The Chemistry of Gold Extraction, Ellis Horwood, British. Miller, J.D., Wan, R.Y. and Diaz, X., (2005). Preg-robbing Gold Ores. Development of Minerals Processing, Chapter 38. Elsevier. Rawlings, D.E., 1998. Industrial practice and the biology of leaching of metals from ores. Journal of Industrial Microbiology and Biotechnology 20, 268–274. Sand, W., Gehrke, T., Hallmann, R., Schippers, A., 1995. Sulphur chemistry, biofilm, and the (in)direct attack mechanism - a critical evaluation of bacterial leaching. Appl. Microbiol. Biotechnol. 43, 961–966. Sand, W., Gehrke, T., Jozsa, P., Schippers, A., 2001. (Bio)chemistry of bacterial leachingdirect vs. indirect bioleaching. Hydrometallurgy 59, 159-175. Schultze-Lam, S, Fortin, D., Davis, B.S. and Beveridge, T.J., 1996. Mineralization of bacterial surfaces, Chemical Geology, Vol.132, p. 171-181. Vilcáez, J., Inoue, C., (2008). Bioleaching of chalcopyrite with thermophiles:Temperature– pH–ORP dependence. Int. J. Mineral Proc.88 (1-2): 37-44.
ACCEPTED MANUSCRIPT
AC
CE P
TE
D
MA
NU
SC R
IP
T
Wang, A., Zhang, Y., Liu, H., 2000. The property of carboniferous species and its effect on leaching of gold in Dongbeizhai Gold Mine. Multipurpose Utilization of Mineral Resources 3, 4-8 (in Chinese) Yang, F., Xu, X., Zhao, J., Liang, Z., 2003. Experimental study on bacterial oxidation of carbon-bearing higharsenic refractory gold concentrates. Gold, 24(4): 37-39 (in Chinese). Yang, H., Liu, Q., Song, X., Dong, J., 2013. Research status of carbonaceous matter in carbonaceous gold ores and bio-oxidation pretreatment. Trans. Nonferrous Met. Soc. China 23, 3405-3411. Yang, Y. Xie, Z., Xu, B., Li. Q, and Jiang T., 2015. Gold Extraction from a High Carbon Low-Grade Refractory Gold Ore by Flotation-Roasting-Leaching Process. Rare Metal Technology, Published online, 20 February 2015, 63-70.
ACCEPTED MANUSCRIPT
AC
CE P
TE
D
MA
NU
SC R
IP
T
Highlight: • Biooxidation (BIOX) of refractory gold concentrates using iron- and sulfur-oxidizing mixotrophic bacterium of SKC1 and sulfur-oxidizing mixotrophic bacterium of SKC2 was able to increase gold extraction during cyanidation • The highest gold extraction from high sulfidation concentrate of 91.4%, was achieved by BIOX with SKC2 for 14 days. • The SKC1 and SK2 bacteria are able to generate EPS that forms Fe(III)-EPS anionic complex which prevents passivation of sulfide minerals • The bacteria were able to treat both sulfide type of refractory gold concentrate and preg-robbing type at neutral pH.
S0304-386X(16)30756-3 doi:10.1016/j.hydromet.2016.10.018 HYDROM 4454
To appear in:
Hydrometallurgy
Received date: Revised date: Accepted date:
2 April 2016 20 September 2016 17 October 2016
Please cite this article as: Mubarok, M.Z., Winarko, R., Chaerun, S.K., Rizki, I.N., Ichlas, Z.T., Improving Gold Recovery from Refractory Gold Ores Through Biooxidation using Iron-Sulfur-Oxidizing/Sulfur-Oxidizing Mixotrophic Bacteria, Hydrometallurgy (2016), doi:10.1016/j.hydromet.2016.10.018
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Improving Gold Recovery from Refractory Gold Ores Through Biooxidation using Iron-Sulfur-Oxidizing/Sulfur-Oxidizing Mixotrophic Bacteria
T
M.Z. Mubaroka*, R. Winarkoa, S.K. Chaeruna,b, I.N. Rizkic and Z. T. Ichlasa Department of Metallurgical Engineering, Faculty of Mining and Petroleum Engineering, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, Indonesia
b)
Geomicrobiology-Biomining & Biocorrosion Laboratory, Microbial Culture Collection Laboratory, Biosciences and Biotechnology Research Center, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung 40132, Indonesia
c)
GeoAssay Metallurgical Laboratory, Geoservices Company, Cikarang Barat 17530, Indonesia
NU
SC R
IP
a)
MA
*Corresponding author: [email protected], +62 (22) 2502239
AC
CE P
TE
D
Abstract The effect of biooxidation (BIOX) pre-treatment of refractory gold concentrates using mixotrophic and chemolithotrophic bacteria on the gold extraction during cyanidation of the concentrates at neutral pH was studied. A series of biooxidation and cyanidation of the biooxidized concentrates was carried out using three strains of bacteria; two mixotrophic bacteria of SKC1 and SKC2 and a chemolithotrophic bacteria Acidithiobacillus ferroxidans (AC). Two distinct types of refractory gold concentrates with high sulfur content (i.e. S>20%) and low sufur content (i.e. S<5%) were used. The experimental results showed that biooxidation with any of the three bacteria generally showed positive effect on the gold extraction. The highest gold extraction (91.4%) from high sulfur-concentrate was achieved by BIOX with SKC2 for 14 days. It was 18% higher than the extraction level of direct cyanidation of the untreated concentrate. Biooxidation with AC, on the other hand, resulted in only slight increase of gold extraction from both concentrates due to the high solution pH of >5.0 which is not a suitable living environment for the bacteria. Cyanide consumption for cyanidation of the biooxidized concentrates with both mixotrophic bacteria was significantly increased due to reactions of sulfur and iron species that were precipitated during the biooxidation step with cyanides to form thiocyanate and ferrocyanide ions. Keywords: biooxidation, mixotrophic bacteria, gold, refractory ore, preg-robbing, neutral pH Introduction Cyanidation is the most common method for gold extraction from its ore or concentrate. However, direct cyanidation is only effective for treating non-refractory ores. Direct cyanidation of refractory ores would result in inadequate gold recoveries due to various reasons. The inadequate recoveries can be associated with the presences of fine gold entrapped within sulfide minerals and/or carbonaceous materials in the ores (Aylmore and Jaffer, 2012, Yang et.al., 2015). The fine gold particles which are locked inside non-porous sulfide minerals cannot be extracted due to hindrance of the reaction between gold and cyanide ions, while the presence of naturally occurring carbonaceous matters (i.e. elemental
ACCEPTED MANUSCRIPT
T
carbons, organic carbons or hydrocarbons) can lead to adsorption of the dissolved aurocyanide complex ions (Miller et.al., 2005, Helm et.al., 2009, Yang et.al, 2013). This phenomenon is referred to as gold loss due to preg-robbing mechanism. Principally, in order to achieve high gold recovery from refractory ores, ore pre-treatment is required prior to cyanidation leaching.
NU
SC R
IP
The application of bacterial oxidation (biooxidation) for pre-treatment of refractory sulfide gold concentrates is now considered as a proven commercial technique. The advantages of this method compared to roasting or pressure oxidation are that this method is relatively easier to operate and control, less expensive as it does not require high energy costs or expensive autoclaves, and more environmentally-friendly (Brierley 2013, Brierley, 2008; Cui and Zhang, 2008). Biooxidation method uses the bacteria to oxidize sulfide minerals to liberate gold particles from the sulfide matrix that making them readily accessible to the cyanidation stage.
D
MA
The oxidation mechanisms can take place directly or indirectly. Direct mechanism involves physical contact between the bacteria and the sulfide minerals, such as pyrite (FeS2), pyrrhotite and other iron(II) sulfide (FeS), arsenopyrite (4FeAsS) and chalcopyrite (CuFeS2), which then react with dissolved oxygen to convert sulfide- sulfur to sulfate or elemental sulfur according to Equation (1) through (4): (1)
4FeS + 8O2 + 2H2SO4 bacteria 4Fe2+ + 6SO42- + 4H+
(2)
CE P
TE
2FeS2 + 7O2 + 2H2O bacteria 2Fe2+ + 4SO42- + 4H+
4FeAsS + 13O2 + 6H2O bacteria 4Fe2+ + 4SO42- + 12H++ 4AsO4
(3)
CuFeS2 + O2 + 2H2SO4 bacteria Fe2+ + Cu2+ + 2SO42- + 2S + 2H2O
(4)
AC
Indirect mechanism involves oxidation-reduction cycle of ferrous and ferric ions in mineralsolution interface during BIOX process according the following reaction: (Sand et.al., 2001). Fe2+ + ¼O2 + H+ bacteria Fe3+ + ½H2O
(5)
The ferric ion generated by Reaction (5) further plays a role in subsequent oxidation of metal(II) sufide (MS) into its divalent ions and elemental sulfur according to the following reaction: MS + 2Fe3+ M2+ + 2Fe2+ + S˚
(6)
Several microorganisms such as heterotrophic bacteria and fungi are capable in passivation of the surface of carbonaceous matters or decompose them (Yang et al., 2013). In the passivation mechanism, the surface of carbonaceous matters is coated with extracellular polymeric substance (EPS) produced by the microorganism (Wang et.al., 2000; Yang et.al, 2003, Fu et.al., 2010). Several types of fungi can secrete several enzymes that can break the structure of the carbonaceous matters (Amankwah et al., 2005). Brierley and Kulpa (1992) reported that heterotrophic bacteria belonged to the family of Pseudomonas are able to
ACCEPTED MANUSCRIPT passivate the surface of carbonaceous matters and have beneficial effect in increasing gold extraction efficiency from carbonaceous preg-robbing ore.
NU
SC R
IP
T
Researches on biooxidation of refractory gold ores containing both sulfide minerals and carbonaceous materials are generally conducted in two steps, in which the sulfide minerals are oxidized biologically at the first step and then the presence carbonaceous matters is resolved at the second step (Amankwah et al., 2005). The following bacteria are commonly used in biooxidation processes: Acidithiobacillus ferrooxidans, Acidithiobacillus thiooxidans and Leptospirillum ferrooxidans (Brierley and Brans, 1994; Brierley, 1995; Hackl, 1997; Rawlings, 1998). These bacteria are chemolitotrophs and are not resistant to organic substances. Unlike chemolithotrophic bacteria, mixotrophic bacteria have the ability to use both organic and inorganic substances as their carbon sources. These bacteria are also able to grow on wider pH range than chemolithotrophic bacteria. Several types of mixotrophic bacteria have the ability to oxidize iron and/or sulfur (i.e. sulfide-sulfur) for obtaining their energy.
CE P
TE
D
MA
In the present study, two mixotrophic bacteria SKC1 and SKC2 and a chemolithothropic bacterium Acidithiobacillus ferrooxidans (AC), were used for biooxidation of two types of refractory gold concentrates with distinct characteristics. The first type concentrate has high sulfur content, while the second one has low-sulfur content with preg-robbing tendency. The effect of biooxidation with these strains on gold extraction as well as on cyanide and lime consumption was studied. Control experiments without inoculation of the bacteria were also carried out to determine the effect of the dissolved ions in the medium on the cyanidation efficiency.
AC
Materials and Methods Samples of Gold Concentrates Refractory gold concentrates used in the investigations were obtained by flotation of highsulfidation and low sulfidation gold ores originally from Sumatra and Sulawesi Region, in Indonesia. The concentrate with high sulfur content (sample A) contained 20.55% of total sulfur, while low-sulfur concentrate (sample B) contained 5.05% of total sulfur. Each sample was milled with rod mill and sieved to have particle size distribution of P80 -74 µm. The result of semi-quantitative X-ray diffraction (XRD) of the ore concentrates is shown in Table 1. Chemical composition of the concentrate samples is presented in Table 2. Table 1. Semi-quantitative XRD analyses of the gold concentrate samples. Mineral Pyrite Feldspare Quartz Dolomite Calcite Gypsum Mica
Weight percentage (%) Sample A Sample B 30.5 5.1 31.0 8.9 20.2 37.6 6.0 4.6 15.7 1.1 3.4 4.4 21.3
ACCEPTED MANUSCRIPT
IP SC R
Sample B (%) 4.56 0.11 0.11 0.02 0.74 4.79 0.31 0.37 24.45 2.48 18.0 5.0 1.85
NU
Sample A (%) 15.32 0.10 0.02 0.10 0.55 0.82 0.09 0.36 19.68 7.16 46.0 20.55 0.01
MA
Element Fe As Pb Cu Al Ca K Mg Si Au (g/t) Ag (g/t) Total S Total C
T
Table 2. Elemental analyses of the gold concentrate samples.
AC
CE P
TE
D
Bacterial culture and growth media Iron- and sulfur-oxidizing mixothrophic bacteria (designated by SKC1) and sulfur-oxidizing mixotrophic bacteria (designated by SKC2) were used in the present study. SKC1 is the Gram-negative mixotrophic bacterium, which has the abilities of producing EPS (extracellular polymeric substances) and oxidizing iron and sulfur, but its ability to oxidize iron is greater than sulfur. It was isolated from a mine area of nickel laterite ore in South Sulawesi, Indonesia. SKC2 is the Gram-negative bacterium, which was isolated from the sediments of the Domas crater of Tangkuban Perahu Mountain in, Bandung, West Java, Indonesia. It is capable of oxidizing sulfur and growing well in Luria–Bertani (LB) and Febroth medium. Growth of both bacteria occurred at pH 2-9 with optimum pH of 3. Chemolithothrophic bacterium of Acidithiobacillus ferroxidans (designated by AC) was also used as comparison. Each bacterium was cultured in a 500-mL Schott bottle that was shaken on a rotary shaker at 180 rpm and room temperature for 5 days. The three bacteria were growth in different media. AC was cultured in a modified iron-broth medium containing 0.5 g/L MgSO4.7H2O, 3.0 g/L (NH4)2SO4, 0.1 g/L KCl, 0.5 g/L K2HPO4 and 0.01 g/L Ca(NO3)2 supplemented with 15.0 g/L of FeSO4.7H2O and 5.0 g/L of Na2S2O3.5H2O as the sources of energy of the bacteria. Bacterium of SKC-1 was also cultured in a modified iron-broth medium but with the addition of 0.5 g/L tryptone. Bacterium of SKC-2 was cultured in Luria-Bertani broth containing 10 g/L tryptone, 5 g/L yeast extract, 10 g/L NaCl, 5 g/L Na2S2O3.5H2O and 2 g/L FeSO4.7H2O. All materials, except for FeSO4.7H2O and Na2S2O3.5H2O, were dissolved in sterilized distilled water before addition into the medium which was already sterilized. The medium used in the BIOX step was different with the medium used during culturing. The amounts of FeSO4.7H2O and Na2S2O3.5H2O added into the medium for AC and SKC-1 during biooxidation experiments were reduced to 10 g/L and 1 g/L, respectively, as the concentrates themselves already contained sulfide minerals. For SKC-2, only 1 g/L of
ACCEPTED MANUSCRIPT FeSO4.7H2O was added into the medium. The initial pHs of all biooxidation media were adjusted to 3.0 with concentrated sulfuric acid. All inorganic reagents used were of Analytical Reagent (AR) grade.
MA
NU
SC R
IP
T
Biooxidation Procedure The biooxidation experiments were carried out using 500-mL Schott bottle as the reactor with working volume of 350 mL. Non-sterilized solid sample of 60 g and 270 mL of sterilized solution media were placed in the reactor and the pH of the slurry was measured prior to inoculation. Bacterial culture of 30 mL volume was then added into the reactor for inoculation and to achieve a slurry density of 20% (w/v). The slurry was shaken on a rotary shaker under shaking speed of 180 rpm at room temperature for 14 days. pH of the slurry was not adjusted or maintained at a certain value. Control tests were also carried out at the same conditions but without bacterial inoculation for comparison and for determining the effect of medium solution used on the oxidation of the sulfide minerals. During each test, the pH values, oxidation reduction potential (ORP) and dissolved iron concentration were measured every 24 hours. Sampling for dissolved iron analysis was carried out by taking a portion of slurry which was then centrifuged at 10,000 rpm for 10 minutes to separate the solid and aqueous phases.
AC
CE P
TE
D
Bottle Roll Cyanidation Test The biooxidized concentrate samples were leached with a cyanide solution at 40% solid percentage, free cyanide (CN-) concentration of 1000 mg/L, pH 10-10.5 and retention time 48 hours. Solution samples were taken at 2, 4, 8, 24 and 48 hours for measurements of CNconcentration and the concentrations of the dissolved gold and silver. The cyanide concentration was maintained at 1000 mg/L with the addition of solid NaCN. The pH was maintained by lime addition. After 48 hours, the residue was washed and dried for gold and silver analysis. Direct cyanidation tests were also carried out to the concentrates without pretreatment to estimate the efficiency of biooxidation. Preg-robbing Test using Barrick Goldstrike Mines Incorporated (BGMI) Method BGMI tests were performed to determine the preg-robbing tendency of the ore concentrates. A total carbon of concentrate B (1.85%) is an initial indication for preg-robbing tendency of this concentrate. The test was carried out by mixing 10 g of concentrate with 20 mL of standard gold solution containing 3 mg/L of gold. The mixture was then agitated with a mechanical stirrer for 15 minutes. The residue was then filtered and separated from the aqueous phase and the filtrate was analyzed by Atomic Absorption Spectrophotometry (AAS) to determine the gold concentration in the aqueous phase. The indication of preg-robbing was shown by the reduction of gold concentration in the aqueous phase after the test is finished (Goodall, et al. 2005). Analytical Method The elemental composition of the mineral was measured using Inductively Coupled Plasma (ICP MS) and X-ray Fluorescence (XRF) analyses. Semi-quantitative mineral analysis was performed by X-Ray Diffraction (XRD). Mineragraphy analysis was performed using optical microscopy. Prior to the optical microscopy analysis, the concentrate samples were
ACCEPTED MANUSCRIPT
SC R
IP
T
concentrated via dense media separation technique. The values of pH and ORP during BIOX tests were measured using pH/ORP meter equipped with a pH/ORP probe. Dissolved iron concentration in aqueous solution was measured using AAS. Total sulfur and carbon concentrations in the sample were analyzed using LECO®sulfur/carbon analyzer. Gold and silver concentrations in solid residues were measured via fire assay finished by ICP analyses. During cyanidation, cyanide concentration in the aqueous solution was measured by titration using AgNO3 as titrant and rhodamine as indicator while the dissolved gold and silver were measured using AAS.
D
MA
NU
Results and Discussion Sample Characterization Mineragraphy analysis results of thin polished section of concentrate samples are depicted in Figure 1. It is revealed that very fine gold grains (<10 m) were encapsulated in the sulfide minerals. Observation of sample A showed that some gold particles were present inside chalcopyrite and chalcocite minerals. Observation of sample B showed that some gold particles were associated with chalcocite. In addition, gold particles were also found in the form of electrum. Semi-quantative XRD analysis revealed that pyrite and quartz were predominant minerals in the sample A. In sample B, on the other hand, only quartz was predominant, while sulfide minerals content was low (5.1%).
Chalcopyrite
CE P
TE
(a)
AC
Chalcocite
Gold 10m (b)
Gold
10m
Electrum Chalcocite
Figure 1. Micrographs of (a) sample
A and (b) sample B.
ACCEPTED MANUSCRIPT
NU
SC R
IP
T
Biooxidation of the Concentrates Profiles of the slurry pH and Eh during biooxidation tests of Sample A (high sulfur content) are shown in Figure 2a and 2b, respectively. The pH of the biooxidation media, which were initially 3.0, increased to higher values immediately after addition of the concentrates due to the presence of acid-consuming minerals. Profiles of the slurry pH and Eh during biooxidation tests of Sample B (low sulfur content) are shown in Figure 3a and 3b, respectively. The content of carbonates in sample B in the forms of dolomite and calcite (~20%) was higher than that of sample A (~6%) and led to the higher slurry pH of the sample B. The reduction of slurry pH during biooxidation was significant as SKC1 were used in the BIOX of both sample A and Sample B. The slurry pH of BIOX sample A with AC decreased from 6.3 to 5.8 which occurred after day-8, while BIOX of sample B using SKC1 demonstrated pH reduction from 7.2 to 6.3 after day-6. The reduction of pH was caused by H+ generation from oxidation reactions of the sulfide minerals according to Reaction (1) and (2). 8
MA
(a) 7 6
D
4
TE
pH
5
3
A-SKC1
CE P
2
A-SKC2
1
A-AC
0
AC
0
1
2
3
4
5
6
7
8
9
10
8
9
10
Time (days)
600
(b)
500
Eh (mV)
400 300 200 100 0 A-SKC1
A-SKC2
A-AC
-100 0
1
2
3
4
5
6
7
Time (days) Figure 2. Profiles of (a) pH alteration during BIOX of Sample A using SKC-1, SKC-2 and AC bacteria (b) Eh alteration during BIOX of Sample A using SKC-1, SKC-2 and AC bacteria
ACCEPTED MANUSCRIPT 8.00
(a) 7.00 6.00
T
4.00
IP
pH
5.00
B-SKC1
2.00
B-SKC2 1.00
B-AC
0.00 0
1
2
3
4
5
SC R
3.00
6
7
8
9
10
NU
Time (days) 700
B-SKC1
500
D
400
A-AC
300
TE
Eh (mV) vs. SHE
B-SKC2
MA
(b)
600
200
CE P
100 0
-100
AC
0
1
2
3
4
5
6
7
8
9
10
Time (days)
Figure 3. Profiles of (a) pH alteration during BIOX of Sample B using SKC-1, SKC-2 and AC bacteria (b) Eh alteration during BIOX of Sample B using SKC-1, SKC-2 and AC bacteria
Profiles of dissolved iron concentrations as a function of time in each biooxidation test are illustrated in Figure 4. Redox potential and dissolved iron concentration in each biooxidation test showed varying trend as can be seen in Figure 3 and 4, respectively. The alteration trend of redox potential was different for each of the three bacteria. The redox potential in biooxidation test using AC was relatively constant for both sample A and B after day-2. Analysis of the dissolved iron concentration during BIOX of Samples A and B using AC bacteria detected no dissolved iron in the solution after day-2 and onward. Redox potential in the biooxidation tests with SKC-1 showed staggering trend, while those with SKC2 showed decreasing trend. In the BIOX test using SKC1, oxidization of iron performs more predominantly than the oxidation of sulfur. The high pH values resulted in unstable iron ions in the solution that leads to iron precipitation. However, SKC1 is able to generate high amounts of EPS (SK Chaerun, unpublished data 2014) that has an ability to form Fe(III)-EPS anionic complex and avoid iron precipitation. Consequently, the redox
ACCEPTED MANUSCRIPT
SC R
IP
T
potential values in A-SKC1 and B-SKC1 tests were showing staggering trend during the BIOX process. During day-0 to day-6, iron concentrations in A-SKC1 and B-SKC1 experiments showing decreasing trend because SKC1 was still adapting to the environment. After 14 days of biooxidation, iron concentration was increased to 17.5 mg/L and 42.2 mg/L for A-SKC1 and B-SKC1, respectively. At this condition, the BIOXs of samples A and B using SKC 1 bacteria tend to oxidize sulfur than iron. The iron concentrations during BIOX test with SKC2 were showing increasing trend although the pH of the solutions were higher than 5.0. This was also due to generation of EPS by SKC2 bacteria that were able to maintain iron in its ionic state.
TE
D
MA
NU
Cyanidation Test Results Profiles of cumulative gold extraction from concentrate samples A and B for various pretreatment conditions and for direct cyanidation are presented in Figure 5.a and Figure 5.b, respectively. The direct cyanidation results showed that gold extraction from sample A and B were 73.0% and 47.8%, respectively. These low gold extraction levels give strong indication that the concentrates used were of refractory types. BGMI tests showed that sample A was not showing preg-robbing characteristics. In contrast, sample B tends to be preg-robbing, as the gold concentration in the aqueous phase of sample B was reduced after the equilibration. Hence, the gold extraction from sample B was lower than that from sample A due to loss of dissolved aurocyanide complex adsorbed by carbonaceous matters present is sample B, in addition to entrapment of fine gold particles inside the sulfide minerals.
AC
CE P
The biooxidized concentrates were leached in a cyanide solution containing 1000 mg/L of CN- for 48 hours. Biooxidation with SKC2 was able to increase the gold extraction from concentrate sample A to 91.4% which was 18.4% higher than gold extraction from untreated concentrate. On the other hand, biooxidation of sample A with SKC1 slightly reduced gold extraction from 73% (without pre-treatment) to 67.8%. As has been earlier discussed, although both bacteria are able to oxidize the sulfide minerals, SKC1 preferentially oxidize iron than sulfur. During biooxidation of the sulfide mineral, sulfur species formed on the surface of mineral which cannot be oxidized properly would result in passivation of the mineral. Hence, even the gold particles have already been liberated; it may not be leached effectively. The preg-robbing test with BGMI Method revealed the preg-robbing tendency of the concentrate B from significant reduction of gold concentration in solution after 15 minute (i.e. from 3 mg/l downs to 2.14 mg/l or 29%). Similar phenomenon did not occur for the preg-robbing test using sample A. As has been indicated by the results of total carbon analysis, total carbon content of sample B is significantly higher than sample A (Table 1). Moreover, concentrate B has significant content (21.3%) of mica muscovite which is belongs to clay minerals that can also have preg-robbing properties.
ACCEPTED MANUSCRIPT
300
A-AC
250
B-AC
T
200
IP
150 100
SC R
Fe concentration (mg/L)
350
50 0 0
2
4
6
NU
Time (days) (a) 1600
A-SKC1
MA
1200 1000
D
800 600 400
CE P
200
B-SKC1
TE
Fe concntration (mg/L)
1400
0
2
4
6 8 Time (days)
10
12
14
10
12
14
(b)
60 Fe concntration (mg/L)
AC
0
A-SKC2 B-SKC2
50 40 30 20 10 0 0
2
4
6 8 Time (days)
(c) Figure 4. Dissolved iron concentration profiles versus biooxidation time of sample A and B using (a) pure culture AC (A-AC and B-AC), (b) iron- and sulfur- oxidizing mixotrophic bacteria of SKC1 (A-SKC1 and BSKC1) and (c) sulfur-oxidizing mixotrophic bacteria of SKC2 (A-SKC2 and B-SKC2).
ACCEPTED MANUSCRIPT 100
(a)
80 70
50
A-Direct cyanidation
40
A-AC
30
A-SKC1
20
A-SKC2
10
A-Control
0 12
24 Time (hours)
100
80 70
D
60
48
MA
(b)
90
36
NU
0
TE
50 40 30
CE P
%Cumulative Au extracted
IP
T
60
SC R
%Cumulative Au extracted
90
20 10
B-Direct cyanidation B-AC B-SKC1 B-SKC2 B-Control
0
AC
0
12
24
36
48
Time (hours)
Figure 5. Profiles of the cumulative gold extraction from cyanidation of (a) sample A and (b) sample B with biooxidation using pure culture AC (A-AC and B-AC), iron- and sulfur- oxidizing mixotrophic bacteria SKC1 (A-SKC1 and B-SKC1) and sulfur-oxidizing mixotrophic bacteria SKC2 (A-SKC2 and B-SKC2), control tests without bacterial inoculation (A-Control and B-Control) and direct cyanidation
As shown in Figure 5.a, cyanidation of the biooxidized sample A has lower extraction slopes than the direct cyanidation which indicated that the rate of gold dissolution of the biooxidation products was slower. This was associated with re-dissolution of the precipitated sulfur and iron species that were precipitated during the biooxidation step. The dissolution reactions for gold and silver was then competing with the reactions of thiocyanate and ferrocyanide formations, and hence lowering the dissolution rate of gold. Biooxidation of sample B with SKC1 improved gold extraction level from 47.8% (without pretreatment) to 62.8%. SKC1 bacterium has an ability to use both organic and inorganic carbon as their carbon sources that leads to reduction of preg-robbing effect of carbonaceous
ACCEPTED MANUSCRIPT
T
matters in Sample B. However, part of gold is still un-exposed and encapsulated within the sulfide minerals. Biooxidation of sample B with SKC2 slightly increases gold extraction of un-oxidized concentrate from 47.8% to 51.3%. This bacterium was also able to consume carbonaceous matters in sample B due to its mixotrophic nature but not as effective as SKC1. Moreover, SKC2 also oxidized sulfide minerals in sample B as it also did to sample A.
NU
SC R
IP
The cyanidation of the biooxidized concentrated with AC for sample A and B showed nearly similar results to that of direct cyanidation. The gold extraction percentage for sample A and B were 73.0% and 50.3%, respectively. The gold extraction for the control tests were 71.8% and 54.6% for sample A and sample B, respectively. These control tests were carried out using similar biooxidation medium of the BIOX tests using AC bacteria. Insignificant improvements of gold recoveries from sample A and sample B which were pre-treated with AC bacteria indicated inappropriate growth of AC bacteria during the course of the biooxidation process.
AC
CE P
TE
D
MA
Cyanide and Lime Consumption NaCN and lime consumptions for cyanidation of the biooxidized concentrates with mixotrophic bacteria was higher than that of direct cyanidation for both samples as illustrated in Figure 6. NaCN consumption for cyanidation of sample A was increased from 3.72 kg/ton to 9.99 kg/ton and 8.37 kg/ton when the concentrate samples were pre-treated with SKC1 and SKC2, respectively. This was also similar for sample B, in which NaCN consumption was increased from 3.01 kg/ton to 7.77 kg/ton and 6.66 kg/ton, for the BIOX pretreatments using SKC1 and SKC2, respectively. Higher NaCN consumption was due to the addition of sulfur and iron nutrient during the biooxidation process which was precipitated during BIOX and re-dissolved in the cyanidation step. The excess of sulfur reacts with cyanide ions to form thiocyanate anion complexes, while dissolved iron reacts with cyanides ions to form ferrocyanide complexes. Lime consumptions for cyanidation of the biooxidized concentrates with SKC2 were decreased from 3.99 kg/ton to 2.87 kg/ton for sample A and from 3.25 kg/ton to 1.7 kg/ton for sample B. Reducing of lime consumption for cyanidation of bioxidized concentrates by SKC1 and SKC 2 might be in correlation with hydroxyl (OH-) ions generation and ammonification (i.e. the production of NH3) by bacterial cells. Since the concentrates contain significant amount of silica, an increase in solution pH in experimental systems with silica may occur through bacterial production of CO2 with concomitant release of OH-, creating a highly alkaline microenvironment around active cells (Schultze Lam, Fortin, Davis and Beveridge, 1996). During ammonification by bacterial cells, hydrogen ion is consumed when amino acids from dead cells are metabolized, thereby forming NH3 (Lee and Beveridge, 2001). The increased pH remained stable due to the presence of silica which presumably supported the buffering capacity of the solution (Chaerun, Tazaki, Asada and Kogure, 2005).
ACCEPTED MANUSCRIPT 12
Sample A
NaCN
Lime
10
T
6
IP
kg/ton
8
2 0 A-Direct A-Control cyanidation
A-AC
9 8
Sample B
kg/ton
6 5 4
Lime
D
3
TE
2
0
NaCN
A-SKC2
MA
7
1
A-SKC1
NU
(a)
SC R
4
CE P
Direct B-Control cyanidation
B-AC
B-SKC1
B-SKC2
AC
(b) Figure 6. Cyanide and lime consumptions during direct cyanidation and cyanidation of the biooxidized concentrates (a) sample A and (b) sample B with pure culture AC (A-AC and B-AC), iron- and sulfur- oxidizing mixotrophic bacteria SKC1 (A-SKC1 and B-SKC1) and sulfur-oxidizing mixotrophic bacteria SKC2 (A-SKC2 and B-SKC2).
Conclusions The present study demonstrated that biooxidation of the refractory gold concentrates using iron- and sulfur-oxidizing mixotrophic bacterium of SKC1 or sulfur-oxidizing mixotrophic bacterium of SKC2 was able to increase gold extraction during cyanidation. The bacteria were able to treat both sulfide type of refractory gold concentrate and preg-robbing type at neutral pH. The SKC1 and SK2 bacteria are able to generate EPS that forms Fe(III)-EPS anionic complex which prevents passivation of sulfide minerals caused by iron and sulfur precipitation on the surface of the minerals during the BIOX process. However, cyanide consumption of the biooxidized products with mixotrophic bacteria was increased due to reactions of precipitated sulfur and iron during cyanidation with cyanide ions to form thiocyanate and ferrocyanide ions. Biooxidation mechanisms using SKC1 and SKC2 bacteria were dissimilar and still poorly understood. More comprehensive analysis of the formed EPS substances and morphology analysis of the biooxidation products need to be further studied.
ACCEPTED MANUSCRIPT
SC R
IP
T
Acknowledgements This research work was supported by the Geomicrobiology-Biomining & Biocorrosion Laboratory, Biosciences and Biotechnology Research Center of Institut Teknologi Bandung. We thank Mr. Wiku Padmonobo, Manager of Geo-metallurgy Department of PT Geoservice, for providing gold concentrate sample and analytical supports as well as permission to publish the research result.
AC
CE P
TE
D
MA
NU
References Aymore, M.G. and Jaffer, A., 2012. Evaluating process options for treating some refractory ores. ALTA 2012 International Gold Conference, May 31-June 1 2012, Burswood Convention Centre, Perth, Western Australia Amankwah, R.K., Yen, W.T., Ramsay, J.A., 2005. A two-stage bacterial pretreatment process for double refractory gold ore. Minerals Engineering 18(1): 103-108. Brierley, J., Kulpa, C.F., 1992. Microbial consortium treatment of refractory precious metal ores: USA, 5127942. Brierley, C.L., 1995. Bacterial oxidation. Engineering and Mining Journal, 196, 42–44. Brierley, C.L., Brans, R., 1994. Selection of Bactech’s thermophilic biooxidation process for Youanmi Mine. In: Biomine ’94, Australian Mineral Foundation, Glenside, South Australia, pp. 5.1– 5.7. Brierley, C.L., 2008. How will biomining be applied in future, Transactions of Nonferrous Metal Society of China 18, 1302-1310. Brierley, C.L., 2013. Mining Biotechnology: Research to Commercial Development and Beyond, Biomining: Theory, Microbes and Industrial Processes, 3-17. Celep, O., Alp, I., Deveci, H., 2011. Improved gold and silver extraction from a refractory antimony ore by pretreatment with alkaline sulphide leach. Hydrometallurgy 105, 234239. Chaerun, S.K., Tazaki,K., Asada, R. and Kogure, K, 2005. Interaction between clay minerals and hydrocarbon-utilizing indigenous microorganisms in high concentrations of heavy oil: implications for bioremediation, Clay Minerals Vol. 40, p.105-114. Chowdhury, F., Ojumu, T.V., 2014. Investigation of ferrous-iron biooxidation kinetics by Leptospirillum ferriphilum in a novel packed-column bioreactor: Effects of temperature and jarosite accumulation. Hydrometallurgy 141, 36-42. Cui, J., Zhang, L.,2008. Metallurgical recovery of metals fromelectronic waste: A review, Journal of Hazardous Materials 158, 228-256. Diao, M., Nguyen, T.A.H., Taran, E., Mahler, S., Nguyen, A.V., 2014. Differences in adhesion of A. thiooxidans and A. ferrooxidans on chalcopyrite as revealed by atomic force microscopy with bacterial probes. Frizan V., Giaveno A., Chiacchiarini P., Donati E., 2003. Bioleaching of Argentinean sulfide ores using pure and mixed cultures. 15th International Biohydrometallurgy Symposium (IBS 2003)
ACCEPTED MANUSCRIPT
AC
CE P
TE
D
MA
NU
SC R
IP
T
Fu, K., Lin, H., Luo, D., Jiang, W., Zeng, P., 2014. Comparison of bioleaching of copper sulphides by Acidithiobacillus ferrooxidans. African Journal of Biotechnology 13, 664672 Fu, Y., Yang, H., Cui, R., Fan, Y., Wang, C., 2010. Role of extracellular polymers of bacteria during arsenic-bearing gold concentrate bio-oxidation process. The Chinese Journal of Nonferrous Metals, 20(10): 2057-2062 (in Chinese) Goodall, W.R., Leatham, J.D., Scales, P.J., 2005. A new method for determination of pregrobbing in gold ores. Minerals Engineering 18, 1135-1141. Gu, S., Su, L., Chen, M., Sun X., Zhou, H., 2010. Bio-leaching effects of Leptospirillum ferriphilum on the surface chemical properties of pyrite. Mining Science and Technology 20, 287-291. Hackl, R.P., 1997. Commercial applications of bacterial-mineral interactions. In: McIntosh, J.M., Groat, L.A. (Eds.), Biological- Mineralogical Interactions. Mineralogical Association of Canada, pp. 143–167. Hansford, G.S., Chapman, J.T., 1992. Batch and continuous biooxidation kinetics of a refractory gold-bearing pyrite concentrate. Minerals Engineering, Vol. 5, No. 6, pp. 597-612. Harris, D.C., 1997. Quantitative Analytical Chemistry. W.H. Freeman and Company, New York, NY. Helm, M., Vaughan J., Staunton, W.P. and Avraamides, J., (2009): An investigation of the carbonaceous component of preg-robbing gold ores. Proceedings of World Gold Conference 2009, 139-144. Jonglertjunya, W., 2003. Bioleaching of chalcopyrite. Thesis. Department of Chemical Engineering School of Engineering, The University of Birmingham. Lee, J.U. and Beveridge, T.J., 2001. Interaction between iron and Pseudomonas aeruginosa biofilms attached to Sepharose surfaces, Chemical Geology Vol. 180, p. 67-80. Livesey-Goldblatt, E., Norman, P., Livesey-Goldblatt, D.R., 1983. Gold recovery from arsenopyrite/pyrite ore by bacterial leaching and cyanidation. In: Rossi, G. and Torma, A.E. (Eds.), Recent Progress in Biohydrometallurgy. Assoc. Mineraria Sarda, Iglesias, pp. 627-641. Marsden, J.O., House, C.I., 1992. The Chemistry of Gold Extraction, Ellis Horwood, British. Miller, J.D., Wan, R.Y. and Diaz, X., (2005). Preg-robbing Gold Ores. Development of Minerals Processing, Chapter 38. Elsevier. Rawlings, D.E., 1998. Industrial practice and the biology of leaching of metals from ores. Journal of Industrial Microbiology and Biotechnology 20, 268–274. Sand, W., Gehrke, T., Hallmann, R., Schippers, A., 1995. Sulphur chemistry, biofilm, and the (in)direct attack mechanism - a critical evaluation of bacterial leaching. Appl. Microbiol. Biotechnol. 43, 961–966. Sand, W., Gehrke, T., Jozsa, P., Schippers, A., 2001. (Bio)chemistry of bacterial leachingdirect vs. indirect bioleaching. Hydrometallurgy 59, 159-175. Schultze-Lam, S, Fortin, D., Davis, B.S. and Beveridge, T.J., 1996. Mineralization of bacterial surfaces, Chemical Geology, Vol.132, p. 171-181. Vilcáez, J., Inoue, C., (2008). Bioleaching of chalcopyrite with thermophiles:Temperature– pH–ORP dependence. Int. J. Mineral Proc.88 (1-2): 37-44.
ACCEPTED MANUSCRIPT
AC
CE P
TE
D
MA
NU
SC R
IP
T
Wang, A., Zhang, Y., Liu, H., 2000. The property of carboniferous species and its effect on leaching of gold in Dongbeizhai Gold Mine. Multipurpose Utilization of Mineral Resources 3, 4-8 (in Chinese) Yang, F., Xu, X., Zhao, J., Liang, Z., 2003. Experimental study on bacterial oxidation of carbon-bearing higharsenic refractory gold concentrates. Gold, 24(4): 37-39 (in Chinese). Yang, H., Liu, Q., Song, X., Dong, J., 2013. Research status of carbonaceous matter in carbonaceous gold ores and bio-oxidation pretreatment. Trans. Nonferrous Met. Soc. China 23, 3405-3411. Yang, Y. Xie, Z., Xu, B., Li. Q, and Jiang T., 2015. Gold Extraction from a High Carbon Low-Grade Refractory Gold Ore by Flotation-Roasting-Leaching Process. Rare Metal Technology, Published online, 20 February 2015, 63-70.
ACCEPTED MANUSCRIPT
AC
CE P
TE
D
MA
NU
SC R
IP
T
Highlight: • Biooxidation (BIOX) of refractory gold concentrates using iron- and sulfur-oxidizing mixotrophic bacterium of SKC1 and sulfur-oxidizing mixotrophic bacterium of SKC2 was able to increase gold extraction during cyanidation • The highest gold extraction from high sulfidation concentrate of 91.4%, was achieved by BIOX with SKC2 for 14 days. • The SKC1 and SK2 bacteria are able to generate EPS that forms Fe(III)-EPS anionic complex which prevents passivation of sulfide minerals • The bacteria were able to treat both sulfide type of refractory gold concentrate and preg-robbing type at neutral pH.