- Email: [email protected]
Depression of pyrite in the flotation of high pyrite low-grade lead–zinc ore using Acidithiobacillus ferrooxidans
Minerals Engineering 23 (2010) 10–16
Contents lists available at ScienceDirect
Minerals Engineering journal homepage: www.elsevier.com/locate/mineng
Depression of pyrite in the flotation of high pyrite low-grade lead–zinc ore using Acidithiobacillus ferrooxidans J.V. Mehrabani a,*, M. Noaparast a, S.M. Mousavi b,c,**, R. Dehghan d, E. Rasooli a, H. Hajizadeh e a
School of Mining Engineering, College of Engineering, University of Tehran, Tehran, Iran Biotechnology Group, Chemical Engineering Department, Tarbiat Modares University, Tehran, Iran c Department of Chemical Technology, Lappeenranta University of Technology, Lappeenranta, Finland d School of Mining and Metallurgical Engineering, Yazd University, Yazd, Iran e Research and Development Center, Kooshk Lead and Zinc Mine, Yazd, Iran b
a r t i c l e
i n f o
Article history: Received 16 May 2009 Accepted 25 August 2009 Available online 9 October 2009 Keywords: Bacteria Froth flotation Sulfide ores Sodium cyanide Process optimization
a b s t r a c t In this research work, selective depression of pyrite in the flotation of a low-grade lead–zinc ore containing 31% pyrite was investigated in the absence and presence of Acidithiobacillus ferrooxidans. Pyrite was significantly depressed with these bacteria using the optimum dosage of reagents. Sphalerite recovery and Zn grade in the obtained sphalerite concentrate were both enhanced by bacteria. The results of bioflotation experiments showed that bacterial depression of pyrite is very sensitive to the concentration of other flotation reagents. Least significance difference (LSD) bars were used to study the significance of the factors under study with a confidence interval of 90%. Under similar conditions, A. ferrooxidans was seen to increase the Zn grade and separation efficiency more effectively than sodium cyanide. Flotation kinetic studies confirmed a considerable decrease in the pyrite kinetic rate constant in the presence of A. ferrooxidans. Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction The beneficiation of complex base-metal sulfide ores is generally based on the selective production of zinc, lead, and copper concentrates from which the respective metals are extracted using metallurgical processes (Carta et al., 1980). Separation of sphalerite through copper activation becomes complicated when other minerals within the pulp are inadvertently activated along with the sphalerite. Pyrite (FeS2) is one mineral that responds to copper activation and can be floated together with sphalerite (Wills and Napier-Munn, 2006). Complete elimination of iron sulfides (e.g., pyrite, pyrrhotite) from zinc concentrates is economically attractive from the angle of subsequent smelting (Gaudin, 1957). This goal can be achieved by floatation in alkaline solutions (lime) using highly selective inorganic modifiers such as cyanides, sulfites, and ferrocyanides in combination with zinc sulfate (Fuerstenau et al., 1985; Shen et al., 1998). Cyanides have been one of the most commonly used depressants; however, their use has raised much concern in regards to environmental issues. Additionally, depletion of available
* Corresponding author. Tel.: +98 21 82084397; fax: +98 21 88008838. ** Corresponding author. Address: Biotechnology Group, Chemical Engineering Department, Tarbiat Modares University, Tehran, Iran. Tel.: +98 21 82884917; fax: +98 21 82883381. E-mail addresses: [email protected] (J.V. Mehrabani), [email protected] (S.M. Mousavi). 0892-6875/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.mineng.2009.08.008
easy-to-process mineral resources will most likely lead to a search for more advanced solutions to the problem of beneficiation of some refractory ores in cases where conventional flotation or flocculation approaches yield poor results (Carta et al., 1980). The utility of microorganisms in mineral beneficiation has been recently elucidated. Recent developments in biotechnology hold promise for processing of such difficult-to-treat ores as well as for safeguarding the environment. Bioflotation is a relatively new method for processing ores; it is defined as ‘‘the selective separation of commercial gangue ores through interactions with microorganisms” (Deo and Natarajan, 1997). Compared to conventional inorganic reagents such as cyanides, hydrosulfides, dichromate, etc., bacteria are non-toxic and environmentally benign. Many investigations have suggested that certain types of bacteria, such as Acidithiobacillus ferrooxidans, may prevent flotation of certain minerals, such as pyrite (Sharma and Hanumantha, 2001). A. ferrooxidans, a commonly implicated autotrophic, acidophilic, and mesophilic microorganism utilizes sulfur, thiosulfate, and iron as energy sources. These bacteria are Gram-negative and shaped as small rods with dimensions 0.5 by 1–3 lm, occur singly or occasionally in pairs, and have been extensively utilized in mineral bioprocessing (Donati and Sand, 2007). The selective separation of chalcopyrite, sphalerite, or arsenopyrite from pyrite has been studied in the presence of A. ferrooxidans (Chandraprabha et al., 2004a,b; Deshpande et al., 2001, 2004). These papers discuss the utility of A. ferrooxidans for selective flotation of minerals from pyrite. With extraction using
11
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
bacterial cells, pyrite is depressed even in the presence of a potassium isopropyl xanthate collector. The effect of L. ferrooxidans on the floatability of chalcopyrite, sphalerite, and pyrohotite was also investigated using xanthate as a collector (Pecina et al., 2009). In this research, the chalcopyrite flotation rate was significantly increased in the presence of L. ferrooxidans due to the formation of hydrophobic species. It was concluded that L. ferrooxidans causes superficial changes mainly due to oxidation of minerals. The floatabilities of five sulfide minerals (pyrite, chalcocite, molybdenite, millerite, and galena) were examined in the presence of A. ferrooxidans by Nagaoka and co-workers (1999). It was observed that pyrite was significantly depressed by the bacterium, while the floatability of other sulfide minerals was not affected. It was postulated that the suppression of pyrite floatability was caused by profuse bacterial addition to pyrite surfaces (Nagaoka et al., 1999). The majority of bioflotation studies have been carried out on the micro-scale using pure minerals, but there are few studies on the primary ores (Hosseini et al., 2005; Kolahdoozan et al., 2004; Yüce et al., 2006). In the present research, the efficiencies of A. ferrooxidans and the chemical depressant NaCN were compared for selective depression of pyrite in the flotation of a primary Pb–Zn ore. Moreover, selective separation of sphalerite and pyrite was investigated in the absence and presence of A. ferrooxidans and NaCN. 2. Materials and methods 2.1. Ore characterization A low grade Pb–Zn ore sample, containing 31% pyrite and 44% dolomite and assayed at 2.34% Pb, 6.91% Zn and 15.36% Fe, was prepared from the Kooshk mine in Yazd province, Iran, and used as the primary ore in flotation and bioflotation experiments. Mineralogical and elemental compositions of the ore were studied using semi-quantitative X-ray diffraction (SQXRD) and atomic absorption spectroscopy (AAS) techniques, respectively. 2.2. Bacterial culture Pure strains of A. ferrooxidans and Acidithiobacillus thiooxidans, isolated from the acidic water drainage of Sarcheshmeh copper mine (Iran), were used in this study. The bacteria were grown in the laboratory using 9 k medium (3 g/l (NH4)2SO4, 0.5 g/l MgSO4 7H2O, 0.5 g/l K2HPO4, 0.1 g/l KCl and pH = 1.85), and were cultured by inoculating 10 ml of pure strain of the bacterial cells into the medium. Potassium nitrate was used to maintain ionic strength. The cultures were incubated at 32 °C in a rotary shaker maintained at 160 rpm. Microorganisms were adapted with 50 g/l original ore. A. ferrooxidans and A. thiooxidans were grown separately on ferrous sulfate and elemental sulfur, respectively. The culture solution was filtered through Whatman filter paper to remove the suspended solids. The cells were separated from the medium using the biological filter paper (0.42 lm) and then suspended in distilled water. During the experiments, the cell number in the solution was estimated by direct counting, using a Thoma chamber of 0.1 mm depth and 0.0025 mm2 area with an optical microscope. 2.3. Flotation and bioflotation experiments Flotation and bioflotation experiments were performed in a 1.5 l Denver cell running at 850 rpm using 300 g ore sample (25% pulp density) with a size of 95 lm (d90). Industrial grade potassium ethyl xanthate (PEX) and potassium amyl xanthate (PAX) were
used as galena and sphalerite collectors, respectively. Copper sulfate and sodium hydroxide, which were used for sphalerite activation and pH adjustment, were of analytical grade. Four control factors and their levels used in the bioflotation experiments are presented in Table 1. Each of these factors was varied in two levels, including collector dosage, activator dosage, NaCN dosage, and volume of bacterial solution (A. ferrooxidans). Fig. 1 describes the baseline of the flotation and bioflotation experiments. In all experiments, organic materials in the ore were pre-floated using 120 g/t methyl isobutyl carbonyl (MIBC) and 250 g/t diesel oil. For this purpose, the pulp was conditioned for 2 min at pH = 7– 7.5 followed by froth collection for 4 min, and the impeller speed was 850 rpm. Pulp pH was adjusted to 9.5 and 11 in the conditioning stage of galena and sphalerite, respectively. In bioflotation experiments, in which bacteria were used as the pyrite depressant, the minerals were conditioned after pre-floating with 300 ml bacteria solution for 20 min, using an impeller speed of 120 rpm. The bacterial population in the solution was counted at about 3 107 cells/ml. In the sphalerite frothing stage, the flotation froth was collected for 7.5 min. According to Table 2, nineteen flotation experiments in a full factorial design (H1–H16) with the center points (H17–19) were carried out. These experiments are as follows: – – – –
Four experiments without pyrite depressant. Four experiments using NaCN as pyrite depressant. Four experiments using A. ferrooxidans as pyrite depressant. Four experiments using both NaCN and A. ferrooxidans as pyrite depressant. – Three experiments using the center points of the factors to estimate operator error in the tests.
2.4. Separation efficiency Separation efficiency was calculated as an indicator of the metallurgical performance of the sphalerite flotation process using the equation:
SE ¼ 100C
mðc f Þ ; f ðm f Þ
ð1Þ
where C is the weight percent of the feed to the concentrate, m is theoretical zinc content of the sphalerite mineral, c is Zn grade of sphalerite concentrate, and f is Zn grade in the original feed ore sample. 2.5. Least significant difference (LSD) LSD is a numerical value that can be used as a benchmark for comparing treatment means. The LSD is calculated as a comparative tool with:
LSD ¼ t s
pffiffiffiffiffiffiffiffiffiffiffiffi ð2=nÞ;
ð2Þ
where t is a factor that depends on the desired confidence and the degrees of freedom for estimation of error, s is standard deviation, and n is the sample size. The t value obtained from the t-distribution table under the desired conditions was 2.92 (Anderson and Table 1 Studied factors and levels in flotation and bioflotation experiments. Level
A: PAX (g/t)
B: CuSO4 (g/t)
C: NaCN (g/t)
D: bacterial solution (ml)
High Low Center
140 280 210
700 1400 1050
0 100 50
0 300 150
12
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
Fig. 1. Baseline for experimental procedure.
Whitcomb, 2000). When the LSD is exceeded, the means are considered to be significantly different. Before analysis of the experimental results, LSD values for sphalerite and pyrite recovery were calculated. The standard devi-
ation in the recoveries due to the errors in sample analysis and flotation experiments are presented in Table 3. For estimation of the errors in sample analysis, four products of an experiment (including the concentrate from the pre-flotation stage, galena concen-
Table 2 Experimental plan for flotation tests and results in sphalerite concentrate. Std
A
B
C
D
Zn (%)
Fe (%)
Sphalerite recovery (%)
Pyrite recovery (%)
Zn SE (%)
H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19
140 280 140 280 140 280 140 280 140 280 140 280 140 280 140 280 210 210 210
700 700 1400 1400 700 700 1400 1400 700 700 1400 1400 700 700 1400 1400 1050 1050 1050
0 0 0 0 100 100 100 100 0 0 0 0 100 100 100 100 50 50 50
0 0 0 0 0 0 0 0 300 300 300 300 300 300 300 300 150 150 150
15.22 11.52 11.76 9.57 18.63 13.12 18.16 18.61 20.84 11.85 15.04 10.63 20.12 18.20 11.67 19.15 15.56 17.42 16.28
14.23 14.44 22.49 22.88 15.58 16.51 11.81 10.98 11.52 15.20 20.14 18.94 10.55 14.33 11.94 12.69 13.51 14.40 15.16
65.91 64.39 73.77 70.69 77.20 72.53 69.22 76.11 74.03 70.91 73.19 72.95 73.11 76.12 71.36 72.57 76.91 79.39 75.10
38.11 52.98 53.02 59.62 25.31 32.88 24.02 24.62 23.52 49.79 41.11 54.22 20.53 22.81 21.07 23.58 29.94 25.59 28.41
37.59 28.40 36.06 28.01 53.12 41.05 47.21 52.29 56.98 32.78 42.72 32.11 53.53 52.06 49.08 51.51 47.83 49.98 49.90
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
13
trate, sphalerite concentrate, and final tailings) were analyzed three times on different days. Afterwards, the recoveries were calculated and the standard deviations were estimated. Three bioflotation experiments including H17, H18 and H19 were carried out on three different days under exactly the same conditions and the products were analyzed to determine the experimental errors. In the 90% confidence interval, the LSD values for sphalerite and pyrite recoveries were calculated as 7.39% and 7.1%, respectively (Table 3). The LSD values for Fe and Zn grades in the sphalerite concentrate were also calculated as 1.81% and 1.03%, respectively. As an example of using the LSD value in analysis of the experimental results, sphalerite recovery was increased from 64% to 74% by changing the collector dosage. Therefore, comparing the amount of increase (10%) with the LSD value of sphalerite recovery (7.39%), it can be postulated that the effect of collector dosage is significant. LSD values were indirectly considered in analysis of all results of flotation and bioflotation experiments.
3. Results and discussion The recovery values for sphalerite and pyrite minerals, together with Zn separation efficiency (SE) and Zn and Fe grade in the sphalerite concentrate, were calculated as the process responses. The experimental conditions and the results of flotation and bioflotation experiments are presented in Table 2.
3.1. Effect of collector and activator 3.1.1. In absence of depressant The effect of the interaction between collector and activator dosages in the absence of any pyrite depressant on the pyrite recovery is shown in Fig. 2a. From this figure, it is apparent that pyrite recovery was significantly increased (according to the LSD bar), at a low level of activator dosage (700 g/t), by increasing the collector dosage. At a high level of activator concentration (1400 g/t), the increase in pyrite recovery with variation of the collector dosage was not significant, because the LSD bars cover each other. At a low level of activator and collector dosage, only 38% pyrite was floated, while doubling of the concentrations of reagents caused about 52–59% of pyrite to be floated. Therefore, using a low level of the collector at pH 11, pyrite recovery was significantly enhanced with an increase in CuSO4 dosage. However, batch flotation tests on Red Dog ore as reported by (Dichman and Finch, 2001) showed that upon addition of copper, pyrite recovery was decreased significantly. It can be understood from Fig. 2b that changing the collector dosage at the designed point did not improve sphalerite recovery, but with a doubling of CuSO4 at a low level of collector, sphalerite recovery was enhanced significantly from 65.93% to 73.77%. Comparing the results of experiments H1–H4 in Table 2, which were carried out in the absence of depressant, it can be concluded that the highest selective separation between sphalerite and pyrite was obtained using 140 g/t collector and 700 g/t CuSO4 in experiment H1.
Table 3 Errors in sphalerite and pyrite recoveries. Source
Sample analysis Flotation and bioflotation Total
Standard deviations in recovery (%) Sphalerite
Pyrite
0.92 2.18 3.10
0.78 2.20 2.98
Fig. 2. Interaction effect of collector and activator dosage on (a) pyrite recovery and (b) sphalerite recovery.
3.1.2. In the presence of NaCN It is fully clear from Table 2 (H5–H9) that at a low level of activator dosage (700 g/t), pyrite recovery was significantly increased by increasing the collector dosage, while at a high level of activator concentration (1400 g/t), the effect of the collector was negligible. However, pyrite recovery became relatively more stable with changes in the collector and activator dosages in the presence of NaCN, as compared with experiments H1–H4. At a low level of CuSO4, the Zn grade decreased from 18.63% to 13.61% with changes in the collector dosage (from 140 to 280). The Fe grade in the concentrate was significantly decreased when the CuSO4 concentration was doubled.
3.1.3. In the presence of bacteria Fig. 3a shows that sphalerite recovery remained relatively stable with changes in collector and activator dosages in the presence of bacteria. However, Fig. 3a illustrates that at the higher concentrations of the activator and the collector (experiments H10– H12), pyrite was floated at a factor of two times greater than the results obtained in test H9. Because of the negative effect of collector and activator dosage increases, an decrease in Zn grade and increase in Fe grade can be observed from Fig. 3b. Comparing the results of pyrite recovery in experiments H9– H12 with the results of experiments H5–H8, it is obvious that pyr-
14
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
3.2. Comparison between depressant efficiency of bacteria and NaCN In previous discussions, the optimum process conditions used to minimize pyrite flotation were determined in the presence of each depressant and also without any pyrite depressant. These conditions were achieved in experiments H1, H5, H9, and H13. All of these experiments were carried out using low levels of flotation reagents (140 g/t PAX and 700 g/t CuSO4). The results of these four experiments are compared in Fig. 4. The efficiency of NaCN on pyrite depression can be evaluated by comparing the results of experiments H1 and H5. Sphalerite recovery was increased from 65.91% to 77.21% in the presence of sodium cyanide, while pyrite recovery was decreased from 38.1% to 25.3%. Comparing the results of experiments without depressant and with bacteria, it is clear that pyrite was significantly depressed in the presence of A. ferrooxidans, and recovery decreased from 38% to 23%. However, sphalerite recovery was improved from 64.91% to 74.03%. In contrast, the Zn grade in the sphalerite concentrate increased from about 15.22% to 20.84%, and the Fe grade was decreased to 11.52 when using bacteria. It is apparent from Fig. 4 that
Fig. 3. Effect of A. ferrooxidans on (a) sphalerite and pyrite recoveries and (b) Zn and Fe grades in the sphalerite concentrate.
ite recovery was influenced more considerably by changes in flotation reagents when bacteria were used as the depressant. In this category of experiments, the highest sphalerite recovery was obtained with the least pyrite recovery in the H9 test. In this condition, sphalerite and pyrite recoveries were reached to 74.03% and 23.52%, respectively, and Zn and Fe grade were reached to 20.74% and 11.52%, respectively. 3.1.4. In the presence of both NaCN and bacteria In experiments H13–H16, both bacteria and NaCN were used as pyrite depressants. It is apparent from the results in Table 2 that, in this case, changing the collector or activator concentration does not affect the sphalerite and pyrite recoveries at the significance level under study. Comparing the results of experiments H13–H16 with the experiments in which only bacteria or NaCN were used as the depressant, the synergistic effects of two depressants on both pyrite recoveries and Fe grades can be observed. The smallest amounts of pyrite recovery and lowest Fe grade were also achieved in experiment H13, in which both bacteria and NaCN were used together with low levels of other flotation reagents.
Fig. 4. Effect of different depressants on (a) sphalerite and pyrite recoveries and Zn separation efficiency (SE), and (b) Zn and Fe grades.
15
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
Zn separation efficiency was improved when either bacteria or NaCN were used as the pyrite depressant. Meanwhile, the highest separation efficiency was achieved in the presence of bacteria. Since A. ferrooxidans obtain their energy by oxidizing iron ions and elemental sulfur, the cells might have strongly adsorbed on the pyrite surface. A. ferrooxidans can affect mineral surfaces by a direct mechanism; the bacteria eliminate the occurrence of oxidized sulfur which has hydrophobic properties and induces a higher floatability for minerals, so that the hydrophobicity of pyrite is decreased. In other words, the bacteria can suppress floatability, and the postulated mechanism of this suppression is an increase in surface hydrophilicity due to adhesion of bacterial cells (Ohmura et al., 1993). The efficiencies of bacteria and NaCN in pyrite depression can be compared using the results of experiments H5 and H9 (Fig. 4). For pyrite depression, it is clear that A. ferrooxidans was as effective as sodium cyanide, and pyrite recoveries were 25.3% and 23.5% when NaCN and bacteria were used as the depressants, respectively. Moreover, the Zn grade in the sphalerite concentrate was increased from 18.63% to 20.84% and the Fe grade decreased from 15.58% to 11.52%. These changes are both significant, referring to the LSD values of grades. Under these conditions, both sphalerite and pyrite recoveries were approximately the same in the presence of NaCN and A. ferrooxidans. From observation of the same recoveries and different grades, it can be understood that less gangue minerals appeared in the sphalerite concentrate in the presence of A. ferrooxidans. These results illustrate the capability of A. ferrooxidans for depression of pyrite and other gangue minerals in comparison
to NaCN. The results of experiment H13, in which both bacteria and NaCN were used, are similar to the results of H9. 3.3. Flotation kinetics It is widely accepted that most flotation systems can be described by a first-order reaction rate equation (Eq. (3)).
R ¼ 1 ekt ;
ð3Þ
where R is the cumulative recovery, t and k are the time and kinetic constant of flotation. The constant k depends on a large number of variables, which include factors related to ore characteristics like mineralogy and factors defined through the flotation medium such as the type and quantity of reagents, among others (Kelebek and Nanthakumar, 2007). Fig. 5 shows that the flotation of sphalerite and pyrite follows first-order kinetics in the presence and absence of A. ferrooxidans with 140 g/t PAX and 700 g/t CuSO4. The results also show that the pyrite flotation rate was cut to half with the addition of bacteria. In contrast, the sphalerite kinetic constant was increased. 3.4. Validation experiments For confirmation of the bacteria’s effect, two experiments were repeated at the optimum conditions of reagent with bacteria. The procedure in these experiments was exactly the same as in previous tests; however, in the new tests, the frothing time was decreased from 7.5 to 6.5 min One test was carried out with A. ferrooxidans and the next with A. thiooxidans. Results are shown in Table 4. In the validation experiments, in addition to confirmation of the pyrite depression capability of bacteria in the sphalerite concentrate, the Zn grade surged from 20.84% to 26.21% and pyrite recovery declined from 23.56% to 15.825%. 4. Conclusions In the present work, 19 experiments were designed, and flotation and bioflotation tests were carried out in the presence and absence of pyrite depressant. The following major conclusions can be made based on this study:
Fig. 5. Effect of A. ferrooxidans on kinetic constants of sphalerite and pyrite flotation in the sphalerite concentrate at optimum reagent dosage. Kbsp and Kbpy are kinetic constants for sphalerite and pyrite in the presence of bacteria, respectively, while Ksp and Kpy are kinetic constants of sphalerite and pyrite in the absence of bacteria, respectively.
1. Maximum pyrite depression in the absence of any depressant was obtained using 140 g/t collector and 700 g/t CuSO4. 2. Sphalerite and pyrite recoveries were stable with changes in the concentrations of collector and activator when NaCN was used as a pyrite depressant. 3. In the presence of A. ferrooxidans, pyrite recovery was decreased from 38.11% to 23.52%, while sphalerite recovery and Zn grade were enhanced from 65.91% to 74.03% and from 15.22% to 20.84%, respectively. 4. Sphalerite and pyrite recoveries were the same when NaCN and bacterial depressants were compared. However, a 3% increase in Zn grade and a 4% decrease in Fe grade were observed when the bacteria were added as the pyrite depressant at optimum conditions.
Table 4 Validation experiments at optimum conditions. PAX (g/t)
140 140 140
CuSO4 (g/t)
700 700 700
Bacteria
A. ferrooxidans A. ferrooxidans A. thiooxidans
Frothing time (min)
7.5 6.5 6.5
Grade (%)
Recovery (%)
Zn
Fe
Sphalerite
Pyrite
20.84 26.21 24.32
11.52 11.94 13.57
74.03 74.68 74.29
23.52 15.82 18.73
16
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
5. In bioflotation experiments, the process responses were very sensitive to the concentrations of other flotation reagents. Therefore, reagent dosage is a significant factor affecting the bioflotation process, a factor that has been neglected in previously reported bioflotation studies. 6. The pyrite flotation rate constant decreased significantly when A. ferrooxidans interacted with the flotation pulp for 20 min. 7. Validation experiments with different frothing times showed that in the presence of A. ferrooxidans, 74% of sphalerite was floated and the Zn grade surged to 26.21%, while the recovery of pyrite was only 15.82%. Similar results were also obtained using A. thiooxidans.
Acknowledgement Authors would like to thank Professor K.A. Natarajan for his helpful comments on this work. References Anderson, M.J., Whitcomb, P.J., 2000. DOE simplified: practical tools for effective experimentation. Productivity Inc. Carta, M., Ghiani, M., Rossi, G., 1980. Beneficiation of a complex sulphide ores by an integrated process of flotation and bioleaching. In: Complex sulphide ores. Institute of Mining and Metallurgy, London, pp. 178–185. Chandraprabha, M.N., Natarajan, K.A., Modak, J.M., 2004a. Selective separation of pyrite and chalcopyrite by biomodulation. Colloids and Surfaces B: Biointerfaces 37, 93–100. Chandraprabha, M.N., Natarajan, K.A., Somasundaran, P., 2004b. Selective separation of arsenopyrite from pyrite by biomodulation in the presence of Acidithiobacillus ferrooxidans. Journal of Colloid and Interface Science 276, 323–332. Deo, N., Natarajan, K.A., 1997. Interaction of Bacillus polymyxa with some oxide minerals with reference to mineral beneficiation and environmental control. Minerals Engineering 10, 1339–1354. Deshpande, R.J., Subramanian, S., Natarajan, K.A., 2004. Surface-chemical studies on pyrite and arsenopyrite using Acidithiobacillus ferrooxidans. In: Rao, G.V., Misra,
V.N. (Eds.), Proceedings of the International Seminar on Mineral Processing Technology. Allied Publishers Limited, New Delhi, pp. 668–675. Deshpande, R.J., Subramanian, S., Natarajan, K.A., 2001. Effect of bacterial interaction with Thiobacillus ferrooxidans on the floatability of sphalerite and pyrite. In: Padmanabhan, N.P.H., Sreenivas, T., Srinivas, K. (Eds.), Proceedings of the International Seminar on Mineral Processing Technology. Allied Publishers Limited, New Delhi, pp. 159–161. Dichman, T.K., Finch, J.A., 2001. The role of copper ions in sphalerite–pyrite flotation selectivity. Minerals Engineering 14 (2), 217–225. Donati, E.R., Sand, W., 2007. Microbial Processing of Metal Sulfides. Springer. pp. 273-275. Fuerstenau, M.C., Miller, J.D., Kuhn, M.C., 1985. Chemistry of Flotation. AIME, New York. Gaudin, A.M., 1957. Flotation. Mapel Press Company, New York. pp. 436–453. Hosseini, T.R., Kolahdoozan, M., Tabatabaei, Y.S.M., Oliazadeh, M., Noaparast, M., Eslami, A., Manafi, Z., Alfantazi, A., 2005. Bioflotation of Sarcheshmeh copper ore using Thiobacillus ferrooxidans bacteria. Minerals Engineering 18, 371–374. Kelebek, S., Nanthakumar, B., 2007. Characterization of stockpile oxidation of penthlandite and pyrrhotite through kinetic analysis of their flotation. International Journal of Mineral Processing 84, 69–80. Kolahdoozan, M., Tabatabaei Yazdi, S.M., Yen, W.T., Hosseini Tabatabaei, R., Shahverdi, A.R., Oliazadeh, M., Noaparast, M., Eslami, A., Manafi, Z., 2004. Bioflotation of the low grade Sarcheshmeh copper sulfide. Transactions of the Indian Institute of Metals 57 (5), 485–490. Nagaoka, T., Ohmura, N., Saiki, H., 1999. A novel mineral flotation process using Thiobacillus ferrooxidans. Applied Environmental Microbiology 65, 3588–3593. Ohmura, N., Kitamura, K., Saiki, H., 1993. Mechanism of microbial flotation using Thiobacillus ferrooxidans for pyrite suppression. Biotechnology and Bioengineering 41, 671–676. Pecina, E.T., Rodriguez, M., Castillo, P., Diaz, V., Orrantia, E., 2009. Effect of Leptospirillum ferrooxidans on the flotation kinetics of sulphide ores. Minerals Engineering 22, 462–468. Sharma, P.K., Hanumantha, K.R., 2001. Role of heterotrophic bacteria in the bioflotation of some sulfide minerals. Mineral Biotechnol, 67–81. Shen, W.Z., Fornasiero, D., Ralston, J., 1998. Effect of collectors conditioning pH and gases in the separation of sphalerite from pyrite. Minerals Engineering 11, 145– 158. Wills, B.A., Napier-Munn, T.J., 2006. Mineral Processing Technology: An Introduction to the Practical Aspects of ore Treatment and Mineral Recovery, 7th ed. Butterworth, Heinemann. Yüce, A.E., Mustafa Tarkan, H., Zeki Do, M., 2006. Effect of bacterial conditioning and the flotation of copper ore and concentrate. African Journal of Biotechnology 5, 448–452.
Contents lists available at ScienceDirect
Minerals Engineering journal homepage: www.elsevier.com/locate/mineng
Depression of pyrite in the flotation of high pyrite low-grade lead–zinc ore using Acidithiobacillus ferrooxidans J.V. Mehrabani a,*, M. Noaparast a, S.M. Mousavi b,c,**, R. Dehghan d, E. Rasooli a, H. Hajizadeh e a
School of Mining Engineering, College of Engineering, University of Tehran, Tehran, Iran Biotechnology Group, Chemical Engineering Department, Tarbiat Modares University, Tehran, Iran c Department of Chemical Technology, Lappeenranta University of Technology, Lappeenranta, Finland d School of Mining and Metallurgical Engineering, Yazd University, Yazd, Iran e Research and Development Center, Kooshk Lead and Zinc Mine, Yazd, Iran b
a r t i c l e
i n f o
Article history: Received 16 May 2009 Accepted 25 August 2009 Available online 9 October 2009 Keywords: Bacteria Froth flotation Sulfide ores Sodium cyanide Process optimization
a b s t r a c t In this research work, selective depression of pyrite in the flotation of a low-grade lead–zinc ore containing 31% pyrite was investigated in the absence and presence of Acidithiobacillus ferrooxidans. Pyrite was significantly depressed with these bacteria using the optimum dosage of reagents. Sphalerite recovery and Zn grade in the obtained sphalerite concentrate were both enhanced by bacteria. The results of bioflotation experiments showed that bacterial depression of pyrite is very sensitive to the concentration of other flotation reagents. Least significance difference (LSD) bars were used to study the significance of the factors under study with a confidence interval of 90%. Under similar conditions, A. ferrooxidans was seen to increase the Zn grade and separation efficiency more effectively than sodium cyanide. Flotation kinetic studies confirmed a considerable decrease in the pyrite kinetic rate constant in the presence of A. ferrooxidans. Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction The beneficiation of complex base-metal sulfide ores is generally based on the selective production of zinc, lead, and copper concentrates from which the respective metals are extracted using metallurgical processes (Carta et al., 1980). Separation of sphalerite through copper activation becomes complicated when other minerals within the pulp are inadvertently activated along with the sphalerite. Pyrite (FeS2) is one mineral that responds to copper activation and can be floated together with sphalerite (Wills and Napier-Munn, 2006). Complete elimination of iron sulfides (e.g., pyrite, pyrrhotite) from zinc concentrates is economically attractive from the angle of subsequent smelting (Gaudin, 1957). This goal can be achieved by floatation in alkaline solutions (lime) using highly selective inorganic modifiers such as cyanides, sulfites, and ferrocyanides in combination with zinc sulfate (Fuerstenau et al., 1985; Shen et al., 1998). Cyanides have been one of the most commonly used depressants; however, their use has raised much concern in regards to environmental issues. Additionally, depletion of available
* Corresponding author. Tel.: +98 21 82084397; fax: +98 21 88008838. ** Corresponding author. Address: Biotechnology Group, Chemical Engineering Department, Tarbiat Modares University, Tehran, Iran. Tel.: +98 21 82884917; fax: +98 21 82883381. E-mail addresses: [email protected] (J.V. Mehrabani), [email protected] (S.M. Mousavi). 0892-6875/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.mineng.2009.08.008
easy-to-process mineral resources will most likely lead to a search for more advanced solutions to the problem of beneficiation of some refractory ores in cases where conventional flotation or flocculation approaches yield poor results (Carta et al., 1980). The utility of microorganisms in mineral beneficiation has been recently elucidated. Recent developments in biotechnology hold promise for processing of such difficult-to-treat ores as well as for safeguarding the environment. Bioflotation is a relatively new method for processing ores; it is defined as ‘‘the selective separation of commercial gangue ores through interactions with microorganisms” (Deo and Natarajan, 1997). Compared to conventional inorganic reagents such as cyanides, hydrosulfides, dichromate, etc., bacteria are non-toxic and environmentally benign. Many investigations have suggested that certain types of bacteria, such as Acidithiobacillus ferrooxidans, may prevent flotation of certain minerals, such as pyrite (Sharma and Hanumantha, 2001). A. ferrooxidans, a commonly implicated autotrophic, acidophilic, and mesophilic microorganism utilizes sulfur, thiosulfate, and iron as energy sources. These bacteria are Gram-negative and shaped as small rods with dimensions 0.5 by 1–3 lm, occur singly or occasionally in pairs, and have been extensively utilized in mineral bioprocessing (Donati and Sand, 2007). The selective separation of chalcopyrite, sphalerite, or arsenopyrite from pyrite has been studied in the presence of A. ferrooxidans (Chandraprabha et al., 2004a,b; Deshpande et al., 2001, 2004). These papers discuss the utility of A. ferrooxidans for selective flotation of minerals from pyrite. With extraction using
11
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
bacterial cells, pyrite is depressed even in the presence of a potassium isopropyl xanthate collector. The effect of L. ferrooxidans on the floatability of chalcopyrite, sphalerite, and pyrohotite was also investigated using xanthate as a collector (Pecina et al., 2009). In this research, the chalcopyrite flotation rate was significantly increased in the presence of L. ferrooxidans due to the formation of hydrophobic species. It was concluded that L. ferrooxidans causes superficial changes mainly due to oxidation of minerals. The floatabilities of five sulfide minerals (pyrite, chalcocite, molybdenite, millerite, and galena) were examined in the presence of A. ferrooxidans by Nagaoka and co-workers (1999). It was observed that pyrite was significantly depressed by the bacterium, while the floatability of other sulfide minerals was not affected. It was postulated that the suppression of pyrite floatability was caused by profuse bacterial addition to pyrite surfaces (Nagaoka et al., 1999). The majority of bioflotation studies have been carried out on the micro-scale using pure minerals, but there are few studies on the primary ores (Hosseini et al., 2005; Kolahdoozan et al., 2004; Yüce et al., 2006). In the present research, the efficiencies of A. ferrooxidans and the chemical depressant NaCN were compared for selective depression of pyrite in the flotation of a primary Pb–Zn ore. Moreover, selective separation of sphalerite and pyrite was investigated in the absence and presence of A. ferrooxidans and NaCN. 2. Materials and methods 2.1. Ore characterization A low grade Pb–Zn ore sample, containing 31% pyrite and 44% dolomite and assayed at 2.34% Pb, 6.91% Zn and 15.36% Fe, was prepared from the Kooshk mine in Yazd province, Iran, and used as the primary ore in flotation and bioflotation experiments. Mineralogical and elemental compositions of the ore were studied using semi-quantitative X-ray diffraction (SQXRD) and atomic absorption spectroscopy (AAS) techniques, respectively. 2.2. Bacterial culture Pure strains of A. ferrooxidans and Acidithiobacillus thiooxidans, isolated from the acidic water drainage of Sarcheshmeh copper mine (Iran), were used in this study. The bacteria were grown in the laboratory using 9 k medium (3 g/l (NH4)2SO4, 0.5 g/l MgSO4 7H2O, 0.5 g/l K2HPO4, 0.1 g/l KCl and pH = 1.85), and were cultured by inoculating 10 ml of pure strain of the bacterial cells into the medium. Potassium nitrate was used to maintain ionic strength. The cultures were incubated at 32 °C in a rotary shaker maintained at 160 rpm. Microorganisms were adapted with 50 g/l original ore. A. ferrooxidans and A. thiooxidans were grown separately on ferrous sulfate and elemental sulfur, respectively. The culture solution was filtered through Whatman filter paper to remove the suspended solids. The cells were separated from the medium using the biological filter paper (0.42 lm) and then suspended in distilled water. During the experiments, the cell number in the solution was estimated by direct counting, using a Thoma chamber of 0.1 mm depth and 0.0025 mm2 area with an optical microscope. 2.3. Flotation and bioflotation experiments Flotation and bioflotation experiments were performed in a 1.5 l Denver cell running at 850 rpm using 300 g ore sample (25% pulp density) with a size of 95 lm (d90). Industrial grade potassium ethyl xanthate (PEX) and potassium amyl xanthate (PAX) were
used as galena and sphalerite collectors, respectively. Copper sulfate and sodium hydroxide, which were used for sphalerite activation and pH adjustment, were of analytical grade. Four control factors and their levels used in the bioflotation experiments are presented in Table 1. Each of these factors was varied in two levels, including collector dosage, activator dosage, NaCN dosage, and volume of bacterial solution (A. ferrooxidans). Fig. 1 describes the baseline of the flotation and bioflotation experiments. In all experiments, organic materials in the ore were pre-floated using 120 g/t methyl isobutyl carbonyl (MIBC) and 250 g/t diesel oil. For this purpose, the pulp was conditioned for 2 min at pH = 7– 7.5 followed by froth collection for 4 min, and the impeller speed was 850 rpm. Pulp pH was adjusted to 9.5 and 11 in the conditioning stage of galena and sphalerite, respectively. In bioflotation experiments, in which bacteria were used as the pyrite depressant, the minerals were conditioned after pre-floating with 300 ml bacteria solution for 20 min, using an impeller speed of 120 rpm. The bacterial population in the solution was counted at about 3 107 cells/ml. In the sphalerite frothing stage, the flotation froth was collected for 7.5 min. According to Table 2, nineteen flotation experiments in a full factorial design (H1–H16) with the center points (H17–19) were carried out. These experiments are as follows: – – – –
Four experiments without pyrite depressant. Four experiments using NaCN as pyrite depressant. Four experiments using A. ferrooxidans as pyrite depressant. Four experiments using both NaCN and A. ferrooxidans as pyrite depressant. – Three experiments using the center points of the factors to estimate operator error in the tests.
2.4. Separation efficiency Separation efficiency was calculated as an indicator of the metallurgical performance of the sphalerite flotation process using the equation:
SE ¼ 100C
mðc f Þ ; f ðm f Þ
ð1Þ
where C is the weight percent of the feed to the concentrate, m is theoretical zinc content of the sphalerite mineral, c is Zn grade of sphalerite concentrate, and f is Zn grade in the original feed ore sample. 2.5. Least significant difference (LSD) LSD is a numerical value that can be used as a benchmark for comparing treatment means. The LSD is calculated as a comparative tool with:
LSD ¼ t s
pffiffiffiffiffiffiffiffiffiffiffiffi ð2=nÞ;
ð2Þ
where t is a factor that depends on the desired confidence and the degrees of freedom for estimation of error, s is standard deviation, and n is the sample size. The t value obtained from the t-distribution table under the desired conditions was 2.92 (Anderson and Table 1 Studied factors and levels in flotation and bioflotation experiments. Level
A: PAX (g/t)
B: CuSO4 (g/t)
C: NaCN (g/t)
D: bacterial solution (ml)
High Low Center
140 280 210
700 1400 1050
0 100 50
0 300 150
12
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
Fig. 1. Baseline for experimental procedure.
Whitcomb, 2000). When the LSD is exceeded, the means are considered to be significantly different. Before analysis of the experimental results, LSD values for sphalerite and pyrite recovery were calculated. The standard devi-
ation in the recoveries due to the errors in sample analysis and flotation experiments are presented in Table 3. For estimation of the errors in sample analysis, four products of an experiment (including the concentrate from the pre-flotation stage, galena concen-
Table 2 Experimental plan for flotation tests and results in sphalerite concentrate. Std
A
B
C
D
Zn (%)
Fe (%)
Sphalerite recovery (%)
Pyrite recovery (%)
Zn SE (%)
H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12 H13 H14 H15 H16 H17 H18 H19
140 280 140 280 140 280 140 280 140 280 140 280 140 280 140 280 210 210 210
700 700 1400 1400 700 700 1400 1400 700 700 1400 1400 700 700 1400 1400 1050 1050 1050
0 0 0 0 100 100 100 100 0 0 0 0 100 100 100 100 50 50 50
0 0 0 0 0 0 0 0 300 300 300 300 300 300 300 300 150 150 150
15.22 11.52 11.76 9.57 18.63 13.12 18.16 18.61 20.84 11.85 15.04 10.63 20.12 18.20 11.67 19.15 15.56 17.42 16.28
14.23 14.44 22.49 22.88 15.58 16.51 11.81 10.98 11.52 15.20 20.14 18.94 10.55 14.33 11.94 12.69 13.51 14.40 15.16
65.91 64.39 73.77 70.69 77.20 72.53 69.22 76.11 74.03 70.91 73.19 72.95 73.11 76.12 71.36 72.57 76.91 79.39 75.10
38.11 52.98 53.02 59.62 25.31 32.88 24.02 24.62 23.52 49.79 41.11 54.22 20.53 22.81 21.07 23.58 29.94 25.59 28.41
37.59 28.40 36.06 28.01 53.12 41.05 47.21 52.29 56.98 32.78 42.72 32.11 53.53 52.06 49.08 51.51 47.83 49.98 49.90
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
13
trate, sphalerite concentrate, and final tailings) were analyzed three times on different days. Afterwards, the recoveries were calculated and the standard deviations were estimated. Three bioflotation experiments including H17, H18 and H19 were carried out on three different days under exactly the same conditions and the products were analyzed to determine the experimental errors. In the 90% confidence interval, the LSD values for sphalerite and pyrite recoveries were calculated as 7.39% and 7.1%, respectively (Table 3). The LSD values for Fe and Zn grades in the sphalerite concentrate were also calculated as 1.81% and 1.03%, respectively. As an example of using the LSD value in analysis of the experimental results, sphalerite recovery was increased from 64% to 74% by changing the collector dosage. Therefore, comparing the amount of increase (10%) with the LSD value of sphalerite recovery (7.39%), it can be postulated that the effect of collector dosage is significant. LSD values were indirectly considered in analysis of all results of flotation and bioflotation experiments.
3. Results and discussion The recovery values for sphalerite and pyrite minerals, together with Zn separation efficiency (SE) and Zn and Fe grade in the sphalerite concentrate, were calculated as the process responses. The experimental conditions and the results of flotation and bioflotation experiments are presented in Table 2.
3.1. Effect of collector and activator 3.1.1. In absence of depressant The effect of the interaction between collector and activator dosages in the absence of any pyrite depressant on the pyrite recovery is shown in Fig. 2a. From this figure, it is apparent that pyrite recovery was significantly increased (according to the LSD bar), at a low level of activator dosage (700 g/t), by increasing the collector dosage. At a high level of activator concentration (1400 g/t), the increase in pyrite recovery with variation of the collector dosage was not significant, because the LSD bars cover each other. At a low level of activator and collector dosage, only 38% pyrite was floated, while doubling of the concentrations of reagents caused about 52–59% of pyrite to be floated. Therefore, using a low level of the collector at pH 11, pyrite recovery was significantly enhanced with an increase in CuSO4 dosage. However, batch flotation tests on Red Dog ore as reported by (Dichman and Finch, 2001) showed that upon addition of copper, pyrite recovery was decreased significantly. It can be understood from Fig. 2b that changing the collector dosage at the designed point did not improve sphalerite recovery, but with a doubling of CuSO4 at a low level of collector, sphalerite recovery was enhanced significantly from 65.93% to 73.77%. Comparing the results of experiments H1–H4 in Table 2, which were carried out in the absence of depressant, it can be concluded that the highest selective separation between sphalerite and pyrite was obtained using 140 g/t collector and 700 g/t CuSO4 in experiment H1.
Table 3 Errors in sphalerite and pyrite recoveries. Source
Sample analysis Flotation and bioflotation Total
Standard deviations in recovery (%) Sphalerite
Pyrite
0.92 2.18 3.10
0.78 2.20 2.98
Fig. 2. Interaction effect of collector and activator dosage on (a) pyrite recovery and (b) sphalerite recovery.
3.1.2. In the presence of NaCN It is fully clear from Table 2 (H5–H9) that at a low level of activator dosage (700 g/t), pyrite recovery was significantly increased by increasing the collector dosage, while at a high level of activator concentration (1400 g/t), the effect of the collector was negligible. However, pyrite recovery became relatively more stable with changes in the collector and activator dosages in the presence of NaCN, as compared with experiments H1–H4. At a low level of CuSO4, the Zn grade decreased from 18.63% to 13.61% with changes in the collector dosage (from 140 to 280). The Fe grade in the concentrate was significantly decreased when the CuSO4 concentration was doubled.
3.1.3. In the presence of bacteria Fig. 3a shows that sphalerite recovery remained relatively stable with changes in collector and activator dosages in the presence of bacteria. However, Fig. 3a illustrates that at the higher concentrations of the activator and the collector (experiments H10– H12), pyrite was floated at a factor of two times greater than the results obtained in test H9. Because of the negative effect of collector and activator dosage increases, an decrease in Zn grade and increase in Fe grade can be observed from Fig. 3b. Comparing the results of pyrite recovery in experiments H9– H12 with the results of experiments H5–H8, it is obvious that pyr-
14
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
3.2. Comparison between depressant efficiency of bacteria and NaCN In previous discussions, the optimum process conditions used to minimize pyrite flotation were determined in the presence of each depressant and also without any pyrite depressant. These conditions were achieved in experiments H1, H5, H9, and H13. All of these experiments were carried out using low levels of flotation reagents (140 g/t PAX and 700 g/t CuSO4). The results of these four experiments are compared in Fig. 4. The efficiency of NaCN on pyrite depression can be evaluated by comparing the results of experiments H1 and H5. Sphalerite recovery was increased from 65.91% to 77.21% in the presence of sodium cyanide, while pyrite recovery was decreased from 38.1% to 25.3%. Comparing the results of experiments without depressant and with bacteria, it is clear that pyrite was significantly depressed in the presence of A. ferrooxidans, and recovery decreased from 38% to 23%. However, sphalerite recovery was improved from 64.91% to 74.03%. In contrast, the Zn grade in the sphalerite concentrate increased from about 15.22% to 20.84%, and the Fe grade was decreased to 11.52 when using bacteria. It is apparent from Fig. 4 that
Fig. 3. Effect of A. ferrooxidans on (a) sphalerite and pyrite recoveries and (b) Zn and Fe grades in the sphalerite concentrate.
ite recovery was influenced more considerably by changes in flotation reagents when bacteria were used as the depressant. In this category of experiments, the highest sphalerite recovery was obtained with the least pyrite recovery in the H9 test. In this condition, sphalerite and pyrite recoveries were reached to 74.03% and 23.52%, respectively, and Zn and Fe grade were reached to 20.74% and 11.52%, respectively. 3.1.4. In the presence of both NaCN and bacteria In experiments H13–H16, both bacteria and NaCN were used as pyrite depressants. It is apparent from the results in Table 2 that, in this case, changing the collector or activator concentration does not affect the sphalerite and pyrite recoveries at the significance level under study. Comparing the results of experiments H13–H16 with the experiments in which only bacteria or NaCN were used as the depressant, the synergistic effects of two depressants on both pyrite recoveries and Fe grades can be observed. The smallest amounts of pyrite recovery and lowest Fe grade were also achieved in experiment H13, in which both bacteria and NaCN were used together with low levels of other flotation reagents.
Fig. 4. Effect of different depressants on (a) sphalerite and pyrite recoveries and Zn separation efficiency (SE), and (b) Zn and Fe grades.
15
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
Zn separation efficiency was improved when either bacteria or NaCN were used as the pyrite depressant. Meanwhile, the highest separation efficiency was achieved in the presence of bacteria. Since A. ferrooxidans obtain their energy by oxidizing iron ions and elemental sulfur, the cells might have strongly adsorbed on the pyrite surface. A. ferrooxidans can affect mineral surfaces by a direct mechanism; the bacteria eliminate the occurrence of oxidized sulfur which has hydrophobic properties and induces a higher floatability for minerals, so that the hydrophobicity of pyrite is decreased. In other words, the bacteria can suppress floatability, and the postulated mechanism of this suppression is an increase in surface hydrophilicity due to adhesion of bacterial cells (Ohmura et al., 1993). The efficiencies of bacteria and NaCN in pyrite depression can be compared using the results of experiments H5 and H9 (Fig. 4). For pyrite depression, it is clear that A. ferrooxidans was as effective as sodium cyanide, and pyrite recoveries were 25.3% and 23.5% when NaCN and bacteria were used as the depressants, respectively. Moreover, the Zn grade in the sphalerite concentrate was increased from 18.63% to 20.84% and the Fe grade decreased from 15.58% to 11.52%. These changes are both significant, referring to the LSD values of grades. Under these conditions, both sphalerite and pyrite recoveries were approximately the same in the presence of NaCN and A. ferrooxidans. From observation of the same recoveries and different grades, it can be understood that less gangue minerals appeared in the sphalerite concentrate in the presence of A. ferrooxidans. These results illustrate the capability of A. ferrooxidans for depression of pyrite and other gangue minerals in comparison
to NaCN. The results of experiment H13, in which both bacteria and NaCN were used, are similar to the results of H9. 3.3. Flotation kinetics It is widely accepted that most flotation systems can be described by a first-order reaction rate equation (Eq. (3)).
R ¼ 1 ekt ;
ð3Þ
where R is the cumulative recovery, t and k are the time and kinetic constant of flotation. The constant k depends on a large number of variables, which include factors related to ore characteristics like mineralogy and factors defined through the flotation medium such as the type and quantity of reagents, among others (Kelebek and Nanthakumar, 2007). Fig. 5 shows that the flotation of sphalerite and pyrite follows first-order kinetics in the presence and absence of A. ferrooxidans with 140 g/t PAX and 700 g/t CuSO4. The results also show that the pyrite flotation rate was cut to half with the addition of bacteria. In contrast, the sphalerite kinetic constant was increased. 3.4. Validation experiments For confirmation of the bacteria’s effect, two experiments were repeated at the optimum conditions of reagent with bacteria. The procedure in these experiments was exactly the same as in previous tests; however, in the new tests, the frothing time was decreased from 7.5 to 6.5 min One test was carried out with A. ferrooxidans and the next with A. thiooxidans. Results are shown in Table 4. In the validation experiments, in addition to confirmation of the pyrite depression capability of bacteria in the sphalerite concentrate, the Zn grade surged from 20.84% to 26.21% and pyrite recovery declined from 23.56% to 15.825%. 4. Conclusions In the present work, 19 experiments were designed, and flotation and bioflotation tests were carried out in the presence and absence of pyrite depressant. The following major conclusions can be made based on this study:
Fig. 5. Effect of A. ferrooxidans on kinetic constants of sphalerite and pyrite flotation in the sphalerite concentrate at optimum reagent dosage. Kbsp and Kbpy are kinetic constants for sphalerite and pyrite in the presence of bacteria, respectively, while Ksp and Kpy are kinetic constants of sphalerite and pyrite in the absence of bacteria, respectively.
1. Maximum pyrite depression in the absence of any depressant was obtained using 140 g/t collector and 700 g/t CuSO4. 2. Sphalerite and pyrite recoveries were stable with changes in the concentrations of collector and activator when NaCN was used as a pyrite depressant. 3. In the presence of A. ferrooxidans, pyrite recovery was decreased from 38.11% to 23.52%, while sphalerite recovery and Zn grade were enhanced from 65.91% to 74.03% and from 15.22% to 20.84%, respectively. 4. Sphalerite and pyrite recoveries were the same when NaCN and bacterial depressants were compared. However, a 3% increase in Zn grade and a 4% decrease in Fe grade were observed when the bacteria were added as the pyrite depressant at optimum conditions.
Table 4 Validation experiments at optimum conditions. PAX (g/t)
140 140 140
CuSO4 (g/t)
700 700 700
Bacteria
A. ferrooxidans A. ferrooxidans A. thiooxidans
Frothing time (min)
7.5 6.5 6.5
Grade (%)
Recovery (%)
Zn
Fe
Sphalerite
Pyrite
20.84 26.21 24.32
11.52 11.94 13.57
74.03 74.68 74.29
23.52 15.82 18.73
16
J.V. Mehrabani et al. / Minerals Engineering 23 (2010) 10–16
5. In bioflotation experiments, the process responses were very sensitive to the concentrations of other flotation reagents. Therefore, reagent dosage is a significant factor affecting the bioflotation process, a factor that has been neglected in previously reported bioflotation studies. 6. The pyrite flotation rate constant decreased significantly when A. ferrooxidans interacted with the flotation pulp for 20 min. 7. Validation experiments with different frothing times showed that in the presence of A. ferrooxidans, 74% of sphalerite was floated and the Zn grade surged to 26.21%, while the recovery of pyrite was only 15.82%. Similar results were also obtained using A. thiooxidans.
Acknowledgement Authors would like to thank Professor K.A. Natarajan for his helpful comments on this work. References Anderson, M.J., Whitcomb, P.J., 2000. DOE simplified: practical tools for effective experimentation. Productivity Inc. Carta, M., Ghiani, M., Rossi, G., 1980. Beneficiation of a complex sulphide ores by an integrated process of flotation and bioleaching. In: Complex sulphide ores. Institute of Mining and Metallurgy, London, pp. 178–185. Chandraprabha, M.N., Natarajan, K.A., Modak, J.M., 2004a. Selective separation of pyrite and chalcopyrite by biomodulation. Colloids and Surfaces B: Biointerfaces 37, 93–100. Chandraprabha, M.N., Natarajan, K.A., Somasundaran, P., 2004b. Selective separation of arsenopyrite from pyrite by biomodulation in the presence of Acidithiobacillus ferrooxidans. Journal of Colloid and Interface Science 276, 323–332. Deo, N., Natarajan, K.A., 1997. Interaction of Bacillus polymyxa with some oxide minerals with reference to mineral beneficiation and environmental control. Minerals Engineering 10, 1339–1354. Deshpande, R.J., Subramanian, S., Natarajan, K.A., 2004. Surface-chemical studies on pyrite and arsenopyrite using Acidithiobacillus ferrooxidans. In: Rao, G.V., Misra,
V.N. (Eds.), Proceedings of the International Seminar on Mineral Processing Technology. Allied Publishers Limited, New Delhi, pp. 668–675. Deshpande, R.J., Subramanian, S., Natarajan, K.A., 2001. Effect of bacterial interaction with Thiobacillus ferrooxidans on the floatability of sphalerite and pyrite. In: Padmanabhan, N.P.H., Sreenivas, T., Srinivas, K. (Eds.), Proceedings of the International Seminar on Mineral Processing Technology. Allied Publishers Limited, New Delhi, pp. 159–161. Dichman, T.K., Finch, J.A., 2001. The role of copper ions in sphalerite–pyrite flotation selectivity. Minerals Engineering 14 (2), 217–225. Donati, E.R., Sand, W., 2007. Microbial Processing of Metal Sulfides. Springer. pp. 273-275. Fuerstenau, M.C., Miller, J.D., Kuhn, M.C., 1985. Chemistry of Flotation. AIME, New York. Gaudin, A.M., 1957. Flotation. Mapel Press Company, New York. pp. 436–453. Hosseini, T.R., Kolahdoozan, M., Tabatabaei, Y.S.M., Oliazadeh, M., Noaparast, M., Eslami, A., Manafi, Z., Alfantazi, A., 2005. Bioflotation of Sarcheshmeh copper ore using Thiobacillus ferrooxidans bacteria. Minerals Engineering 18, 371–374. Kelebek, S., Nanthakumar, B., 2007. Characterization of stockpile oxidation of penthlandite and pyrrhotite through kinetic analysis of their flotation. International Journal of Mineral Processing 84, 69–80. Kolahdoozan, M., Tabatabaei Yazdi, S.M., Yen, W.T., Hosseini Tabatabaei, R., Shahverdi, A.R., Oliazadeh, M., Noaparast, M., Eslami, A., Manafi, Z., 2004. Bioflotation of the low grade Sarcheshmeh copper sulfide. Transactions of the Indian Institute of Metals 57 (5), 485–490. Nagaoka, T., Ohmura, N., Saiki, H., 1999. A novel mineral flotation process using Thiobacillus ferrooxidans. Applied Environmental Microbiology 65, 3588–3593. Ohmura, N., Kitamura, K., Saiki, H., 1993. Mechanism of microbial flotation using Thiobacillus ferrooxidans for pyrite suppression. Biotechnology and Bioengineering 41, 671–676. Pecina, E.T., Rodriguez, M., Castillo, P., Diaz, V., Orrantia, E., 2009. Effect of Leptospirillum ferrooxidans on the flotation kinetics of sulphide ores. Minerals Engineering 22, 462–468. Sharma, P.K., Hanumantha, K.R., 2001. Role of heterotrophic bacteria in the bioflotation of some sulfide minerals. Mineral Biotechnol, 67–81. Shen, W.Z., Fornasiero, D., Ralston, J., 1998. Effect of collectors conditioning pH and gases in the separation of sphalerite from pyrite. Minerals Engineering 11, 145– 158. Wills, B.A., Napier-Munn, T.J., 2006. Mineral Processing Technology: An Introduction to the Practical Aspects of ore Treatment and Mineral Recovery, 7th ed. Butterworth, Heinemann. Yüce, A.E., Mustafa Tarkan, H., Zeki Do, M., 2006. Effect of bacterial conditioning and the flotation of copper ore and concentrate. African Journal of Biotechnology 5, 448–452.