Appl Microbiol Biotechnol (2013) 97:2735–2742 DOI 10.1007/s00253-012-4116-y
ENVIRONMENTAL BIOTECHNOLOGY
Reactive oxygen species generated in the presence of fine pyrite particles and its implication in thermophilic mineral bioleaching G. C. Jones & R. P. van Hille & S. T. L. Harrison
Received: 13 February 2012 / Revised: 16 April 2012 / Accepted: 16 April 2012 / Published online: 16 May 2012 # Springer-Verlag 2012
Abstract In the tank bioleaching process, maximising solid loading and mineral availability, the latter through decreasing particle size, are key to maximising metal extraction. In this study, the effect of particle size distribution on bioleaching performance and microbial growth was studied through applying knowledge based on medical geology research to understand the adverse effects of suspended fine pyrite particles. Small-scale leaching studies, using pyrite concentrate fractions (106–75, 75–25, −25 μm fines), were used to confirm decreasing performance with decreasing particle size (D50 <40 μm). Under equivalent experimental conditions, the generation of the reactive oxygen species (ROS), hydrogen peroxide and hydroxyl radicals from pyrite was illustrated. ROS generation measured from the different pyrite fractions was found to increase with increasing pyrite surface area loading (1.79–74.01 m2 L−1) and Fe2+ concentration (0.1–2.8 gL−1) in solution. The highest concentration of ROS was measured from the finest fraction of pyrite (0.85 mM) and from the largest concentration of Fe2+ (0.78 mM). No ROS was detected from solutions containing only Fe3+ under the same conditions tested. The potential of ROS to inhibit microbial performance under bioleaching conditions was demonstrated. Pyrite-free Sulfolobus metallicus cultures challenged with hydrogen peroxide (0.5– 2.5 mM) showed significant decrease in both cell growth and Fe2+ oxidation rates within the concentration range 1.5– 2.5 mM. In combination, the results from this study suggest that conditions of large pyrite surface area loading, coupled with high concentrations of dissolved Fe2+, can lead to the G. C. Jones (*) : R. P. van Hille : S. T. L. Harrison Centre for Bioprocess Engineering Research, Department of Chemical Engineering, University of Cape Town, Private Bag, Rondebosch 7701, Cape Town, South Africa e-mail: Gavin.Jones@uct.ac.za
generation of ROS, resulting in oxidative stress of the microorganisms. Keywords Mineral bioleaching . Reactive oxygen species . Particle size . Mineral sulphide . Pyrite
Introduction Bioleaching has become an increasingly popular process option for metal extraction from refractory mineral sulphidecontaining ores and concentrates. Thermophilic (65–80 °C) bioleaching technologies, using iron- and sulphur-oxidising Archaea, offer improved rate and extent of leaching as a competitive advantage over moderate thermophilic and mesophilic systems for refractory sulphide minerals such as chalcopyrite. BioCOPTM, commissioned in 2003 at Chuquicamata, Chile, is an example of a demonstration-scale high temperature, stirred tank bioleaching process (Batty and Rorke 2005). These thermophilic systems are recognised as being more susceptible to the microbial stress that mineral solids can induce, constraining bioleaching performance. Sulfolobus metallicus has been identified as an important thermophilic species in industrial and laboratory hightemperature mineral bioleaching systems (Plumb et al. 2002; Hallberg and Johnson 2001). Pyrite (FeS2) is often associated with base metal-containing concentrates (Jones et al. 2011a) and may be a host mineral bearing gold in refractory gold concentrates. S. metallicus and pyrite have been used as a simplified model system with which to investigate thermophilic microbial stress response to differing mineral feed pulp densities and particle size distributions. The effect of varying the initial pyrite particle size, measured as a median particle size (D50), from 202.1 to 6.4 μm, at a constant pulp density of 3 % (w/v), on planktonic cell growth of S.
2736
Appl Microbiol Biotechnol (2013) 97:2735–2742
metallicus and the associated pyrite oxidation rates reported by Harrison’s laboratory is summarised in Fig. 1. Maxima in cell growth and leaching performance were obtained at a particle size of approximately 40 μm, in good agreement with Acevedo et al. (2004). The decrease in performance with finer particle sizes (<40 μm) is more marked compared with the more gradual decrease in performance with coarser particle sizes (>110 μm). The gradual decrease in performance observed at coarser particles size distributions may be the effect of reduced surface area and associated mineral availability or the increased impact of particle–cell–particle collisions causing increased physical stress. Susceptibility to the latter is exacerbated with Archaea which lack the structural peptidoglycan component of the cell walls found in bacteria (Woese et al. 1978; Harrison et al. 2003b; Clark and Norris 1996). Decreasing the initial particle size distribution below a D50 of approximately 40 μm has been reported to have an adverse effect on S. metallicus, resulting in marked decrease in pyrite oxidation and planktonic cell growth rates (Nemati et al. 2000; Acevedo et al. 2004; Gericke and Pinches 1999). At a nominal particle size of 6.4 μm, no pyrite biooxidation was observed, and microbial cells were undetectable 24 h after inoculation, suggesting cell lysis. While the improvement in rate of mineral leaching associated with decreasing particle size in the range above a D50 of 40 μm can be attributed to increasing surface area and associated mineral availability (Nemati et al. 2000), the factors controlling the decrease in leaching rate on decrease of the particle size below this critical value and on increasing solids loading are not clear. Factors investigated to influence leaching performance owing to their dependence on solid loading and particle size included suspension viscosity, available surface area, frequency and momentum of particle collisions, energy dissipation, bubble–particle interaction and gas–liquid mass transfer; however, definitive correlation was not found (Harrison et al. 2001; Sissing and Harrison 2003). Most of the S. metallicus–pyrite model experiments conducted have focused on analysing bioleaching performance data from 0.05
0.10
0.04
0.08 0.03 0.06 0.02 0.04
Growth rate (h-1)
Pyrite oxidation rate (g L-1h-1)
0.12
0.01
0.02 0.00
0.00 0
50
100
150
200
D50 particle size (µm)
Fig. 1 Maximum pyrite oxidation and planktonic cell growth rates of fractionated pyrite using S. metallicus in stirred tank reactors 3 % (w/v) (Nemati et al. 2000; Harrison et al. 2001)
experiments in which either pulp density or particle size are varied individually. Both these variables affect the total pyrite surface area loading (SAL) in suspension. Valencia and Acevedo (2009) studied SAL as possibly being the overall rate controlling factor. The results for iron and sulphate productivities showed that the bioleaching of pyrite by S. metallicus depended not only on the surface area but also on solid concentration and particle size, and no increase in rates were measured for finer pyrite fractions (75–38 and −38 μm). The mechanism of this effect was speculated to be due to difficulties in cell attachment to fine particles and metabolic stress. The field of medical geology has identified and implicated the generation of toxic reactive oxygen species (ROS) such as hydrogen peroxide (H2O2), superoxide anion radicals (O2•−) and hydroxyl radicals (•OH) from pyrite particles as a causative agent in coal-induced pulmonary disease common in coal miners (Cohn et al. 2006; Schoonen et al. 2006). The deleterious effect of ROS, as powerful oxidising agents, on cells has been rigorously studied with respect to oxidative stress (Halliwell 2007; Imlay 2003). Pyrite in solution forms ROS (Schoonen et al. 2006) by the ironcatalysed Haber–Weiss reactions (1 and 2). The H2O2 formed can react with Fe2+ in a solution to form ·OH via the Fenton reaction (3): py Fe2þ þ O2 ! Fe3þ þ ðO2 Þ
ð1Þ
py Fe2þ þ ðO2 Þ þ 2Hþ ! Fe3þ þ H2 O2
ð2Þ
Fe2þ þ H2 O2 ! OH þ OH þ Fe3þ
ð3Þ
Observations have indicated that the formation of ROS in solution is directly proportional to the SAL of pyrite suspended in solution (Cohn et al. 2005). The upregulation of the putative antioxidant protein alkyl hydroperoxide reductase (Burton et al. 1995; Bathe and Norris 2007) has been observed in S. metallicus cells after switching from elemental and reduced sulphur to ferrous iron and pyrite substrates. Additionally, the inorganic polyphosphate-based copper resistance mechanism (Remonsellez et al. 2006; Orell et al. 2010) has also been identified as being important in preventing oxidative stress induced indirectly in S. metallicus cells by challenging with high concentrations of dissolved copper. The formation of ROS from pyrite in bioleaching systems may have a direct oxidative stress effect on microorganisms. This proposed oxidative stress mechanism would be distinctly different from that of ROS produced endogenously through cellular respiration. This effect can result in oxidative stress conditions caused by external stress factors, such as high copper concentrations, leading to the
Appl Microbiol Biotechnol (2013) 97:2735–2742
2737
disruption of cellular respiration pathways or the diminished functioning of antioxidant systems. In this study, the underlying cause of reduced thermophilic bioleaching performance, under non-optimal mineral solids loading conditions, is sought through further fundamental understanding of the physicochemical factors affecting S. metallicus. The objective of this study was to explore the interrelationship between the performance of pyrite leaching in the presence of fine pyrite particles, ROS generation in the presence of pyrite particulates and the impact of ROS on microbial performance. The impact of particle size distribution of bioleaching performance was demonstrated in shake flasks. ROS generation from these pyrite fractions was quantified as a function of SAL. Thirdly, bioleaching cultures were challenged with ROS, as predominately hydrogen peroxide, in the absence of pyrite solids to establish a correlation to the concentration of ROS required for inhibited cell growth and ferrous iron oxidation to occur.
Material and methods Microorganism S. metallicus (DSM 6482; Nemati et al. 2000) was used for all shake flask experiments. The following basal salt medium was used: (NH4)2SO4 (1.30 gL−1), KH2PO4 (0.28 g L−1), MgSO4·7H2O (0.25 gL−1) and CaCl2·2H2O (0.07 g L−1), supplemented with yeast extract (0.20 g L−1) and K2S4O6 (0.15 gL−1). The pH was adjusted to 1.6 with H2SO4. Two stock cultures were adapted and maintained by repeated sub-culturing in shake flasks on 1 % (w/v) pyrite concentrate or 2.8 gL−1 Fe2+ (pyrite free) in basal salt medium, respectively. These were used as inocula for shake flask experiments. Mineral concentrate A non-gold-containing pyrite concentrate, produced for research purposes, was used in this study. Three fractions were prepared by wet screening (106–75, 75–25 and −25 μm), and a fines fraction was prepared by gravity separation of the −25μm fraction. Fractions were washed with ethanol and air dried Table 1 Fractionated pyrite sample grade, surface area and particle size distribution
BET Brunauer, Emmett and Teller surface area a
Subfraction prepared from −25 μm fraction via gravity sedimentation
Fraction mesh size (μm)
106–75 75–25 −25 −25 finesa
Pyrite (w/w %)
94.2 97.9 93.6 94.7
before being stored in a vacuum dessicator to prevent surface oxidation. The four particle size fractions were analysed in terms of pyrite grade, particle size and particle surface area (Table 1). Mineralogical composition of the fractions was quantified by XRD analysis using a PANalyticalX’Pert Pro powder diffractometer with X’Celerator and the relative phases estimated using the Rietveld method (Autoquan programme). The pyrite grade differed by less than 5 % (w/w) between fractions and the iron content varied from 42.4 to 44.4 % (w/w). Minor minerals associated with the pyrite concentrate were chalcopyrite (0.1–0.7 % w/w) and quartz (0.5–3.9 % w/w). Volumetric particle size distributions were determined using laser diffraction (Malvern Mastersizer 2000). Surface area was measured using the five-point N2 adsorption Brunauer, Emmett and Teller method (Micromeritics TriStar). Detection of reactive oxygen species A modified method using a fluorescent probe (3′-(p-aminophenyl) fluorescein, (APF) Invitrogen) was used to measure ROS, according to Jones et al. (2011b). ROS was measured from Fe2+-containing solutions (0.06–2.79 gL−1) and pyrite suspensions incubated in 50 mL Erlenmeyer flasks for 24 h in an orbital shaker incubator (180 rpm). ROS quantification experiments were performed at 65 °C in basal salt medium (pH 1.6). ROS data are presented as hydrogen peroxide reaction equivalents. Standard curve was generated by reacting known concentrations of H2O2, peroxidase enzyme and APF probe. Shake flask experiments Pyrite samples were bioleached in duplicate for up to 35 days using an initial 3 % (w/v) solids loading in a 300mL culture volume in 500 mL Erlenmeyer flasks. Flasks were inoculated with 1×106 cells/mL using the pyrite adapted stock culture and incubated at 65 °C on a rotary shaking incubator (180 rpm). On sampling, the following were measured: redox potential (Crison ELP 21), pH (Metrohm 704), total soluble iron and Fe2+ measured colorimetrically using the 1,10-phenanthroline method (APHA 1998). The effect of hydrogen peroxide stress was studied in the absence of pyrite with a starting cell concentration of 1×106 cells/mL. The H2O2 was applied at the beginning of the
BET (m2 g−1)
0.06 0.12 1.12 2.47
Particle size distribution (volume % below diameter μm) D10
D20
D50
D80
D90
64.2 41.3 6.7 2.9
78.5 51.6 12.3 6.1
109.0 76.1 27.9 15.9
148.0 109.4 75.7 29.4
171.5 130.1 115.5 38.6
2738
Appl Microbiol Biotechnol (2013) 97:2735–2742
experiment (0.5–2.5 mM final) in the presence of 1.68 gL−1 Fe3+ for 1 h. Thereafter, 1.68 gL−1 Fe2+ was added to the flasks. Samples were taken at a given time interval to monitor ferrous iron oxidation and cell growth (Thoma counting chamber, Olympus BX-40 microscope). The loss of water due to evaporation was compensated by adding sterilised distilled water gravimetrically.
Results Bioleaching of pyrite as a function of particle size Thermophilic leaching performance in shake flask culture is presented as a function of the particle size of the pyrite concentrate in Fig. 2. For the two coarser fractions (106–75 and 75–25 μm), the rate and extent of bioleaching were similar. After 30 days, the maximum total soluble iron concentrations measured corresponded to a leaching extent of 93 and 98 % w/ w for 106–75 and 75–25 μm fractions, respectively. A decrease in pH was measured to a final value of 0.9. Redox potential remained above 500 mV throughout the experiment and the maximum Fe2+ concentrations measured were 0.58 and 0.65 gL−1 for 106–75 and 75–25 μm fractions, respectively. On leaching the −25-μm fraction (D50, 28 μm), a delay in the onset of iron solubilisation and decrease in pH were observed. The total soluble iron measured was 0.46 gL−1 after 6 days, compared to approximately 2.45 gL−1 for both coarser fractions. The redox potential decreased to 392 mV and the pH remained constant at 1.6 at the end of the lag period. After the initial lag period, there was a rapid increase in bioleaching, evident as an increase in iron solubilisation
Quantification of ROS generation from suspended pyrite fractions The presence of microbial cells was found to interfere with the measurement of ROS directly from bioleaching solutions. This was postulated to result from hydroxyl radicals reacting with cell biomass and being transformed into other free radical species or due to the cells inherent ability to generate and scavenge ROS directly. The concentration of ROS generated from contacting pyrite fractions with acidic solution for 24 h, in the absence of cells, is presented as a function of SAL in Fig. 3. The pyrite SAL ranged from 1.8 to 74.0 m2 L−1 across the size fractions used. The ROS measured increased proportionally from 0.09 to 0.85 mM
b 14
2.5
Fraction (µm)
12
106-75
10
75-25
8
-25
6
-25 f ines
Fe2+(g L-1)
Total soluble Fe (g L-1)
a
4
1.5 1.0 0.5
2 0
2.0
0
10
20
30
0.0
40
0
10
Time (d)
20
30
40
30
40
Time (d)
d 700 650 600 550 500 450 400 350 300
1.8 1.6 1.4
pH
c REDOX (mV)
Fig. 2 Thermophilic shake flask bioleaching of fractionated pyrite 3 % (w/v). a Total soluble iron concentration; b ferrous iron concentration; c redox potential and d pH
to approximately 13 gL−1 after 26 days, corresponding to 94.1 % iron leached and an increase in redox to 609 mV. A decrease in soluble iron observed after 26 days is attributed to precipitated ferric iron. A maximum Fe2+ concentration of 1.03 gL−1 was measured at 18 days for the −25-μm pyrite fraction, indicating that the mineral leaching rate exceeded the ferrous iron oxidation rate over the period 0–18 days. For the −25-μm fines fraction (D50, 16 μm), insignificant bioleaching activity was observed. Only 2.5 gL−1 total soluble iron was measured after 34 days, corresponding to 19.0 % w/w of the total iron leached, and the final pH of 1.45 showed reduced acid formation. For the duration of the experiment, the soluble iron present was predominantly as Fe2+, corresponding to a redox potential of approximately 375 mV. This confirmed that any biological oxidation activity occurring was negligible, consistent with the low level of mineral leaching achieved.
1.2 1.0
0
10
20
Time (d)
30
40
0.8
0
10
20
Time (d)
Appl Microbiol Biotechnol (2013) 97:2735–2742
2739
1.0 Fraction (µm)
0.9
106-75
ROS (mM)
0.8
75-25
0.7
-25
0.6
-25 fines
0.5 0.4 0.3 0.2 0.1 0.0 40
20
0
60
80
SAL (m2L-1)
Fig. 3 ROS generation measured after 24 h as a function of surface area loading (SAL) from 3 % (w/v) pyrite fractions. Triplicate experiments, with standard error bars presented
with increasing SAL of the pyrite. The offset of the fitted trend line from the origin indicated an additional factor influencing ROS generation. The ferrous iron concentration in the pyrite slurries at 24 h was found to increase in solution from 0.1 to 0.7 g L −1 with increasing pyrite surface area loading. Therefore, the effect of Fe2+ concentration on ROS generation was studied under the same experimental conditions in the absence of pyrite particles and is presented in Fig. 4. Upon increasing Fe2+ concentration across the range 0.0–2.8 gL−1, the ROS measured increased to 0.78 mM for 2.8 gL−1 Fe2+. On repeating the experiment under the same conditions with the exception of Fe2+ being replaced by Fe3+, no ROS was detected at any Fe3+ concentration used for the 24-h incubation period. Ferric iron interference of the fluorescence assay was excluded by the positive control (H2O2 + Fe3+). Therefore, the dissolved ferrous iron concentration was found to be an important secondary factor contributing towards ROS generation, independent of the presence of pyrite particles. Effect of hydrogen peroxide on S. metallicus The S. metallicus stock culture adapted and maintained on 2.8 gL−1 Fe2+ (pyrite free) was used to investigate the response
ROS (mM)
1.0 0.9
Fe2+
0.8
Fe 3+
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
Concentration (g L-1)
Fig. 4 ROS generation measured after 24 h from initial dissolved Fe2+ and Fe3+ concentrations (pyrite free), performed at pH 1.6 and at 65 °C
to oxidative stress induced by the addition of hydrogen peroxide (0.5–2.5 mM). To ensure that cultures were challenged with the full dose of hydrogen peroxide, the cultures were stressed in the presence of Fe3+ (1.68 gL−1) for 1 h. Fe3+ iron was selected to avoid oxidising Fe2+ via the Fenton reaction (reaction 3). Following this, an equal concentration of Fe2+ was added, and the rate of Fe2+ oxidisation by cells was used to provide a measure of the response to the induced oxidative stress. The Fe2+ oxidation and cell growth are shown as a function of time in Fig. 5. The maximum Fe2+ oxidation and specific growth rates are presented as a function of oxidative stress in terms of H2O2 concentration in Fig. 6. For cultures exposed to 0.5 and 1.0 mM H2O2, the maximum oxidation rates were 0.095 and 0.085 gL−1 h−1, respectively, compared to 0.104 gL−1 h−1 for the control experiment. Maximum growth rates for the control, 0.5 and 1.0 mM H2O2 flasks were 0.182, 0.127 and 0.127 h−1, respectively. A significant lag period of 44 h was observed for the 1.5-mM H2O2 flask; after which the majority (±80 %) of the Fe2+ was oxidised at a maximum rate of 0.053 gL−1 h−1 with an associated maximum specific growth rate of 0.111 h−1. Increasing the H2O2 concentration to 2.0 and 2.5 mM resulted in a further decrease in iron oxidation rates to 0.049 and 0.010 gL−1 h−1, respectively. The net specific growth rate in the 2.0-mM flask was 0.022 h−1. No cells were detected in the presence of 2.5 mM H2O2, below detection limit of 3× 105 cells mL−1, after 44 h, indicating cell lysis had occurred.
Discussion Analysis of the pyrite fractions prepared showed that the two finest fractions gave D50 values below 40 μm. The surface area of the four size distributions prepared ranged from 0.06 to 2.47 m2 g−1 with the largest surface area being measured for the finest pyrite fraction. Previously, it has been shown that the highest Fe2+ oxidation rate and specific growth rates in the thermophilic bioleaching system were observed around a D50 of 40 μm (Nemati et al. 2000; Harrison et al. 2003a; Acevedo et al. 2004). A series of shake flask bioleaching experiments were performed to determine whether a similar decrease in leaching performance was observed in the presence of the two finest pyrite fractions. Poor bioleaching performance was observed with the −25 μm (D50, 27.9 μm) and pyrite fines (D50, 15.9 μm) fractions compared with the two coarser fractions (106– 75 μm, D50 of 109 μm and 75–25 μm, D50 of 76 μm), in agreement with the results reported above. Bioleaching performance was measured as a function of the rate of increase in soluble iron concentration. However, it should be noted that ferric iron precipitation, confirmed by visual observation, occurred mostly towards the end of the experiments for 106–75, 75–25 and −25 μm fractions. With the exception of the pyrite fines, the extent of iron leached from the pyrite
2740
Appl Microbiol Biotechnol (2013) 97:2735–2742
a
b
Fe2+(g L-1)
2.1 1.8 1.5 1.2 H 2O2 (mM) 0.0 0.5 1.0 1.5 2.0 2.5
0.9 0.6 0.3 0.0
0
10
20
30
40
Time (h)
fractions was in the range of 93–98 %. The error in the soluble iron concentrations, attributed to iron precipitation, is therefore likely to be less than 10 %. No visible iron precipitation was observed for the pyrite fines experiment. A significant lag period of 6 days was observed for the −25 μm fraction, indicative of culture adaptation. After this lag period, a rapid increase in the rate of iron dissolution and decrease in pH were observed. No bioleaching was observed for the fines fraction. The microbial oxidation of ferrous iron to the ferric iron oxidant is a critical reaction in the bioleaching of metal sulphide minerals. The indirect microbial oxidation of pyrite has been described by the following reactions (Sand et al. 2001): FeS2 þ 14Fe3þ þ 8H2 O ! 15Fe2þ þ 2SO4 2 þ 16Hþ ð5Þ 4Fe2þ þ O2 þ 4Hþ ! 4Fe3þ þ 2H2 O
ð6Þ
In the absence of microbial ferrous iron oxidation (i.e. inhibited microbial oxidation), ferrous iron accumulated in the solution. This was most noticeable for the pyrite fines fraction, displaying the highest SAL and ferrous iron concentration. Under the same experimental conditions used, the concentration of ROS formed in pyrite suspensions was
Max. Fe2+ox. rate (g L-1h-1)
0.10
0.16
0.08 0.12 0.06 0.08 0.04 0.04
0.02 0.00 0.0
0.5
1.0
1.5
2.0
2.5
Max. growth rate (h-1)
0.20
0.12
0.00 3.0
H2O2(mM)
Fig. 6 Effect of hydrogen peroxide concentration on maximum Fe2+ oxidation and growth rates
50
60
70
Cell counts (x107cell mL-1)
Fig. 5 Fe2+ oxidation (a) and cell count (b) data of cultures stressed with 0.0–2.5 mM H2O2. Starting cultures of approximately 1×106 cells mL−1 challenged with H2O2 in 1.68 gL−1 Fe3+ for 1 h prior to the addition of 1.68 gL−1 Fe2+
10.0
1.0
0.1
0.0 0
10
20
30
40
50
60
70
Time (h)
observed to be a function of the SAL. The generation of hydrogen peroxide in pyrite–water slurries under ambient conditions was also found to be surface area dependent using leuco crystal violet as a probe in medical geology applications (Cohn et al. 2005). The fluorescent method used in this study to measure ROS, using 3'-(p-aminophenyl) fluorescein, has been reviewed and found to be the most sensitive and accurate probe for use in these type of studies (Cohn et al. 2008). The accumulation of ROS, under the experimental conditions used, indicated that ROS forming reactions at the pyrite surface occurred at a much faster rate compared to ROS consuming reactions over the 24-h period. Ferrous iron in the presence of dissolved oxygen has also been shown to generate hydroxyl radicals (Schoonen et al. 2006; Cohn et al. 2009) proportionally with Fe2+ concentration at neutral pH and ambient temperatures. Dissolved ferrous iron was measured after suspension of pyrite fractions in solution. The dissolution of iron oxides present on the surface of the pyrite particles provided the source of this dissolved iron. This occurs with exposure to air during sample preparation (Murphy and Strongin 2009) and possibly by the oxidation of pyrite by hydrogen peroxide formed in solution (Schoonen et al. 2010). A second ROS-generating reaction mechanism involved Fe2+ reaction with dissolved oxygen in the absence of pyrite surfaces. The generation of ROS from Fe2+ was concentration dependent, consistent with that found previously under neutral conditions (Schoonen et al. 2006). The direct measurement of ROS concentration in the presence of microorganisms implicated in bioleaching was not consistent, owing to interference due to microbes. The estimated range for the two finest pyrite fractions was 1.0–2.0 mM, based on the sum of expected ROS generation values from both the initial pyrite surface area loading and the maximum ferrous iron concentrations measured during the bioleaching experiments. The generation of ROS, from both proposed mechanisms, is potentially compounded in industrial tank bioleaching systems, where microorganisms are subjected to very high pyrite surface area loadings and high dissolved iron concentrations in the presence of dissolved oxygen. In the absence of pyrite, S. metallicus was stressed with hydrogen peroxide in the presence of ferric iron for 1 h. Both cell growth and ferrous iron oxidation rates decreased
Appl Microbiol Biotechnol (2013) 97:2735–2742
with increasing peroxide concentration, indicating that hydrogen peroxide had a dose-dependent inhibitory effect on S. metallicus. Significant inhibition was observed at 1.5– 2.5 mM hydrogen peroxide. This concentration range corresponded to the estimated range for the two finest pyrite fractions (1.0–2.0 mM) used in the bioleaching experiments. This suggests that the generation and accumulation of ROS from pyrite fines and ferrous iron may be responsible for an oxidative stress effect under conditions of high solid loading, fine particle size or both. The findings of this study are pertinent when considering chalcopyrite thermophilic bioleaching technologies. Chalcopyrite has also been shown to generate ROS (Jones et al. 2011b). Recent developments into chalcopyrite bioleaching have shown the need to operate at finer particle sizes (<10 μm) and lower (±420 mV) redox potentials (Gericke et al. 2008). These operating conditions are needed to overcome the recalcitrant nature of chalcopyrite (Gericke et al. 2010). The use of novel thermophilic mixed cultures (Plumb et al. 2002) and culture adaptation techniques (Astudillo and Acevedo 2009; Harrison et al. 2003a) has also been researched. Despite these developments, the use of thermophiles has been demonstrated to be effective only up to the maximum pulp density range of 10–12 % (w/v). A sufficient explanation of this limitation has not been provided. The extension of a similar ROS study to include chalcopyrite bioleaching systems would be insightful. In this study, the decrease in thermophilic bioleaching performance in the presence of fine pyrite particles (D50 <40 μm) is confirmed. It is postulated that the mechanism of this reduced bioleach performance is via the formation and accumulation of toxic ROS in pyrite suspensions, leading to microbial oxidative stress and bioleaching inhibition. Acknowledgments The authors gratefully acknowledge the financial support from the Department of Science and Technology and the National Research Foundation, South Africa through the SARChI Research Chair in Bioprocess Engineering. The authors would also like to acknowledge Dr. Chris Bryan for his valuable contribution to the compilation of this paper.
References Acevedo F, Gentina JC, Valencia P (2004) Optimization of pulp density and particle size in the biooxidation of a pyritic gold concentrate by Sulfolobus metallicus. World J Microbiol Biotechnol 20:865 APHA (1998) Standard methods for the analysis of water wastewater, 20th edn. American Public Health Association, Washington, DC Astudillo C, Acevedo F (2009) Adaptation of Sulfolobus metallicus to high pulp densities in the biooxidation of a flotation gold concentrate. Hydrometall 92:11–15 Bathe S, Norris PR (2007) Ferrous iron- and sulfur-induced genes in Sulfolobus metallicus. Appl Environ Microbiol 73:2491–2497
2741 Batty JD, Rorke GV (2005) Development and commercial demonstration of the BioCOPTM thermophile process. Proceedings of the 16th International Biohydrometallurgy Symposium, Cape Town, South Africa Burton NP, Williams TD, Norris PR (1995) A potential antioxidant protein in a ferrous iron-oxidizing Sulfolobus species. FEMS Microbiol Lett 134:91–95 Clark DA, Norris PR (1996) Oxidation of mineral sulphides by thermophilic microorganisms. Min Eng 9:1119–1125 Cohn CA, Pak A, Schoonen M, Strongin DR (2005) Quantifying hydrogen peroxide in iron-containing solutions using leuco crystal violet. Geochem Transact 6:47–52 Cohn CA, Laffers R, Simon S, ORiordan T, Schoonen M (2006) Role of pyrite in formation of hydroxyl radicals in coal: possible implications for human health. Part Fibre Toxicol 3:16 Cohn CA, Simon SR, Schoonen M (2008) Comparison of fluorescencebased techniques for the measurement of hydroxyl radicals in particle suspensions. Part Fibre Toxicol 5:2 Cohn C, Pedigo C, Hylton S, Simon S, Schoonen M (2009) Evaluating the use of 3′-(p-aminophenyl) fluorescein for determining the formation of highly reactive oxygen species in particle suspensions. Geochem Transact 10:8 Gericke M, Pinches A (1999) Bioleaching of copper sulphide concentrate using extreme thermophilic bacteria. Min Eng 12:893–904 Gericke M, Muller HH, van Staden PJ, Pinches A (2008) Development of a tank bioleaching process for the treatment of complex Cupolymetallic concentrates. Hydrometall 94:23–28 Gericke M, Govender Y, Pinches A (2010) Tank bioleaching of lowgrade chalcopyrite concentrates using redox control. Hydrometall 104:414–419 Hallberg KB, Johnson DB (2001) Biodiversity of acidophilic prokaryotes. Adv Appl Microbiol 49:37–84 Halliwell B (2007) Biochemistry of oxidative stress. Biochem Soc Trans 35:1147–1150 Harrison STL, Sissing A, Illing S, Nemati M (2001). Effect of mineral particulate parameters on the kinetics of thermophilic bioleaching. Proceedings of the 14th International Biohydrometallurgy Symposium, Ouro Preto, Brazil Harrison STL, Sissing A, Raja S, Nemati M (2003a) Identifying and quantifying biological responses of Sulfolobus to high pulp densities in the slurry bioleach reactor. Proceedings of the International Mineral Processing Congress, Cape Town, South Africa Harrison STL, Sissing A, Raja S, Nemati M (2003b) Solids loading in the bioleach slurry reactor: mechanisms through which particulate parameters influence slurry bioreactor performance. Proceedings of the 15th International Biohydrometallurgy Symposium, Athens, Greece Imlay JA (2003) Pathways of oxidative damage. Annu Rev Microbiol 57:395–418 Jones GC, van Hille RP, Harrison STL (2011a) Generation of reactive oxygen species from sulfide concentrates as a function of mineralogy. Proceedings of the 19th International Biohydrometallurgy Symposium, Changsha, China Jones GC, Corin KC, van Hille RP, Harrison STL (2011b) The generation of toxic reactive oxygen species (ROS) from mechanically activated sulphide concentrates and its effect on thermophilic bioleaching. Min Eng 24:1198–1208 Murphy R, Strongin DR (2009) Surface reactivity of pyrite and related sulfides. Surf Sci Rep 64:1–45 Nemati M, Lowenadler J, Harrison STL (2000) Particle size effects in bioleaching of pyrite by acidophilic thermophile Sulfolobus metallicus (BC). Appl Microbiol Biotechnol 53:173–179 Orell A, Navarro CA, Arancibia R, Mobarec JC, Jerez CA (2010) Life in blue: copper resistance mechanisms of bacteria and Archaea used in industrial biomining of minerals. Biotechnol Adv 28:839–848 Plumb JJ, Gibbs B, Stott MB, Robertson WJ, Gibson JAE, Nichols PD, Watling HR, Franzmann PD (2002) Enrichment and characterisation
2742 of thermophilic acidophiles for the bioleaching of mineral sulphides. Min Eng 15:787–794 Remonsellez F, Orell A, Jerez CA (2006) Copper tolerance of the thermoacidophilic archaeon Sulfolobus metallicus: possible role of polyphosphate metabolism. Microbiol 152:59–66 Sand W, Gehrke T, Jozsa P-G, Schippers A (2001) (Bio)chemistry of bacterial leaching—direct vs indirect bioleaching. Hydrometall 59:159–175 Schoonen MA, Cohn CA, Roemer E, Laffers R, Simon SR, O'Riordan T (2006) Mineral-induced formation of reactive oxygen species. Rev Mineral Geochem 64:179–221
Appl Microbiol Biotechnol (2013) 97:2735–2742 Schoonen MAA, Harrington AD, Laffers R, Strongin DR (2010) Role of hydrogen peroxide and hydroxyl radical in pyrite oxidation by molecular oxygen. Geochim Cosmochim Acta 74:4971–4987 Sissing AA and Harrison STL (2003) Thermophilic mineral bioleaching performance: a compromise between maximising mineral loading and maximizing microbial growth and activity. JS Afr Inst Min Metall 1-4 Valencia P, Acevedo F (2009) Are bioleaching rates determined by the available particle surface area concentration? World J Microbiol Biotechnol 25:101–106 Woese CR, Magnum IJ, Fox GE (1978) Archaebacteria. J Mol Evol 11:245–252