Rates of Arsenopyrite Oxidation by Oxygen and Fe(III) - American

Aug 10, 2007 - PRC, Department of Geological Engineering, Montana Tech of. The University of Montana, Butte, Montana 59701, Institute of Mineral ...
1 downloads 0 Views 471KB Size
Environ. Sci. Technol. 2007, 41, 6460-6464

Rates of Arsenopyrite Oxidation by Oxygen and Fe(III) at pH 1.8-12.6 and 15-45 °C YUNMEI YU,† YONGXUAN ZHU,† ZHENMIN GAO,† C H R I S T O P H E R H . G A M M O N S , * ,‡ A N D DENXIAN LI§ State Key Laboratory of Ore Deposit Geochemistry, Institute of Geochemistry, Chinese Academy of Sciences, Guiyang, 550002, PRC, Department of Geological Engineering, Montana Tech of The University of Montana, Butte, Montana 59701, Institute of Mineral Resources, Chinese Academy of Geological Sciences, Beijing, 100037, PRC

The oxidation rate of arsenopyrite by dissolved oxygen was measured using a mixed flow reactor at dissolved O2 concentrations of 0.007-0.77 mM, pH 1.8-12.6, and temperatures of 15-45 °C. As(III) was the dominant redox species (>75%) in the experimental system, and the As(III)/As(V) ratio of effluent waters did not change with pH. The results were used to derive the following rate law expression (valid between pH 1.8 and 6.4): r ) 10(-2211(57)/T(mO2)0.45(0.05, where r is the rate of release of dissolved As in mol m-2 s-1 and T is in Kelvin. Activation energies (Ea) for oxidation of arsenopyrite by O2 at pH 1.8 and 5.9 are 43 and 57 kJ/mol, respectively, and they compare to an Ea value of 16 kJ/mol for oxidation by Fe(III) at pH 1.8. Apparent As release rates passed through a minimum in the pH range 7-8, which may have been due to oxidation of Fe2+ to hydrous ferric oxide (HFO) with attenuation of dissolved As onto the freshly precipitated HFO.

Introduction Arsenic is a very toxic element which, besides being carcinogenic, can damage the nervous and digestive systems and cause a variety of skin diseases. Arsenic occurs dominantly as inorganic As(III) (arsenite) and As(V) (arsenate) ions in natural waters. The health risk is particularly high if arsenite is dominant, as this form of dissolved arsenic is 60 times more toxic than arsenate (1). Although some arsenic pollution is associated with anthropogenic input, elevated concentrations of arsenic can also occur naturally, for example, from the weathering of arsenic-bearing sulfide minerals, such as arsenopyrite, orpiment, and realgar. Arsenic occurs in many ore deposits in a variety of mineral forms. Arsenopyrite (FeAsS) is usually present in Carlin-type gold deposits, but is less common than arsenian pyrite or the simple As-sulfides, orpiment and realgar (2). Arsenopyrite becomes increasingly abundant in higher temperature “mesothermal” gold deposits, such as Homestake, South Dakota (3). Weathering of arsenopyrite and other sulfide minerals * Corresponding author phone: 406 496-4763; fax: 406 496-4260; e-mail: [email protected]. † Chinese Academy of Sciences. ‡ Montana Tech. § Chinese Academy of Geological Sciences. 6460

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 18, 2007

by oxygen-rich surface waters leads to acid mine drainage and release of elevated dissolved arsenic to down-gradient waters. Therefore, experimental studies are needed to understand the oxidation rates and mechanisms of these arsenic-rich sulfide minerals to provide a theoretical basis for effective policies for the management of natural water contaminated by arsenic. Lengke and Tempel (4-6) studied the kinetics of orpiment and realgar oxidation by dissolved oxygen (DO) at circumneutral pH and 25-40 °C. The oxidation rates were found to increase with increasing pH values and the authors pointed out that prevention of acid mine drainage by mixing sulfiderich waste rock with lime or limestone may actually enhance the rate of release of As to the environment. Rimstidt et al. (7), Ruitenberg et al. (8), and Yu et al. (9) investigated the rate of oxidation of arsenopyrite in acidic solutions. However, relatively little information is available on the rates and mechanisms of arsenopyrite oxidation at near-neutral or alkaline pH. Walker et al. (10) examined the rate of oxidation of arsenopyrite at near-neutral pH and 25 °C, and found the rates to be essentially independent of both DO concentration (in the range 0.007-0.53 mM) and pH (6.3-6.7). However, the pH range over which Walker et al. (10) performed their experiments was quite narrow. Furthermore, they did not examine the effect of changing temperature. In this paper, we report the results of kinetic experiments using a mixed flow reactor designed to investigate the rate and mechanisms of aqueous oxidation of arsenopyrite by DO over a wide range of pH, temperature, and DO concentrations. The results are discussed in terms of their significance to natural environments where weathering of arsenopyrite may be a concern.

Experimental Methods Experiments were conducted on arsenopyrite extracted from a gold-bearing quartz vein deposit located in Guizhou Province, China. The concentrations (wt %) of As, Fe, and S in the arsenopyrite based on electron microprobe analyses were 40.6, 36.4, and 22.6%, respectively (average of 10 analyses). In addition to the major elements, trace quantities of Au, Ag, Si, Ca, Mg, and Mn were detected. Samples were crushed and ground to a grain size of 60-80 mesh and handpicked under the microscope to yield a concentrate containing ∼96 vol % arsenopyrite. The arsenopyrite was then washed in alcohol for 10 min, ultrasonically cleaned three times for 1 min to remove fine silty grains coating the surface of the crystals, washed with distilled water three times, washed with acetone twice, dried, and stored in a desiccator for later use. Twenty grams of arsenopyrite were prepared in this manner. The average specific surface area of the unreacted arsenopyrite was 0.153 m2 g-1 as determined by a Quantachrome NOVA 1000 BET specific surface area and aperture distribution analyzer. Solutions of variable DO concentration (0.0-0.775 mM) were prepared by saturating feed solutions with either highly pure oxygen (99.99%), pure nitrogen, or a mixture of the two gases. The concentration of DO in the reacted waters was measured using a polarographic soluble oxygen analyzer (Leici type JPB-601, manufactured in Shanghai) at the end of the each experiment. To avoid introduction of CO2 into the solutions the inlet gases were passed over solid NaOH before entering the feed solutions. A few experiments were run at low pH with Fe+3 (0.116 × 10-4 m) as the main oxidant. The initial pH of the feed solution was adjusted by adding small amounts of superpure sulfuric acid or analytical grade sodium carbonate and sodium bicarbonate. 10.1021/es070788m CCC: $37.00

 2007 American Chemical Society Published on Web 08/10/2007

FIGURE 2. Diagram showing changes in total As and As(III)/As(V) speciation with time in a preliminary set of experiments run at 35 °C, pH 1.8, DO ) 0.25 mM, and Fe(III) ) 0.98 mM. Steady state was achieved after approximately 100 min. FIGURE 1. Experimental design. The experiments were conducted using a pneumatically mixed-flow reactor (Figure 1). This approach facilitated determination of reaction rates, as these can be determined directly from analyses of the effluent composition once the system has reached a steady state. The mixed-flow reactor used in this study was designed and developed at the Institute of Geochemistry at the Chinese Academy of Sciences (Guiyang), and it is similar to the apparatus described by Wollast and Chou (11). Additional details of the experimental design were described in ref 12. One set of experiments was conducted at a constant temperature (35 °C) and pH to evaluate the relationship between oxidation rate and varying DO concentration. In another set of experiments, DO concentration was held constant at 0.25 mM, and pH was varied between 1.8 and 10.1. A third set of experiments was conducted at temperatures of 15, 25, 35, and 45 °C to examine the temperature dependence of the reaction rate. In most experiments of this study, the mass of arsenopyrite used was around 0.2-0.5 grams and the flow rate was 5-8 mL/min. This ensured that the concentration of As in the effluent was high enough to be measured accurately and that an experiment could be completed in a day. The duration of the experiments varied between 6 and 8 h, depending on the flow rate, and was set at 3 τ, where τ refers to the average retention time (hours) of water in the reactor. After 2 τ, samples of water exiting the reactor were collected and filtered every 0.5-1 h until the end of the experiment. The samples were analyzed for total As, As(III), total Fe, and Fe2+. At the start of each experiment, a feed solution of 800 mL was introduced into the reactor from a constant-level bottle. The reactor was then placed in a constant temperature bath ((0.05 °C) and the flow of the solution and gas was initiated. The experiments were conducted at atmospheric pressure. The progress of the reaction was monitored by measuring the redox potential of the solution using a 501 ORP combination platinum/KCl/Ag-AgCl electrode. When the temperature in the reactor reached the required value, and the redox potential remained constant, the experiment was started by adding a predetermined amount of arsenopyrite to the reactor. A preliminary experiment showed that steadystate As concentrations were achieved at the effluent to the reactor after approximately 100 min (Figure 2). After completion of an experiment, the mineral grains were washed with distilled water and acetone, dried in air, and then stored in a desiccator.

TABLE 1. Experimental Data for Oxidation of Arsenopyrite (This Study) rate log mol -2 m s-1

pH s.u.

DO log m

-8.48 5.90 -3.12 -8.52 5.90 -3.11 -8.62 5.90 -3.30 -8.99 5.90 -3.60 -8.78 5.90 -3.60 -9.50 5.90 -4.81 -9.40 5.90 -4.66 -9.31 5.90 -5.03 -8.79 1.80 -3.00 -8.62 2.92 -3.00 -8.47 4.05 -3.00 -8.59 5.90 -3.00 -9.18 7.85 -3.00 -9.04 10.11 -3.00 -8.57 12.60 -3.00 -8.88 2.00 -3.60 -8.53 4.15 -3.60 -8.72 5.90 -3.60

rate temp log mol -2 °C m s-1 35 35 35 35 35 35 35 35 35 35 35 35 35 35 35 35 35 35

pH s.u.

DO log m

-8.76 6.54 -3.60 -9.52 7.71 -3.60 -9.40 7.80 -3.60 -9.26 8.58 -3.60 -8.90 10.10 -3.60 -9.33 1.80 -3.60 -8.95 1.80 -3.60 -8.79 1.80 -3.60 -8.56 1.80 -3.60 -9.38 5.90 -3.60 -8.94 5.90 -3.60 -8.59 5.90 -3.60 -8.42 5.90 -3.60 -8.20 1.80 -8.10 1.80 -7.98 1.80 -7.94 1.80

temp °C

Fe(III) log m

35 35 35 35 35 15 25 35 45 15 25 35 45 15 25 35 45

-4.94 -4.94 -4.94 -4.94

The total As and As(III) concentrations were measured by hydride generation atomic fluorescence spectroscopy (13) using an AFS-920 instrument. The detection limit for arsenic was 0.01 µg/L. Standard solutions for As(III) were prepared using As2O3 dissolved in NaOH solution. Concentrations of Fe2+ and total iron (TFe) were determined spectrophotometrically using the 1,10-phenanthroline method (12). To minimize oxidation of Fe(II) and As(III), all samples were analyzed for metal valence state within 24 h of sample collection. The rate of oxidation of arsenopyrite in the mixed-flow reactor was quantified using the following relationship (14):

r ) (Cx - Co)V/A

(1)

where r is the reaction rate (mol m-2 s-1), Cx is the steadystate concentration (mol kg-1) of an element in the effluent solution, Co is the concentration of this element in the feed solution (mol kg-1), V is the mass flow rate (kg s-1), and A is the area of mineral surface exposed to the solution (m2). For all of the experiments of this study, r was calculated based on the rate of release of total dissolved As (sum of As(III) and As(V)) to solution. Any minor amount of As introduced to the system during preparation of the oxidizing solution was subtracted as the term Co. The reactor was VOL. 41, NO. 18, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6461

FIGURE 3. The effect of DO concentration on the rate of arsenic release during arsenopyrite oxidation at 35 °C and pH 5.9.

FIGURE 5. The effect of temperature on the rate of arsenic release during arsenopyrite oxidation at pH 1.8 and 5.9, using either DO (0.25 mM) or Fe(III) (0.012 mM) as the primary oxidant. The slopes of the linear regressions were used to estimate activation energies for each set of conditions (see text).

cleaned carefully with 15% hot HNO3 solution for 12 h and rinsed several times with distilled water before its next use because we found that dissolved As can be adsorbed on the wall of the reactor and then released again during the next experiment. Following this procedure, the concentration of dissolved As in the reactor in the absence of arsenopyrite was reduced to less than 1 µg/L. The flow rate was determined by the average of three measurements conducted at the beginning, middle, and end of each experiment.

in this study, particularly for the DO-only experiments, are consistent with a surface reaction being the rate-limiting step as opposed to diffusion of solutes within the mixedflow chamber. Dissolved arsenic was primarily As(III) in the experiments of this study using DO as the main oxidant, in agreement with our previous results at low pH (9). The average As(III)/total As ratio of the effluent solutions was 0.81 ( 0.05, and showed no variation with pH. Taken as a whole, we feel that the role of As(V) in the experiments discussed in the present study is subordinate to that of As(III), and that the rate-limiting mechanisms for the oxidative dissolution of arsenopyrite should be written with As(III) compounds as both reactant and product species. This differs somewhat from previous studies of orpiment and amorphous As2S3 oxidation, in which the ratio of As(III) to total dissolved As varied from 0.51 to 0.68 (4). The reasons for the higher dissolved As(III)/As(V) ratios accompanying oxidation of arsenopyrite (this study) as opposed to orpiment (4) are not known.

Results

Discussion

Experimental results discussed in this paper are summarized in Table 1. Relationships between As release rate and DO concentration, pH, and temperature are shown in Figures 3, 4, and 5, respectively. Measured As release rates increased with increase in DO concentration (Figure 3), and were similar at low pH (10), but passed through a minimum at weakly alkaline pH (between pH 7 and 8) (Figure 4). As release rates also increased with increasing temperature (Figure 5), although the temperature dependence was less steep when Fe(III) as opposed to DO was used as the main oxidant. The slopes of the linear regressions in Figure 5 were used to derive activation energies (Ea) for the rate-limiting step for arsenopyrite oxidation. The Ea values so obtained are 16 kJ/mol and 43 kJ/mol for oxidation at pH 1.8 by Fe(III) and DO, respectively, and 57 kJ/mol for oxidation by DO at pH 5.9. We know of no published studies on the temperature dependence of arsenopyrite oxidation rates to compare with these results. However, Lengke and Tempel (4) calculated a nearly identical Ea of 59 kJ/mol for oxidation of crystalline orpiment at near-neutral pH, and Schoonen et al. (15) reported an Ea of roughly 60 kJ/mol for oxidation of pyrite at pH 2-6. The relatively high activation energies obtained

Rate Law and Comparison with Previous Work. The program Matlab was used to fit the experimental data of this study to a number of model rate law equations. The first attempt was to fit data collected at 35 °C and at pH < 6 to an equation of the form log r ) k + xpH + y(log mO2). However, no significant dependence on pH was found for the data collected in the pH range 1.8-5.9. Next, all data in Table 1 collected at pH < 6 (excluding experiments where Fe(III) was added) were regressed vs both temperature and DO concentration to obtain the following relationship:

FIGURE 4. The effect of pH on the rate of arsenic released during arsenopyrite oxidation at 35 °C and 0.25 mM or 1.0 mM dissolved O2. The apparent oxidation rates pass through a minimum near pH 7.5-8.0. Oxidation rates of orpiment from Lengke and Tempel (4) are shown for comparison.

6462

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 18, 2007

r ) 10(-2211(57)/T(mO2)(0.45(0.05)

(2)

where r is the dissolved As release rate (mol m-2 s-1) and T is in Kelvin. Because of the complex relationship between pH and As release rate at pH above 6 (Figure 4), no attempt was made to derive rate laws in the higher pH range. Also, as discussed below, there is a high likelihood that the pH neutral to alkaline experiments were complicated by precipitation of hydrous ferric oxide (HFO), which introduces considerable uncertainty into any theoretical interpretation of the experimental reaction rates. Although we have no direct experimental evidence for HFO precipitation, a single sample

Stoichiometry of Arsenopyrite and Orpiment Oxidation. The overall equation describing the first step in the oxidation of arsenopyrite by O2 can be expressed as follows:

FeAsS(s) + 3⁄2H2O + 11⁄4O2(aq) ) Fe2+ + H3AsO3(aq) + SO42- (3) This equation is balanced with respect to protons, and therefore should not result in any change in pH of the weathering solution. However, at pH values above about 5, aqueous Fe2+ will rapidly oxidize to Fe3+, even in the absence of bacteria, and will then undergo hydrolysis to precipitate some form of hydrous ferric oxide (HFO). The overall arsenopyrite oxidation reaction then becomes incongruent, and may be written as follows: FIGURE 6. The dependence of arsenopyrite oxidation rates as a function of DO concentration based on the results of this study (pH 5.9, T ) 35 °C) and those of Walker et al. (10) (pH 6.3 to 6.7, T ) 25 °C). Also shown (dashed line) are the predicted rates from this study at 25 °C, based on eq 2.

collected at pH 7.8 without filtration yielded a 12% higher As concentration than a parallel sample which was filtered. The presence of suspended forms of As is consistent with the hypothesis that some As released from arsenopyrite oxidation was lost to sorption onto HFO. The derived reaction order (0.45) with respect to mO2 in eq 2 is similar to the reported dependence of pyrite oxidation rate on (mO2)0.5 obtained by previous workers (16, 17). However, this result is in marked disagreement with the results of Walker et al. (10), who found no dependence of arsenopyrite oxidation rate on mO2 (Figure 6). Also, the As release rates obtained in the present study are greater than those of Walker et al. (10), even after our data are corrected to 25 °C by eq 2 (Figure 6). The reasons for these discrepancies are not known, although differences in pH between the two studies could be a contributing factor. Whereas the data in Figure 6 from this study were collected at pH 5.9, Walker et al. conducted their experiments at pH 6.3-6.7. Our results (Figure 4) do show a decrease in As release rates above pH 6.5. The experiments of ref 10 were performed over too small a pH range to define a clear correlation between arsenopyrite oxidation rate and pH. However, the results of Feng et al. (18) and Craw et al. (19) are in general agreement with our observation (Figure 4) that the rate of arsenopyrite oxidation passes through a minimum at pH’s between 7 and 8. Feng et al. (18) measured the in situ oxidation rate of chalcopyrite, galena, pyrite, and arsenopyrite in the natural weathering environment with a corrosion current method. These authors concluded that the oxidation rates for pyrite and arsenopyrite passed through a minimum at a pH of 6.9, and were considerably faster at pH 9.2 and 11. Craw et al. (19) observed the oxidation of arsenopyrite in the field and conducted experiments simulating arsenopyrite oxidation in the laboratory. Craw and co-workers found that the dissolved arsenic concentrations were at a minimum at pH 5-8, and increased at more acidic or more alkaline conditions (19). It is interesting to compare the rate of oxidation of arsenopyrite with simple As-sulfide minerals. Oxidation rates for orpiment reported by Lengke and Tempel (4) increased steadily with increase in pH over the complete pH range (6.6-8.3) of their experiments (Figure 4). The reaction rates for orpiment are actually very similar to those for arsenopyrite based on the results of this study at pH > 7.5, but they diverge quickly at lower pH (Figure 4). Possible reasons for this discrepant behavior are discussed in the following section.

FeAsS(s) + 4H2O + 3O2(aq) ) Fe(OH)3(s) +

H3AsO3(aq) + SO42- + 2H+ (4)

The results of Walker et al. (10) are consistent with reaction 4, since the release rates of dissolved As were faster than those of dissolved Fe in their experiments. In reaction 4, each mole of arsenopyrite oxidized produces two moles of protons, and the reaction is, therefore, acid-generating. Additional acid may be produced if arsenite is oxidized to an anionic form of arsenate:

H3AsO3(aq) + 1⁄2O2 ) HAsO42- + 2H+

(5)

H3AsO3(aq) + 1⁄2O2 ) H2AsO4- + H+

(6)

Although reactions 5 and 6 were relatively unimportant on the time scale of the experiments of this study (t ) several hours), they may progress to completion in the weathering environment (t ) days to months) such that dissolved arsenate becomes the dominant species (20). However, in other cases, particularly where soil is completely saturated, As(III) may persist as the dominant redox state in solution (21). At pH > 5, dissolved ferric iron is highly insoluble, and therefore should not be available as a reactant except for the brief amount of time elapsed between the oxidation of Fe2+ to Fe3+ and the subsequent hydrolysis of aqueous Fe3+ to form Fe(OH)3(s). Precipitation of HFO via reaction 4 will cause adsorption of dissolved As, especially if the latter is present in the As(V) valence state (22-23). However, because the solubility of Fe(OH)3(s) increases quickly with decrease in pH, dissolved ferric iron may persist in low pH solutions. In this case, oxidation of arsenopyrite can occur via both the O2 pathway (reaction 2) or the Fe(III) pathway. The latter reaction can be written as follows:

FeAsS(s) + 7H2O + 11Fe3+ ) 12Fe2+ + H3AsO3(aq) +

SO42- + 11H+ (7)

Because reaction 7 consumes a large quantity of Fe3+, the supply of this reactant will quickly be depleted unless it can be regenerated by reoxidation of Fe2+ to Fe3+. In a flowthrough experimental system where ferric sulfate is added to the feed solution, the supply of Fe3+ is continuously replenished. However, in a field situation the above reaction would eventually stop, unless DO is present to reoxidize Fe2+ to Fe3+:

Fe2+ + H+ + 1⁄4 O2 ) Fe3+ + 1⁄2 H2O

(8)

Many studies have shown that the abiotic rate of Fe2+ oxidation via reaction 8 is extremely slow at low pH, but is VOL. 41, NO. 18, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

6463

catalyzed many orders of magnitude by acid-tolerant bacteria such as Acidithiobacillus ferrooxidans (24-25). Compared to arsenopyrite, the overall oxidation reaction for orpiment is relatively simple, and can be written as follows:

As2S3(s) + 6H2O + 6O2(aq) ) 2H3AsO3(aq) + 3SO42- +

6H+ (9)

Because orpiment contains no iron, there is no possibility of involvement of Fe3+ as an oxidant species in laboratory studies of this mineral. However, in a natural setting it is likely that some Fe-bearing minerals such as pyrite will be present to serve as a source of Fe3+. If so, and if pH is < 5, then reaction 9 can be rewritten with Fe3+ as the main oxidant species as follows:

As2S3(s) + 18H2O + 24Fe3+ ) 2H3AsO3(aq) + 24Fe2+ +

3SO42- + 30H+ (10)

The above discussion helps to interpret the results of the laboratory experiments of this study, as well as those of previous workers. For example, our observation that arsenopyrite oxidation rates pass through a minimum at pH 7-8 (Figure 4) may reflect a switch in the main oxidant species from aqueous Fe(III) (at lower pH) to dissolved O2 (at higher pH). Alternatively, Fe(III) may be the initial oxidant at all pH values and that the amphoteric behavior of the reaction rates simply reflects the lower solubility of Fe(OH)3(s) at neutral pH, with concomitant sorption of dissolved As onto HFO. Following either hypothesis, it makes sense that orpiment shows a much simpler pH-dependence of laboratory oxidation rates (Figure 4) since there is no possibility for any Fe3+ species as either a product or reactant in the artificial experimental system. Finally, the fact that abiotic oxidation of Fe2+ to Fe3+ is very slow at low pH may explain the fact that the arsenopyrite oxidation rates in this study were slightly lower at pH < 3 when compared to the rates at pH 5-7 (Figure 4). Although the experiments were not completely sterile, it is likely that the population of Fe-oxidizing bacteria were much lower than would be expected in the natural environment.

Acknowledgments This project was financially supported by the National Natural Science Foundation of China (grant no.40373041) and State Key Laboratory of Ore Deposit Geochemistry, Institute of Geochemistry, Chinese Academy of Sciences. Guiyang. We thank Prof. Wang Fuya, who determined the specific surface area of arsenopyrite charges, and Prof. Zhu Jianming, who supported the equipment for determining the concentration of dissolved arsenic collected from effluent solutions of the experiments. We thank Don Rimstidt and two anonymous reviewers for their very thorough and constructive reviews. We also thank Curtis Link (Montana Tech), who performed the Matlab statistical analysis.

Literature Cited (1) Bottomly, D. J. Origins of some arseniferous groundwater in Nova Scotia and New Brunswick. Can. J. Hydrol. 1984, 69, 223257. (2) Hofstra, A. H.; Cline, J. S. Characteristics and models for Carlintype gold deposits. In Gold in 2000, Reviews in Economic Geology; Hagemann, S. G., Brown, P. E., Eds.; Society of Economic Geologists: Littleton, CO, 2000; Vol. 13, pp 163-220. (3) Slaughter, A. L. The Homestake Mine. In Ore Deposits of the United States, 1933-1967; Ridge, J. R., Ed.; American Institute of Mining Metallurgical Engineers: New York, 1968; Vol. 2, pp 1436-1459.

6464

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 18, 2007

(4) Lengke, M. F.; Tempel, R. N. Reaction rate of natural orpiment oxidation at 25 to 40 °C and pH 6.8 to 8.2 and comparison with amorphous As2S3 oxidation. Geochim. Cosmochim. Acta. 2002, 66, 3281-3291. (5) Lengke, M. F.; Tempel, R. N. Natural realgar and amorphous AsS oxidation kinetics. Geochim. Cosmochim. Acta. 2003, 67, 859-871. (6) Lengke, M. F.; Tempel, R. N. Geochemical modeling of arsenic sulfide oxidation kinetics in a mining environment. Geochim. Cosmochim. Acta. 2005, 69, 341-356. (7) Rimstidt, J. D.; Chermak, J. A.; Gagen, P. M. Rate of reaction of galena, sphalerite, chalcopyrite, and arsenopyrite with Fe(III) in acidic solutions. In Environmental Geochemistry of Sulfide Oxidation; Alpers, C. N., Blowes, D. W., Eds.; American Chemical Society: Washington DC, 1994; pp 2-13. (8) Ruitenberg, R.; Hansford, G.; Reuter, M.; Breed, A. The ferric leaching kinetics of arsenopyrite. Hydrometallurgy 1999, 52, 37-53. (9) Yu, Y.; Zhu, Y.; Williams-Jones, A. E.; Gao, Z.; Li, D. A kinetic study of the oxidation of arsenopyrite in acidic solutions: implications for the environment. Appl. Geochem. 2004, 19, 435-444. (10) Walker, F. P.; Schreiber, M. E.; Rimstidt, J. D. Kinetics of arsenopyrite oxidative dissolution by oxygen. Geochim. Cosmochim. Acta. 2006, 70, 1668-1676. (11) Wollast, R.; Chou, L.; Kinetic study of the dissolution of albite with a continuous flow-through fluidized bed reactor. In The Chemistry of Weathering, NATO Science Series C; Drever, J. I., Ed.; Kluwer Academic Publishers: Hingham, MA, 1985; Vol. 149. (12) Yu, Y.; Zhu, Y.; Gao, Z. Kinetic experiment on the oxidation of arsenopyrite in acidic solutions-(1) Experimental method and part of results. Acta Mineral. Sinica (in Chinese) 2000, 20 (4), 390-396. (13) Featherstone, A. M.; Butler, E. C. V.; O’Grady, B. V.; Michel, P. Determination of arsenic species in sea-water by hydride generation atomic fluorescence spectroscopy. J. Anal. At. Spectrom. 1998, 13, 1355-1360. (14) Zhang, S.; Li, T.; Wang L. The principle of geochemically dynamic reactor and the equation of velocity determination. Geol.Geochem. 1997, 1, 53-58 (in Chinese). (15) Schoonen, M.; Elsetinow, A.; Borda, M.; Strongin, D. Effect of temperature and illumination on pyrite oxidation between pH 2 and 6. Geochem. Trans. 2000, 4, DOI: 10.1039/b004044o. (16) McKibben, M. A.; Barnes, H. L. Oxidation of pyrite in low temperature acidic solutions: Rate laws and surface textures. Geochim. Cosmochim. Acta. 1986, 50, 1509-1520. (17) Williamson, M. A.; Rimstidt, J. D. The kinetics and electrochemical rate-determining step of aqueous pyrite oxidation. Geochim. Cosmochim. Acta. 1994, 58, 5443-5454. (18) Feng, Q.; Xu, S.; Chen, J. Study of the oxidation kinetics of sulfides. J. Cent. South Inst. Mining Metall. 1993, 24, 31-35. (19) Craw, D.; Falconer, D.; Youngson, J. H. Environmental arsenopyrite stability and dissolution: theory, experiment, and field observations. Chem. Geol. 2003, 199, 71-82. (20) Bowell, R. J.; Morley, N. H.; Din, V. K. Arsenic speciation in soil porewaters from the Ashanti Mine, Ghana. Appl. Geochem. 1994, 9, 15-22. (21) Salzsauler, K. A.; Sidenko, N. V.; Sherriff, B. L. Arsenic mobility in alteration products of sulfide-rich, arsenopyrite-bearing mine wastes, Snow Lake, Manitoba, Canada. Appl. Geochem. 2005, 20, 2303-2314. (22) Oscarson, D. W.; Huang, P. M.; DeFosse, C.; Herbillon, A. Oxidative power of Mn(IV) and Fe(III) oxides with respect to As(III) in terrestrial and aquatic environments. Nature 1981, 291, 50-51. (23) Stollenwerk, K. Geochemical processes controlling transport of arsenic in groundwater. A review of adsorption. In Arsenic in Groundwater; Welch, A., Ed.; Kluwer Academic Publishers: Boston, 2003; pp 67-100. (24) Singer, P. C.; Stumm, W. Acidic mine drainage: the ratedetermining step. Science 1970, 167, 1121-1123. (25) Millero, F. J.; Sotolongo, S.; Izaguirre, M. The oxidation kinetics of Fe(II) in seawater. Geochim. Cosmochim. Acta. 1987, 51, 793801.

Received for review April 3, 2007. Revised manuscript received June 27, 2007. Accepted July 5, 2007. ES070788M