Iron(II)-Catalyzed Oxidation of Arsenic(III) - American Chemical Society

Manhattan College, Manhattan College Parkway, Riverdale,. New York 10471, and Department of Civil and Environmental. Engineering, University of Delawa...
0 downloads 0 Views 187KB Size
Download PDF
Environ. Sci. Technol. 2005, 39, 9217-9222

Iron(II)-Catalyzed Oxidation of Arsenic(III) in a Sediment Column KEVIN J. BISCEGLIA,§ KEVIN J. RADER,‡ RICHARD F. CARBONARO,† K E V I N J . F A R L E Y , †,| JOHN D. MAHONY,† AND D O M I N I C M . D I T O R O * ,‡,| Civil and Environmental Engineering Department, Manhattan College, Manhattan College Parkway, Riverdale, New York 10471, and Department of Civil and Environmental Engineering, University of Delaware, Newark, Delaware 19716

Arsenic contamination in aquatic systems is a worldwide concern. Understanding the redox cycling of arsenic in sediments is critical in evaluating the fate of arsenic in aquatic environments and in developing sediment quality guidelines. The direct oxidation of inorganic trivalent arsenic, As(III), by dissolved molecular oxygen has been studied and found to be quite slow. A chemical pathway for As(III) oxidation has been proposed recently in which a radical species, Fe(IV), produced during the oxidation of divalent iron, Fe(II), facilitates the oxidation of As(III). Rapid oxidation of As(III) was observed (on a time scale of hours) in batch systems at pH 7 and 7.5, but the extent of As(III) oxidation was limited. The Fe(II)-catalyzed oxidation of As(III) is examined in a sediment column using both computational and experimental studies. A reactive-transport model is constructed that incorporates the complex kinetics of radical species generation and Fe(II) and As(III) oxidation that have been developed previously. The model is applied to experimental column data. Results indicate that the proposed chemical pathway can explain As(III) oxidation in sediments and that transport in sediments plays a vital role in increasing the extent of As(III) oxidation and efficiency of the Fe(II) catalysis.

Introduction In sediments, arsenic exists primarily as inorganic trivalent arsenic, As(III), and pentavalent arsenic, As(V) (1-4). Typically As(V) dominates in the shallow, oxic layer of sediments whereas As(III) species are present in the anoxic layers below (1, 5, 6). Cycling of arsenic between the oxic and anoxic layers and the release of arsenic to the overlying water is controlled by a series of complex reactive-transport processes. Of particular interest in the overall cycling of arsenic is the oxidation of As(III) to As(V) in the oxic sediment layer. The direct oxidation of As(III) by molecular oxygen has been found to be quite slow in aqueous solutions with a half-life around 1 year (7). Several processes have been suggested in accelerating oxidation rates. For example, bacterial oxidation of As(III) has been studied (8, 9) and has been invoked to * Corresponding author phone: (302) 831-4092; fax: (302) 8313640; e-mail: [email protected]. † Manhattan College. ‡ University of Delaware. § Department of Geography and Environmental Engineering, Johns Hopkins University, Baltimore, MD 21218. | HydroQual, 1200 MacArthur Boulevard, Mahwah, NJ 07430. 10.1021/es051271i CCC: $30.25 Published on Web 10/20/2005

 2005 American Chemical Society

explain arsenic oxidation in natural systems (10, 11). The abiotic oxidation of As(III) at the surface of manganese dioxide particles has also been found to proceed relatively rapidly (12-14). Recently, the oxidation of As(III) by various free radical species has been reported (15-19). In particular, Hug et al. (16, 19) have studied the influence of iron redox chemistry on the oxidation of As(III) in oxic systems. In their proposed kinetic mechanism, radical species generated during the oxidation of divalent iron, Fe(II), by molecular oxygen are capable of facilitating As(III) oxidation over a wide range of pH values in the absence of light. A single addition of Fe(II) to an oxygenated As(III) solution resulted in rapid oxidation of As(III) to As(V) (i.e., oxidation occurred on a time scale of hours at pH 7 and 7.5). The oxidation of As(III), however, was incomplete with only 25-30% of the As(III) being oxidized even though Fe(II) was in excess (19). Multiple Fe(II) additions to an As(III) solution in the form of stepwise iron spiking resulted in nearly complete As(III) oxidation (16). This result suggests that in situations where continual oxidation of Fe(II) occurs, As(III) oxidation could potentially proceed to completion. Iron is a common constituent of sediments and its redox cycling has been studied extensively (20, 21). In sediments with an oxic overlying water column, Fe(II) diffuses from the anoxic sediment layer to the oxic-anoxic interface within the sediment where it is rapidly oxidized to Fe(III). It is therefore plausible to expect that Fe(II)-catalyzed oxidation of As(III) will occur in sediments via the mechanism proposed by Hug and Leupin (19), and that transport of Fe(II) from the anoxic layers to the oxic-anoxic interface by diffusion will be important in determining the overall extent of As(III) oxidation. The purpose of this research is to examine the behavior of the Fe(II)-catalyzed oxidation of As(III) mechanism in a simplified synthetic sediment. A laboratory sand column is used to examine the influence of transport and redox kinetics on the extent of arsenic oxidation and the potential release of arsenic to the overlying water.

Background As(III) Oxidation Mechanism. The Fe(II)-catalyzed As(III) oxidation mechanism is described in detail by Hug and Leupin (19) (see Figure S1 in the Supporting Information for a graphical representation). A brief description follows. The sequence begins with a series of one-electron transfers in which molecular oxygen is reduced to superoxide and hydrogen peroxide as Fe(II) is oxidized to Fe(III). The reaction of Fe2+ and Fe(OH)+ with hydrogen peroxide occurs producing the intermediates INT and INT-OH, respectively. The nature of the intermediates, specifically the redox state of the iron present, is not known although several possibilities have been suggested (19). These two species interconvert rapidly enough such that they are in equilibrium. According to Hug and Leupin (19), each of the two intermediates produces a different species capable of oxidizing arsenic. At low pH ( 5.24), an Fe(IV) species, is of greater importance. The Fe(IV) species is less prone to scavenging by dissolved organic species than the hydroxyl radical (19). Another sequence of one-electron transfers follows the formation of the Fe(IV) species. As(III) is oxidized to As(IV) as Fe(IV) is reduced to Fe(III), and then As(IV) is oxidized to As(V) as molecular oxygen is reduced to superoxide. AlterVOL. 39, NO. 23, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9217

TABLE 1. TICKET Parametersa Sediment Column Properties segments overlying water sediment segment area (dm2) segment height (dm): overlying water sediment sand bulk density (g/cm3) sand porosity diffusion coefficient (dm2/d) Initial Conditions simulation time pH dissolved oxygen (µM): overlying water sediment As(III) concentration (µM): overlying water sediment Fe(II) concentration (µM): overlying water sediment a

1 50 0.636 0.629 0.02 2.63 0.38 0.01 17 days 8.3 (fixed) 250 (fixed) 0 0 179 0 1790

All concentrations are in µmol/Lporewater.

FIGURE 1. Sediment apparatus for column experiments. natively, the Fe(IV) produced from INT-OH can also be reduced by Fe(II) to form Fe(III) as Fe(II) is oxidized to Fe(III). The effects of Fe(II) on the overall process are difficult to predict qualitatively. Higher concentrations of Fe(II) will likely increase the rate of Fe(IV) production. However, higher concentrations of Fe(II) will also inhibit the extent of As(III) oxidation through competition for the Fe(IV) species.

Materials and Methods Batch and Sediment Column Modeling. Initial simulations for As(III) oxidation in batch systems were made using ACUCHEM (22). The input file is supplied in the Supporting Information of ref 19. Subsequent model simulations for iron-arsenic behavior in laboratory sand sediment columns were performed using the Tableau Input Coupled Kinetic Equilibrium Transport (TICKET) model (23). For TICKET simulations, equilibrium and kinetic reactions are taken from Hug and Leupin (19). These equilibrium and kinetics reactions are incorporated into TICKET using the tableau approach (24), similar to that of the chemical equilibrium program MINEQL+ (25). Additions to the equilibrium reaction set are made to account for adsorption of iron and arsenic onto sand and adsorption of arsenic onto precipitated iron(III) oxides (modeled as hydrous ferric oxide, HFO) (26). In the TICKET simulations, it is assumed that species adsorbed to sand and HFO do not oxidize. The kinetic reactions from Hug and Leupin (19) are used with minimal alteration. The Supporting Information contains a table of the equilibrium and kinetic reactions with their associated constants as well as sample TICKET tableaux. A list of TICKET parameters is provided in Table 1. The sediment/overlying water system of the laboratory sediment column is modeled using 51 segments. The values of bulk density and porosity are set at typical values for sand. Transport in the TICKET simulations is assumed to occur solely through diffusion. For simplicity, a single diffusion coefficient is applied to all dissolved species. The diffusion coefficient used is approximately equal to the molecular diffusivity of oxygen as corrected for tortuosity according to Archie’s Law (27). Solid species such as sand and Fe(III) solids, as well as any other species adsorbed to these species, do not diffuse. 9218

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 23, 2005

Experimental Sediment Column Apparatus. The apparatus (Figure 1), constructed entirely of extruded Lucite plexiglass (AIN Plastics, Inc., Mount Vernon, NY), consists of a piston surrounded by a 1/8-in.-thick cylindrical casing (inner diameter 9 cm). The piston head has three countersunk rubber O-rings to provide a watertight seal and prevent the piston from slipping. Above the piston is a 3-cm layer of fluorocarbon-filled poly(dimethylsiloxane) (product ST-140, Sil-Tech Corp., Tecumseh, MI), which also helps to provide a watertight seal and prevents the sediment media from jamming the piston as the sediment is extruded for sampling, and a Lucite support plate upon which the sediment rests. Sediment Column Preparation. For each column, 1500 g of silica-based sand (Sand type III, Clifford W. Estes Co. Inc., Totowa, NJ) was rinsed thoroughly with distilled water to remove very fine particles, soaked for 36 h in a 1 M HCl acid bath, rinsed with 17-MΩ‚cm Millipore water to raise the pH to approximately 5.5, and dried at 103 °C for 48 h. For the sediment porewater, 200 mL of 0.179 mM As(III) was prepared in 0.05 M TRIS (2-amino-2-(hydroxymethyl)1,3-propanediol) buffer (reagent ACS grade, CAS 77-86-1, Fisher Scientific, Hampton, NH) at pH 8.3 using Fisher Scientific certified sodium arsenite, NaAsO2 (CAS 7784-46-5, Fisher Chemicals, Fairlawn, NJ). This solution was then purged with oxygen-scrubbed N2 gas for 180 min, and FeSO4‚ 7H2O (reagent ACS grade, CAS 7720-78-7, Fisher Chemicals) was added to make the Fe(II) concentration 1.79 mM. The solution was then purged for an additional 30 min, and a 5-mL sample was removed and analyzed to determine initial concentrations in the porewater. To prepare the sediment column, the porewater solution was slowly transferred into the column casing and 1000 g of the acid-washed sand was slowly poured into the porewater solution under continual mixing and N2 gas purging. For the overlying water, 400 mL of 0.05 M TRIS buffer at pH 8.3 (that had been purged with oxygen-scrubbed N2 gas for 240 min) was then gently pipetted into the column over the sand and porewater. A 1-mm-thick Teflon slip was placed over the sand during pipetting to prevent mixing between the overlying and porewaters. The column casing was then covered completely in aluminum foil to prevent lightcatalyzed reactions, and the overlying water was aerated with

FIGURE 2. ACUCHEM (22) time series output showing the extent of As(III) oxidation in a batch system under different initial Fe(II)/ As(III) molar ratios at pH 8.3. The initial concentration of As(III) is 0.179 mM. Kinetic rates are taken from Hug and Leupin (19). an aquarium pump connected to a particulate filter and water trap. Sediment Column Sampling. Periodically, 10-mL samples of the overlying water were taken, passed through a 0.1-µm Poretics syringe filter, and analyzed for As(III), total arsenic, and total iron according to the methods described below. After 17 days, the porewater of the column was sampled at 1-cm intervals (see Supporting Information for details). Each porewater sample was passed through a 0.1-µm Poretics syringe filter, and the pH was recorded. The samples were then analyzed for As(III), total arsenic, and total iron according to the methods described below. Immediately after porewater sampling, the overlying water of the sediment column was carefully removed. The sediment was slowly extruded from the column by raising the piston. As the sediment protruded above the casing it was removed at 0.5-cm intervals using a rubber squeegee and collected for analysis. Solids analysis and a 48-hour 2 M HCl extraction were performed on each 0.5-cm section (see Supporting Information for details). The extraction was filtered with a 0.1-µm Poretics syringe filter, and the filtrate was analyzed for total arsenic and total iron according to the methods described below. Arsenic and Iron Analysis. All samples were acidified to 2 M with HCl immediately. As(III) measurements were made immediately after sampling to minimize oxidation. Total arsenic and total iron measurements were made within 2 weeks of sampling. Samples were refrigerated prior to analysis. As(III) measurements were made using anodic stripping voltammetry with a thin-film gold electrode and the TraceDetect Nano-Band Explorer (28). Total arsenic and total iron measurements were performed using inductively coupled plasma-atomic emission spectrometry. The concentration of As(V) was obtained by subtracting the As(III) concentration from the total arsenic concentration.

Results and Discussion Batch Model. Batch simulations were performed with ACUCHEM to evaluate As(III) oxidation under chemical conditions corresponding to the column experiment. The initial As(III) concentration and pH (179 µM and pH ) 8.3, respectively) were chosen to reflect those used in the column experiment. The initial Fe(II) concentration was set at 10, 100, and 1000 times the initial molar As(III) concentration. For all molar ratios examined, the initial rate of As(III) oxidation is predicted to be rapid (Figure 2). However, the reaction terminates prior to complete As(III) oxidation. There is only a slight increase in the extent of As(III) oxidation (7.7% to 15%) as the Fe(II)/As(III) molar ratio is increased

from 10 to 1000. Model results indicate that increased initial Fe(II) concentrations sustain the Fe(IV) concentration at levels required to oxidize As(III) for longer periods of time (data not shown). However, the higher concentration of Fe(II) results in increased competition for the available Fe(IV). These opposing effects are responsible for the observed minor increases in As(III) oxidized associated with large increases in initial Fe(II). Thus in systems where Fe(II) is added in a single addition only a limited amount of As(III) oxidation is possible. Column Model without Adsorption. To explore the application of Hug and Leupin’s (19) kinetic mechanism in sediment systems, preliminary TICKET model simulations were performed using parameters that reflect the conditions of the column experiment (Table 1). In these initial simulations, adsorption of dissolved species to solid species was not considered. As indicated in Table 1, the model column has a constant dissolved oxygen concentration of 250 µM in the overlying water (depth ) 0 cm). As time progresses, oxygen diffuses from the overlying water into the sediment segments which are initially anoxic (Figure 3a). Concomitantly, Fe(II) and As(III) are oxidized and the oxygen is quickly depleted. As dissolved oxygen penetrates into the sediment layers, the oxidation of Fe(II) produces the transient Fe(IV) species (Supporting Information, Table S1, Kinetic Reactions B-F). The Fe(IV) species oxidizes additional Fe(II) to Fe(III) (Supporting Information, Table S1, Kinetic Reactions G-H). The resulting Fe(III) from these reaction sequences precipitates as iron hydroxide in the oxic portion of the sediment (Supporting Information, Table S1, Kinetic Reactions M-N). Diffusion transports more Fe(II) to the oxic-anoxic interface (Figure 3b) where it is oxidized, and the concentration of Fe(III) precipitate, which cannot diffuse, builds up in the oxic layers (Figure 3c). Thus, the model reproduces the high particulate iron concentrations typically found at the oxicanoxic interface (20). Diffusion also transports As(III) to the oxic-anoxic interface where the Fe(IV) species oxidizes it to As(IV) (Figure 3d). This arsenic species is then oxidized by dissolved oxygen to As(V) (Figure 3e). The location of the As(V) peak in the vertical profile follows the progression of the oxic-anoxic interface over time (Figure 3a). The As(V) produced is transported to the overlying water and the deeper sediment layers by diffusion. The behavior of total sediment arsenic is presented in Figure 3f. In the absence of any partitioning, total sediment arsenic is the sum of the soluble As(III) and As(V) concentrations or simply the total soluble arsenic. There is a soluble arsenic flux to the overlying water that is mostly As(V) (Figure 4). Initially, the gradient in As(III) between the porewater and overlying water is large (Figure 3d). However, the oxidation of As(III) to As(V) rapidly reduces this gradient, decreasing the flux of As(III) out of the sediment. As a result, As(III) transport between the porewater and overlying water rapidly approaches steady-state (Figure 4). Initially, As(V) accumulates in the porewater near the top of the column, establishing a large gradient between the porewater and overlying water (Figure 3e). The flux of As(V) to the overlying water is maintained during the simulation time as As(III) oxidation occurs through the column. Soluble iron does not accumulate to an appreciable amount in the oxic overlying water. Any soluble iron, which is essentially all Fe(II), that diffuses to the overlying water is oxidized very quickly and precipitates. As the Fe(II) in the sediment is rapidly oxidized and forms an immobile precipitate, the supply of soluble iron to the overlying water is terminated. This TICKET simulation indicates that of the total amount of As(III) and Fe(II) initially present in the sediment/overlying water system, 52.5% of the As(III) and 62.1% of the Fe(II) VOL. 39, NO. 23, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9219

FIGURE 3. Depth profiles obtained from the column model without adsorption. The various series indicate output from different times during the 17-day simulation. Overlying water concentrations are plotted at a depth of zero. Model input conditions: pH ) 8.3, initial porewater As(III) concentration ) 0.179 mM, initial porewater Fe(II) concentration ) 1.79 mM. All output concentrations are expressed as µmol/ Lporewater.

FIGURE 4. Overlying water time series obtained from the column model without adsorption. were oxidized in 17 days. This compares to the batch system results where, for the same initial Fe(II)/As(III) ratio, 100% of the Fe(II) and only 7.69% of the As(III) was oxidized. The iron in the column is much more efficient in catalyzing As(III) oxidation indicating the significance of transport on the extent of Fe(II)-catalyzed oxidation of As(III) in sediments. Comparison of Experimental Results to Column Model without Adsorption. The data from the sand column experiment are shown in Figures 5 and 6 (data points). Some of the trends noted in the initial column modeling results (Figure 3) are reflected in the experimental data. Both the column model output (Figure 3d, e) and the column data (Figure 5c, d) indicate that, in the porewater, As(V) is the most prevalent arsenic species in the shallow sediment layers while As(III) is the most prevalent species in the deeper sediment layers. The experimental soluble iron data below 4 cm (Figure 5a) have a shape similar to that of the corresponding model data (Figure 3b), suggesting an upward flux of Fe(II) is present in the experimental column. The increase in soluble iron above a depth of 4 cm is unexpected and we will return this point below. There are a few discrepancies between the column model without adsorption and the experimental data that point to necessary model refinements. First, the magnitudes of the experimental soluble iron and soluble As(III) concentrations in the bottom portion of the column (Figure 5a, c) are significantly less than the initial column model results (Figure 3b, d). This suggests that partitioning of these two chemicals to the sand media, the only sorbent in the anoxic portion of 9220

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 23, 2005

the column, must be included in the model formulation. Second, unlike the model results (Figure 3c), the experimental data (Figure 5b) do not show a large accumulation of total sediment iron in the upper portion of the column. As will be shown, adsorption of Fe(II) to sand and its effect in retarding Fe(II) transport from the anoxic sediment layers is in large part responsible for this behavior. Finally, the magnitude of the porewater As(V) as well as the total arsenic and As(V) fluxes to the overlying water are greater in the column model without adsorption (Figures 3e and 4) than in the experimental data (Figures 5d, 6b, and 6d). This indicates that As(V) partitioning to solids in the oxic sediment layers is necessary. The parallel behavior of measured total sediment iron and arsenic (Figure 5b and e) at depths less than 3 cm suggests that adsorption of arsenic to Fe(III) solids is occurring in the experimental column. Therefore, adsorption of As(V) (and As(III)) to precipitated Fe(III) are considered below. Comparison of Experimental Results to Column Model with Adsorption. In refining the column model, only minimal modifications were made. No changes were made to the kinetic parameters of the mechanism. The model results are shown in Figures 5 and 6 (lines). The Fe(II) and As(III) sand partition coefficients were calibrated in TICKET by fitting the model to the experimental porewater data at the bottom of the column (depth >5 cm) (Figure 5a, c). It is assumed that all dissolved iron in this region of the column is Fe(II). Figure 5a and b show the iron output from the column model with adsorption. The adsorption of Fe(II) to sand lowers the soluble iron concentrations in the sediment, and, as a result, iron transport to the upper layers of the column is retarded. Most of the initial iron in the bottom of the column remains there, adsorbed to immobile sand. This dramatically reduces the spike in total sediment iron near the top of the column (Figure 5b) relative to that present in the column model without adsorption (Figure 3c) producing a more reasonable fit to the experimental data. Although the model provides a good description of soluble iron concentrations at the bottom of the column (Figure 5a), it cannot account for the observed soluble iron concentrations near the top of the column nor can it explain the observed soluble iron concentrations in the overlying water (Figure

FIGURE 5. Depth profiles for 17-day experiment (data points) and 17-day simulation using the column model with adsorption (lines). Model input conditions: pH ) 8.3, initial porewater As(III) concentration ) 0.179 mM, initial porewater Fe(II) concentration ) 1.79 mM. Adsorption of arsenic to iron is formulated in the model using literature hydrous ferric oxide (HFO) parameters (26). All output concentrations are expressed as µmol/Lporewater. Overlying water concentrations are plotted at a depth of zero.

FIGURE 6. Overlying water time series for 17-day experiment (data points) and 17-day simulation using the column model with adsorption (lines). Adsorption of arsenic to iron is formulated using literature hydrous ferric oxide (HFO) parameters (26). 6a) discussed below. Although there is no direct proof, this iron is likely nonfilterable colloidal Fe(III). This species is not specifically accounted for in the current model formulation. The kinetic mechanism proposed by Hug and Leupin (19) does not account for the intricacies of Fe(III) precipitate formation and aging, particularly in the presence of TRIS, which has been shown to slow Fe(III) precipitation (29, 30). Incorporation of additional kinetic reactions to simulate Fe(III) precipitate formation through colloidal intermediates would be required to model these observations. The column model with adsorption describes the total sediment arsenic reasonably well (Figure 5e). Unlike the

model without adsorption (Figure 3f), the model with adsorption shows an accumulation of total sediment arsenic in the upper (oxic) sediment layers caused by adsorption of As(III) and As(V) to precipitated Fe(III) (HFO). This behavior is more consistent with the experimental total sediment arsenic data (Figure 5e). With adsorption of As(III) to sand and HFO included, the model also fits the experimental soluble As(III) data quite well (Figure 5c). The soluble As(V) model results (Figure 5d), which far exceeded the experimental data in the model simulation without adsorption (Figure 3e), are now in the range of the experimental data. As with the soluble iron concentrations (Figure 5a), measured soluble As(V) concentrations at the top of the column are unexpectedly high and may be associated with nonfilterable colloidal Fe(III). The overlying water concentration data for iron and arsenic are shown in Figure 6. The experimental arsenic and iron data at t ) 0 (Figure 6a and b) indicate that some mixing between the overlying water and porewater occurred during column preparation. The relative magnitudes of these initial overlying water concentrations of iron and arsenic are consistent with the initial concentrations in the sediment porewater. It is suspected that the iron released at the beginning of the experiment was rapidly oxidized and formed colloidal Fe(III). Adsorption in the model decreases the amount of soluble As(V) released to the overlying water (Figures 6d and 4). However, the adsorption of As(III) to sand and HFO has a negligible effect on the release of As(III) (Figures 6c and 4). A comparison of results from column model simulations without adsorption and with adsorption shows that adsorption affects the overall rate and extent of As(III) oxidation. The total amounts of As(III) and Fe(II) oxidized in the sediment/overlying water system in 17 days are 52.5% for As(III) and 62.1% for Fe(II) in the column model without adsorption and 26.4% for As(III) and 37.1% for Fe(II) in the column model with adsorption. Adsorption of As(III) and Fe(II) lowers the porewater concentrations of these species, slowing their oxidation kinetics and retarding their diffusive transport. As a result, the efficiency with which Fe(II) catalyzes As(III) oxidation is reduced. This can be quantified by calculating the molar ratio of total mass As(III) oxidized to total mass Fe(II) oxidized (in sediment/overlying water system VOL. 39, NO. 23, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

9221

in 17 days). This ratio is 8.45 × 10-2 for the column model without adsorption and 7.10 × 10-2 for the column model with adsorption. For both cases, the ratio is much greater than the value for the batch model (7.69 × 10-3 moles As(III) oxidized/moles Fe(II) oxidized). This indicates the significant role that the transport plays in determining the efficiency of Fe(II)-catalyzed As(III) oxidation. The column model with adsorption was formulated using an established set of equilibrium and kinetic reactions for Fe(II)-catalyzed As(III) oxidation mechanism. The only additions were those necessary to describe the partitioning of arsenic and iron in a sediment environment. In light of this, the reasonable fit of the model to the data (Figures 5 and 6), particularly the arsenic data, suggests that this mechanism is capable of explaining the oxidation of As(III) in iron-rich sediment systems. However, it is clear that more work needs to be done. To more directly assess the dynamic behavior of the proposed mechanism and provide further verification of its applicability to sediment systems, iron and arsenic sediment profiles must be generated at various time intervals and compared to model results. In addition, the potential effect of surface oxidation reactions must be investigated. Finally, research is needed to elucidate the effects of various sediment characteristicsspH, reduced sulfur content, and dissolved organic matter contentson the proposed Fe(II)-catalyzed As(III) oxidation mechanism.

Acknowledgments This research was funded by the National Institute of Environmental Health Sciences through a Superfund Basic Research Program Grant (P42ES10344). K.J.B. and K.J.R. contributed equally to this paper. We thank Joseph Nemesh for his help in measuring arsenic.

Supporting Information Available A graphical representation of the proposed Fe(II)-catalyzed As(III) oxidation mechanism (Figure S1); table of the equilibrium and kinetic reactions used in the TICKET model (Table S1); sample TICKET equilibrium and kinetic tableaux with descriptions (Tables S2 and S3); description of how TICKET quantifies adsorption of arsenic onto iron(III) oxide (Table S4 and Figure S2); description and schematic of the porewater sampler (Figure S3); and description of the solid analysis and acid extraction procedure. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Neff, J. M. Ecotoxicology of arsenic in the marine environment. Environ. Toxicol. Chem. 1997, 16, 917-927. (2) Tye, C. T.; Haswell, S. J.; O’Neill, P.; Bancroft, K. C. C. Highperformance liquid chromatography with hydride generation/ atomic absorption spectrometry for the determination of arsenic species with application to some water samples. Anal. Chim. Acta 1985, 169, 195-200. (3) Reimer, K. J.; Thompson, J. A. J. Arsenic speciation in marine interstitial water. The occurrence of organoarsenicals. Biogeochemistry 1988, 6, 211-237. (4) Masscheleyn, P. H.; Delaune, R. D.; Patrick, W. H., Jr. Arsenic and selenium chemistry as affected by sediment redox potential and pH. J. Environ. Qual. 1991, 20, 522-527. (5) Moore, J. N.; Ficklin, W. H.; Johns, C. Partitioning of arsenic and metals in reducing sulfidic sediments. Environ. Sci. Technol. 1988, 22, 432-437. (6) Aggett, J.; Kriegman, M. R. The extent of formation of arsenic(III) in sediment interstitial waters and its release to hypolimnetic waters in Lake Ohakuri. Water Res. 1998, 22, 407-411. (7) Eary, L. E.; Schramke, J. A. Rates of inorganic oxidation reactions involving dissolved oxygen. In Chemical Modeling in Aqueous Systems II; Melchior, D. C., Bassett, R. L., Eds.; ACS Symposium

9222

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 23, 2005

(8) (9)

(10) (11)

(12) (13) (14) (15) (16)

(17) (18) (19)

(20) (21) (22) (23)

(24) (25) (26) (27) (28) (29)

(30)

Series; American Chemical Society: Washington, DC, 1990; Vol. 416, pp 379-396. Oremland, R. S.; Stolz, J. F. The ecology of arsenic. Science 2003, 300, 939-944. Nicholas, D. R.; Ramamoorthy, S.; Palace, V.; Spring, S.; Moore, J. N.; Rosenzweig, R. F. Biogeochemical transformations of arsenic in circumneutral freshwater sediments. Biodegradation 2003, 14, 123-137. Wilkie, J. A.; Hering, J. G. Rapid oxidation of geothermal arsenic(III) in streamwaters of the Eastern Sierra Nevada. Environ. Sci. Technol. 1998, 32, 657-662. Langner, H. W.; Jackson, C. R.; McDermott, T. R.; Inskeep, W. P. Rapid oxidation of arsenite in a hot spring ecosystem, Yellowstone National Park. Environ. Sci. Technol. 2001, 35, 3302-3309. Scott, M. J.; Morgan, J. J. Reactions at oxide surfaces. 1. Oxidation of As(III) by synthetic birnessite. Environ. Sci. Technol. 1995, 29, 1898-1905. Manning, B. A.; Fendorf, S. E.; Bostick, B.; Suarez, D. L. Arsenic(III) oxidation and arsenic(V) adsorption reactions on synthetic birnessite. Environ. Sci. Technol. 2002, 36, 976-981. Tournassat, C.; Charlet, L.; Bosbach, D.; Manceau, A. Arsenic(III) oxidation by birnessite and precipitation of manganese(II) arsenate. Environ. Sci. Technol. 2002, 36, 493-500. Emett, M. T.; Khoe, G. H. Photochemical oxidation of arsenic by oxygen and iron in acidic solutions. Water Res. 2001, 35, 649-656. Hug, S. J.; Canonica, L.; Wegelin, M.; Gechter, D.; von Gunten, U. Solar Oxidation and removal of arsenic at circumneutral pH in iron containing waters. Environ. Sci. Technol. 2001, 35, 21142121. Lee, H.; Choi, W. Photocatalytic oxidation of arsenite in TiO2 suspension: Kinetics and mechanisms. Environ. Sci. Technol. 2002, 36, 3872-3878. Voegelin, A.; Hug, S. J. Catalyzed oxidation of arsenic(III) by hydrogen peroxide on the surface of ferrihydrite: An in situ ATR-FTIR study. Environ. Sci. Technol. 2003, 37, 972-978. Hug, S. J.; Leupin, O. Iron-catalyzed oxidation of arsenic(III) by oxygen and by hydrogen peroxide: pH-dependent formation of oxidants in the Fenton reaction. Environ. Sci. Technol. 2003, 37, 2734-2742. Stumm, W.; Morgan, J. J. Aquatic Chemistry: Chemical Equilibria and Rates in Natural Waters, 3rd ed.; John Wiley & Sons: New York, 1996. Di Toro, D. M. Sediment Flux Modeling; Wiley-Interscience: New York, 2001. Braun, W.; Herron, J. T.; Kahaner, D. K. Acuchem: A computer program for modeling complex chemical reaction systems. Int. J. Chem. Kinet. 1988, 20, 51-62. Miller, B. E. Development of a general aquatic multicomponent reactive transport computer model, with application to a wetland sediment. Doctoral thesis, Clemson University, South Carolina, 1997. Morel, F. M. M.; Hering, J. G. Principles and Applications of Aquatic Chemistry; John Wiley & Sons: New York, 1993. Schecher, W. D.; McAvoy, D. C. MINEQL+: A chemical equilibrium modeling system, Version 4.0 for Windows; Environmental Research Software: Hallowell, ME, 1998. Dixit, S.; Hering, J. G. Comparison of arsenic(V) and arsenic(III) sorption onto iron oxide minerals: Implications for arsenic mobility. Environ. Sci. Technol. 2003, 37, 4182-4189. Berner, R. A. Early Diagenesis: A Theoretical Approach; Princeton University Press: Princeton, NJ, 1980. TraceDetect. Users’ Guide: TraceDetect Nano-Band Explorer, Version 2.5; Seattle, WA, 2001. Von Gunten, U.; Schneider, W. Primary products of the oxygenation of iron(II) at an oxic-anoxic boundary: Nucleation, aggregation, and aging. J. Colloid Interface Sci. 1991, 145, 127139. Taillefert, M.; Bono, A. B.; Luther, G. W. Reactivity of freshly formed Fe(III) in synthetic solutions and (pore)waters: Voltammetric evidence of an aging process. Environ. Sci. Technol. 2000, 34, 2169-2177.

Received for review July 1, 2005. Revised manuscript received September 9, 2005. Accepted September 14, 2005. ES051271I