Removal of Arsenic from High Ionic Strength Solutions: Effects of Ionic

Membrane concentrate disposal: practices and regulation Denver, CO., 2001; USBR Desalination and Water Purification Research and Development Program ...
0 downloads 0 Views 901KB Size
Download PDF
Environ. Sci. Technol. 2008, 42, 3797–3802

Removal of Arsenic from High Ionic Strength Solutions: Effects of Ionic Strength, pH, and preformed versus in situ formed HFO KENNETH L. MERCER* AND JOHN E. TOBIASON Civil and Environmental Engineering, University of Massachusetts at Amherst, 18 Marston Hall, Amherst, Massachusetts, 01003-9293

Received December 1, 2007. Revised manuscript received February 13, 2008. Accepted March 1, 2008.

Arsenic sorption to hydrous ferric oxide (HFO) is an effective treatment method for removing dissolved arsenic from fresh drinking water sources. However, detailed information is limited regarding arsenic removal from solutions of high ionic strength such as brackish groundwater, seawater, or highpressure membrane process residuals. Bench-scale treatment experiments were conducted exploring arsenic removal from simple solutions with ionic strengths ranging from 0.008 to 1.5 M by addition of ferric chloride followed by solid/liquid separation (microfiltration or ultrafiltration). Arsenic removal from these solutions during in situ iron precipitation was approximately 90% at Fe:As molar ratios of 10 to 15 and >95% for Fe:As molar ratios greater than 20. Arsenic removal at iron doses of 10-6 to 10-4 mol-Fe/L improved when pH was lowered from 8 to less than 6.5 at ionic strength 0.2 M; this improvement was not as significant at ionic strength 0.7 M. Arsenic removal diminished when alkalinity was increased from 400 to 1,400 mg/L as calcium carbonate; however, arsenic removal at the higher alkalinity improved when pH was lowered from approximately 8 to less than 7. Arsenic removal with preformed HFO solids and subsequent microfiltration was significantly less than that observed with in situ HFO precipitation. Increased removal by in situ precipitation compared to that of preformed solids is explained by an increased number of adsorption sites due to uptake during iron oxy-hydroxide polymerization as well as an increase in surface area resulting in diminished surface charge effects. Model simulations of arsenic uptake by in situ precipitation adequately captured these effect by changing the model parameters used to model arsenic uptake by preformed HFO, specifically the total number of surface sites and surface area.

Introduction Arsenic removal from high ionic strength solutions (∼0.2–1.5 M) is important to several situations encountered in potable water treatment. Arsenic is known to be rejected by nanofiltration (NF) and reverse osmosis (RO) membranes (1–7) and is therefore concentrated in membrane process residuals (concentrate) if initially present in the raw water source. Accumulation of pollutants such as arsenic in membrane * Corresponding author fax: 413-545-2202; e-mail: kmercer@ ecs.umass.edu. 10.1021/es702946s CCC: $40.75

Published on Web 04/11/2008

 2008 American Chemical Society

concentrate is of concern to the water treatment industry as implementation of NF or RO membrane processes has been limited or even ruled out in some cases due to concerns over residuals management (8, 9). Removal of arsenic from membrane concentrate may increase the number of feasible alternatives for its disposal, reuse, and/or resource recovery. Ferric Iron in Water Treatment. Ferric iron salts such as ferric chloride are typically used in water treatment as coagulants to destabilize and subsequently aggregate suspended, colloidal, and dissolved matter (10). At oversaturated concentrations, hydrolyzed iron species precipitate and aggregate into amorphous structures with high porosity known as hydrous ferric oxide (HFO). HFO provides active sites for sorption of dissolved species by surface complexation or ligand exchange (11–13). The kinetics of iron hydrolysis in saline waters have been reported to be fast with second order rate constants on the order of 4 × 107 M-1s-1 at pH 8.1 (14); this rate may be enhanced by the presence of colloids and particles acting as nuclei for polymerization (15). Further description of iron solubility in high ionic strength environments is included in the Supporting Information. Ionic Strength Effects. Ionic strength effects can impact particle aggregation by influencing electrostatic interactions. Most particles in natural waters have negatively charged surfaces (11) which must be balanced locally by counterions in the adjacent solution to maintain overall electroneutrality. The region adjacent to a charged surface is classically treated as a double layer consisting of a tightly bound layer (Stern layer) and a diffuse layer occupying the region between the Stern layer and the electrically neutral bulk solution (16). When similarly charged particles approach each other, their surface induced electric potential fields and associated diffuse layers interact, creating a repulsive force that can overcome the attractive van der Waals force and keep the particles separated. Increased ionic strength has been shown to reduce electrostatic repulsion and thereby increase particle aggregation in both natural waters (17) and laboratory simulations (18). Arsenic Removal. Arsenic is a naturally occurring carcinogenic metalloid (19). Fresh water arsenic concentrations are typically less than 10 µg/L (20), however, levels in groundwater in the United States can range from 10 to 120 µg/L (21, 22). Arsenic in natural waters is commonly encountered in two oxidation states: As(V) or arsenate and As(III) or arsenite (23, 24). Arsenite is more mobile and toxic than arsenate; however, arsenate species have been shown to be more readily removed than arsenite (25, 26), and water treatment for arsenic removal often involves preoxidation (27, 28), where conventional oxidants in water treatment (e.g., ozone, chlorine, and permanganate) are capable of rapidly oxidizing arsenite to arsenate (29). Removal of arsenic by treatment with metal salts and subsequent solid/liquid separation has been studied extensively for treatment of fresh waters (29–33), where uptake by iron precipitates has been more effective than that of aluminum precipitates (34–36). Several researchers (37, 38) have reported effective removal of arsenic at Fe:As mole ratios of 7 to 8. There is little to no reported difference in arsenic removal using ferric chloride or ferric sulfate, however, polymeric ferric salts have shown higher removals for comparable molar iron doses (34). Arsenate has been shown to specifically adsorb to HFO, and although the mechanism of adsorption is not fully agreed upon, several researchers (39, 40) have shown strong evidence of a bidentate, corner sharing, AsO4-FeOOH inner-sphere VOL. 42, NO. 10, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3797

surface complex, although a monodentate complex may also exist (41). Arsenate is thought to bind via a ligand-exchange mechanism to surface hydroxyl functional groups on iron solids. In addition to adsorption, surface precipitation may also be an important mechanism for removal of dissolved arsenic (42), although some researchers have indicated that only adsorption is important (40, 43). Treatment of waste streams with iron salts for arsenic removal has been explored previously. Clifford et al. (44) found that a wastewater with 90 mg/L of arsenic and 50,000 mg/L TDS required a Fe:As mole ratio of 12 to decrease the arsenic concentration by 90% and a Fe:As mole ratio of 20 to decrease the arsenic concentration to less than 1.5 mg/L. Another study (45) explored arsenic removal by ferric chloride treatment of residuals from several full-scale processes including desalting membranes; this study found that use of ferric chloride resulted in 80–99% arsenic removal over the range of residual qualities tested. Arsenate sorption to iron hydroxide surfaces has been shown to increase with decreasing pH (26, 29); such behavior has been explained by decreased competition of hydroxide ions for active HFO surface sites (30). In seawater matrices, arsenate sorption may be inhibited because of ion pairing and complexation with seawater ions (46). Increasing ionic strength from that of fresh to seawater was found to decrease arsenic adsorption onto activated alumina, carbon, and bauxite (47). The displacement of arsenic by bicarbonate/carbonate may be important in waters with high alkalinity (48). Bicarbonate was shown by some researchers to affect arsenic removal during in situ iron precipitation (49) or adsorption to performed iron solids (50). However, other researchers have found only a slight decrease in arsenic removal capacity attributable to inorganic carbon (32, 51). Sulfate has been reported to have little influence on arsenic sorption to HFO for pH > 5 (29, 51). Additionally, silicate (51) and phosphate (52) have been shown to negatively impact arsenate adsorption during coprecipitation with iron.

Experimental Section The composition of the saline solutions used for these experiments was chosen to represent simplified NF or RO concentrates. Specific ratios of sodium chloride and sodium bicarbonate were used to create solutions at ionic strengths ranging from 0.008 to 1.5 M with alkalinity of 400 or 1,400 mg/L as CaCO3. Solutions with low ionic strength (∼0.008 M) consisted of only deionized (DI) water and sodium bicarbonate without the addition of sodium chloride. Reagent grade chemicals or higher quality were used for all experiments. Arsenic was added to the simulated concentrate to achieve approximately 500 ppb As(V) from a stock solution of Na2HAsO4 · 7H2O (Alfa Aesar, Ward Hill, MA). Arsenic addition (as As(V)) was always made prior to the addition of ferric chloride or preformed ferric solids. Treatment experiments were conducted using a jar-stirrer (Phipps and Bird; Richmond, VA) with stainless steel mixer blades (1.9 cm × 5.1 cm) and 600-mL glass beakers. Rapid mixing was performed using a magnetic stirrer and lasted three minutes from the time of ferric chloride or preformed HFO addition. After rapid mix the sample was mixed at 40 rpm for thirty minutes under the jar-stirrer. For the solutions under consideration in this study, arsenic uptake by HFO reached pseudosteady state in approximately ten minutes (see Supporting Information). Replicate experiments were performed for each experimental series. After mixing, microfiltered samples were prepared using a 0.22 µm nitrocellulose membrane syringe filter (Millipore; Billerica, MA). At least 10 mL of sample water was passed through the microfilter before collecting 15 mL for analysis. 3798

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 10, 2008

Ultrafiltered samples were prepared using an Amicon Stirred Cell model 8200 (Millipore; Billerica, MA) with a 63.5 mm diameter YM30 regenerated cellulose filter (Millipore; Billerica, MA) with 30,000 nominal molecular weight limit (NMWL). At least 100 mL of sample water was passed through the ultrafilter before collecting 15 mL for analysis. Ferric chloride solutions were made the day of the experiments from FeCl3•6H2O (Fisher Chemical Co., Fairlawn, NJ). A 0.1 M as Fe solution was continuously mixed in a closed vessel for one hour prior to making diluted iron stock solutions (0.01 and 0.001 M as Fe) which were used within 1 h of creation. Preformed hydrous ferric oxide solids were prepared by addition of ferric chloride to DI water to obtain a specific iron concentration. Sodium hydroxide was added until the pH stabilized at 8.0. The resulting suspensions were mixed for at least two hours and dosed volumetrically to achieve a known iron concentration. Fresh suspensions were prepared prior to each experiment. Arsenic and iron were measured with a Perkin-Elmer Sciex ELAN DRC-e ICP Mass Spectrometer (Perkin-Elmer; Waltham, MA) equipped with a 600-µL flow injection (FI) system. FI was used to minimize the solids loading to the ICP-MS cones, thereby reducing the cleaning requirements. No attempt was made to determine arsenic speciation. Data was collected in the Data Only method in counts/second; reported concentrations are based on the peak maximum after subtracting background levels. Calibration curves were made for each experiment using a matrix resembling that of the samples. Aqueous standard solutions were prepared on the day of treatment using the salt matrix prior to arsenic addition. Coefficient of determination for linear regression fits of the calibration data were always greater than 0.99. The method detection limit of this method was 1 ppb As. Arsenic analysis was carried out within one week of sample collection. Samples were diluted on a mass basis to maintain the total salt content at less than 1%. Concentrations were determined by accounting for all dilutions on a mass basis. At least one arsenic spike recovery was made with each round of sample tests; the measured spike recovery averaged within 6 ppb of the calculated value for spiked arsenic levels ranging from 50 to 200 ppb. Modeling. Mechanistic surface complexation models (SCMs) have been widely used to describe the interactions between aqueous solutes and solid surfaces. The underlying physical/chemical phenomena leading to adsorption are described in these models using equilibrium adsorption reactions that place adsorbed ions in specific electrostatic planes within an electrical double layer (EDL) adjacent to the mineral surface (53). In essence, surface complexation is a competition between H+, OH-, metals, and ligands for surface active sites. In the case of HFO, surface charge is considered to develop via chemical reactions at amphoteric hydroxyl groups (’FeOH) which comprise the reactive surface sites (54). Calculation of dissolved and adsorbed speciation was made using the USGS model PHREEQC version 2.14.00 with the Lawrence Livermore National Laboratory (LLNL) database (55). PHREEQC uses ion-association and extended Debye–Hückel expressions based on fits for major ions using chloride mean-salt activity-coefficient data to account for the nonideality of aqueous solutions. Pertinent modeling parameters including As(V) speciation and pertinent surface complexation reactions are included in the Supporting Information.

Results and Discussion Ionic Strength and pH Effects. The effect of addition of 10-6 to 10-3 M ferric iron on the removal of arsenic at several ionic strengths is shown in Figure 1. In all cases arsenic

FIGURE 1. Arsenic concentration in filtrate and final pH after treatment of solutions with varying ionic strengths (0.008 to 1.5 M) as a function of iron dose (alkalinity 400 mg/L as CaCO3; MF: 0.22 µm filtration, UF: 30,000 NMWL). removal increases with ferric dose, with 90% removal occurring at Fe:As molar ratios of approximately 10 to 15 and >95% removal occurring at Fe:As molar ratios greater than 20. The pH was not controlled and decreased with increasing iron dose. There was little difference between the removals achieved at ionic strengths from 0.2 to 1.5 M following either microfiltration (MF) or ultrafiltration (UF, results not shown for these ionic strengths). Arsenic removal at ionic strength 0.008 M was less than that observed at the higher ionic strengths following solid/ liquid separation by MF or UF. At the higher ionic strengths, reduced electrostatic repulsions between colloids can increase opportunities for particle contacts and subsequent aggregation allowing for larger particle sizes and therefore more effective solid/liquid separation. At ionic strength 0.008 M, electrostatic interactions may have prevented effective floc formation and lower subsequent removal following filtration. This explanation is supported by the results for ultrafiltered samples at ionic strength 0.008 M, which show lower arsenic levels than those of the microfiltered samples at intermediate iron doses, yet higher levels than those for ionic strengths of 0.2–1.5 M. Additionally, iron levels following MF or UF at the higher ionic strengths were negligible, while iron levels at ionic strength 0.008 M depended on separation method. Negligible levels were observed following ultrafiltration but up to 1.4 mg-Fe/L were found in the microfiltered samples (see Supporting Information), meaning iron particles and associated arsenic were passing through the microfilter. This finding may have implications on membrane selection for removal of arsenate from main process treatment trains by iron coagulation and subsequent low pressure membrane filtration.

FIGURE 2. Arsenic concentration in filtrate and final pH after treatment of solutions with 0.2 M ionic strength as a function of ferric iron dose (alkalinity 400 mg/L as CaCO3; MF: 0.22 µm filtration). The effect of solution pH is shown in Figure 2 and Figure 3. Figure 2 shows that at ionic strength 0.2 M, lowering pH improved arsenic removal for comparable iron doses. To achieve >90% arsenic removal without initial pH adjustment (pH0 8.1) required an Fe:As ratio of ∼15; in contrast, the Fe:As ratio for similar arsenic removal was ∼9 at pH0 7.3 and ∼ 6 at pH0 6.2. However, Figure 3 shows that at ionic strength 0.7 M, only a slight increase in arsenic removal was observed when the pH was lowered from 8.0 to 6.5 for solutions with alkalinity of 400 mg/L as CaCO3; the Fe:As ratio to achieve at least 90% removal was approximately the same regardless of initial pH. One possible explanation for this is that the final pH at ionic strength 0.2 M was almost 0.5 pH units lower than at ionic strength 0.7 M for iron doses where arsenic removal is different. Figure 3 also shows pH buffering effects of alkalinity, which, when increased from 400 to 1,400 mg/L as CaCO3 without initial pH adjustment, diminished the removal of arsenic for similar iron doses. In addition to diminished competition from OH-, decreased pH should increase the positive charge of the iron hydroxide surface thereby improving removal of anionic species such as arsenate even as its speciation changes. The high alkalinity solution required a Fe:As ratio of more than 30 to achieve greater than 90% arsenic removal without acid addition. However, when the pH of the high alkalinity solution was lowered prior to ferric chloride addition, arsenic removal was similar that observed at the lower alkalinity level, with Fe:As ratio of ∼10 resulting in >90% arsenic removal. In Situ Precipitation and Adsorption to Preformed Solids. Experiments were made to compare arsenic uptake by preformed HFO solids to removal via in situ iron preVOL. 42, NO. 10, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3799

FIGURE 3. Arsenic concentration in filtrate and final pH after treatment of solutions with 0.7 M ionic strength as a function of ferric iron dose (alkalinity 400 or 1,400 mg/L as CaCO3; MF: 0.22 µm filtration).

FIGURE 4. Arsenic concentration in filtrate after treatment by in situ precipitation (model: 0.700 mol-sites/mol Fe, 5.3 × 104 m2/ mol) or adsorption to preformed solids (model: 0.700 mol-sites/ mol Fe, 5.3 × 104 m2/mol) for solutions with I ) 0.2 M as a function of ferric dose. cipitation. The results in Figure 4 show less arsenic removal following microfiltration when using preformed solids as compared to in situ precipitation. Potential explanations for increased removal by in situ precipitation include direct precipitation, surface precipitation, solid-solution formation, or an increase in the number of adsorption sites provided by polymerizing iron oxy-hydroxide molecules during in situ precipitation. 3800

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 10, 2008

Differences between arsenic adsorption to iron oxyhydroxide solids during in situ precipitation (also referred to as coprecipitation) as compared to addition of preformed solids have been reported by several researchers. Hering et al. (29) reported discrepancies between model predictions for in situ precipitation of arsenate using ferric chloride and hypothesized that these differences resulted from surface charge effects. Holm (49) also reported differences between results from in situ precipitation experiments and surface complexation model predictions, but explained the differences as resulting from a greater sorption capacity for in situ precipitation (assumed to be 3.8 times greater) than that of preformed HFO. Direct precipitation of arsenic-iron solids has been shown by EXAFS analysis to be negligible with no evidence of subsequent surface precipitation or formation of a new solidsolution at near-neutral to alkaline pH (40, 43). Surface precipitation was reported only under acidic conditions (pH 3–5) (56). The log solubility product (log Ksp) of amorphous ferric arsenate has been reported to be in the range of -23.0 to -24.45 (57, 58). For the system conditions explored in the present study, direct iron-arsenic precipitation was not predicted to occur using PHREEQC model calculations. Fuller et al. (43) also observed significantly higher arsenic uptake during in situ precipitation in comparison to adsorption to preformed solids. In this study, however, arsenic was reported to be coordinated by surface sites before crystal growth with ratios of approximately 0.7 mol-As/mol-Fe for in situ precipitation versus 0.25 for preformed solids. Another study reported that experiments using in situ precipitation showed from 1 to 3 orders of magnitude difference in arsenic adsorption compared to adsorption onto preformed solids (38). Adsorption during iron hydrolysis and precipitation has been hypothesized to halt the structuring of iron oxyhydroxide, effectively poisoning the crystallization processes (43). When sorbing ions are present during the polymerization/precipitation of iron, they may adsorb to crystallite surfaces before aggregates are formed, thereby increasing the number of apparent adsorption sites. This may explain the greater arsenic removal by in situ precipitation compared to that observed with preformed HFO found in the present study. PHREEQC was used to simulate the effects of iron addition on arsenic removal. To model the observed differences between in situ iron precipitation and preformed HFO solids, the number of surface sites for the in situ precipitation case was changed from 0.205 total mol site/mol iron (used for preformed HFO) to 0.7 mol site/mol iron as indicated by Fuller et al. (43). With this change incorporated into the model, the surface area was used as a fitting parameter to fine-tune the model fits (see Supporting Information). Justification for increasing the surface area follows from the same arguments for increasing the surface site density for in situ ferric iron precipitation versus preformed HFO. Based on model fits of the in situ results, the surface areas for modeling arsenic removal via in situ precipitation at ionic strengths 0.2 and 0.7 M were chosen to be 5 × 10 (5) and 2 × 10 (5) m2/mol Fe, respectively (see Supporting Information); the results of the model simulations are shown in Figure 4 and Figure 5. Even though particle aggregation and ultimate size are not parameters included in the model, the model output describes the experimental results reasonably well for both in situ iron precipitation and adsorption to preformed solids. Changes in pH due to iron addition were modeled reasonably well, within 0.2 pH units (model results not shown). The results presented in this paper show that removal of arsenic (as arsenate) from NF or RO membrane concentrate should differ from treatment of fresh water even before

FIGURE 5. Arsenic concentration in filtrate after treatment by in situ precipitation (model: 0.700 mol-sites/mol Fe, 5.3 × 104 m2/ mol) or adsorption to preformed solids (model: 0.700 mol-sites/ mol Fe, 5.3 × 104 m2/mol) for solutions with I ) 0.7 M as a function of ferric dose. additional complexities are considered (i.e., antiscalants interactions, inhibited precipitation of solids phase, etc.) The higher ionic strength of concentrate should lend itself to effective removal of arsenic where optimal removals may be achieved by in situ iron precipitation followed by microfiltration. Adjustment of pH, which may be accomplished during pretreatment to prevent membrane scaling or fouling, should improve arsenic removal from the concentrate depending on the final pH of the residual stream.

Acknowledgments This research was funded in part through a joint fellowship from the National Water Research Institute (NWRI-USA) and Cargill, Inc. Additional funding has been provided by the Robert and Patricia Switzer Foundation and the Perrell Scholarship Endowment in Environmental Engineering from the University of Massachusetts at Amherst.

Supporting Information Available Details regarding iron solubility calculations as a function of ionic strength, kinetics of arsenic uptake by is-situ iron precipitation, surface complexation modeling parameters and parameter optimization and selection. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Nguyen, T. V.; Vigneswaran, S.; Ngo, H. H.; Pokhrel, D.; Viraraghavan, T. Specific treatment technologies for removing arsenic from water. Engineering in Life Sciences 2006, 6 (1), 86–90. (2) Saitua, H.; Campderros, M.; Cerutti, S.; Padilla, A. P. Effect of operating conditions in removal of arsenic from water by nanofiltration membrane. Desalination 2005, 172 (2), 173–180. (3) Brandhuber, P.; Amy, G. Alternative methods for membrane filtration of arsenic from drinking water. Desalination 1998, 117 (1–3), 1–10. (4) Waypa, J. J.; Elimelech, M.; Hering, J. G. Arsenic removal by RO and NF membranes. Journal American Water Works Association 1997, 89 (10), 102–114. (5) Vrijenhoek, E. M.; Waypa, J. J. Arsenic removal from drinking water by a “loose” nanofiltration membrane. Desalination 2000, 130 (3), 265–277. (6) Seidel, A.; Waypa, J. J.; Elimelech, M. Role of charge (Donnan) exclusion in removal of arsenic from water by a negatively charged porous nanofiltration membrane. Environ. Eng. Sci. 2001, 18 (2), 105–113. (7) Urase, T.; Oh, J.; Yamamoto, K. Effect of pH on rejection of different species of arsenic by nanofiltration. Desalination 1998, 117 (1–3), 11–18.

(8) Mickley, M. Membrane concentrate disposal: practices and regulation Denver, CO., 2001; USBR Desalination and Water Purification Research and Development Program Report No. 69. (9) Truesdall, J.; Mickley, M.; Hamilton, R. Survey of Membrane Drinking-Water Plant Disposal Methods. Desalination 1995, 102 (1–3), 93–105. (10) Crittenden, J., et al. MWH - Water Treatment Principles and Design; John Wiley and Sons: Hoboken, NJ: 2005. (11) Stumm, W.; Sigg, L.; Sulzberger, B. Chemistry of the solid-water interface: processes at the mineral-water and particle-water interface in natural systems; Wiley: New York: 1992, 428. (12) Dzombak, D.; Morel, F. Surface Complexation Modeling: Hydrous Ferric Oxide; John Wiley and Sons: Hoboken, NJ: 1990. (13) Anderson, M. A.; Tejedortejedor, M. I.; Stanforth, R. R. Influence of Aggregation on the Uptake Kinetics of Phosphate by Goethite. Environ. Sci. Technol. 1985, 19 (7), 632–637. (14) Rose, A. L.; Waite, T. D. Kinetics of hydrolysis and precipitation of ferric iron in seawater. Environ. Sci. Technol. 2003, 37 (17), 3897–3903. (15) Pham, A. N.; Rose, A. L.; Feitz, A. J.; Waite, T. D. Kinetics of Fe(III) precipitation in aqueous solutions at pH 6.0–9.5 and 25 degrees C. Geochim. Cosmochim. Acta 2006, 70 (3), 640–650. (16) Letterman, R. D.; American Water Works Association. Water quality and treatment: a handbook of community water supplies; McGraw-Hill:New York: 1999. (17) Edzwald, J. K.; Upchurch, J. B.; Omelia, C. R. Coagulation in Estuaries. Environ. Sci. Technol. 1974, 8 (1), 58–63. (18) Sato, D.; Kobayashi, M.; Adachi, Y. Capture efficiency and coagulation rate of polystyrene latex particles in a laminar shear flow: Effects of ionic strength and shear rate. Colloids and Surfaces A-Physicochemical and Engineering Aspects 2005, 266 (1–3), 150–154. (19) Chen, C.; Lin, L. Human Carcinogenicity and Atherogenicity Induced by Chronic Exposure to Inorganic Arsenic; In Anonymous Arsenic in the Environment; Part II: Human Health and Ecosystem Effects; John Wiley and Sons: NY: 1994. (20) Smedley, P. L.; Kinniburgh, D. G. A review of the source, behaviour and distribution of arsenic in natural waters. Appl. Geochem. 2002, 17 (5), 517–568. (21) Chen, A. S. C.; Fields, K. A.; Sorg, T. J.; Wang, L. L. Field evaluation of As removal by conventional plants. Journal American Water Works Association 2002, 94 (9), 64–77. (22) Watt, C.; Le, X. C. Arsenic speciation in natural waters. Biogeochemistry of Environmentally Important Trace Elements 2003, 835, 11–32. (23) Anderson, L. C. D.; Bruland, K. W. Biogeochemistry of Arsenic in Natural-Waters - the Importance of Methylated Species. Environ. Sci. Technol. 1991, 25 (3), 420–427. (24) Korte, N. E.; Fernando, Q. A Review of Arsenic(iii) in Groundwater. Critical Reviews in Environmental Control 1991, 21 (1), 1–39. (25) Arienzo, M.; Adamo, P.; Chiarenzelli, J.; Bianco, M. R.; De Martino, A. Retention of arsenic on hydrous ferric oxides generated by electrochemical peroxidation. Chemosphere 2002, 48 (10), 1009–1018. (26) Pierce, M. L.; Moore, C. B. Adsorption of Arsenite and Arsenate on Amorphous Iron Hydroxide. Water Res. 1982, 16 (7), 1247– 1253. (27) Lee, Y.; Um, I. H.; Yoon, J. Arsenic(III) oxidation by iron(VI) (ferrate) and subsequent removal of arsenic(V) by iron(III) coagulation. Environ. Sci. Technol. 2003, 37 (24), 5750–5756. (28) Lytle, D. A.; Sorg, T. J.; Snoeyink, V. L. Optimizing arsenic removal during iron removal: Theoretical and practical considerations. Journal of Water Supply Research and Technology-Aqua 2005, 54 (8), 545–560. (29) Hering, J. G.; Chen, P. Y.; Wilkie, J. A.; Elimelech, M.; Liang, S. Arsenic removal by ferric chloride. Journal American Water Works Association 1996, 88 (4), 155–167. (30) Gulledge, J. H.; OConnor, J. T. Removal of Arsenic (V) from Water by Adsorption on Aluminum and Ferric Hydroxides. Journal American Water Works Association 1973, 65 (8), 548– 552. (31) Hering, J. G.; Chen, P. Y.; Wilkie, J. A.; Elimelech, M. Arsenic removal from drinking water during coagulation. Journal of Environmental Engineering-Asce 1997, 123 (8), 800–807. (32) Wilkie, J. A.; Hering, J. G. Adsorption of arsenic onto hydrous ferric oxide: Effects of adsorbate/adsorbent ratios and cooccurring solutes. Colloids and Surfaces A-Physicochemical and Engineering Aspects 1996, 107, 97–110. VOL. 42, NO. 10, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3801

(33) Scott, K. N.; Green, J. F.; Do, H. D.; McLean, S. J. Arsenic Removal by Coagulation. Journal American Water Works Association 1995, 87 (4), 114–126. (34) Fan, M. H.; Brown, R. C.; Sung, S. W.; Huang, C. P.; Ong, S. K.; van Leeuwen, J. Comparisons of polymeric and conventional coagulants in arsenic(V) removal. Water Environ. Res. 2003, 75 (4), 308–313. (35) Cheng, R. C.; Wang, H. C.; Beuhler, M. D. Enhanced Coagulation for Arsenic Removal. Journal American Water Works Association 1994, 86 (9), 79–90. (36) McNeill, L. S.; Edwards, M. Predicting as removal during metal hydroxide precipitation. Journal American Water Works Association 1997, 89 (1), 75–86. (37) Nenov, V.; Dimitrova, N.; Dobrevsky, I.; Rands, D. G. Effective Precipitation of Arsenic from Aqueous-Solution by Iron(III) Sulfate. Zeitschrift Fur Wasser- Und Abwasser-Forschung-Journal for Water and Wastewater Research-Acta Hydrochimica Et Hydrobiologica 1992, (1), 14–17. (38) Jia, Y. F.; Demopoulos, G. P. Adsorption of arsenate onto ferrihydrite from aqueous solution: Influence of media (sulfate vs nitrate), added gypsum, and pH alteration. Environ. Sci. Technol. 2005, 39 (24), 9523–9527. (39) Sherman, D.; Randall, S. Surface complexation of arsenic(V) to iron(III) (hydro)oxides: Structural mechanism from ab initio molecular geometries and EXAFS spectroscopy. Geochim. Cosmochim. Acta 2003, 67 (22), 4223. (40) Waychunas, G. A.; Rea, B. A.; Fuller, C. C.; Davis, J. A. SurfaceChemistry of Ferrihydrite 0.1. Exafs Studies of the Geometry of Coprecipitated and Adsorbed Arsenate. Geochim. Cosmochim. Acta 1993, 57 (10), 2251–2269. (41) Waychunas, G. A.; Davis, J. A.; Fuller, C. C. Geometry of Sorbed Arsenate on Ferrihydrite and Crystalline FeOOH - Reevaluation of Exafs Results and Topological Factors in Predicting Sorbate Geometry, and Evidence for Monodentate Complexes. Geochim. Cosmochim. Acta 1995, 59 (17), 3655–3661. (42) Aggett, J.; Roberts, L. S. Insight into the Mechanism of Accumulation of Arsenate and Phosphate in Hydro Lake-Sediments by Measuring the Rate of Dissolution with Ethylenediaminetetraacetic Acid. Environ. Sci. Technol. 1986, 20 (2), 183–186. (43) Fuller, C. C.; Davis, J. A.; Waychunas, G. A. Surface-Chemistry of Ferrihydrite 0.2. Kinetics of Arsenate Adsorption and Coprecipitation. Geochim. Cosmochim. Acta 1993, 57 (10), 2271– 2282. (44) Clifford, D. Ion Exchange and Inorganic Adsorption; In Anonymous AWWA Water Quality and Treatment; McGraw-Hill:NY: 1999.

3802

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 10, 2008

(45) MacPhee, M. J.; Charles, G. E.; Cornwell, D. A. Treatment of Arsenic Residuals from Drinking Water Removal Processes 2001; EPA 600/R-01/033. (46) Fitzpatrick, A. J. Adsorption of Arsenate and Phosphate on Gibbsite from Artificial Seawater. 1998. (47) Gupta, S. K.; Chen, K. Y. Arsenic Removal by Adsorption. Journal Water Pollution Control Federation 1978, 50 (3), 493–506. (48) Appelo, C. A. J.; Van der Weiden, M. J. J.; Tournassat, C.; Charlet, L. Surface complexation of ferrous iron and carbonate on ferrihydrite and the mobilization of arsenic. Environ. Sci. Technol. 2002, 36 (14), 3096–3103. (49) Holm, T. R. Effects of CO32-/bicarbonate, Si, and PO43- on arsenic sorption to HFO. Journal American Water Works Association 2002, 94 (4), 174–181. (50) Arai, Y.; Sparks, D. L.; Davis, J. A. Effects of dissolved carbonate on arsenate adsorption and surface speciation at the hematitewater interface. Environ. Sci. Technol. 2004, 38 (3), 817–824. (51) Meng, X. G.; Bang, S.; Korfiatis, G. P. Effects of silicate, sulfate, and carbonate on arsenic removal by ferric chloride. Water Res. 2000, 34 (4), 1255–1261. (52) Roberts, L. C.; Hug, S. J.; Ruettimann, T.; Billah, M.; Khan, A. W.; Rahman, M. T. Arsenic removal with iron(II) and iron(III) waters with high silicate and phosphate concentrations. Environ. Sci. Technol. 2004, 38 (1), 307–315. (53) Tadanier, C. J.; Eick, M. J. Formulating the charge-distribution multisite surface complexation model using FITEQL. Soil Sci. Soc. Am. J. 2002, 66 (5), 1505–1517. (54) Dzombak, D. A.; Morel, F. M. M. Adsorption of Inorganic Pollutants in Aquatic Systems. Journal of Hydraulic EngineeringAsce 1987, 113 (4), 430–475. (55) Parkhurst, D.; Appelo, C. User’s Guide to PHREEQC (Version 2) A Computer Program for Speciation, Batch-Reaction, OneDimensional Transport, and Inverse Geochemical Calculations; U.S. Dept. of the Interior, U.S. Geological Survey: Denver, CO: 1999. (56) Jia, Y. F.; Xu, L. Y.; Fang, Z.; Demopoulos, G. P. Observation of surface precipitation of arsenate on ferrihydrite. Environ. Sci. Technol. 2006, 40 (10), 3248–3253. (57) Langmuir, D.; Mahoney, J.; Rowson, J. Solubility products of amorphous ferric arsenate and crystalline scorodite (FeAsO4: 2H2O) and their application to arsenic behavior in buried mine tailings. Geochim. Cosmochim. Acta 2006, 70 (12), 2942–2956. (58) Lumsdon, D. G.; Meeussen, J. C. L.; Paterson, E.; Garden, L. M.; Anderson, P. Use of solid phase characterization and chemical modeling for assessing the behavior of arsenic in contaminated soils. Appl. Geochem. 2001, 16 (6), 571–581.

ES702946S