Sorption Mechanisms of Arsenate during Coprecipitation with

Dec 16, 2009 - were compiled using the diffused layer model (DLM). The. DLM is one of the most general and simple surface com- plexation models and ca...
0 downloads 0 Views 601KB Size
Download PDF
Environ. Sci. Technol. 2010, 44, 638–643

Sorption Mechanisms of Arsenate during Coprecipitation with Ferrihydrite in Aqueous Solution CHIHARU TOKORO,* YOHEI YATSUGI, HAJIME KOGA, AND SHUJI OWADA Department of Creative Science and Engineering, Faculty of Science and Engineering, Waseda University, Tokyo 169-8555, Japan

Received January 12, 2009. Revised manuscript received November 24, 2009. Accepted December 2, 2009.

Dilute arsenate (As(V)) coprecipitation by ferrihydrite was investigated to determine if treatment of acid mine drainage containing dilute As(V) using coprecipitation is feasible. The sorption density obtained at pH 5 and 7 was nearly identical when As(V) was coprecipitated with ferrihydrite, while it was higher at pH 5 when As(V) was adsorbed on the ferrihydrite. The high sorption density of As(V) to ferrihydrite in coprecipitation with 1-h reaction time suggested that coprecipitation occurs via both adsorption and precipitation. Furthermore, the relationship between residual As(V) and sorption density revealed a BETtype isotherm, with a transition point from a low residual As(V) concentration to a high residual As(V) concentration being observed for all initial As(V) concentrations between 0.15 and 0.44 mmol/dm3 when the initial molar ratio was 0.56 at pH 5 and 7. X-ray diffraction and the ζ potential revealed that the transition point from surface complexation to precipitation was obtained when the initial As/Fe ratio was 0.4 or 0.5. When dilute As(V) was coprecipitated with ferrihydrite at pH 5 and 7, it was primarily adsorbed as a surface complex when the initial molar ratio was As/Fe < 0.4, while a ferric arsenate and surface complex was formed when this ratio was g 0.4.

Introduction Arsenic is toxic to animals, including humans, and longterm exposure to arsenic via drinking-water causes cancers of the skin, lungs, urinary bladder, and kidney, as well as other effects on the skin such as changes in pigmentation and thickness (1). Elevated levels of arsenic are often present in the environment as a result of the weathering and dissolution of minerals as well as in response to numerous anthropogenic sources, including the discharge of mine wastes and coal fly ash and the use of pesticides that contain arsenic (2, 3). Indeed, the potential for arsenic exposure affects many people, as evidenced by several million people in Bangladesh and India being at risk of exposure via drinking water obtained from contaminated groundwater sources (3, 4). Acid mine drainage (AMD) containing arsenic is a global environmental problem (5, 6) that also affects Japan, which has many abandoned or closed mines that have been generating AMD for the last few decades (7-9). Several such facilities in Japan produce AMD that contains dilute concentrations of arsenic that exceed Japanese effluent standards * Corresponding author phone: +81(3)5286-3320; fax: +81(3)52863491; e-mail: [email protected]. 638

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 2, 2010

(0.1 mg/dm3) (10). For example, AMD from the abandoned Horobetsu sulfur mine, which is the largest AMD source in Japan, contains approximately 10 mg/dm3 arsenic. The conventional method for treating AMD is the addition of a source of alkalinity to increase the pH, which results in the precipitation of heavy metal pollutants (11). Arsenate (As(V)) or arsenite (As(III)), which are the predominant inorganic arsenic species found in natural aquatic systems (12), also coprecipitate with the hydroxides of heavy metals such as ferrihydrite or aluminum hydroxide (13). Once a mine begins to generate AMD, it generally must be treated indefinitely. However, the disposal of sludge following AMD treatment is also becoming an environmental problem due to the potential for toxic substances in the sludge to be redissolved. As a result, it is difficult to develop sustainable methods for the treatment of AMD. This work was conducted to investigate the mechanisms by which dilute As(V) coprecipitates with ferrihydrite in aqueous solutions. Many studies that have involved multiple experimental methodologies and techniques have been conducted to evaluate the interaction between anions and ferrihydrite. In addition, several studies have investigated the bonding mechanism responsible for the adsorption of anions onto ferrihydrite, goethite, or lepidocrocite using spectroscopic methods (14-17). However, few studies have been conducted to evaluate the mechanisms responsible for coprecipitation, in which the precipitation of ferrihydrite and adsorption of As(V) simultaneously develop. Coprecipitation reactions depend on the conditions of the solutions, (e.g., pH, ionic strength, the presence of coexisting ions) and the surface conditions (e.g., crystal states or the forms of minerals). However, to date, the role that these conditions play in the coprecipitation of anions and ferrihydrite has not been quantitatively assessed because the coprecipitation mechanisms have not been well described. Coprecipitation occurs as a result of a combination of different mechanisms including adsorption, precipitation, storage, or solid solution. In the present work, the coprecipitation mechanisms were divided into two categories: surface complexation and surface precipitation. Surface complexation was considered to be the two-dimensional adsorption of As(V) onto the surface of ferrihydrite, whereas surface precipitation was used to describe the threedimensional uptake of As(V) into ferrihydrite. Surface precipitation involves ternary adsorption, storage or solid solution that occurs beyond the thermodynamic bulk precipitation range. Recent studies have suggested that the interaction between arsenate or phosphate and goethite in aqueous solution includes surface precipitation as well as surface complexation under certain conditions (18-20). Moreover, Violante et al. (21-23) investigated the mechanism of As(V)-Fe, As(V)-Al, and As(V)-Fe-Al coprecipitation after 30 or 210 days of aging at 50 °C. The present study was conducted to provide a fundamental assessment of the surface precipitation that occurs during As(V) coprecipitation with ferrihydrite without any aging or heating. When conducting such an assessment, it is important to identify the transition point at which surface complexation changes to surface precipitation because the efficiency of the removal of As(V) from wastewater or the redissolution of As(V) from sludge differs for these mechanisms. The objective of this study was to identify the transition point of As(V) adsorption and surface precipitation in aqueous systems in which the initial As/Fe molar ratios were varied from 0.13 to 11 at pH 5 or 7. To accomplish this, we conducted batch experiments of coprecipitation or adsorption with 10.1021/es902284c

 2010 American Chemical Society

Published on Web 12/16/2009

ferrihydrite. The reaction time was fixed at one hour, which is the time commonly used to treat wastewater that contains AMD. We then compared the results and found that an isotherm formed during coprecipitation when the initial As(V) concentration was fixed at 0.15, 0.30, or 0.45 mmol/dm3. A concentration of As(V) of 0.15 mmol/dm3 corresponds to 10 mg/dm3 arsenic, which represented the total arsenic concentration in the AMD from the abandoned Horobetsu sulfur mine. Moreover, we measured the ζ potential of As(V) coprecipitated ferrihydrite formed when the initial Fe(III) concentration was fixed at 0.18 mmol/dm3 (10 mg/dm3 ferric) and the initial As(V) concentration was varied from 0.022 to 0.90 mmol/dm3. We also evaluated the ferrihydrite coprecipitated under these conditions by X-ray diffraction. It should be noted that nitric acid was used as the electrolyte in this work instead of sulfuric acid, which is often present in AMD. Nitric acid was used because the objective of this study was to assess the mechanism by which coprecipitation of As(V) with ferrihydrite occurs without considering the effects of any other anions. Jia et al. (19) proposed that the difference in the effect of sulfuric and nitric acid on the As(V) coverage density was negligible during its adsorption to goethite, but that the sulfate ions incorporated or adsorbed into ferrihydrite were more easily displaced by As(V) as pH increased.

Materials and Methods Standards and Reagents. All chemicals and solutions used in this study were of analytical grade and were purchased from Kanto Chemicals, Inc., Japan. The As(V) and Fe(III) solutions were prepared from Na2HAsO4 · 7H2O and Fe(NO3)3 · 9H2O, respectively. For all experiments, the pH and ionic strength were adjusted by the addition of 0.05 M HNO3, KOH, and KNO3. Specifically, the pH was fixed at 5 or 7 and the ionic strength was fixed at 0.05. All experiments were conducted at 25 °C. In addition, all experiments were conducted at least in triplicate and the error was confirmed to be within 1%. Coprecipitation Experiments. The coprecipitation experiments involved the formation of ferrihydrite particles in the presence of As(V). To accomplish this, the Fe(NO3)3 · 9H2O and As(V) solutions were initially combined in 0.5 dm3 of deionized (DI) water to adjust the initial As(V)/Fe(III) molar ratio to the target level, after which the pH and ionic strength were adjusted to the target levels. The suspension was then agitated using a magnetic stirrer under pH control for one hour, which was accomplished by adding a few drops of KOH by hand. One hour was used because it is known to be the reaction time required to treat wastewater that contains AMD and sufficient for the initial rapid coprecipitation reaction between As(V) and ferrihydrite to be completed, whereas only slow and small changes in pH occur after one hour. The suspension was then centrifuged for 20 min at 5000 rpm and filtered through a 3 kDa membrane filter. Next, the As(V) concentration in the supernatant solution was analyzed by ICP-AES using a SPS-4000 atomic emission spectrometer (Seiko Instruments, Japan) with a Hydride Generator Accessory (HYD-10, Seiko Instruments, Japan). For comparison with the adsorption experimental results or preparation of the coprecipitation isotherm, the initial As(V) concentration was fixed at 0.15, 0.30, or 0.45 mmol/ dm3 and the initial Fe(III) concentration was varied to adjust the initial As/Fe molar ratio from 0.28-11. Conversely, for measurement of the ζ potential and XRD analysis, the initial Fe(III) concentration was fixed at 0.18 mmol/dm3 and the initial As(V) concentration was varied to adjust the initial As/Fe molar ratio from 0.13-5. In addition, the filter residue of ferrihydrite with no As(V) was freeze-dried at -45 °C for 24 h and then ground using an agate mortar, after which the surface area was determined

to be 276 m2/g by BET-N2 adsorption analysis (Accelerated Surface Area and Porosimetry System, Micromeritics, USA). We also measured the surface area of As(V) coprecipitated ferrihydrite and found it to be 276 ( 10 m2/g. Violante et al. (22) showed that As(V) coprecipitated ferrihydrite obtained at different pH values and As/Fe molar ratios and aged 30 or 120 days at 50 °C had different surface areas, whereas the surface area of samples that had not been aged did not vary when different pH values or initial As/Fe molar ratios were used. However, it is important to note that the actual specific surface area of ferrihydrite should be larger than the value determined by BET-N2 adsorption analysis due to the porous structure of ferrihydrite (24). Adsorption Experiments. In this work, simple adsorption experiments using prepared ferrihydrite were conducted, and the results were then compared with those of the coprecipitation experiments. In the adsorption experiments, the initial As(V) concentration was fixed at 0.15 mmol/dm3 and the initial Fe(III) concentration was varied to adjust the initial As/Fe molar ratio to 0.28-11. The adsorption experiments involved the formation of ferrihydrite particles and the separate adsorption of As(V). To accomplish this, the Fe(NO3)3 · 9H2O and As(V) solutions were initially prepared separately in 0.5 dm3 of deionized (DI) water to give twice the target concentration of Fe(III) and As(V). The pH and ionic strength were then adjusted to the target levels by the addition of HNO3 and KOH. Next, equal amounts of the Fe(III) solution and As(V) solution were combined and agitated using a magnetic stirrer under pH control for one hour, which was accomplished by adding a few drops of KOH. The solid/ liquid separation of the solutions and analysis of the concentration of As(V) in the supernatant were then conducted as described in the coprecipitation experiment section. XRD Analysis. The powder XRD patterns were obtained using a copper target (Cu KR), a crystal graphite monochromator, and a scintillation detector. The equipment was operated at 40 kV and 30 mA by step-scanning from 2° to 80° 2θ at increments of 0.02° 2θ and a scan speed of 2°/min. A crystal sample holder was used and the diffractograms were not corrected by background diffraction. The filter residue was freeze-dried at -45 °C for 24 h prior to analysis. Poorly crystalline ferric arsenate was synthesized as a reference material to be compared with the ferrihydrite coprecipitated with As(V) using the procedure described by Jia et al. (18), with slight modification. Briefly, a mixture of 0.02 M As(V) and 0.02 M Fe(III) was adjusted from an initial pH of 1.3 to pH 1.8 using KOH solution and then maintained at that pH for 1 h. Next, the solid product was separated by filtration, washed with HNO3-acidified water (pH 2), and freeze-dried at -45 °C. ζ Potential Measurements. All samples subjected to ζ potential measurements were prepared in a glovebox that was purged with N2 gas using N2 purged DI water to exclude the effects of CO2. Measurement of the ζ potential was conducted using an electrophoresis light scattering spectrophotometer (ELS-8000, Otsuka Electronics, Japan). To accomplish this, prior to solid/liquid separation, suspensions of the products of the coprecipitation experiments were dispersed in an ultrasonic bath for 5 min and then rapidly analyzed by spectrophotometry.

Results and Discussion Comparison of Coprecipitation and Adsorption Characteristics. Figures 1 and 2 show a comparison of the removal ratios of As(V) in the coprecipitation and adsorption experiments conducted at pH 5 and 7. In these experiments, the initial concentration of As(V) was 0.15 mmol/dm3 (10 mg/ dm3 arsenic), which is approximately the same concentration of As(V) in AMD from the abandoned Horobetsu sulfur mine. VOL. 44, NO. 2, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

639

FIGURE 1. Comparison of the coprecipitation and adsorption experimental results at pH 5. The initial As(V) concentration was 0.15 mmol/dm3 and the ionic strength was fixed at 0.05 using KNO3. The solid line was calculated using Dzombak and Morel’s model (21). The data label is the initial As/Fe molar ratio from 0.28 to 11.

FIGURE 2. Comparison of the coprecipitation and adsorption experimental results at pH 7. The initial As(V) concentration was 0.15 and the ionic strength was fixed at 0.05 using KNO3. The solid line was calculated from Dzombak and Morel’s model (21). The data label is the initial As/Fe molar ratio from 0.28 to 11. The As(V) in a solution containing an initial concentration of As(V) of 0.15 mmol/dm3 could be reduced to less than the regulated concentration of Japanese effluent standard (0.1 mg/dm3) by the addition of solution containing 0.45 mmol/ dm3 (30 mg/dm3) Fe(III) at pH 5 or 7. In addition, more As(V) was removed during the coprecipitation experiment than the adsorption experiment, regardless of the pH. The higher removal that was observed in the coprecipitation experiments when compared with the simple adsorption experiments indicates that the coprecipitation of As(V) with ferrihydrite involves more than adsorption. Moreover, in the adsorption experiments, the As(V) removal by a given dosage of Fe(III) was larger at pH 5 than at pH 7. One reason for this was that the surface of ferrihydrite has a stronger positive charge at pH 5 than at pH 7. However, there was little difference in the As(V) removal at any given dosage of Fe (III) at pH 7 and pH 5 in the coprecipitation experiments. Dzombak and Morel have compiled a set of thermodynamic data to describe the adsorption of several ions, including As(V), onto hydro ferric oxide (HFO). Their data were compiled using the diffused layer model (DLM). The DLM is one of the most general and simple surface complexation models and can be used to estimate surface complexation within a set of quasi thermodynamic constants (24). In addition, it is well-known that Dzombak and Morel’s model can be used to describe the adsorption of As(V) to HFO (25). The solid lines in Figures 1 and 2 show the curves 640

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 2, 2010

FIGURE 3. Relationship between the sorption density and the residual As(V) concentration at pH 5 during As(V) coprecipitation with ferrihydrite. The initial concentration of As(V) was fixed at 0.15, 0.30, or 0.45 mmol/dm3 and the Fe(III) dosage was adjusted to attain the target As/Fe molar ratio of 0.28-11. The ionic strength was fixed at 0.05 using KNO3. The data label is the initial As/Fe molar ratio.

FIGURE 4. Relationship between the sorption density and the residual As(V) concentration at pH 7 during As(V) coprecipitation with ferrihydrite. The initial concentration of As(V) was fixed at 0.15, 0.30, or 0.45 mmol/dm3 and the Fe(III) dosage was adjusted to attain the target As/Fe molar ratio of 0.28-11. The ionic strength was fixed at 0.05 using KNO3. The data label is the initial As/Fe molar ratio. generated using the DLM with Dzombak and Morel’s parameters. Our adsorption experimental results agreed well with the results calculated at pH 5 and 7. We previously confirmed that there was little difference between the removal of Cr(VI) by coprecipitation and adsorption using ferrihydrite (26), which suggests that the coprecipitation of Cr(VI) with ferrihydrite occurs almost entirely via surface complexation. However, in the present study, the removal of As(V) by coprecipitation with ferrihydrite was greater than the removal by adsorption, which indicates that factors in addition to surface complexation are involved in coprecipitation. These findings likely occurred because As(V) has a higher affinity for ferrihydrite than Cr(VI). In the As(V) coprecipitation experiments, As(V) adsorption and ferrihydrite precipitation occurred simultaneously and As(V) was removed by precipitation with ferrihydrite in addition to adsorption. In other words, a surface precipitation was formed because of the high affinity of freshly precipitated ferrihydrite for As(V). Relationship between Sorption Density and Residual As(V) Concentration. Figures 3 and 4 show the sorption density of the As(V) coprecipitation experiments for ferrihydrite obtained at pH 5 and 7. In these experiments, the initial concentration of As(V) was fixed at 0.15, 0.30, or 0.45 mmol/dm3, while the Fe(III) concentrations were changed

from 0.042 to 5.0 mmol/dm3 to give an initial target As/Fe molar ratio of 0.28 to 11. In the treatment of wastewater, especially AMD containing Fe(II), it is possible for this high As(V)/Fe(III) molar ratio to be present if the Fe(II) is oxidized slowly to Fe(III) while in the presence of As(V). As(V) coprecipitation with ferrihydrite with a 1-h reaction time produced a BET-type isotherm for all initial concentrations of As(V) evaluated in this study. We previously found that Cr(VI) coprecipitation with ferrihydrite, which occurred almost entirely as a result of surface complexation, produced a Langmuir-type isotherm and was not affected by the initial concentration of Cr(VI) (26). In addition, it has often been proposed that As(V) adsorption to ferrihydrite or goethite produces a Langmuir or Freundlich-type isotherm (27). The BET-type isotherm of As(V) observed in this study indicates that the mechanism by which coprecipitation of As(V) with ferrihydrite occurs when there is a 1-h reaction time involves more than simple adsorption onto the surface of the ferrihydrite. Specifically, these findings indicate that some three-dimensional uptake, such as precipitation or the formation of several layers of absorbate, occurs. The density of the As(V) sorption to ferrihydrite showed a steep increase when there was a low As(V) concentration in solution under 0.003 mmol/dm3. This steep increase was followed by a gradual increase as the residual As(V) concentration increased, which was followed by a second steep increase when there was a sufficiently high As(V) concentration in solution. In this study, a low As(V) concentration represents a high concentration of Fe(III) or a low molar ratio of As/Fe. This is because the As(V) sorption density was obtained by changing the Fe(III) dosage while maintaining a constant initial concentration of As(V). As shown in Figures 3 and 4, the gradual increase was obtained at an absorption density of around 0.8 mol-As/mol-Fe for pH 5 and 7, regardless of the initial As(V) concentration. The transition from a steep increase to a gradual increase occurred when the molar ratio of As/Fe was 0.56, while the transition from the gradual increase to the second steep increase occurred when the molar ratio of As/Fe was approximately 5.6, regardless of the pH and initial As(V) concentration. XRD Measurements. Figures 5 and 6 show a comparison of the XRD patterns of the coprecipitated As(V) with ferrihydrite as a function of the initial molar ratio of As/Fe at pH 5 and 7. In these experiments, the initial concentration of Fe(III) was fixed at 0.18 mmol/dm3 and the concentration of As(V) was changed to attain the target initial As/Fe molar ratio of 0.13-5. Products produced by As(V)-ferrihydrite coprecipitation or adsorption are poorly crystalline, which makes it difficult to characterize their arsenate phases using XRD. However, Jia et al. (18, 19) proposed that surface precipitation could be detected based on peak shifts in the XRD spectra from ferrihydrite to poorly crystalline ferric arsenate. As shown in Figures 5 and 6, the poorly crystalline ferric arsenate showed two broad XRD bands, one at 28° 2θ (d ≈ 3.2 Å) and a second small peak at 58° 2θ (d ≈ 1.6 Å). However, the XRD spectrum obtained for As/Fe ) 0 (no As(V)) showed two broad bands at 34° 2θ (d ≈ 2.6 Å) and 61° 2θ (d ≈ 1.5 Å), which were the same positions as observed for the two-line ferrihydrite. These differences in peak positions between the poorly crystalline ferric arsenate and two-line ferrihydrite can be used to distinguish between precipitation of ferric arsenate and As(V) adsorbed ferrihydrite in reactions produced using different initial As/Fe molar ratios. As shown in Figures 5 and 6, when the coprecipitated products were generated using an initial As/Fe ratio of 0.13 or 0.25, the XRD patterns were almost identical to those of ferrihydrite, but the two peaks corresponding to ferrihydrite became broader and weaker at pH 5 and 7. This could have occurred because the presence of an increasing As(V)

FIGURE 5. Comparison of the XRD patterns of As(V) coprecipitated ferrihydrite formed at pH 5 with those of the reference materials (ferrihydrite, poorly crystalline ferric arsenate). The initial concentration of Fe(III) was fixed at 0.18 mmol/dm3 and the As(V) dosage was adjusted to attain the target As/Fe molar ratio of 0.13 to 5.

FIGURE 6. Comparison of the XRD patterns of As(V) coprecipitated ferrihydrite formed at pH 7 with those of the reference materials (ferrihydrite, poorly crystalline ferric arsenate). The initial concentration of Fe(III) was fixed at 0.18 mmol/dm3 and the As(V) dosage was adjusted to attain the target As/Fe molar ratio of 0.13 to 5. concentration promoted the formation of ferrihydrite with lower crystallinity and finely reduced particles. In these cases, the mechanism of coprecipitation of As(V) is predominantly adsorption to the surface of ferrihydrite. We also evaluated the XRD spectrum of As(V) adsorbed ferrihydrite at different pH values and initial As/Fe molar ratios and found that they were all the same as the spectrum of ferrihydrite, whereas VOL. 44, NO. 2, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

641

the two peaks corresponding to ferrihydrite became broader and weaker as the initial As/Fe molar ratio increased. Conversely, when the initial As/Fe ratio was 0.5, the first peak of ferrihydrite became broader and shifted toward that of poorly crystalline ferric arsenate, whereas the XRD peaks formed when the initial As/Fe ratio was >0.5 were located at the same positions as those of poorly crystalline ferric arsenate. In these cases, the mechanism of coprecipitation of As(V) is predominantly precipitation of ferric arsenate, which would involve ternary adsorption, storage, or solid solution that occurs beyond the thermodynamic bulk precipitation range. The broader and weaker peaks in the XRD spectra obtained during the coprecipitation experiments indicate that the precipitates are a mixture of poorly crystalline ferric arsenate and As(V) adsorbed ferrihydrite. However, the results of these XRD measurements revealed that when the initial As/Fe ratio was 0.4, the sorption isotherm was changing from a steep increase to a gradual increase, which indicated that the mechanism of coprecipitation of As(V) was changing from adsorption to precipitation at this point. Jia et al. (18) proposed that surface precipitation of ferric arsenate was present on 10 g/dm3 As(V) adsorbed to ferrihydrite that was aged for two weeks at pH 3-5. However, they confirmed that no ferric arsenate was produced at pH 8. Despite this finding, in the present study, dilute As(V) (e.g., an initial As/Fe molar ratio of 0.4 corresponding to 0.072 mmol/dm3 (5.4 mg/dm3) As(V)) that coprecipitated with ferrihydrite showed precipitation of ferric arsenate at pH 5 and 7 without aging. The differences in the results of these studies may have occurred because the reaction in our experiments did not involve simple adsorption onto prepared ferrihydrite, but instead involved coprecipitation in which the precipitation of ferrihydrite and adsorption of As(V) occurred simultaneously. Because the pH was adjusted from acidic pH to pH 7 via pH 5 and the reaction time was fixed at one hour, the coprecipitation reaction would not reach an equilibrium condition and transient uptake of As(V) to fresh ferrihydrite would not occur in our coprecipitation experiments. ζ Potential Measurements. Li and Stanforth (20) proposed that adsorption and surface precipitation of phosphate with goethite can be distinguished by the relationship between ζ potential and adsorption density. Specifically, they stated that surface precipitation resulted in a smaller negative surface charge than simple adsorption. Therefore, to evaluate the mechanism by which As(V) coprecipitates with ferrihydrite, we evaluated the relationship between the ζ potential of As(V) that coprecipitated ferrihydrite and the As(V) sorption density of ferrihydrite under constant pH. Figure 7 shows the ζ potential of As(V) coprecipitated ferrihydrite as a function of the sorption density of As(V) to ferrihydrite at pH 5 and 7. In these experiments, the initial concentration of Fe(III) was fixed at 0.18 mmol/dm3 and the As(V) concentration was changed to attain the target initial As/Fe molar ratio of 0.13-5. A high sorption density represents a high concentration of As(V), which results in a high initial molar ratio of As/Fe. In general, a linear relationship between the ζ potential and the As(V) sorption density would be expected for simple adsorption at a given surface pH (20). We also measured the ζ potential of As(V) adsorbed ferrihydrite and confirmed the linear relationship between the ζ potential and the As(V) sorption density without any breaking point. However, the curve describing the relationship between ζ potential and As(V) sorption in the present study decreased rapidly when the sorption density was low, but changed much less rapidly at higher sorption densities, regardless of the pH. This break in the slope of the curves between the ζ potential and sorption 642

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 2, 2010

FIGURE 7. Relationship between the ζ potential and the As(V) sorption density at pH 5 and 7 during As(V) coprecipitation with ferrihydrite. The initial concentration of Fe(III) was fixed at 0.18 mmol/dm3 and the As(V) dosage was adjusted to attain the target As/Fe molar ratio of 0.13 to 2. The ionic strength was fixed at 0.05 using KNO3. The data label is the initial As/Fe molar ratio. density occurred when the initial molar ratio of As/Fe was 0.3 or 0.4 at pH 5 and 7. These findings suggest that the dominant mechanism in the surface by which As(V) is coprecipitated was simple adsorption when a low initial As/ Fe molar ratio was used and surface precipitation when a high initial As/Fe molar ratio was used. It should be noted that the slope of the curve describing the ζ potential versus the sorption density did not vary with pH at the lower and higher As/Fe molar ratio, as shown in Figure 7. These results suggest that the speciation of As(V) adsorbed onto the surface ferrihydrite is almost the same at both pHs. Jia et al. (19) proposed that As(V) was adsorbed as a bidentate surface complex of both protonated tFeO2As(O)(OH)- and unprotonated tFeO2As(O)22- at pH 7, but that ferric arsenate and the aforementioned surface complex presented on ferrihydrite at pH 5. In the present study, since the reaction time was fixed at one hour and As(V) coprecipitation did not reach an equilibrium condition, we confirmed the formation of ferric arsenate at both pH 5 and 7 when a higher initial molar ratio of As/Fe was used. Thermodynamically, unsaturated conditions were used for the precipitation of ferric arsenate in the bulk solution phase in the present experiment (28). Jia et al. (19) also detected surface precipitation when As(V) adsorbed to ferrihydrite with aging under unsaturated conditions and concluded that the initial uptake of As(V) to ferrihydrite by surface complexation was followed by a transition to ferric arsenate formation on the surface of ferrihydrite. They further proposed that the process of surface precipitation of As(V) adsorbed to ferrihydrite involved the slow dissolution of ferrihydrite, as well as the ternary complexation of Fe3+ and subsequent precipitation of As(V). In the present work, the ferrihydrite did not necessarily need to be dissolved because the precipitation of ferrihydrite and adsorption of As(V) developed simultaneously. As shown in Figures 1 and 2, more As(V) uptake to ferrihydrite occurred at pH 7 than at pH 5, whereas more surface complexation occurred at pH 5 than at pH 7. This may indicate that more Fe3+ could be adsorbed at pH 7 than at pH 5 because of the formation of a negative charge on the surface at pH 7 (Figure 7). In addition, more coprecipitation of As(V) at pH 7 may have occurred because the coprecipitation reaction could not reach equilibrium and the transient uptake of As(V) would not be desorbed from the coprecipitated ferrihydrite since the pH was adjusted from acidic pH to pH 7 via pH 5 and the reaction time was fixed at one hour in our experiments. It should be noted that the initial molar ratio of As/Fe at the shifting point of the isotherm shown in Figures 3 and 4, as well as the XRD patterns shown in Figures 5 and 6 and

the changing point of the ζ potential slope was 0.4 or 0.5. These results demonstrate that surface complexation and surface precipitation can be distinguished based on the shape of isotherms formed by the coprecipitation of As (V) and ferrihydrite. In this work, we discussed the mechanism of dilute As(V) coprecipitation by ferrihydrite using a 1-h reaction time for a wide range of initial As/Fe molar ratios up to 11. We confirmed that As(V) coprecipitation by ferrihydrite achieved an efficient As(V) immobilization when high initial As/Fe molar ratios (As/Fe > 0.4) were used. In general, the total As concentration do not exceed the total Fe concentration in environmental systems. However, especially in AMD, As(V) often coexists with a small amount of Fe(III) and a high As/ Fe molar ratio is realized because As(III) and Fe(II) are the predominant species in pretreatment AMD and Fe(II) is oxidized to Fe(III) at a much slower rate than As(III) is oxidized to As(V) because several minerals or bacteria in pretreatment AMD accretes As(III) oxidation (7, 29-32). We believe that this slow oxidation process of Fe(II) was favorable for efficient As(V) immobilization by ferrihydrite.

(14)

(15)

(16)

(17)

(18)

(19)

Acknowledgments This study was partially supported by the Japanese Ministry of Education, Culture, Sports, Science and Technology, Grantin-Aid for Young Scientists (B), A08205300, 2008, and by a Waseda University Grant for Special Research Projects, 2008.

(20)

(21)

Literature Cited (1) Ghosha, P.; Banerjeea, M.; Giri, A. K.; Raya, K. Toxicogenomics of arsenic: Classical ideas and recent advances. Mutat. Res. 2008, 659, 293–301. (2) Nriagu, J. O. Arsenic in the Environment: Part I, Advances in Environmental Science and Technology 26; John Wiley & Sons: New York, 1994. (3) Singh, N.; Kumar, D.; Sahu, A. P. Arsenic in the environment: Effects on human health and possible prevention. J. Environ. Biol. 2007, 28, 359–365. (4) Ahmed, F. A. An overview of arsenic removal technologies in Bangladesh and India. Proceedings of BUET-UNU International Workshop on Technologies for Arsenic Removal from Drinking Water; Dhaka, Bangladesh, May 5-7, 2001; pp 251-269. (5) Dold, B. Sustainability in metal mining: from exploration, over processing to mine waste management. Rev. Environ. Sci. Biotechnol. 2008, 7, 275–285. (6) Liang, H. C.; Thomson, B. M. Minerals and mine drainage. Water Environ. Res. 2008, 80, 1481–1509. (7) Masuda, N.; Hashimoto, K.; Asano, H.; Matsushima, E.; Yamaguchi, S. Test results of a newly proposed neutralization process to reduce and utilize the sludge. Miner. Eng. 2008, 21, 310–316. (8) Herrera, P.; Uchiyama, H.; Igarashi, T.; Asakura, K.; Ochi, Y.; Ishizuka, F.; Kawada, S. Acid mine drainage treatment through a two-step neutralization ferrite-formation process in northern Japan: Physical and chemical characterization of the sludge. Miner. Eng. 2007, 20, 1309–1314. (9) Herrera, P.; Uchiyama, H.; Igarashi, T.; Asakura, K.; Ochi, Y. Treatment of acid mine drainage through a ferrite formation process in central Hokkaido, Japan: Evaluation of dissolved silica and aluminium interference in ferrite formation. Miner. Eng. 2007, 20, 1255–1260. (10) Japanese Ministry of Economy, Trade and Industry; Kougai Boushi no Gijutsu to Houki, Suishitu (in Japanese); Maruzen: Tokyo, 2006, pp 593-926. (11) Kalin, M.; Fyson, A.; Wheeler, W. N. The chemistry of conventional and alternative treatment systems for the neutralization of acid mine drainage. Sci. Total Environ. 2006, 366, 395–408. (12) Sadiq, M. Arsenic chemistry in soils: An overview of thermodynamic predictions and field observations. Water Air Soil Pollut. 1997, 93, 117–136. (13) Tokoro, C.; Shimizu, T.; Hirai, K.; Badulis, G. C.; Sasaki, H. As(V) removal by Fe(III), Al or Pb salts and rapid solid/liquid separation

(22)

(23)

(24) (25)

(26)

(27)

(28) (29)

(30)

(31)

(32)

in wastewater containing dilute arsenic: A fundamental study for efficient treatment of wastewater containing dilute arsenic (Part 1) (in Japanese). J. MMIJ 2005, 121, 399–406. Farquhar, M. L.; Charnock, J. M.; Livens, F. R.; Vaughan, D. J. Mechanisms of arsenic uptake from aqueous solution by interaction with goethite, lepidocrocite, mackinawite, and pyrite: An x-ray absorption spectroscopy study. Environ. Sci. Technol. 2002, 36, 1757–1762. Lenoble, V.; Bouras, O.; Deluchat, V.; Serpaud, B.; Bollinger, J. C. Arsenic adsorption onto pillared clays and iron oxides. J. Colloid Interface Sci. 2002, 255, 52–58. Waychunas, G. A.; Rea, B. A.; Fuller, C. C.; Davis, J. A. Surface chemistry of ferrihydrite; Part 1, EXAFS studies of the geometry of coprecipitated and adsorbed arsenate. Geochim. Cosmochim. Acta 1993, 57, 2251–2269. Waychunas, G. A.; Davis, J. A.; Fuller, C. C. Geometry of sorbed arsenate on ferrihydrite and crystalline FeOOH: Re-evaluation of EXAFS results and topological factors in predicting sorbate geometry, and evidence for monodentate complexes. Geochim. Cosmochim. Acta 1995, 59, 3655–3661. Jia, Y.; Xu, L.; Fang, Z.; Demopoulos, G. P. Observation of surface precipitation of arsenate on ferrihydrite. Environ. Sci. Technol. 2006, 40, 3248–3253. Jia, Y.; Xu, L.; Wang, X.; Demopoulos, G. P. Infrared spectroscopic and X-ray diffraction characterization of the nature of adsorbed arsenate on ferrihydrite. Geochim. Cosmochim. Acta 2007, 71, 1643–1654. Li, L.; Stanforth, R. Distinguishing adsorption and surface precipitation of phosphate on goethite (R-FeOOH). J. Colloid Interface Sci. 2000, 230, 2–21. Violante, A.; Ricciardella, M.; Gaudio, S. D.; Pigna, M. Coprecipitation of arsenate with metal oxides: Nature, mineralogy, and reactivity of aluminum precipitates. Environ. Sci. Technol. 2006, 40, 4961–4967. Violante, A.; Gaudio, S. D.; Pigna, M.; Ricciardella, M.; Banerjee, D. Coprecipitation of arsenate with metal oxides. 2. Nature, mineralogy, and reactivity of iron(III) precipitates. Environ. Sci. Technol. 2007, 41, 8275–8280. Violante, A.; Pigna, M.; Gaudio, S. D.; Cozzolino, V.; Banerjee, D. Coprecipitation of arsenate with metal oxides. 2. Nature, mineralogy, and reactivity of iron(III)- aluminum precipitates. Environ. Sci. Technol. 2009, 43, 1515–1521. Dzombak, D. A.; Morel, F. M. M. Surface Complexation Modeling; John Wiley & Sons: New York, 1990, pp 191-238. Tokoro, C.; Yatsugi, Y.; Sasaki, H.; Owada, S. A quantitative modeling of co-precipitation phenomena in wastewater containing dilute anions with ferrihydrite using a surface complexation model. Resour. Recycl. 2008, 55, 3–8, 2008. Tokoro, C.; Yatsugi, Y.; Otsuki, A.; Owada, S.; Sasaki, H. Coprecipitation phenomena in wastewater treatment of dilute toxic anions. Proceedings of XXIV Mineral Processing Congress; Beijing, China, September 24-28, 2008; pp 3809-3814. Hsia, T. H.; Loa, S. L.; Lid, C. F.; Lee, D. Y. Characterization of arsenate adsorption on hydrous iron oxide using chemical and physical methods. Colloids Surf., A. 1994, 85, l–7. Dove, P. M.; Rimstidt, J. D. The solubility and stability of scorodite, FeAsO4 · 2H2O. Am. Mineral. 1985, 70, 838–844. Morin, G.; Juillot, F.; Casiot, C.; Bruneel, O.; Personne, J. C.; Elbaz-Poulichet, F.; Leblanc, M.; Ildefonse, P.; Galas, G. Bacterial formation of tooeleite and mixed arsenic(III) or arsenic(V)iron(III) gels in the Carnoule`s acid mine drainage, France. A XANES, XRD, and SEM study. Environ. Sci. Technol. 2003, 37, 1705–1712. Barrett, J.; Ewart, D. K.; Hughes, M. N.; Poole, R. K. Chemical and biological pathways in the bacterial oxidation of arsenopyrite. FEMS Microbiol. Rev. 1993, 11, 57–62. Badulis, G. C.; Tokoro, C.; Sasaki, H. Sludge generation in the treatment of acid mine drainage (AMD) by high-density sludge (HDS) recycling methodsoptimum neutralization process of Horobetsu AMD (first paper). J. MMIJ 2006, 122, 406–414. Tomioka, Y.; Hiroyoshi, N.; Tsunekawa, M. Recent topics on environmental contamination caused by minerals containing arsenic and its remediationsArsenic dissolution and stabilization in mine tailing dumpss(in Japanese). Resour. Recycl. 2005, 52, 145–150.

ES902284C

VOL. 44, NO. 2, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

643