ATR-FTIR Studies on the Nature of Surface Complexes and

Apr 3, 2009 - Adrian Adamescu , Ian P. Hamilton , and Hind A. Al-Abadleh. The Journal of Physical Chemistry A 2014 118 (30), 5667-5679. Abstract | Ful...
0 downloads 0 Views 577KB Size
Download PDF
Environ. Sci. Technol. 2009, 43, 3142–3147

ATR-FTIR Studies on the Nature of Surface Complexes and Desorption Efficiency of p-Arsanilic Acid on Iron (Oxyhydr)oxides MATTHEW CHABOT, TUAN HOANG, AND HIND A. AL-ABADLEH* Chemistry Department, Wilfrid Laurier University, Waterloo, Ontario N2L 3C5, Canada

Received November 10, 2008. Revised manuscript received January 20, 2009. Accepted February 10, 2009.

The fate of organoarsenicals introduced to the environment through the application of arsenic-contaminated manure has attracted considerable attention after the recent implementation of the latest maximum contaminant level (MCL) of total arsenic in drinking water by the U.S. Environmental Protection Agency (EPA). We report herein detailed spectroscopic analysis of the surface structure of p-arsanilic acid (p-AsA) adsorbed on Fe-(oxyhydr)oxides using attenuated total internal reflectance Fourier transform infrared spectroscopy (ATRFTIR). Spectra of p-AsA(ads) were collected in situ as a function of pH and ionic strength and using D2O at 298 K in flow mode. Results indicate the formation of inner-sphere complexes, which are likely monodentate and become protonated under acidic pH(D). We also examined the desorption efficiency of p-AsA(ads) due to flowing electrolyte and phosphate solutions as low as 0.1 mol/m3 (3 ppm P) by collecting ATR-FTIR spectra as a function of time. Our results suggest that aqueous phosphate is an efficient desorbing anion of p-AsA(ads), which has implications on its bioavailability and mobility in geochemical environments.

Introduction With the recent implementation of total arsenic MCL in drinking water (10 µg/L) by the U.S. EPA (1), there are concerns about the fate, transport, and bioavailability of arsenic compounds introduced to the environment from anthropogenic sources. More specifically, the application of As-contaminated poultry litter on agricultural lands is of environmental concern in light of several studies that showed that biogeochemical transformations of organoarsenicals under anaerobic environments result in the formation of the more toxic inorganic arsenate [iAs(V)] and arsenite [iAs(III)] [see references in (2)]. These studies also showed rapid flushing of organoarsenicals to the subsurface during rain storm events. For accurate assessment of the environmental risks of the above practice, factors that control the fate of organoarsenicals used as feed additives in the poultry industry, such as their interactions with soil components, should be investigated. It is well established that the interaction of nutrients and pollutants with soil particles controls their transport and bioavailability (3). Despite the fact that this process is a surface * Corresponding author phone: (519) 884-0710, ext. 2873; fax: (519) 746-0677; e-mail: [email protected]. 3142

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 9, 2009

phenomena occurring at the water/solid interface, measurements of the partitioning of chemical species to common soil components are conducted using ex situ bulk techniques to derive binding constants and maximum adsorption capacities of soil components. These measurements have a number of limitations that include (1) the lack of surface sensitivity needed to extract information about the binding mechanism and the nature of surface complexes, (2) the need to chemically “digest” samples to extract surface-bound species into the aqueous phase, which are then analyzed for “total” concentration of a specific element, and (3) the ex situ and off-line analysis of surface-bound species, which makes it challenging to draw conclusions on the nature of surface complexes in equilibrium with the aqueous phase, as it is well-established that changes in the hydration conditions of the surface result in changing the geometry of weakly adsorbed surface-bound species (4). Clearly, there is a need for monitoring interfacial phenomena of environmental processes with surface sensitivity, in situ and in real time, and under environmentally representative conditions that resemble geochemical environments. Little is known about the adsorption behavior of aromatic organoarsenicals and the reactive centers on the surfaces of soil responsible for adsorption, whether it is the mineral or the natural organic phase. Brown et al. investigated roxarsone (ROX) adsorption and biotransformation in soil samples collected from an agricultural area known for intense poultry production. They concluded that ROX adsorption on soils can be described by both weak and strong interactions (5). We recently utilized surface-sensitive ATR-FTIR to study surface interaction of p-AsA (4-aminophenylarsonic acid, C6H8NO3As) with Fe-(oxyhydr)oxides (2). p-AsA is one of the most commonly used aromatic organoarsenicals in poultry feed, and its structure resembles HAPA (4-hydorxy3-aminophenylarsonic acid, a stable degradation product of ROX in anaerobic environments). We measured the IR spectra of surface complexes, adsorption thermodynamics, and maximum adsorption capacities of hematite, maghemite, and goethite to p-AsA. The Langmuir adsorption model was used to model the experimental data, and binding constants were incorporated into the simple Kd model to predict the transport of this organoarsenical in Fe-rich soils relative to other As(V)-containing compounds. We concluded that p-AsA is more mobile than methylated and inorganic arsenic species. Our studies also indicated that due to the irreversible nature of p-AsA binding to nanoscale Fe-(oxyhydr)oxide particles, nanoparticles transport should also be included in pollutant transport models. Our in situ and surface-sensitive spectroscopic studies reported herein aim at gaining insight into (1) the nature of p-AsA surface complexes from pH- and ionic strength (I)dependent studies and (2) the desorption behavior of adsorbed p-AsA due to inorganic ligands ubiquitous in geochemical environments. The effect of pH and I on the ligand exchange mechanism for Fe-(oxyhydr)oxides may cause shifts in the vibrational modes assigned to surface species as commonly found for the adsorption of oxyanions (6). The involvement of protons in the binding process can be indirectly probed with ATR-FTIR using D2O as the solvent (6). Due to their high water solubility, irrigation and rain events would mobilize aromatic organoarsenicals, which were detected in the drainage water from contaminated fields. Desorption due to phosphate is also likely to occur given that water-extractable phosphorus is the predominant form of P in broiler litter (56-77%) (7). Batch studies on the desorption efficiency of iAs(V) and iAs(III) from Fe-containing 10.1021/es803178f CCC: $40.75

 2009 American Chemical Society

Published on Web 04/03/2009

FIGURE 1. ATR-FTIR absorption spectra of p-AsA(ads) on (a) r-Fe2O3, (b) γ-Fe2O3, and (c) r-FeOOH as a function of pH at equilibrium with 1 mM of p-AsA(aq) of I ) 0.001, 0.01, and 0.1 M, respectively (left to right). samples due to phosphate and other ubiquitous inorganic anions indicate that aqueous phosphate causes the most significant release of inorganic arsenic species (8, 9). Our studies reported herein are the first ATR-FTIR investigations to examine the desorption efficiency of an As(V)-containing organoarsenical due to phosphate. We use the results to formulate equations for the surface complexation of p-AsA on Fe-(oxyhydr)oxides and predict conditions that can result in increasing the bioavailability and mobility of p-AsA.

Experimental Section Chemicals. Stock solutions of p-AsA (99%, Sigma-Aldrich, used as received) were prepared according to the procedure described previously (2). The ionic strength of the p-AsA solutions was adjusted using KCl (99.5%, EM Science). Solutions of p-AsA were also prepared in D2O (99.9 atom % D, Sigma) by dissolving p-AsA in NaOD (40 wt % solution, Sigma) and then lowering the pD using DCl (35 wt % solution, Sigma). Caution: p-AsA is highly toxic via inhalation and skin contact and is a carcinogen. Part of the desorption behavior experiments was conducted using stock solutions of phosphate (99%, BDH) adjusted to pH ) 7 and I ) 0.01 M. The Fe-(oxyhydr)oxides used herein are hematite (RFe2O3, >99.9%, Nanostructured and Amorphous Materials), maghemite (γ-Fe2O3, 99.95%, NanoArc, Alfa Aesar), and goethite (R-FeOOH, >99.9%, Alfa Aesar). These materials were characterized for BET surface area, particle shape and size, and isoelectric points (2). The detailed experimental procedure of film deposition on the ATR internal reflection element (IRE) was described in the Supporting Information of our earlier publication (2). ATR-FTIR Experiments. Details on the collection of ATRFTIR spectra were previously described (2). Briefly, FTIR spectra were collected using a HATRPlus accessory (Pike Technologies) installed in a Nicolet 8700 FTIR spectrometer (Thermo Instruments) equipped with a MCT detector. The 100 µL ATR flow cell houses a 60° ZnSe crystal as the IRE (80 × 10 × 4 mm) on which the Fe-(oxyhydr)oxide films were

directly deposited. The effective angle of incidence and number of reflections were calculated to be 73° and 3, respectively. All spectra were collected at 4 cm-1 resolution by averaging 300 scans. ATR-FTIR single-beam spectra were collected as a function of I and pH(D) by flowing 1 mM p-AsA(aq)atarateof1mL/minacrossagivenFe-(oxyhydr)oxide film using Tygon tubes (0.8 mm i.d., Maserflex) and a compact pump (Masterflex L/S). This concentration of p-AsA(aq) is below the detection limit of our ATR-IRE (8 mM) and results in about 0.5 monolayer of p-AsA(ads) at pH 7 (2). Hence, the contribution of p-AsA(aq) to the IR signal assigned to p-AsA(ads) is negligible. Desorption behavior experiments by electrolyte and disodium hydrogen phosphate solutions were conducted at pH 7 and I ) 0.01 M. Details on the experimental procedure for data collection and processing are provided in the Supporting Information.

Results and Discussion ATR-FTIR Spectra of p-AsA(ads) as a Function of pH and I. Figure 1 shows representative ATR-FTIR spectra of pAsA(ads) on Fe-(oxyhydr)oxides, collected in situ and at equilibrium with 1 mM p-AsA(aq) as a function of pH and I. Infrared absorbances due to symmetric (vs) and asymmetric stretching (vas) vibrations of the -AsO3 moiety in p-AsA are located in the 700-950 cm-1 spectral range, and the number of resolved components is sensitive to the protonation state and binding to surfaces. Useful information can be extracted from the net absorbance of the most intense spectral component and the location of the resolved spectral components shown in Figure 1. The net absorbance, which is proportional to the maximum surface coverage of p-AsA(ads) (Sp-AsA), shows stronger dependency on pH than on I. This is expected since the protonation state of p-AsA(aq) and a given Fe-(oxyhydr)oxide film depends on pH, which affects Sp-AsA. The pKa values of -AsO3H2 in p-AsA(aq) are 4.1 and 9.2 (10), and the point of zero charge of Fe-(oxyhydr)oxides used in these studies is around 9 for R-Fe2O3 and R-FeOOH and 8 for γ-Fe2O3 (2). Sp-AsA is quantified according to VOL. 43, NO. 9, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3143

FIGURE 3. (Left) ATR-FTIR absorption spectra p-AsA(ads) on (a) r-Fe2O3, (b) γ-Fe2O3, and (c) r-FeOOH as a function of pH at equilibrium with 1 mM of p-AsA(aq) of at I ) 0.01 M (solid lines). Dotted lines represent spectra recorded in D2O at I ) 0.01 M. (Right) Difference between the normalized absorbance spectra shown in the left panel. Spectra are offset for clarity. FIGURE 2. pH envelopes of p-AsA adsorption (1 mM) on (a) r-Fe2O3, (b) γ-Fe2O3, and (c) r-FeOOH as a function of I at 298 K. Experimental data (filled markers) are the baseline-corrected ATR absorbance at 838 ( 2 cm-1, which is used to calculate Sp-AsA. Error bars are (σ from averaging 3-4 experiments. Solid lines are a guide to the eye. S(molecules · cm- 2) ) A(λ)/[ε(λ)N 2dp(λ)FbulkSABET], where A(λ) is the baseline-corrected absorbance at 838 ( 2 cm-1, the rest of the parameters were defined previously, and their values were provided in the Supporting Information of ref (2). Figure 2 shows pH envelopes obtained from a series of ATR-FTIR spectra that include those shown in Figure 1. As expected, the minimum Sp-AsA are observed for pH > 9, irrespective of I. The maximum Sp-AsA is observed between pH 4 and 5.5 for both R-Fe2O3 and γ-Fe2O3, whereas for R-FeOOH, Sp-AsA decreases as the pH is raised above 4.5. The location of the resolved spectral components shown in Figure 1 is independent of pH and I and is (2 cm-1 from those recorded at pH 7 (2), suggesting the formation of inner-sphere surface complexes characterized by an intense component at 838 ( 2 cm-1. Components < 800 cm-1 are most intense using R-FeOOH as adsorbent rather than using R- and γ-Fe2O3. Low-frequency components between 805 and 817 cm-1 were observed for adsorbed iAs(V) on Fe-(oxyhydr)oxides and assigned to bidentate inner-sphere complexes aided by results from X-ray absorption studies (4, 11). Components observed between 837 and 878 cm-1 were assigned to H-bonded complexes. By analogy, intense components at 838 ( 2 and < 800 cm-1 are assigned to v(AsOFe) of inner-sphere p-AsA(ads). The relatively small absorbances at 891 and 877 cm-1 using R-Fe2O3 as an adsorbent and 870 cm-1 using γ-Fe2O3 and R-FeOOH indicate the existence of -AsO3H- groups that increase in concentration with lowering pH. These components are observed in the ATR-FTIR spectra of crystalline p-AsA (2) due to the involve3144

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 9, 2009

ment of the -AsO3H- group in hydrogen bonding with neighboring molecules in the crystalline structure. The increase in intensity of these spectral components with decreasing pH is consistent with the increase in Sp-AsA of innersphere p-AsA(ads) whose arsonic group becomes protonated. In the pH range 4-8, the arsonic group in p-AsA(aq) is singly protonated (i.e., -AsO3H-) and the concentration of tFeOH and tFeOH2+ sites is larger than tFeO- sites. The singly protonated arsonic group could be uncomplexed or H bonded to hydroxylated sites or adsorbed water. In order to examine the role of protons in the inner-sphere binding of p-AsA to Fe-(oxyhydr)oxide films, we collected spectra of p-AsA(ads) using D2O instead of H2O as described in the following section. ATR-FTIR Spectra of p-AsA(ads) on Fe-(Oxyhydr)oxides Using D2O. The left panel of Figure 3 shows representative ATR-FTIR spectra of p-AsA(ads) on Fe-(oxyhydr)oxides, which were recorded in situ and at equilibrium with 1 mM p-AsA(aq) in D2O (dotted lines) and H2O (solid lines) at I ) 0.01 M. In general, using D2O as a solvent to examine the nature of oxyanion surface complexes results in red shifting the frequency or lowering the intensities of some spectral features, suggesting the presence of protons (6). A frequency red shift of 12-14 cm-1 was observed for p-AsA(aq) in D2O relative to H2O [see Figure S1, Supporting Information, and ref (2)]. Spectra collected for p-AsA(ads) on R-Fe2O3 in the pH(D) range 4-10 show structured broad absorptions centered at 840 cm-1, with the persistence of features at 793 and 883 cm-1, assigned to v(AsO3) in the inner-sphere complex. This observation suggests a reduction in the intensity or number of spectral components contributing to the broad absorption feature. New spectral features at ca. 854 cm-1 in D2O and 877 cm-1 in H2O increase in intensity with lowering pH(D). This observation suggests an increase in the protonation state of inner-sphere p-AsA(ads) occurring simultaneously with increasing Sp-AsA. In order to extract the

FIGURE 4. Desorption behavior of p-AsA(ads) on r-Fe2O3 (θ ≈ 0.5 ML) due to flowing H2O(l) and HPO42-(aq) at pH ) 7 and I ) 0.01 M: (a) ATR-FTIR absorption spectra as a function of desorption time, (b) reference ATR-FTIR absorption spectra for phosphate adsorption on r-Fe2O3 at pH ) 7 and I ) 0.01 M, and (c) time profile of the baseline-corrected ATR absorbance at 795 and 1045 cm-1 assigned to adsorbed p-AsA and phosphate, respectively. Error bars are (σ from averaging two experiments. exact location of new spectral features, each spectrum shown in the left panel of Figure 3 was normalized to the absorbance at 838 ( 2 cm-1, followed by subtracting spectra recorded at pH(D) > 4 from those at pH(D)4 collected using the same solvent (e.g., spectrum at pD4-spectrum at pD10). Resultant difference spectra are shown in Figure 3 (right panel). For R-Fe2O3, it is clear that protonated inner-sphere p-AsA(ads) complex that exists at acidic pH(D) gives rise to spectral components at 764 and 860 cm-1 in D2O and 870 cm-1 in H2O. There is also a component at 920 cm-1 in D2O that increases in intensity with lowering pD, and its interpretation is provided below. The above analysis was also applied to the spectra collected on γ-Fe2O3 and R-FeOOH as shown in Figure 3. Using γ-Fe2O3 as an adsorbent, results suggest the formation of inner-sphere p-AsA(ads) at basic pH whose arsonic group becomes protonated with lowering pH(D). This type of surface complexes gives rise to difference spectral features at ca. 785 and 871 cm-1 in D2O and 877 cm-1 in H2O (right panel) becoming more pronounced at lower pH(D) relative to basic pH(D). Using R-FeOOH as an adsorbent, spectra collected in H2O (left panel) show that the broad absorption feature centered at 800 cm-1 at basic pH splits into a minimum of two spectral components at ca. 789 and 838 cm-1 and becomes more intense with decreasing pH. This is also accompanied by the appearance of new features at ca. 769 and 876 cm-1 in H2O. This behavior was observed as a function of increasing Sp-AsA at pH 7 (2). When D2O is used as the solvent, there is little or no shift observed in the frequencies of components observed using H2O. The most obvious difference in the case of D2O is the reduction of intensity observed around 800 cm-1, which can be attributed to the disappearance of a component in that region, or a reduction in the width of features at 789 and 838 cm-1. The resultant difference of normalized spectra reveals the growth of features at 764, 787, 849, 874, and 908 cm-1 in D2O and 764, 849, and 870 cm-1 in H2O (right panel). The spectral analysis provided above suggests that the structure of p-AsA(ads) at basic pH(D) is likely a monodentate complex (As-OFe) that is stabilized by resonance between uncomplexed (AsdO) and (As-O-) moieties in the molecule, which explains the intense component at 838 cm-1 that lies between 818(827) and 870(858) cm-1, characteristic of -AsO32- and -AsO3H(D)- of p-AsA(aq) (2) (also see Figure S1, Supporting Information). This assignment is consistent

with the fact that binding of -AsO32- to a metal cation causes a smaller blue shift in v(AsO3) than binding to H+ (4). Lowintensity components < 800 cm-1 increase in intensity as a result of further reduction in the symmetry of the monodentate inner-sphere p-AsA(ads) relative to p-AsA(aq). Features at ca. 870 cm-1 that increase in intensity with lowering pH(D) are around the same location as features assigned to v[AsO2H(D)-] in the ATR-FTIR spectra of aqueous monosubstituted organoarsenicals (12) and p-AsA(s) (2). Hence, inner-sphere p-AsA(ads) complexes that exist at acidic pH(D) are likely to be singly protonated, and the -AsOH group may be involved in weak H bonding with surface tFeOH and tFeOH2+ sites dominant at acidic pH(D). Bidendate innersphere p-AsA(ads) complexes are not likely to form because components at ca. 838 cm-1 and < 800 cm-1 [assigned to v(AsOFe) of inner-sphere p-AsA(ads)] and ca. 870 cm-1 [characteristic of -AsO3H-(aq)] grow simultaneously. The apparent insensitivity of the 870 cm-1 component to solvent in Figure 3 is consistent with the persistence of this component in the spectra collected for p-AsA(aq) at pH(D) 7 (Figure S1, Supporting Information). This observation suggests that it originates from the stretching motion of the nonprotonated As-O bond and cannot be interpreted as an indicator for bidentate complexes. The significance of the above results to modeling the surface complexation of p-AsA is provided below. Moreover, features observed in Figure 3 (right panel) at 920 and 908 cm-1 using R-Fe2O3 and R-FeOOH, respectively, are too high in frequency to be assigned to v(AsO3) in p-AsA(ads) and are more pronounced using D2O as a solvent. The former appears as a positive feature in the absorbance and difference spectra using R-Fe2O3. For the R-FeOOH case, the 908 cm-1 feature appears with a negative absorbance [Figure 3 (left panel)]. We attribute these features to the stretching vibrations of surface FeFeOH groups (13, 14), whose density increases in acidic conditions. It is apparent that deuteration of these groups is more efficient under acidic pD, causing a negative feature in that spectral region. Desorption Behavior of p-AsA(ads). Figures 4a and 5a show ATR-FTIR spectra of p-AsA(ads) on R-Fe2O3 and R-FeOOH, respectively, in the spectral range 700-1250 cm-1 as a function of desorption time due to flowing electrolyte and phosphate solutions at pH 7 and I ) 0.01 M. The starting Sp-AsA is about 0.5 monolayers on both Fe-(oxyhydr)oxides. This spectral range shows peaks at 1097 and 1186 cm-1 VOL. 43, NO. 9, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3145

FIGURE 5. Desorption behavior of p-AsA(ads) on r-FeOOH (θ ≈ 0.5 ML) due to flowing H2O(l) and HPO42-(aq) at pH ) 7 and I ) 0.01 M: (a) ATR-FTIR absorption spectra as a function of desorption time, (b) reference ATR-FTIR absorption spectra for phosphate adsorption on r-FeOOH at pH ) 7 and I ) 0.01 M, and (c) time profile of the baseline-corrected ATR absorbance at 787 and 1045 cm-1 assigned to adsorbed p-AsA and phosphate, respectively. Error bars are (σ from averaging two experiments. assigned to v(C-As) and in-plane δ(C-H) of the aromatic ring in p-AsA(ads). As expected, the former is absent and the latter is present at the same location in the IR spectrum of aniline (15). At pH 7, the dominant phosphate species is HPO42-(aq) with a contribution from H2PO4-(aq) (16). In order to quantify the decrease in Sp-AsA due to desorption, it is important to choose the spectral component assigned to v(AsO3) that has the least contribution from adsorbed phosphate. Hence, control experiments were carried out where spectra of adsorbed phosphate on R-Fe2O3 and R-FeOOH were recorded in situ as shown in Figures 4b and 5b, respectively. Our results are in excellent agreement with studies reported on phosphate adsorption on Fe(oxyhydr)oxides using ATR-FTIR (11, 17-19). Observed peaks with frequencies > 900 cm-1 are assigned to v3(PO4), and those observed at 881 and 891 cm-1 are assigned to v1(PO4) of inner-sphere surface complexes in R-Fe2O3 and R-FeOOH, respectively. Because the width of the peak at 1097 cm-1 assigned to p-AsA(ads) is ca. 9 cm-1, it will not significantly contribute to the 1045 cm-1 assigned to v3(PO4). Hence, the latter peak was used to quantify adsorbed phosphate, and spectral components at 787 and 795 cm-1 assigned to p-AsA(ads) on R-Fe2O3 and R-FeOOH, respectively, were used to quantify Sp-AsA as a function of desorption time. Figures 4c and 5c show the temporal decrease in Sp-AsA as a result of flowing electrolyte and phosphate solutions, along with the temporal increase in adsorbed phosphate. Two concentrations of HPO42-(aq) were used, 10-4 and 10-3 M, and each was flown for 90 min starting with the lowest concentration. This data shows a 68% and 34% total decrease in Sp-AsA, respectively, after 180 min of flowing phosphate, relative to Sp-AsA after flowing electrolyte for 90 min. The difference in the desorption kinetics could be due to a different ligand exchange mechanism occurring between phosphate and p-AsA(ads). It could be also due to a porosity difference of the films deposited on the IRE. Since goethite forms acicular crystals and hematite usually forms spheroid ones (20), a large difference in pore geometry is expected. The porosity (i.e., total pore volume) of deposited R-Fe2O3 and R-FeOOH films was estimated from the volume fraction of the solid, Fv, to be 0.51(0.2) and 0.57(0.1), respectively [see the Supporting Information in ref 2]. Given the negligible difference in the calculated porosity of the two films in our experiments, it is likely that the desorption kinetics are 3146

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 9, 2009

controlled by ligand exchange mechanisms, which is substrate dependent (21, 22). Quantifying the rate parameters for the adsorption/desorption of phosphate and p-AsA on Fe-(oxyhydr)oxides is currently underway in our laboratory using ATR-FTIR. Recently, Parikh and co-workers (23) reported a methodology for studying rapid kinetic reactions using ATR-FTIR with the oxidation of iAs(III) via Mn-oxide as an illustration. The implications of the aforementioned results are discussed below. Environmental Significance. Spectroscopic results reported herein are the first to examine the binding of p-AsA to Fe-(oxyhydr)oxides as a function of pH and I, complemented with experiments in D2O to identify the structure of p-AsA(ads). Analysis of spectral data reveals a ligand exchange mechanism expressed using eqs 1 and 2 ≡FeOH + Ar-AsO3H- f [≡FeO-AsO2Ar]-+H2O

(1)

≡FeOH + Ar-AsO3H2 f ≡FeO-AsO(OH)Ar + H2O (2) where Ar is C6H6NH2. Equations 1 and 2 illustrate the formation of p-AsA(ads) under basic/neutral pH and under acidic conditions, respectively. A schematic of proposed p-AsA(ads) is shown in Figure S2, Supporting Information. These equations are consistent with earlier work showing that the adsorption of As oxyanions on Al- and Fe-(oxyhydr)oxides occurs via ligand exchange mechanisms (22, 24). Combined with our previous results (2), these equations may be useful in constructing surface complexation models that incorporate a binding mechanism. Quantum mechanical calculations of the geometry and IR frequencies of p-AsA(ads) on Fe-(oxyhyr)oxide clusters are ongoing in our laboratory to further support the interpretation of spectral data. Moreover, p-AsA buildup on the mineral component of soils may result from frequent application of arseniccontaminated manure. In addition to irrigation and rainfall, the presence of aqueous phosphate with concentrations as low as 10-4 M (3 ppm P) can result in releasing adsorbed p-AsA from iron-rich in neutral soils, hence increasing its bioavailability and aqueous phase mobility within minutes. These results highlight the connectedness of As and P surface chemistry in iron-rich soils and its impact on the bioavailability of compounds containing these elements (25). Hence, it is important to account for the surface chemistry of both elements when designing effective management and soil

remediation strategies in agricultural fields treated with manure-containing phosphorus and arsenic.

Acknowledgments The authors acknowledge funding from a Petro-Canada Young Innovator Award, Wilfrid Laurier University, a Research Corporation Cottrell College Award, and the Canadian Foundation for Innovation. T.H. is grateful for the NSERC Undergraduate Student Research Award program.

Supporting Information Available Details of the experimental procedure, spectra of p-AsA(aq) in D2O, and schematic of p-AsA(ads) on Fe-(oxyhydr)oxides. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) U.S. EPA: Arsenic Rule. http://www.epa.gov/safewater/arsenic/ regulations.html. (2) Depalma, S.; Cowen, S.; Hoang, T. N.; Al-Abadleh, H. A. Adsorption thermodynamics of p-arsanilic acid on iron (oxyhydr)oxides: in-situ ATR-FTIR studies. Environ. Sci. Technol. 2008, 42, 1922–1927. (3) Sparks, D. L. Environmental Soil Chemistry, 1st ed.; Academic Press: San Diego, 1995. (4) Myneni, S. C. B.; Traina, S. J.; Waychunas, G. A.; Logan, T. J. Experimental and theoretical vibrational spectroscopic evaluation of arsenate coordination in aqueous solutions, solids, and at the mineral-water interfaces. Geochim. Cosmochim. Acta 1998, 62, 3285–3300. (5) Brown, B. L.; Slaughter, A. D.; Schreiber, M. E. Controls on roxarsone transport in agricultural watersheds. Appl. Geochem. 2005, 20, 123–133. (6) Sparks, D. L. Metal and oxyanion sorption on naturally occurring oxide and clay mineral surfaces. In Environmental Catalysis; Grassian, V. H., Ed.; Taylor & Francis Group: Boca Raton, 2005; pp 3-36. (7) Toor, G. S.; Peak, J. D.; Sims, J. T. Phosphorus speciation in broiler litter and turkey manure produced from modified diets. J. Environ. Qual. 2005, 34, 687–697. (8) Frau, F.; Biddau, R.; FanFani, L. Effect of major anions on arsenate desorption from ferrihydrite-bearing natural samples. Appl. Geochem. 2008, 23, 1451–1466. (9) Su, C.; Puls, R. W. Arsenate and arsenite removal by zerovalent iron: effects of phosphate, silicate, carbonate, borate, sulfate, chromate, molybdate, and nitrate, relative to chloride. Environ. Sci. Technol. 2001, 35, 4562–4568. (10) Nuallain, C. O.; Cinneide, S. O. Thermodynamic ionization constants of aromatic arsonic acids. J. Inorg. Nucl. Chem. 1973, 35, 2871–2881.

(11) Lefevre, G. In situ Fourier-transform infrared spectroscopy studies of inorganic ions adsorption on metal oxides and hydroxides. Adv. Colloid Interface Sci. 2004, 107, 109–123. (12) Cowen, S.; Duggal, M.; Hoang, T. N.; Al-Abadleh, H. A. Vibrational spectroscopic characterization of some environmentallyimportant organoarsenicals: a guide for understanding the nature of their surface complexes. Can. J. Chem. 2008, 86, 942– 950. (13) Ji, Y.-Q.; Black, L.; Weidler, P. G.; Janek, M. Preparation of nanostructured materials by heterocoagulation interaction of montmorillonite with synthetic hematite particles. Langmuir 2004, 20, 9796–9806. (14) Al-Mashta, F.; Sheppard, N. Infrared study of adsorption on oxygen-covered R-Fe2O3: bands due to adsorbed oxygen and their modification by co-adsorbed hydrogen or water. J. Chem. Soc., Faraday Trans. 1982, 78, 979–989. (15) Silverstein, R. M.; Webster, F. X. Spectrometric Identification of Organic Compounds, 6th ed.; Wiley: New York, 1998. (16) Snoeyink, V. L.; Jenkins, D. Precipitation and Dissolution Water Chemistry; Wiley: New York, 1980; p 243. (17) Tejedor-Tejedor, M. I.; Anderson, M. A. Protonation of phosphate on the surface of goethite as studied by CIR-FTIR and electrophoretic mobility. Langmuir 1990, 6, 602–611. (18) Arai, Y.; Sparks, D. L. ATR-FTIR spectroscopic investigation on phosphate adsorption mechanisms at the ferrihydrite-water interface. J. Colloid Interface Sci. 2001, 241, 317–326. (19) Elzinga, E. J.; Sparks, D. L. Phosphate adsorption onto hematite: an in situ ATR-FTIR investigation of the effects of pH and loading level on the mode of phosphate surface complexation. J. Colloid Interface Sci. 2007, 308, 53–70. (20) Schwertmann, U.; Cornell, R. M. Iron Oxides in the Laboratory: Preparation and Characterization, 2nd ed.; Wiley-VCH: Weinhein, 2000. (21) Stumm, W. The inner-sphere surface complex-a key to understanding surface reactivity. In Aquatic Chemistry-Interfacial and Interspecies Processes; Huang, C. P., O’Melia, C. R., Morgan, J. J., Eds.; American Chemical Society: Washington D.C., 1995; Vol. 244, pp 1-32. (22) Parikh, S. J.; Lafferty, B. J.; Sparks, D. L. An ATR-FTIR spectroscopic approach for measuring rapid kinetics at the mineral/water interface. J. Colloid Interface Sci. 2008, 320, 177– 185. (23) Fukushi, K.; Sverjensky, D. A. A predictive model (ETLM) for arsenate adsorption and surface speciation on oxides consistent with spectroscopic and theoretical molecular evidence. Geochem. Cosmochim. Acta 2007, 71, 3717–3745. (24) Arai, Y.; Sparks, D. L.; Davis, J. A. Arsenate adsorption mechanisms at the allophane-water interface. Environ. Sci. Technol. 2005, 39, 2537–2544. (25) Dixit, S.; Hering, J. G. Comparison of As(V) and As(III) sorption onto iron oxide minerals: implication for arsenic mobility. Environ. Sci. Technol. 2003, 37, 4182–4189.

ES803178F

VOL. 43, NO. 9, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3147