Aluminum Layered Double Hydroxide

Nov 1, 2006 - adjusted to 5.0 using HNO3 or NaOH by pH-stat (Radiometer,. TIM865) and .... The net charge of Li/Al LDH-Cl would switch from positive...
0 downloads 0 Views 186KB Size
Environ. Sci. Technol. 2006, 40, 7784-7789

Arsenate Sorption on Lithium/ Aluminum Layered Double Hydroxide Intercalated by Chloride and on Gibbsite: Sorption Isotherms, Envelopes, and Spectroscopic Studies Y U T I N G L I U , † M I N G K U A N G W A N G , * ,† TSAN YAO CHEN,‡ PO NENG CHIANG,† P A N M I N G H U A N G , †,⊥ A N D J H Y F U L E E § Department of Agricultural Chemistry, National Taiwan University, Taipei, Taiwan, 10617, Department of Engineering and System Science, National Tsing Hua University, Hsin Chu, Taiwan, 30043, National Synchrotron Radiation Center, Hsin Chu, Taiwan, 30076

The objective of this study was to provide fundamental knowledge of arsenate sorption on lithium/aluminum layered double hydroxide intercalated by chloride (Li/Al LDHCl) and further to reveal the contribution of exposed positive charge surface of Li/Al LDH-Cl created by intercalating LiCl into Al(OH)3 layers to arsenate sorption. Therefore, sorption isotherms, envelopes and extended X-ray absorption fine structure (EXAFS) technique were employed to examine the reaction of arsenate on Li/Al LDH-Cl and on gibbsite. Based on an isotherm study, the sorption maximum of Li/Al LDH-Cl for arsenate was approximately six times higher than that of gibbsite. Sorption envelopes of arsenate on Li/Al LDH-Cl displayed a pH-sensitive behavior from pH 4.0 to 7.0, but it was insensitive to pH above pH 7.0, approaching to the pHpzc of Li/Al LDH-Cl (7.22). This transformation with shifted pHs illustrated that there were two types of reaction sites within Li/Al LDH-Cl that participate in arsenate sorption; one is pH-sensitive and the other is not. From EXAFS analysis, arsenate sorbed on Li/Al LDH, reacted not only with Al in the edges of Al(OH)3 layers, but also with Li located in the vacant octahedral sites within Al(OH)3 layers; however, the decreasing intensity of As(V)-Al shells with increasing pH represented there were fewer As(V)-Al complex existed at higher pH, i.e., the complex between arsenate and Al is pHsensitive. The superior sorption capability of Li/Al LDH-Cl to that of gibbsite could be attributed to the intercalated Li cations which served as the permanent sorption sites and made the surface of Al(OH)3 have high affinity to arsenate.

Introduction Soil and sediment polluted by toxic metals pose deleterious effects on the health of ecosystems. Arsenate, for example, * Corresponding author: phone: (0118862) 3366-4808; fax: (0118862) 2366-0751; e-mail: [email protected]. † National Taiwan University. ‡ National Tsing Hua University. § National Synchrotron Radiation Center. ⊥ Present address: Department of Soil Science, University of Saskatoon, Saskatchewan, Canada 57H 4A5. 7784

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 24, 2006

a known carcinogen in humans, is often found in contaminated groundwater as a result of weathering from rocks, industrial waste discharges, and agricultural use of arsenical herbicides and pesticides. Sorption, however, is a major process dictating the distribution of arsenate in the environment, and there has been a wide range of studies of sorption by metal oxides, including aluminum hydroxides, goethite, hematite, lepidocrocite, hydrous ferric oxide (HFO), and birnessite (1). Phyllosilicates, metal (hydr)oxides, and humic substances sorb heavy metals by forming inner- or outersphere sorption complexes, creating important sinks for these metals in ecosystems (2). Although there were previous investigations describing arsenate sorption on gibbsite, the most ubiquitous aluminum hydroxide in soil, it still merits further study (1, 3, 4), since aluminum is one of the most abundant elements on earth. Layered double hydroxides (LDHs), which were reported in the mid 19th century, are mixed-metal hydroxides consisting of octahedral double hydroxyl layers with exposed positive surface charge and interlayer anions. The general known group of LDH is [M1-x2+Mx3+(OH)2] Ax/nn-‚mH2O, where M2+ and M3+ are divalent and trivalent cations, respectively, with an ionic radius similar to that of Mg2+ (5). Intercalation of inorganic or organic ions into gibbsite matrix is the process to construct Li/Al LDHs (6-9). Gibbsite has shown to be intercalated with lithium cations obtained from a range of lithium salts, and these materials readily dehydrate giving highly crystalline intercalates. The crystal structures of these materials display highly ordered lamellar phases consisting of eclipsed [LiAl2(OH)6]+ sheets with sandwiching layers of intercalated anions (7). Li/Al LDHs with a formula of [LiAl2(OH)6]+A-‚mH2O (A- ) Cl-, OH-, NO3-, ect.) have not been well studied until 1977 (10-12). The structure of Li/Al LDH intercalated by chloride (Li/Al LDH-Cl) along the c-axis clearly illustrates the relative arrangements of the cations and anions; Li+ and Cl- ions line up parallel to the c-axis and form an alternating Li+... C-... Li+... chain (7, 8). According to the structure, intercalating Li cations in the octahedral sites of Al(OH)3 would provide a positive surface and serve as permanent sorption sites for arsenate in contrast to Al presented in the edges of Al(OH)3, on which the sorption ability would be weaker as pH is raised. Therefore, the contribution of Li and Al enabling Li/Al LDH-Cl to render arsenate immobile and limit its bioavailability is desirable. The local coordination environments of arsenate sorbed on gibbsite was previously demonstrated using extended X-ray absorption fine structure (EXAFS) spectroscopy, illustrating dominant surface complex of arsenate sorbed on gibbsite was bidentate binuclear (13). By examining net OHrelease stoichiometry, Weerasooriya reported that arsenate forms a bidentate surface complex on gibbsite (4). However, there was no direct information for the atomic structure of arsenate sorption on Li/Al LDH-Cl published. Moreover, the oxyanion sorption by soil minerals under a prescribed set of conditions depends on the pH of the soil solution. The pH value significantly influences the sorption plateau and hydrolysis mechanism of sorption sites. Thus, the primary objective of this research was to study the radical knowledge, including sorption isotherms, envelopes, and mechanism between arsenate and Li/Al LDH-Cl, on an atomic scale. In addition, the sorption behavior of Li/Al LDH-Cl was compared with that of gibbsite to demonstrate the contribution of intercalating LiCl into Al(OH)3 layers to arsenate sorption. The material of Li/Al LDH-Cl can be incorporated in to a filter, barrier or column to remove dissolved arsenate in nature environments and drinking water; however, these 10.1021/es061530j CCC: $33.50

 2006 American Chemical Society Published on Web 11/01/2006

applications must be based on the fundamental sorption study. According to sorption information, the theoretical characteristic, reactive mechanism, and latent efficacy of sorbent can be defined and assessed clearly. The present study seeks to elucidate the fundamental properties of arsenate sorption to Li/Al LDH-Cl and to demonstrate how it may be an effective remover for arsenate; additionally, this conclusion can be treated as the groundwork of Li/Al LDHCl for further use in remedying the As-contaminated environments.

Experimental Section Preparation and Identification of Li/Al LDH-Cl and Gibbsite. The procedure of Li/Al LDH-Cl synthesis was modified from previously reported methods (11, 14). Briefly, 13.5 g of Al foil (Nacalai Tesque, Kyoto, Japan) was dissolved with 2.5 L of 0.6 M NaOH solution in a 5 L Teflon bottle for 1.5 h. The molar ratio of OH to Al was 3. Then 210 g of LiCl (Acros, NJ, U.S.) were added to achieve the final concentration of 2 M. The suspension was stirred for 48 h and the temperature was maintained at 80-90 °C. Suspension of Li/Al LDH-Cl thus formed was adjusted to pH 5.0 with 1 M HCl/1M LiCl solutions and centrifuged at 17 700 × g (Hitachi, 18PR-52) for 10 min. The excess salt of the precipitate was removed by washing with double distilled water (DDW) five times, after which the remaining solids were freeze-dried, instead of air or oven dried, to avoid aggregation. Gibbsite (Al(OH)3) was supplied by Merck, KGaA, Darmstadt, Germany. The specific surface area measurements of Li/Al LDH-Cl and gibbsite were achieved by N2 gas absorption at 78 K and fitted with the BET isothermal equation. Samples were pretreated with evacuation at 373 K and 2 × 10-5 Torr vacuum for 24 h to remove the surface absorption gas and water molecules. The whole measurement was processed on a vacuum system utilizing a precision vacuum gauge (Texas Instruments model 145). X-ray diffraction analysis was processed on a Rigaku Miniflex X-ray diffractometer from 3 to 50°(2θ) at a rate of 1°(2θ) per min. Cu-ΚR radiation used in this study was generated at 30 kV and 10 mA. Point of zero charge (pHPZC) of Li/Al LDH-Cl and gibbsite was examined through measuring zeta potential of suspensions with pH of 3.8∼9.4, adjusted with HNO3 and NaOH solution by a Malvern Zetasizer 3000. Sorption Isotherms. Sorption isotherms were performed at pH 5.0 with arsenate concentrations of 0.02, 0.03, 0.04, 0.05, 0.06, 0.08, 0.1, 0.2, 0.5, and 1 mmol L-1. Na2HAsO4‚7H2O (J.T. Baker, Phillipsburg, NJ, U.S.) was used as the arsenate source. Arsenate solution and suspensions of Li/Al LDH-Cl and gibbsite were prepared in a background electrolyte of 0.1 M NaNO3 solution. The concentrations of stock arsenate solution were equipped and diluted to 0.04, 0.06, 0.08, 0.1, 0.12, 0.16, 0.2, 0.4, 1, and 2 mmol L-1. Twenty mL of 2.5 g L-1 Li/Al LDH-Cl or gibbsite suspensions in 0.1 M NaNO3 were added to the 50 mL polyethylene vessels and agitated by magnetic stirrer all the time. The pHs of suspensions were adjusted to 5.0 using HNO3 or NaOH by pH-stat (Radiometer, TIM865) and maintained constant for one week. Twenty mL of arsenate solution of different concentrations with preadjusted pH was then quickly added to the Li/Al LDH-Cl or gibbsite suspension and reacted at 150 rpm on a rotary shaker. After 4 days, the suspensions were sampled with a syringe and filtered through a 0.2 µm Millipore filter membrane. Total arsenic concentration of the filtrates was determined by inductively coupled plasma atomic emission spectrometry (ICP-AES, Perkin-Elmer, Optima 2000DV). Triplicate runs were carried out for each treatment. Sorption Envelopes. Sorption envelopes were performed in 0.1 M NaNO3 at arsenate concentration of 0.1 mmol L-1. The stock solutions of arsenate were prepared having arsenate concentration of 0.2 mmol L-1. Twenty mL of 2.5 g L-1 Li/Al

LDH-Cl or gibbsite suspension in 0.1 M NaNO3 and 20 mL of stock arsenate solution, both with preadjusted pH, were added into the 50-mL polyethylene vessel simultaneously and agitated at 150 rpm on a rotary shaker for 4 days. Before suspensions and arsenate solution were mixed together, the pH of Li/Al LDH-Cl and gibbsite suspensions and stock arsenate solutions were adjusted to the range of 4.0∼9.0 ( 0.1 with HNO3 or NaOH by pH-stat and maintained constant for one week. At the end of reaction, the suspensions were sampled with a syringe and filtered through a 0.2 µm Millipore filter membrane. Total arsenic concentration of the filtrates was determined by ICP-AES. Triplicate runs were carried out for each treatment. Arsenate EXAFS Analysis. Samples for EXAFS analysis were obtained by mixing 15 mL of Li/Al LDH-Cl or gibbsite suspensions with 15 mL of 0.02 M arsenate solution of pH 5.0, 7.0, and 9.0 (adjusted by NaOH and HNO3 solution) in a 30 mL Teflon centrifuge tube. The suspensions were prepared by stirring 0.5 g of Li/Al LDH-Cl or gibbsite with 15 mL of 0.1 M NaNO3 solution employed in all the samples to control the ionic strength. Initial concentration of arsenate was 0.01 M. After being shaken for 4 days, the suspensions were centrifuged at 17 700 × g for 10 min. The precipitates were washed five times by DDW to remove excess salts and freeze-dried for EXAFS analysis. X-ray absorption spectra (XAS) at the As K-edge (11 867 eV) were collected at Wiggler 20 beam line BL-17C at National Synchrotron Radiation Research Center (NSRRC), Hsin-Chu, Taiwan. The electron storage ring operated at 1.5 GeV with a fixed current of 250 mA. Beam line Wiggler 20 (17C), with energy ranging from 4 to 15 keV, employs a Si (111) doublecrystal monochromator for energy scanning with a resolving power (E/∆E) of 7000 and beam intensity of around 109∼1010 photons per sec. The K-edge spectra of As reacted with Li/Al LDH-Cl and gibbsite were recorded in transmission and fluorescence mode, respectively, and absorbance of the incident X-rays was collected by the ionization chambers It and If (Lytle Detector) (15), separately. In order to attenuate scattered principle energy X-ray from entering the fluorescence detector, soller slits, and an absorbing filter (Ge for As atom) were placed between sample and If detector. All samples were fixed onto an aluminum holder, sealed with Kapton tape and placed at 45° to the X-ray beam. The experiments were carried out from 11 667 to 12 867 eV at ambient temperature. In addition, the Au L3-edge spectrum was monitored by the Ir chamber simultaneously with It and If chambers, serving as the reference to calibrate energy shift due to monochromator drifts. Several scans were processed and averaged on each sample to improve the data quality of XAS spectra. All of the XAS were normalized by AUTOBK2.61 program (16), transformed from electron energy to photon-electron wave vector unit (k, Å-1) and weighted by k3 to generate k3χ(k) spectra for better demonstration of the contribution of different shells of the EXAFS oscillation function. Fourier transformation (FT) was further processed to produce radial structure functions (RSF) within selected k range (Table SI1 of the Supporting Information). The FEFF8.20 program (17) was employed to create theoretical phases and amplitude functions representing photoelectron scattering paths of AsO, As-Li, and As-Al by inputting standard structure parameters of AlAsO4 (18) and LiH2AsO4 (19). The model fitting was conducted by FEFFIT2.32 program (20), yielding the structural parameters including coordination number (CN), interatomic distance (R), and Debye-Waller factor (σ2). The phase and amplitude shift of EXAFS spectra between local structure of experimental sample and theoretical atomic model were adjusted by Debye-Waller factors and amplitude reduction factor. VOL. 40, NO. 24, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7785

FIGURE 1. Sorption isotherms for arsenate reacted with Li/Al LDHCl and with gibbsite at pH 5.0.

TABLE 1. Surface Loading for Arsenate on a Variety of Minerals compound/formula

pH

surface loading (µmol m-2)

refs

Li/Al LDH-Cl/[LiAl2(OH)6]Cl gibbsite/R-Al(OH)3 aluminum oxide/γ-Al2O3 activted alumina/Al2O3‚xH2O ferryhydrite/δ′-FeOOH goethite/R-FeOOH

5.0 5.0 5.5 6.5 4.6 4.0

17.19 0.79 1.40 0.12 9.90 2.26

this study this study 23 32 22 24

Results and Discussion Characteristics of Li/Al LDH-Cl and Gibbsite. X-ray diffraction pattern of Li/Al LDH-Cl synthesized and freeze-dried in this study showed the d values of 7.61, 3.82 and 2.55 Å corresponding to [002], [004, 102], and [006, 110] point diffraction patterns (Figure SI1). The crystallographic data of Li/Al LDH-Cl agreed with that of [LiAl2(OH)6]Cl synthesized by mixing Al with NaOH as well as LiCl (14) and by stirring gibbsite with LiCl (7). Specific surface areas of Li/Al LDH-Cl and gibbsite were 18.76 ( 1.76 m2 g-1 and 65.30 ( 0.45 m2 g-1, respectively. The pHPZC was 7.22 ( 0.09 for Li/Al LDH-Cl and 5.06 ( 0.06 for gibbsite. Sorption Isotherms. The sorption isotherms of arsenate sorbed on Li/Al LDH-Cl and on gibbsite were shown in Figure 1. As can be seen, Li/Al LDH-Cl had a higher capacity for removing arsenate compared to gibbsite. The adherence of the sorption isotherm data to the Langmiur equation was tested graphically, and the correlation coefficients (r2) of the regression was 0.85 and 0.96 for Li/Al LDH-Cl and gibbsite, respectively. Li/Al LDH-Cl had monolayer sorption maximum of 322.58 mmol kg-1 calculated from extrapolation of Langmuir equation, and for gibbsite, it was 51.55 mmol kg-1. The sorption maxima corresponding to surface loading are 17.19 µmol m-2 for Li/Al LDH-Cl and 0.79 µmol m-2 for gibbsite. The surface loadings of arsenate sorbed on different sorbents are listed in Table 1. It is apparent that sorption capacity of Li/Al LDH-Cl was generally superior compared with that of gibbsite and even other oxides (Table 1). Sorption Envelopes. For Li/Al LDH-Cl and gibbsite, there was a general decrease in arsenate with increasing pH (Figure 2). With Li/Al LDH-Cl, the amount of sorbed arsenate decreased approximately from 67% of the original added at pH 4.0-21% at pH 9.0. Arsenate sorption was sensitive to changes in pH between 4.0 and 7.0, but relatively insensitive above pH 7.0. The inflection point of arsenate sorption envelope on Li/Al LDH-Cl occurred at pH 7.0, approaching to the pHpzc of Li/Al LDH-Cl (7.22) and pK2 of arsenate (6.97). The net charge of Li/Al LDH-Cl would switch from positive at pH lower than 7.22 through zero to negative at pH higher 7786

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 24, 2006

FIGURE 2. Sorption envelopes for arsenate reacted with Li/Al LDHCl and with gibbsite. than 7.22; meanwhile, the dominant species of arsenate would shift from H2AsO4- to HAsO42-. Thus, the decrease of sorbed arsenate with raising pH was sound since Coulombic repulsion caused by the increasing negative charge of Li/Al LDH and arsenate would diminish the possibility that arsenate accessed to Li/Al LDH-Cl. At pH 7.0-9.0; however, the decrease of sorbed arsenate was less pH-sensitive than that at pH 4.0-7.0. It is very interesting that the amount of arsenate sorbed on kaolinite, montmorillonite, illite, γ-Al2O3, ferrihydrite and goethite would decrease dramatically with raised pH from 7.0 to 9.0 regardless of the difference in experimental setups of these investigations (21-24), but, on Li/Al LDH-Cl, it would not. Conclusively, we supposed that there were pH-sensitive and less pH-sensitive sorption sites presented on the Li/Al LDH-Cl. For gibbsite, only 17% of the initial added arsenate was sorbed at pH 4.0 and decreased to 10% at pH 9.0 (Figure 2). The pHpzc of gibbsite is 5.06 corresponding to 5.0 presented by Stumm (25). Hingston et al. (26) indicated sorption of incompletely dissociated acids can also take place at pH values which were more alkaline than the pHpzc of sorbent with the pH near a pKa of the acid, where the energy needed to abstract a proton required for the removal of a surface OH-, thus providing a site for the anion from the acid is at a minimum. It is one reason that arsenate could be sorbed by gibbsite ignoring Coulombic repulsion; the other reason is that the inner-sphere complex occurred between arsenate and Al atoms of gibbsite (13). Halter and Pfeifer (27) as well as Weerasooriya et al. (4) reported that almost 99% of arsenate adsorbed on R-Al2O3 and gibbsite below pH 7.5. The experimental conditions employed to study arsenate sorption on R-Al2O3 and on gibbsite were 10 g L-1 R-Al2O3 with 6.7 × 10-5 M arsenate (27); 20 g L-1 gibbsite with 2.67 × 10-6 M arsenate (4), respectively. In our study, the condition of 1.25 g L-1 gibbsite with 1.0 × 10-4 M arsenate was adopted. The discrepancy in percentage of sorbed arsenate was due to the different ratios of suspensions with arsenate concentration. The high concentration ratio of suspensions to arsenate solution would enlarge the percentage of sorbed arsenate because the sorption sites were sufficient for arsenate. Arai et al. (23) illustrated the prevailing configuration between As(V) and aluminum oxide-water interface was inner-sphere complex. Goldberg and Johnston (28) showed that the formation of inner-sphere complex via ligand exchange was predominantly responsible for As(V) adsorption at amorphous aluminum oxide by vibrational spectroscopic study. In this research, EXAFS technique was employed to distinguish the complex mechanism between arsenate and Li/Al LDH-Cl. Arsenate EXAFS Analysis. Figures 3a and 4a showed the k3χ(k) spectra of arsenate sorbed on Li/Al LDH-Cl and on gibbsite at pH 5.0, 7.0, and 9.0, respectively. The conformity between experimental data and fitting models indicated that

TABLE 2. Structure Parameters of Arsenate Sorbed on Li/Al LDH-Cl and on Gibbsite at pH 5.0, 7.0, and 9.0 Derived from EXAFS Analysisa As(V)-O

Li/Al LDHCl pH 5.0 Li/Al LDH-Cl pH 7.0 Li/Al LDH-Cl pH 9.0 Gibbsite pH 5.0 Gibbsite pH 7.0 Gibbsite pH 9.0

CNb

R(Å)c

3.6 ( 0.7 3.5 ( 0.7 3.5 ( 0.7 3.5 ( 0.7 3.4 ( 0.7 3.6 ( 0.7

1.70 ( 0.03 1.70 ( 0.02 1.70 ( 0.02 1.71 ( 0.03 1.70 ( 0.02 1.71 ( 0.03

As(V)-Li σ2(Å2)d

CN

R(Å)

As(V)-Al σ2(Å2)

CN

R(Å)

0.0032 1.2 ( 0.2 2.62 ( 0.03 0.0013 2.0 ( 0.4 3.06 ( 0.02 0.0024 1.1 ( 0.2 2.56 ( 0.05 0.0027 1.7 ( 0.3 3.03 ( 0.01 0.0024 1.0 ( 0.2 2.57 ( 0.04 0.0041 1.6 ( 0.3 3.03 ( 0.01 0.0022 1.9 ( 0.4 3.09 ( 0.05 0.0021 1.7 ( 0.3 3.07 ( 0.04 0.0019 2.0 ( 0.4 3.08 ( 0.05

σ2(Å2)

χν2e

R-factore

0.0085 0.0079 0.0093 0.0056 0.0029 0.0029

33.4 33.3 42.2 18.5 6.4 10.5

0.019 0.018 0.020 0.023 0.024 0.028

a The amplitude reduction factor (S 2) of all shells was fixed to 0.85 (33). b Coordination number. c Interatomic distance. d Debye-Waller factor 0 (disorder parameter). e χν2 ) χ2/ν. χ2 is generally considered the best figure of merit to judge the quality of the fit. ν is the number of degrees of freedom in the fit. R-factor is directly proportional to χ2, and gives a sum-of-squares measure of the fractional misfit.

FIGURE 3. (a) k3χ(k) spectra and (b) RSF profiles of EXAFS signal for arsenate sorbed on Li/Al LDH-Cl at pH 5.0, 7.0, and 9.0. Solid lines and open circles represent fitted and experimental data, respectively (without phase shift correction). these models were proper and reliable for the configuration of samples. Structural parameters obtained from the EXAFS analysis were presented in Table 2. The RSF profiles of arsenate sorbed on Li/Al LDH-Cl and on gibbsite were presented in Figures 3b and 4b, with the peak position correlating with interatomic distance (without phase shift correction). In Figure 3b, the RSF profiles of arsenate sorbed on Li/Al LDH-Cl at pH 5.0, 7.0, and 9.0 revealed three shells of As(V)-O, As(V)-Li, and As(V)-Al. The first shell of As(V)-O at three pHs were simulated with that extracted from AlAsO4, displaying the interatomic distance of 1.70 Å with a CN of 3.5∼3.6 O atoms around the central As. The XANES results (Figure SI2) demonstrated that the reactive arsenic species

was As(V), i.e., H2AsO4- or HAsO42-; therefore, the theoretical CN atoms of As(V)-O shell should be four. The difference of CN atoms between theoretical and fitting results could be explained by the presence of distorted bonding of arsenate sorbed on structure defects of Li/AL LDH-Cl that would result in backscattering noise in RSF, which strongly influenced the EXAFS analysis (29). Based on the As(V)-Li shell extracted from LiH2AsO4 and the theoretical interatomic distance calculated from the supposed sorption complex, the second marked peaks at three pHs were confirmed as As(V)-Li shells with interatomic distance ranged from 2.56∼2.62 Å and the CN of 1.0∼1.2 Li atom. The existence of As(V)-Li shells with equal intensity at three pHs represented the amount of innersphere complex occurred between arsenate and Li was almost the same across the pH values utilized in the present study. The shortest distance of two Li atoms within neighboring octahedral holes was about 5.1 Å, calculated from the structure of Li/Al LDH-Cl presented by Besserguenev et al. (7). The interatomic distance of As(V)-O was 1.70 Å with the O-As-O angle of 102.2∼113.3° (30). Thus, the configuration of binuclear complex existed as two O atoms of arsenate reacted with two Li was not stable since the distance of O-O of arsenate was much shorter than that of Li-Li. Bidentate mononuclear complex, hence, was the dominant configuration of As(V)-Li shell. Besides, using As(V)-O distance of 1.70 Å, arsenate O-O distances of 2.65∼2.79 Å (30) and Li-O distance of 1.95∼1.99 Å (19), the theoretical interatomic distance of bidentate mononuclear complex of As(V)-Li could be estimated as 2.26∼2.60 Å. Combined both of the abovementioned points, the bidentate mononuclear was the most reasonable formation for As(V)-Li shell. The third shells of As(V)-Al with 1.6∼2.0 Al atoms and 3.03∼3.06 Å interatomic distances at three pHs showed that the bidentate binuclear complex was the most pronounced since the distance of 3.03∼3.06 Å corresponded with that of 3.03∼3.41 Å for bidentate binuclear complex of arsenate sorbed on aluminum oxide proposed by Arai et al. (23). Moreover, the amount of complex could be conjectured by the peak intensity of RSF profile; thus, the reducing intensity of As(V)-Al shell with increasing pH could be considered that there were fewer As(V)-Al complex existed at higher pH. Figure 4b illustrates that arsenate sorbed on gibbsite exhibited the same first-shell environment as that on Li/Al LDH-Cl: for all three pHs, 3.4∼3.6 O atoms coordinated the central As atom at a distance of 1.70 Å and the oxidation state maintained at As(V) (Figure SI2). The second shell was As(V)-Al with 1.7∼2.0 Al atoms enclosing the central As atom at the interatomic distance of 3.07∼3.09 Å. Accordingly, bidentate binuclear complex was the prevailing configuration between arsenate and Al within gibbsite. This configuration was in good agreement with that of arsenate sorbed on gibbsite (13) and γ-Al2O3 (23). Mechanism of Arsenate Sorption. From the sorption isotherms, Li/Al LDH-Cl appeared as an excellent sorbent, VOL. 40, NO. 24, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7787

with arsenate would be weaker as pH is raised. Combining the points derived from sorption envelopes and EXAFS analysis, we were convinced that both Li and Al participated in arsenate sorption on Li/Al LDH-Cl below pH 7.0 and caused the pH-sensitive sorption behavior because the contribution of Al to arsenate sorption would decrease with increasing pH. Above pH 7.0, however, Li was the most pronounced reactive site for arsenate and resulted in slightly pH-sensitive sorption behavior since it presented as ionic state in the surface of Li/Al LDH-Cl and would not vary with pH. The macroscopic and spectroscopic experiments were connected to investigate the sorption capability of Li/Al LDHCl and of gibbsite. In contrast to gibbsite, the superior sorption ability of Li/Al LDH-Cl was mainly due to the two different sorption sites on structure. Treatment with lithium chloride leads to intercalation of Li cations into the host structure of Al(OH)3 to form the layered double hydroxide of Li/Al LDHCl. The Li cations occupied the vacant octahedral holes within Al(OH)3 would transform the surface of Al(OH)3 layers into active sorption sites with high affinity for arsenate, resulting from the additional positive charge brought by Li cations and raising the pHpzc of Li/Al LDH-Cl. Our experimental results can shed light on the actual mechanisms of arsenate sorbed on Li/Al LDH-Cl and gibbsite at different pHs. We have demonstrated the intercalation of LiCl into Al(OH)3 layer can improve the sorption capability for anion, and Li/Al LDH may be a potential and low-cost material to remediate environments contaminated by toxic anions.

Acknowledgments

FIGURE 4. (a) k3χ(k) spectra and (b) RSF profile of EXAFS signal for arsenate sorbed on gibbsite at pH 5.0, 7.0, and 9.0. Solid lines and open circles represent fitted and experimental data, respectively (without phase shift correction). superior to gibbsite. Sorption maximum of Li/Al LDH-Cl was approximately 6 times higher than that of gibbsite, although the surface area of gibbsite was larger than that of Li/Al LDHCl. Gibbsite comprises stacks of dioctahedral Al (OH)3 layers. There is no net permanent charge on the surface of gibbsite due to no appreciable isomorphous substitution. The OHgroups on the planar surfaces of gibbsite are fully chargesatisfied and relatively inert; therefore, the surfaces do not participate in ligand exchange reactions. The only active sites for sorption reactions are the pH-dependent edges that have an abundance of undercoordinated O atoms which can never be fully charge-satisfied by the addition or removal of a proton (31). Part of pHs employed in this investigation were above pH 5.0, the pHpzc of gibbsite; hence, the edges would be almost satisfied with negative charge, diminishing the possibility of arsenate accessing to gibbsite. Nevertheless, still some arsenate reacted with gibbsite tightly by ligand exchange mechanism as the results of EXAFS analysis, exhibiting a bidentate binuclear complex. In sorption envelope, the inflection point of pH 7.0 divided arsenate sorption on Li/Al LDH-Cl into a pH-sensitive section below pH 7.0 and a relatively pH-insensitive section above pH 7.0. According to EXAFS analysis, arsenate would react with Li and Al of Li/Al LDH-Cl from pH 5.0 to 9.0. However, the contribution of Al to arsenate sorption diminished above pH 7.0 comparing with that at pH 5.0; meanwhile, the interaction between Li and arsenate still maintained at equal degree. These results demonstrated that the Li located in the vacant octahedral sites within the Al(OH)3 sheets existed as permanent sorption sites for arsenate in contrast to Al presented in the edges of Al(OH)3 layers, on which the reaction 7788

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 24, 2006

We thank the National Science Council, Taiwan for financial support (NSC 91-2313-B-002-361, 92-2313-B-002-090, and 93-2313-B-002-008) of nanoparticle projects. In addition, we especially appreciate the Prof. Chiun-Tih Yeh’s (Department of Chemistry NTHU, Taiwan) assistance in processing the surface area measurements of Li/Al LDH-Cl and of gibbsite.

Supporting Information Available The XRD pattern of Li/Al LDH (Figure SI1), XANES spectra of arsenate reacted with Li/Al LDH-Cl and with gibbsite (Figure SI2), and proposed schematic model for arsenate sorbed on Li/Al LDH-Cl at pH 5.0 (Figure SI3) the justification of XAS sample preparation and used EXAFS fitting parameters (Table SI1). This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Smedley, P. L.; Kinniburgh, D. G. A review of the source, behavior and distribution of arsenic in natural waters. Appl. Geochem. 2002, 17, 517-568. (2) Sparks, D. L.; Scheidegger, A. M.; Strawn, D. G.; Scheckel, K. G. Kinetics and mechanisms of metal sorption at the mineralwater interface. In Mineral-Water Interfacial Reactions: Kinetics and Mechanisms; Sparks, D. L., Grundl, T. J., Eds.; American Chemical Society: Washington, DC, 1999. (3) Weerasooriya, R.; Tobschall, H. J.; Wijesekara, H. K. D. K.; Arachchige, E. K. I. A. U. K.; Pathirathne, K. A. S. On the mechanistic modeling of As(III) adsorption on gibbsite. Chemosphere 2003, 51, 1001-1013. (4) Weerasooriya, R.; Tobschall, H. J.; Wijesekara, H. K. D. K.; Bandara, A. Macroscopic and vibration spectroscopic evidence for specific bonding of arsenate on gibbsite. Chemosphere 2004, 55, 1259-1270. (5) Newman, S. P.; Jones, W. Synthesis, characterization and application of layered double hydroxides containing organic guests. New J. Chem. 1998, 105-115. (6) Twu, J.; Dutta, P. K. Structure and reactivity of oxovanadate anions in layered lithium aluminate materials. J. Phys. Chem. 1989, 93, 7863-7868. (7) Besserguenev, A. V.; Fogg, A. M.; Francis, R. J.; Price, S. J.; O’Hare, D. Synthesis and structure of the gibbsite intercalation compounds [LiAl2(OH)6]X (X ) Cl, Br, NO3) and [LiAl2(OH)6]X‚H2O

(8)

(9)

(10) (11) (12) (13)

(14) (15)

(16) (17) (18) (19) (20) (21)

using synchrotron X-ray and neutron powder diffraction. Chem. Mater. 1997, 9, 241-247. Fogg, A. M.; Dunn, J. S.; Shyu, S. G.; Cary, D. R.; O’Hare, D. Selective ion-exchange intercalation of isomeric dicarboxylate anions into the layered double hydroxide [LiAl2(OH)6]Cl‚H2O. Chem. Mater. 1998, 10, 351-355. Tarasov, K. A.; Isupov, V. P.; Chupakhina, L. E.; O’Hare, D. A time resolved, in-situ X-ray diffraction study of the deintercalation of anions and lithium cations from [LiAl2(OH)6]n X‚qH2O (X ) Cl-, Br-, NO3-, SO42-). J. Mater. Chem. 2004, 14, 1443-1447. Serna, C. J.; White, J. L.; Hem, S. L. Hydrolysis of aluminumtri-(sec-butoxid) in ionic and nonionic media. Clays Clay Miner. 1977, 25, 384-391. Serna, C. J.; Rendon, J. L.; Iglesias, J. E. Crystal-chemical study of layered [Al2Li(OH)6]+X-‚nH2O. Clays Clay Miner. 1982, 30, 180-184. Yang, D. S.; Wang, M. K.; Wang, S. L. Synthesis of Li/Al layered double hydroxide-guest composites under mild acid conditions. Clay Miner. 2004, 39, 115-121. Ladeira, A. C. Q.; Ciminelli, V. S. T.; Duarte, H. A.; Alves, M. C. M.; Ramos, A. Y. Mechanism of anion retention from EXAFS and density functional calculations: Arsenic (V) adsorbed on gibbsite. Geochim. Cosmoschim. Acta. 2001, 65, 1211-1217. Dutta, P. K.; Puri, M. Anion exchange in lithium aluminate hydroxides. J. Phys. Chem. 1989, 93, 376-381. Lytle, F. W.; Greefgor, R. B.; Sandstrom, D. R.; Marques, E. C.; Wong, F.; Spiro, C. L.; Huffman, G. P.; Huggins, F. E. Measurement of soft-X-ray absorption spectra with a fluorescencechamber detector. Nucl. Instrum. Methods Phys. Res., A Accel. Spectrom. Detect. Assoc. Equip. 1984, 226, 542-548. Newville, M. AUTOBK: Automated background removal for XAFS data. In AUTOBK program documentation; Consortium for Advanced Radiation Sources: Chicago, 1995. Ankudinov, A. L.; Ravel, B.; Rehr, J. J.; Conradson, S. D. FEFF8.20. Phys. Rev. B 1998, 58, 7565-7576. Inorganic Crystal Structure Database (ICSD); Kurchhof, A., Pebler, A., Warkenten, E., Eds.; Gmelin-Institut fu ¨ r Anorganische Chemie, Springer-Verlag: Berlin, 1991. Fanchon, E.; Vicat, J.; Oui, D. T.; Boudjada. A. Absolute structure of LiH2AsO4. Acta Crystallogr. C 1987, 43, 1022-1025. Newville, M. FEFFIT: Using FEFF to Model XAFS data. In FEFFIT program Documentation; Consortium for Advanced Radiation Sources: Chicago, 1995. Manning, B. A.; Goldberg, S. Modeling arsenate competitive adsorption on kaolinite, montmorillonite and illite. Clays Clay

Miner. 1996, 44, 609-623. (22) Raven, K. P.; Jain, A.; Loeppert, R. H. Arsenite and arsenate adsorption on ferrihydrite: kinetics, equilibrium, and adsorption envelopes. Environ. Sci. Technol. 1998, 32, 344-349. (23) Arai, Y.; Elzinga, E. J.; Sparks, D. L. X-ray absorption spectroscopic investigation of arsenite and arsenate adsorption at the aluminum oxide-water interface. J. Colloid Interface Sci. 2001, 235, 80-88. (24) Liu, F.; De Cristofaro, A.; Violante, A. Effect of pH, phosphate and oxalate on the adsorption/desorption of arsenate on/from goethite. Soil Sci. 2001, 166, 197-208. (25) Stumm, W. The coordination chemistry of the hydrous oxidewater interface. In Chemistry of the Solid Water Interface; John Wiley & Sons: New York, 1992. (26) Hingston, F. J.; Posner, A. M.; Quirk, J. P. Anion adsorption by goethite and gibbsite I. The role of the proton in determining adsorption envelopes. J. Soil Sci. 1971, 38, 176-192. (27) Halter, W. E.; Pfeifer, H. R. Arsenic (V) adsorption onto R-Al2O3 between 25 and 70 °C. Appl. Geochem. 2001, 16, 793-802. (28) Goldberg, S.; Johnston, C. T. Mechanisms of arsenic adsorption on amorphous oxides evaluated using macroscopic measurements, vibrational spectroscopy, and surface complexation modeling. J. Colloid Interface Sci. 2001, 234, 204-216. (29) Ziegler, F.; Scheidegger, A. M.; Johnson, C. A.; Dahn, R.; Wieland, E. Sorption mechanisms of zinc to calcium silicate hydrate: X-ray absorption fine structure (XAFS) investigation. Environ. Sci. Technol. 2001, 35, 1550-1555. (30) Ferraris, G.; Chiari, G. The crystal structure of Na2HAsO4‚7 H2O. Acta Crystallogr. B 1970, 26, 1574-1583. (31) Huang, P. M.; Wang, M. K.; Ka¨mpf, N.; Schulze, D. G. Aluminum hydroxides. In Soil Mineralogy with Environmental Applications; Dixon, J. B., Schulze, D. G., Eds; Soil Science Society of America Inc., Madison, WI, 2002. (32) Gupta, S. K.; Chen, K. Y. Arsenic removal by adsorption. J. Water Pollut. Control Fed. 1978, 50, 493-506. (33) Foster, A. L.; Brown, G. E., Jr.; Tingle, T. N.; Parks, G. A. Quantitative arsenic speciation in mine tailings using X-ray absorption spectroscopy. Am. Mineral. 1998, 83, 553-568.

Received for review June 28, 2006. Revised manuscript received September 14, 2006. Accepted September 25, 2006. ES061530J

VOL. 40, NO. 24, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7789