Modeling Competitive Adsorption of Arsenate with Phosphate and Molybdate on Oxide Minerals

The mobility of As in soils depends on several factors including redox potential, soil mineralogy, pH, and the presence of other oxyanions that compete with As for soil retention sites. We investigated the effects of pH and competing anions on the adsorption of arsenate [As(V)] on a-FeOOH (goethite) and y-AI(OHb (gibbsite). Batch equilibrium As(V) adsorption experiments were conducted with P and MO as competing anions in order to produce single-anion [As(V), P, and MO] and binary-anion [As(V/P and As(V)/Mo] adsorption envelopes (adsorption vs. solution pH). Arsenate and P single-anion adsorption envelopes were similar with substantial adsorption occurring across a wide pH range, including pH values above the points of zero charge of the oxides. Maximum MO adsorption occurred across a narrower pH range (pH 4-6). On both oxides, equimolar P concentrations decreased As(V) adsorption within the pH range 2 to 11, whereas MO decreased As(V) adsorption only below pH 6. The constant capacitance model was used to predict competitive surface complexation behavior between As(V)/P and As(V)/Mo using intrinsic equilibrium constants [Kd, tint)] optimized from single-anion data. In addition, the model was applied using one-site (monodentate) and two-site (monodentate + bidentate) conceptualizations of the oxide surface. The two approaches gave comparable fits to experimental adsorption data and were consistent with competitive adsorption observed in binary adsorption envelopes. and MO has been described in whole soils with a competitive Freundlich-type isotherm equation (Roy et al., 1986), though the applicability of the model was limited to cases where the As/P and As/Mo equilibrium concentration ratios were >20. Barrow (1974) investigated As(V) and P competitive adsorption in soil and found that, though As(V) desorbed some previously adsorbed P, a substantial portion of the bound P was not displaced by As(V). Based on competitive adsorption between As(V) and P on goethite, Hingston et al. (1971) postulated that the goethite surface contains adsorption sites common to both As(V) and P anions, as well as sites that adsorb either one anion or the other. The ability to predict As(V) adsorption in complicated systems such as whole soils will require quantitative information on the adsorption of As(V) by individual soil minerals in the presence of competing anions. THE REACTIVITY of As(V) with individual soil minerals is important in determining the general mobility of As in whole soils. Arsenate is the predominant inorganic species of As under oxidizing soil conditions (Sadiq et al., 1983; Masscheleyn et al., 1991) and is retained in soils by adsorption reactions (Roy et al., 1986; Goldberg and Glaubig, 1988). Important minerals that control the As(V) adsorption capacity of soils include Fe and Al oxides (Jacobs et al., 1970; Livesey and Huang, 1981; Fuller et al., 1993). Investigations of As(V) adsorption on Fe and Al oxides have generated considerable evidence for the formation of inner-sphere As(V) surface complexes (Hingston et al., 1971; Anderson and Malotky, 1979). Direct spectroscopic evidence for innersphere adsorption of As(V) on Fe oxide has been obtained using EXAFS (Waychunas et al., 1993), energy dispersive analysis of x-rays (EDAX) (Hsia et al., 1994), and infrared spectroscopy (Harrison and Berkheiser, 1982; Lumsdon et al., 1984). Though several studies have investigated As(V) adsorption on oxide minerals (e.g., Hingston et al., 1971; Anderson and Malotky , 1979; Leckie et al., 1980; Pierce and Moore, 1982), only a few studies (e.g., Goldberg, 1986; Belzile and Tessier, 1990) have investigated surface complexation modeling as a means of quantifying the surface reactions of As(V). The CCM of the oxidewater interface (Schindler and Gamsjager, 1972; Hohl and Stumm, 1976; Schindler et al., 1976; Stumm et al., 1976, 1980) has been used to investigate As(V) adsorption on soil materials (Goldberg, 1986; Goldberg and Glaubig, 1988). Good agreement has been found between binding constants for As(V) adsorption on amorphous Fe oxyhydroxide derived from field data and those obtained in the laboratory (Belzile and Tessier, 1990). Arsenate, phosphate, and molybdate are tetrahedral oxyanions (Cotton and Wilkinson, 1980) that can compete for adsorption sites on soil mineral surfaces (Murali and Aylmore, 1983). Hingston (198 1) estimated the mean areas occupied by the AsOd, Pod, and Moo4 tetrahedra on the goethite surface to be 0.61, 0.61, and 0.31 nm2, respectively. Competitive adsorption between As(V), P, The objectives of this study were to investigate the ability of the CCM to predict As(V) competitive adsorption on goethite and gibbsite and to evaluate the ability of the model to describe oxyanion adsorption using both monoand bidentate surface complexes. Experimental single-anion [As(V), P, and MO] adsorption envelopes were generated in the laboratory and used for optimizing intrinsic equilibrium constants. The set of constants that were optimized in single-anion systems were then used as fixed parameters to predict competitive adsorption in a separate set of binary As(V)/P and As(V)/Mo experimental adsorption data. MATERIALS AND METHODS