Modeling Arsenic(III) Adsorption and Heterogeneous Oxidation Kinetics in Soils

Arsenite [As(III>] is a soluble and toxic species of arsenic that can be introduced into soil by geothermal waters, mining activities, irrigation practices, and disposal of industrial wastes. We determined the rates of As(III) adsorption, and subsequent oxidation to arsenate | As(V)| in aerobic soil-water suspensions using four California soils. The rate of As(III) adsorption on the soils was closely dependent on soil properties that reflect the reactivity of mineral surfaces including citrate-dithionite (CD) extractable metals, soil texture, specific surface area, and pH. Heterogeneous oxidation of As(III) to As(V) was observed in all soils studied. The recovery of As(V') from As(III)treated soils was dependent on levels of oxalate-extractable Mn and soil texture. After derivation of rate equations to describe the changes in soluble and recoverable As(III) and As(V) in soil suspensions, soil property measurements were used to normalize the empirically derived rate constants for three soils. The fourth soil, which had substantially different soil properties from the other three soils, was used to independently test the derived soil property-normalized model. The soil property-normalized consecutive reaction model gave a satisfactory description of the trends seen in the experimental data for all four soils. Understanding the effects of soil properties on the kinetics of chemical reactions of As(III) and As(V) in soils will be essential to development of quantitative models for predicting the mobility of As in the field. T OXIDATION of trace metals and metalloid oxyanions by soils has an important role in determining their mobility and toxicity. Reduced species such as As(III), Se(IV), and Cr(III) can be oxidized in soils to produce As(V), Se(VI), and Cr(VI), respectively. In the case of Se and Cr, the oxidized Se(VI) and Cr(VI) species are less strongly adsorbed to soils than Se(IV) and Cr(III) and thus are more mobile and bioavailable. The oxidation of As(III) produces As(V), which is strongly adsorbed in soils and is less toxic than As(III) (Knowles and Benson, 1983). Despite the fact that the reactions of As(III) with soil have been studied, the rates and mechanisms of As(III) adsorption, as well as oxidation to As(V), are still not well understood. The As(III) species can also be present in groundwater (Korte and Bruce A. Manning, Dep. of Chem. and Biochem., San Francisco State Univ., 1600 Holloway Ave., San Francisco, CA 94132; and Donald L. Suarez, USDA-ARS, U.S. Salinity Lab., 450 W. Big Springs Rd., Riverside, CA 92507-4617. Contribution from the USDA-ARS U.S. Salinity Lab. Received 19 Aug. 1998. * Corresponding author ([email protected]). Published in Soil Sci. Soc. Am. J. 64:128-137 (2000). Fernando, 1991), and therefore the stability of As(III) after coming in contact with aerobic soils is of interest in environmental management. Previous work on the reactions of As(III) with soil has focused primarily on As(III) adsorption rather than As(III) oxidation (Manning and Goldberg, 1997a; McGeehan and Naylor, 1994; Elkhatib et al., 1984). Iron oxides have been shown to be the most important mineral component in determining a soil's overall capacity to adsorb As(III) and As(V) (Jacobs et al., 1970; Fordham and Norrish, 1974, 1979; Elkhatib et al., 1984; Livesey and Huang, 1981; Manning and Goldberg, 1997a). It has also been concluded, using x-ray absorption spectroscopy (XAS), that As(V) is specifically adsorbed on synthetic Fe(III) oxide mineral surfaces (Waychunas et al., 1993; Fendorf et al., 1997). Electrophoretic mobility work (Pierce and Moore, 1982) has shown that As(III) is also specifically adsorbed on the Fe(III) oxide surface, and this has now been confirmed using Fourier transform infrared spectroscopy (Sun and Doner, 1996) and XAS (Manning et al., 1998). The oxidation of As(III) to As(V) in soils has not been extensively studied despite the fact that this is an important reaction in the cycling of As in the environment. The oxidation of As(III) by lake sediments has been investigated (Oscarson et al., 1980, 1981) and it was concluded that an abiotic process involving Mn oxide minerals was directly responsible for As(III) oxidation. Heterogeneous oxidation of As(III) by synthetic Mn oxides (Oscarson et al., 1983; Scott and Morgan, 1995; Sun and Doner, 1998), and clay minerals (Manning and Goldberg, 1997b) has been shown, though very little is known about the rates or mechanisms of As(III) oxidation in whole soils. An improved understanding of the rate of As(III) oxidation in whole soil is necessary for the application of predictive models to describe As transport in the field. In addition, linking measurements of important soil properties with a quantitative description of As(III) adsorption and oxidation would improve the predictive capability of reactive transport models. Given the need for a better understanding of the Abbreviations: CD, citrate-dithionite; DI, deionized; HPLCHGAAS, high performance liquid chromatography-hydride generation atomic absorption spectrometry; MeT, total citrate-dithionite extractable metals (Al + Fe + Mn); Mn0x, oxalate-extractable manganese; SA, specific surface area; XAS, x-ray absorption spectroscopy; XRD, x-ray diffraction. MANNING & SAUREZ: MODELING AS(HI) ADSORPTION AND HETEROGENEOUS OXIDATION KINETICS 129 oxidation and adsorption rates of As(III) in soils, the objectives of this study were: (i) to determine the As(III) adsorption and oxidation rates in soil by measuring the As(III)-As(V) speciation and fractionation in As(III)treated soil as a function of time; (ii) to develop a simple, verifiable kinetic rate law for describing the time-dependent reactions of As(III) in soil for eventual use in solute transport models; and (iii) to incorporate soil property measurements into kinetic expressions to improve their general applicability to different soils.