Predicting Molybdenum Adsorption by Soils Using Soil Chemical Parameters in the Constant Capacitance Model

Abstract

cated that Mo adsorption on calcium carbonate was low, reanalysis of their data indicated that adsorption on The constant capacitance model, a chemical surface complexation this CaCO3 at pH 7.1 was comparable in magnitude to model, was applied to Mo adsorption on 36 samples from 27 soil series selected for variation in soil properties. A general regression adsorption on illite at pH 7.0 and three times as great model was developed for predicting soil Mo surface complexation as adsorption on montmorillonite at pH 7.0 and wellconstants from four independently measured soil chemical characteriscrystallized kaolinite at pH 6.8. tics: cation exchange capacity, organic carbon content, inorganic carMolybdenum adsorption on soils and soil minerals bon content, and iron oxide content. The constant capacitance model has been described by various modeling approaches. was well able to predict Mo adsorption on all 36 soil samples. IncorpoSuch models include the empirical Freundlich, Langration of these regression prediction equations into chemical speciamuir, and Temkin adsorption isotherms (Reisenauer et tion–transport models will allow simulation of soil solution Mo conal., 1962; Theng, 1971; Karimian and Cox, 1978; Phelan centrations under diverse environmental and agricultural management and Mattigod, 1984; Xie and MacKenzie, 1991; Goldconditions without requiring soil specific adsorption data and subseberg and Forster, 1998) and surface complexation modquent parameter optimization. els: constant capacitance model (Motta and Miranda, 1989; Goldberg et al., 1996, 1998), triple layer model (Goldberg et al., 1998; Wu et al., 2001), Stern variable M is both a micronutrient essential for surface charge-variable surface potential model (Mcplant growth and a potentially toxic trace eleKenzie, 1983), and CD-MUSIC model (Bourikas et al., ment, especially for grazing animals. Molybdenum defi2001). Parameters from empirical models are only valid ciencies have been reported for many agronomic crops for the conditions under which the experiment was conthroughout the world (Murphy and Walsh, 1972). Moducted. Surface complexation models are chemical modlybdenum occurs in an anionic form that is readily taken els that utilize defined surface species, chemical reacup by forage plants and can accumulate to levels detritions, mass balances, and charge balances and contain mental to grazing ruminant animals (Reisenauer et al., molecular features that can be given thermodynamic 1973). Cattle grazing in some areas of the San Joaquin significance (Sposito, 1983). Valley of California, especially on legumes, were found Chemical modeling of Mo adsorption at the mineral– to be adversely affected by elevated soil Mo content solution interface has been successful using the constant mainly on alkaline soils (Barshad, 1948). On the other capacitance model (Motta and Miranda, 1989; Goldberg hand, soils in northwest Oregon producing pastures with et al., 1996, 1998) and the triple layer model (Goldberg high Mo content are acid with a pH in the range of 4.4 et al., 1998) for oxides, clay minerals, and soils. In these to 5.3 (Kubota et al., 1967). Legumes are efficient Mo studies, Mo adsorption was described as forming monoaccumulators, especially at high soil pH, whereas most dentate surface complexes. In the triple layer modeling grasses and grain crops do not accumulate Mo to toxic studies, Mo adsorption on Fe and Al oxides, kaolinite, levels (O’Connor et al., 2001). Molybdenum exerts its illite, and two soils was best described by an innertoxic effect on grazing cattle by inducing a copper defisphere adsorption mechanism. Molybdenum adsorption ciency that is especially pronounced in the presence of on montmorillonite, on the other hand, was best desulfur. The adverse effects of high Mo can be mitigated scribed by an outer-sphere adsorption mechanism (Goldby Cu supplementation of the animals (O’Connor et berg et al., 1998). Molybdenum adsorption on titanium al., 2001). Careful quantification of soil solution Mo oxide was described with the CD-MUSIC model using concentrations and characterization of Mo adsorption a mixture of monodentate and bidentate surface comreactions on soil mineral surfaces is needed. plexes (Bourikas et al., 2001). Availability of Mo to plants is affected by a variety Although it is chemically reasonable for the dominant of factors including soil solution pH, soil texture, soil solution species to be dominant on the exchange commoisture, temperature, oxide content, organic matter plex, there is no necessary 1:1 correspondence between content, and clay mineralogy (Reisenauer et al., 1973). solution and surface species (Sposito, 1983). The domiMolybdenum becomes more available with increasing nant Mo species in solution is MoO2 4 over most of the solution pH. The dominant Mo adsorbing surfaces in pH range: pKa1 4.00, pKa2 4.24 (Lindsay, 1979). soil are oxides, clay minerals, and organic matter (GoldAttenuated total reflectance Fourier transform infraberg et al., 1996). Although Goldberg et al. (1996) indired and diffuse reflectance infrared Fourier transform spectroscopy have provided direct evidence for the presUSDA-ARS, George E. Brown Jr. Salinity Lab., 450 W. Big Springs Road, Riverside, CA 92507. Contribution from the George E. Brown ence of both monodentate and bidentate Mo surface Jr., Salinity Lab. Received 29 Oct. 2001. *Corresponding author complexes adsorbed on the surface of amorphous Fe (sgoldberg@ussl.ars.usda.gov). hydroxide (Goldberg et al., 1998). However, these studies are not definitive for typical soil conditions because Published in Soil Sci. Soc. Am. J. 66:1836–1842 (2002).