Modelling competitive anion adsorption on oxide minerals and an allophane-containing soil

Can surface complexation constants for anions, drawn from the literature for reference oxides, be combined to describe competitive adsorption in a spodic B horizon sample containing the important adsorbent minerals proto-imogolite allophane and ferrihydrite? To answer this and to derive complexation constants for the corresponding reference oxides, a CD-MUSIC model was used, with arsenate as the sorbing ion. To minimize the interference from competing organic substances, a sample containing little organic matter was used. To describe the adsorption of added arsenate correctly, it was found that competitive interactions from sulphate, silicic acid and phosphate had to be considered. In the model, the specific surface area of singly coordinated AlOH groups of allophane, the sulphate surface complexation constant on allophane, and the total concentration of reactive silicic acid were fitted. All other parameters were fixed using reference oxide values. The results indicated that arsenate, phosphate and silicic acid formed stronger surface complexes on ferrihydrite than on gibbsite or allophane, whereas the reverse was true for sulphate. I conclude that the approach used should provide significant qualitative information on the competitive adsorptive interactions in soils. However, the approach may be impractical for routine simulations and predictions. This is partly due to the uncertainty of the assumption that the properties of allophane and ferrihydrite in real soils can be approximated by those of gibbsite and ferrihydrite synthesized in the laboratory. Another difficulty is that the adsorption of arsenate and phosphate might not reach equilibrium within the limited time of most experiments.