Mineralogical controls on arsenite adsorption onto soils: Batch experiments and model-based quantification.

Abstract The accumulation of arsenic (As) in agrarian soils poses a potential long-term risk to human health, and this accumulation largely depends on the adsorption behavior of As onto soil minerals. This study considered the adsorption of As(III) onto natural soils from the Datong Basin, focusing on the quantification of the adsorption capacities of soil minerals and further the prediction of As(III) adsorption isotherms of the bulk soils. Linear programming calculations show that Fe-bearing minerals, illite, dolomite, and soil organic matter all contribute to As(III) adsorption, on average accounting for 73.9, 11.4, 8.2, and 6.5% of the overall adsorption capacity of soil to As(III), respectively. However, not all the Fe-bearing minerals in soils can adsorb As(III). Evidence from the sequential chemical extractions shows that 90.1% of the soil Fe is associated with silicates (FeSi), while results of the linear programming calculations suggest that FeSi cannot adsorb As(III). Based on the above results, a surface complexation model well predicts the experimental As(III) adsorption isotherms for aeolian and riverine soils. However, the adsorption of As(III) onto lacustrine soils is underestimated in both linear programming calculations and surface complexation modeling. This study highlights the importance of both Fe-bearing minerals and non-Fe minerals for As(III) adsorption and the difference in the adsorption capacity between various soil minerals. It further suggests that more comprehensive considerations are necessary when building a reactive transport model for As(III) in soil systems.