Multicomponent transport of major cations predicted from binary adsorption experiments

Accurate modeling of multicomponent sorption and transport of major cations in subsurface porous media is a prerequisite for predicting complex environmental processes, such as the movement of trace metals in soils and aquifers. In this study, various cation exchange models were compared in their ability to predict ternary Ca–Mg–Na transport in an acidic soil from binary Ca, Mg, and Na adsorption data. A flow-through reactor technique was used to measure binary adsorption isotherms of Ca, Mg, and Na over wide concentration ranges of the adsorptive and the respective background cations. High-resolution transport experiments were conducted in water-saturated chromatographic glass columns. Three sorption models based on cation exchange equations were compared: a 1-site Gaines–Thomas (1-GT), a 1-site Rothmund–Kornfeld (1-RK), and a 3-site Gaines–Thomas (3-GT) model. Although the fit of adsorption data was clearly improved from the 1-GT to the 1-RK to the 3-GT model, transport predictions were overall not improved compared to the 1-GT model. While predictions by the 1-GT and the 3-GT model were virtually identical, predictions by the 1-RK model were partly improved and partly deteriorated. The most simple 1-GT model, therefore, seems to be adequate for predicting multicomponent transport phenomena involving major cations, however, multi-site models may be useful for predicting transport of trace metals in the presence of several major cations. Regardless of the model used, accurate determination of the cation exchange capacity at the pH conditions of interest is extremely critical in cation transport modeling.