Towards building a unified adsorption model for goethite based on direct measurements of crystal face compositions: I. Acidity behavior and As(V) adsorption

Abstract

We have determined the proton, electrolyte (NaNO3) and arsenate affinity constants on the corresponding surface sites of goethite. An empirical optimization approach was applied using the Charge Distribution Multisite Complexation (CD-MUSIC) model, initially, by simulating an extensive proton adsorption data set obtained from four goethites of specific surface areas (SSAs) spanning from 43 to 89 m2/g. Each goethite showed different crystal face compositions, which were measured previously from independent high resolution electron microscopy observations. Only the capacitance value was left as an additional optimizing parameter. For arsenate adsorption, with the above optimized values fixed, the contributing surface arsenate complexes and their respective affinity constants and charge distributions (CD) were also optimized. A unified set of affinity and CD parameters was obtained, where the only resulting variables among goethites were the capacitance 1, the specific surface area and the crystal face contributions (and thus their reactive surface site densities).It was found that the site density parameter is decisive when data show adsorption maxima, such as in the arsenate adsorption isotherms. The self-consistent modeling approach yielded pKa values of singly-coordinated sites of 8.82, and of triply-coordinated sites of 9.65. Three surface arsenate complexes bound only to singly-coordinated sites were found to describe all experimental data, although only two of them were predominant across the pH and As(V) concentration ranges investigated: a bidentate non protonated (BD) complex, and a protonated monodentate (HMD) one. Charge distribution of surface arsenate complexes was found to significantly influence the optimizations, but increased values of optimized capacitance 1 were found to compensate for previously underestimated surface site density values, to yield almost identical surface affinity and CD parameters.Finally, correlations found between surface site density and capacitance-1 parameters with SSA expanded the applicability of the modeling approach presented, to other goethites with different SSAs than the ones investigated here.