Surface Complexation of Ferrous Iron and Carbonate on Ferrihydrite and the Mobilization of Arsenic

Surface complexation models are commonly used to predict the mobility of trace metals in aquifers. For arsenic in groundwater, surface complexation models cannot be used because the database is incomplete. Both carbonate and ferrous iron are often present at a high concentration in groundwater and will influence the sorption of arsenic, but the surface complexation constants are absent in the database of Dzombak and Morel. This paper presents the surface complexation constants for carbonate and ferrous iron on ferrihydrite as derived for the double-layer model. For ferrous iron the constants were obtained from published data supplemented by new experiments to determine the sorption on the strong sites of ferrihydrite. For carbonate the constants were derived from experiments by Zachara et al., who employed relatively low concentrations of carbonate. The double-layer model, optimized for low concentrations, was tested against sorption experiments of carbonate on goethite at higher concentration by Villalobos and Leckie, and reasonable agreement was found. Sorption was also estimated using linear free energy relations (LFER), and results compared well with our derived constants. Model calculations confirm that sorption of particularly carbonate at common soil and groundwater concentrations reduces the sorption capacity of arsenic on ferrihydrite significantly. The displacing effect of carbonate on sorbed arsenate and arsenite has been overlooked in many studies. It may be an important cause for the high concentrations of arsenic in groundwater in Bangladesh. Sediments containing high amounts of sorbed arsenic are deposited in surface water with low carbonate concentrations. Subsequently the sediments become exposed to groundwater with a high dissolved carbonate content, and arsenic is mobilized by displacement from the sediment surface.