High levels of arsenic in groundwater and drinking water are a major health problem. Although the processes controlling the release of As are still not well known, the reductive dissolution of As-rich Fe oxyhydroxides has so far been a favorite hypothesis. Decoupling between arsenic and iron redox transformations has been experimentally demonstrated, but not quantitatively interpreted. Here, we report on incubation batch experiments run with As(V) sorbed on, or co-precipitated with, 2-line ferrihydrite. The biotic and abiotic processes of As release were investigated by using wet chemistry, X-ray diffraction, X-ray absorption and genomic techniques. The incubation experiments were carried out with a phosphate-rich growth medium and a community of Fe(III)-reducing bacteria under strict anoxic conditions for two months. During the first month, the release of Fe(II) in the aqueous phase amounted to only 3% to 10% of the total initial solid Fe concentration, whilst the total aqueous As remained almost constant after an initial exchange with phosphate ions. During the second month, the aqueous Fe(II) concentration remained constant, or even decreased, whereas the total quantity of As released to the solution accounted for 14% to 45% of the total initial solid As concentration. At the end of the incubation, the aqueous-phase arsenic was present predominately as As(III) whilst X-ray absorption spectroscopy indicated that more than 70% of the solid-phase arsenic was present as As(V). X-ray diffraction revealed vivianite Fe(II)₃(PO₄)₂.8H₂O in some of the experiments. A biogeochemical model was then developed to simulate these aqueous- and solid-phase results. The two main conclusions drawn from the model are that (1) As(V) is not reduced during the first incubation month with high Eh values, but rather re-adsorbed onto the ferrihydrite surface, and this state remains until arsenic reduction is energetically more favorable than iron reduction, and (2) the release of As during the second month is due to its reduction to the more weakly adsorbed As(III) which cannot compete against carbonate ions for sorption onto ferrihydrite. The model was also successfully applied to recent experimental results on the release of arsenic from Bengal delta sediments.

Massive deleterious impacts to human health are resulting from the use of arsenic-bearing groundwaters in South-East Asia deltas and elsewhere in the world for drinking, cooking and/or irrigation. In Bangladesh alone, a fifth of all deaths are linked to arsenicosis. In the natural and engineered subsurface environment, the fate of arsenic is, to a large extent, controlled by redox potential, pH, as well as total iron, sulfur and carbonate content, via sorption and coprecipitation on a variety of natural and engineered (nano)particles. In the present article, we address: (1) new insights in the sorption mechanisms of As on Fe(II) and Fe(III) nanophases recognized to play an important role in the microbial cycling of As and Fe; (2) artifacts often encountered in field and laboratory studies of As speciation due to the extreme redox sensitivity of the Fe-As-O-H phases; and (3) as a conclusion, the implications for water treatment. Indeed the specific reactivity of nanoparticles accounts not only for the As bioavailability within soils and aquifers, but also opens new avenues in water treatment.