Subsurface iron and arsenic removal for shallow tube well drinking water supply in rural Bangladesh

Subsurface iron and arsenic removal has the potential to be a cost-effective technology to provide safe drinking water in rural decentralized applications, using existing shallow tube wells. A community-scale test facility in Bangladesh was constructed for injection of aerated water (similar to 1 m(3)) into an anoxic aquifer with elevated iron (0.27 mmol L-1) and arsenic (0.27 mu mol L-1) concentrations. The injection (oxidation) and abstraction (adsorption) cycles were monitored at the test facility and simultaneously simulated in the laboratory with anoxic column experiments. Dimensionless retardation factors (R) were determined to represent the delayed arrival of iron or arsenic in the well compared to the original groundwater. At the test facility the iron removal efficacies increased after every injection-abstraction cycle, with retardation factors (R-Fe) up to 17. These high removal efficacies could not be explained by the theory of adsorptive-catalytic oxidation, and therefore other ((a)biotic or transport) processes have contributed to the system's efficacy. This finding was confirmed in the anoxic column experiments, since the mechanism of adsorptive-catalytic oxidation dominated in the columns and iron removal efficacies did not increase with every cycle (stable at R-Fe = similar to 8). RA, did not increase after multiple cycles, it remained stable around 2, illustrating that the process which is responsible for the effective iron removal did not promote the co-removal of arsenic. The columns showed that subsurface arsenic removal was an adsorptive process and only the freshly oxidized adsorbed iron was available for the co-adsorption of arsenic. This indicates that arsenic adsorption during subsurface treatment is controlled by the amount of adsorbed iron that is oxidized, and not by the amount of removed iron. For operational purposes this is an important finding, since apparently the oxygen concentration of the injection water does not control the subsurface arsenic removal, but rather the injection volume. Additionally, no relation has been observed in this study between the amount of removed arsenic at different molar Fe:As ratios (28, 63, and 103) of the groundwater. It is proposed that the removal of arsenic was limited by the presence of other anions, such as phosphate, competing for the same adsorption sites. (C) 2010 Elsevier Ltd. All rights reserved.