Evaluation of mixed valent iron oxides as reactive adsorbents for arsenic removal

The objective of this research was to determine if Fe(II)-bearing iron oxides generate ferric hydroxides at sufficient rates for removing low levels of arsenic in packed-bed reactors, while at the same time avoiding excessive oxide production that contributes to bed clogging in oxygenated waters. Column experiments were performed to determine the effectiveness of three media for arsenic removal over a range in empty bed contact times, influent arsenic concentrations, dissolved oxygen (DO) levels, and solution pH values. Corrosion rates of the media as a function of the water composition were determined using batch and electrochemical methods. Rates of arsenic removal were first order in the As(V) concentration and were greater for media with higher corrosion rates. As(V) removal increased with increasing DO levels primarily due to faster oxidation of the Fe²⁺ released by media corrosion. To obtain measurable amounts of arsenic removal in 15 mM NaCl electrolyte solutions containing 50 μg/L As(V), the rate of Fe²⁺ released by the media needed to be at least 15 times greater than the As(V) feed rate into the column. In waters containing 30 mg/L of silica and 50 μg/L of As(V), measurable amounts of arsenic removal were obtained only for Fe²⁺ release rates that were at least 200 times greater than the As(V) feed rate. Although all columns showed losses in hydraulic conductivity over the course of 90 days of operation, the conductivity values remained high, and the losses could be reversed by backwashing the media. The reaction products produced by the media in domestic tap water had average As-to-Fe ratios that were ∼25% higher than those for a commercially available adsorbent.