Geochemical controls of the microbially mediated redox cycling of uranium and iron

Abstract

As oxidation state is often a primary determinant of the mobility of uranium in subsurface environments, a comprehensive understanding of the redox cycling of uranium is essential to predict the fate of this contaminant. The potential of iron to serve as both a reductant and an oxidant of uranium suggests that remediation strategies which primarily rely on the reduction of the uranyl ion (U(VI)) to the poorly soluble uranous ion (U(IV)) to immobilize uranium in the solid phase may be either enhanced or hindered depending on the biogeochemical transformations of iron. To identify the geochemical controls of both uranium and iron redox cycling, batch incubations with the model metal-reducing bacterium Shewanella putrefaciens were conducted with either U(VI), ferrihydrite, or both as terminal electron acceptors, and concentrations of dissolved inorganic carbon (DIC), calcium, and silica were varied to alter aqueous uranyl speciation and secondary mineralization pathways of ferrihydrite. The presence of ferrihydrite increased the pseudo-first order rate constant of initial uranium removal from solution in nearly all conditions compared to uranium-only controls due to a combination of abiotic U(VI) reduction by Fe(II) and solid-phase association of uranium with secondary mineralization products of ferrihydrite. Following an initial period of uranium reduction, U(IV) was oxidized by ferrihydrite which led to a rebound in dissolved uranium. Once ferrihydrite was completely consumed by microbial respiration and secondary mineralization, uranium was again removed from solution via reductive precipitation and association with secondary mineralization products. DIC and calcium enhanced the oxidative dissolution of U(IV) solids but did not affect the redox potential of the U(VI)/U(IV) redox couple despite the formation of aqueous uranyl carbonate and calcium-uranyl carbonate species. Instead, removal of dissolved Fe(II) from solution and the accompanied shift in the redox potential of the Fe(III)/Fe(II) redox couple was responsible for the abrupt shift in iron acting as a reductant of U(VI) to an oxidant of U(IV). The removal of dissolved Fe(II) via siderite precipitation at the highest DIC concentration and in the presence of calcium enhanced the oxidation of U(IV), and the presence of silica limited the conversion of ferrihydrite to magnetite and sustained U(IV) oxidation. A kinetic model was developed which could reproduce incubation time series of uranium and iron speciation provided the thermodynamic favorability of abiotic uranium-iron redox cycling was accounted for. The modeling exercise revealed that the non-uraninite U(IV) solids formed in the incubations have a redox potential approximately 84 mV lower than that of amorphous uraninite, highlighting the need for a thermodynamic characterization of non-uraninite U(IV) solids. The results of this study identify dissolved Fe(II) concentration as the primary geochemical control of the role of iron in uranium redox cycling at moderate to high DIC concentrations and emphasize the importance of considering this parameter when designing site-specific, in situ bioremediation strategies.