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

Abstract

Mining and processing of uranium ore, followed by improper disposal and aging nuclear waste infrastructure, have left behind a legacy of uranium contamination across the United States. Uranium bioreduction, an in situ bioremediation strategy which promotes microbial reduction of aqueous U(VI) to insoluble U(IV) solids, has proven successful at decreasing groundwater uranium concentrations below regulatory limits. However, iron oxy(hydr)oxides, ubiquitous components of natural sediments, may either enhance or hinder uranium removal from groundwater depending on whether iron acts as a reductant or an oxidant of uranium, making the success of U(VI) bioreduction difficult to predict. This dissertation investigated the geochemical controls of uranium-iron redox cycling using pure culture bioreduction incubations with Shewanella putrefaciens, a model metal-reducing bacterium able to respire both U(VI) and solid Fe(III) as terminal electron acceptors. Across a range of pH and solution compositions, the rate of U(VI) bioreduction was found to be dependent on the concentration of uranyl non-carbonate species, which represent the most bioavailable and least abundant fraction of U(VI). Dissolved Fe(II) concentrations, via changes in the Fe(OH)3/Fe(II) redox potential, determined the extent of oxidative dissolution of U(IV) solids by the poorly crystalline Fe(III) oxy(hydr)oxide ferrihydrite and controlled the net rate of uranium removal from solution. Aluminum impurities in the ferrihydrite structure enhanced the oxidative dissolution of U(IV) solids and preserved ferrihydrite from consumption by microbial respiration and secondary mineralization, suggesting that impurities likely make natural iron oxy(hydr)oxides more effective and longer-lived oxidants of U(IV) than previously suggested. A biogeochemical kinetic model was specifically developed for this work to identify the complex network of biogeochemical reactions in these incubations using an inverse modeling approach. The model that accounted for aqueous uranyl speciation and the thermodynamic favorability of uranium-iron redox cycling was able to reproduce uranium removal rates across the broad range of geochemical conditions tested in this study. Overall, this body of work identifies geochemical parameters which should be considered when implementing U(VI) bioreduction technologies and advances the ability to predict the fate of uranium in geochemically diverse environments.