DOE/EMSP--73914-Final

Abstract

The purpose of this research was to provide information to DOE on microbiological and geochemical processes underlying the potential use of dissimilatory metal-reducing bacteria (DMRB) to create subsurface redox barriers for immobilization of uranium and other redox-sensitive metal/radionuclide contaminants that were released to the environment in large quantities during Cold War nuclear weapons manufacturing operations. Several fundamental scientific questions were addressed in order to understand and predict how such treatment procedures would function under in situ conditions in the subsurface. These questions revolved the coupled microbial-geochemical phenomena which are likely to occur within a redox barrier treatment zone, and on the dynamic interactions between hydrologic flux and biogeochemical process rates. First, we assembled a robust conceptual understanding and numerical framework for modeling the kinetics of microbial Fe(III) oxide reduction and associated DMRB growth in sediments. Development of this framework is a critical prerequisite for predicting the potential effectiveness of DMRB-promoted subsurface bioremediation, since Fe(III) oxides are expected to be the primary source of electron-accepting capacity for growth and maintenance of DMRB in subsurface environments. We also defined in detail the kinetics of microbial (enzymatic) versus abiotic, ferrous iron-promoted reduction of U(VI) in the presence and absence of synthetic and natural Fe(III) oxide materials. The results of these studies suggest that (i) the efficiency of dissolved U(VI) scavenging may be influenced by the kinetics of enzymatic U(VI) reduction in systems with relative short fluid residence times; (2) association of U(VI) with diverse surface sites in natural soils and sediments has the potential to limit the rate and extent of microbial U(VI) reduction, and in turn modulate the effectiveness of in situ U(VI) bioremediation; and (3) abiotic, ferrous iron (Fe(II)) drive n U(VI) reduction is likely to be less efficient in natural soils and sediments than would be inferred from studies with synthetic Fe(III) oxides. A key implication of these findings is that production of Fe(II)-enriched sediments during one-time (or periodic) stimulation of DMRB activity is not likely to permit efficient long-term abiotic conversion of U(VI) to U(IV) in biogenic redox barriers designed to prevent far-field subsurface U(VI) migration. Instead our results suggest that ongoing DMRB activity will be required to achieve maximal U(VI) reduction efficiency, and emphasize the need for detailed understanding of patterns of DMRB growth, colonization, and maintenance in physically and chemically heterogeneous subsurface environments in order to predict the effectiveness of subsurface U(VI) bioremediation operations. A final ''capstone'' aspect of experimental work on the project was to examine the potential for sustained coupled Fe(III) oxide and U(VI) reduction in experimental flow-through reactor systems (i.e. sediment columns and ''semicontinuous culture'' systems) that are conceptually analogous to hydrologically-open subsurface environments. The results conclusively demonstrated the potential for sustained removal of U(VI) from solution via DMRB activity in excess of the U(VI) sorption capacity of the natural mineral assemblages as determined in abiotic controls. In addition, the abundance of sorbed U(VI) (a potential long-term source of U(VI) to the aqueous phase) was much lower in the biotic vs. abiotic systems. The latter results agree with other project findings which demonstrated the capacity for G. sulfurreducens to reduce sorbed U(VI). Throughout the project we have developed and refined a variety of reaction-based models of coupled Fe(III) oxide/U(VI) reduction, including a generalized model which accounts for the population dynamics of various respiratory microbial functional groups. These models have been employed in numerical simulations of both batch bench- and field-scale systems. Our progress on this front gives us confidence that such models can be successfully applied to field conditions that required large reaction networks and physical heterogeneity. Other project accomplishments included careful examination of thermodynamic and kinetic aspects of U(VI) adsorption onto Fe(III) oxide surfaces in the presence of competing ligands such as carbonate and phosphate, and theoretical assessment of the influence of solid-to-solution ratio the reactive transport of U(VI) and dissolved inorganic carbon in hypothetical groundwater aquifer materials.