Source: Patricia Fox, Jim Davis, Dan Hawkes
Over the last several decades, the U.S. Department of Energy has been very active in seeking ways to clean up radioactive waste—particularly uranium-related waste—at various former U.S. nuclear weapons and energy sites throughout the country. One of these sites, the Uranium Mill Tailings Remedial Action (UMTRA) site at Old Rifle, Colorado, poses particular problems for remediation, in that much of the contamination there lies underwater, within a shallow alluvial aquifer, where uranium is mostly present in an oxidized U(VI) form. Scientists are looking for ways to fix the uranium in place, decreasing its bioavailability (i.e., its potential hazard to plants or animals). To do this precisely, they need to understand uranium’s complex aqueous speciation—how it evolves and changes molecularly within water as the chemistry of the water changes—as well as create reactive transport models that accurately describe and track what happens to uranium in this environment.
As part of DOE’s Integrated Field Research Challenge and Sustainable Systems Scientific Focus Area research, an ESD team of scientists led by Patricia Fox and Jim Davis, and including Mark Conrad (ESD Geochemistry Department Head), Ken Williams, and Phil Long, recently set out to achieve a new understanding of U(VI) at the Old Rifle UMTRA site. They did this by first generating a “surface complexation” model to describe U(VI) adsorption more accurately than was possible before, and then (as a way of testing this model) by conducting field experiments at Rifle in which a bicarbonate solution was injected into the aquifer.
During the field experiment, dissolved uranium concentrations increased up to 2.6 fold over background concentrations in response to the changing water chemistry, specifically in response to increases in calcium and bicarbonate concentrations occurring as a result of the injected solution. Uranium desorption from aquifer sediments was predicted by the surface complexation model due to the formation of highly stable uranium-calcium-bicarbonate aqueous complexes that form under these chemical conditions. In a highly heterogeneous aquifer such as this one, where aquifer sediments are a mixture of everything from micrometer-sized clay particles to sand and even large boulders, investigators must have some information about the distribution of different grain sizes, in order to accurately predict the mobility of uranium in the aquifer. In this study, they found that the highest concentrations of uranium, both solid-bound and dissolved, were correlated with the finer-grained, micrometer-sized particles.
Although the surface complexation model predicted the general trends in uranium release from sediments, the model overpredicted the dissolved uranium concentrations when chemical equilibrium was assumed. However, the model could accurately predict uranium behavior when a rate limitation was included in their reactive transport model—in other words, when U(VI) desorption was not instantaneous, and the dissolved uranium concentrations observed in the groundwater were controlled by the rate of U(VI) desorption (in addition to the aqueous chemical conditions). This was likely because much of the uranium in the solid phase resided in intragranular pore spaces within the sediments, which are physically isolated from the bulk water.
The results of this study (Fox et al., 2012) demonstrate some of the key factors controlling U(VI) mobility in the Old Rifle aquifer—the aqueous chemistry and heterogeneous size-distribution of aquifer sediments both play an important role in U(VI) mobility. Additionally, uranium desorption may be rate-limited during periods of rapid chemical changes, such as those occurring during bioremediation. These results may have an important impact on the success of various remediation techniques.
Caption: Dissolved U(VI) concentrations over time for a well 1.4 m downstream of the injection at 6.40 m below ground surface (bgs).. Dissolved U(VI) data (symbols) are plotted along with model predictions (lines) using an equilibrium assumption (top) and a rate limitation (bottom).
Fox., P.M., J.A. Davis, M.B. Hay, M.E. Conrad, K.M. Campbell, K.H. Williams, and P.E. Long (2012), Rate-limited U(VI) desorption during a small-scale tracer test in a heterogeneous uranium contaminated aquifer. Water Resources Research, 48, W05512; DOI: 10.1029/2011WR011472.