Characterization of uranium-contaminated sediments from beneath a nuclear waste storage tank from Hanford, Washington: Implications for contaminant transport and fate

Abstract

Abstract The concentration and distribution of uranium (U) in sediment samples from three boreholes recovered near radioactive waste storage tanks at Hanford, Washington, USA, were determined in detail using bulk and micro-analytical techniques. The source of contamination was a plume that contained an estimated 7000 kg of dissolved U that seeped into the subsurface as a result of an accident that occurred during filling of tank BX-102. The desorption character and kinetics of U were also determined by experiment in order to assess the mobility of U in the vadose zone. Most samples contained too little moisture to obtain quantitative information on pore water compositions. Concentrations of U (and contaminant phosphate—P) in pore waters were therefore estimated by performing 1:1 sediment-to-water extractions and the data indicated concentrations of these elements were above that of uncontaminated “background” sediments. Further extraction of U by 8 N nitric acid indicated that a significant fraction of the total U is relatively immobile and may be sequestered in mobilization-resistant phases. Fine- and coarse-grained samples in sharp contact with one another were sub-sampled for further scrutiny and identification of U reservoirs. Segregation of the samples into their constituent size fractions coupled with microwave-assisted digestion of bulk samples showed that most of the U contamination was sequestered within the fine-grained fraction. Isotope exchange (233U) tests revealed that ∼51% to 63% of the U is labile, indicating that the remaining fund of U is locked up in mobilization-resistant phases. Analysis by Micro-X-ray Fluorescence and Micro-X-ray Absorption Near-Edge Spectroscopy (μ-XRF and μ-XANES) showed that U is primarily associated with Ca and is predominately U(VI). The spectra obtained on U-enriched “hot spots” using Time-Resolved Laser-Induced Fluorescence Spectroscopy (TRLIFS) provide strong evidence for uranophane-type [Ca(UO2)2(SiO3OH)2(H2O)5] and uranyl phosphate [Ca(UO2)2(PO4)2(H2O)10–12] phases. These data show that disseminated micro-precipitates can form in narrow pore spaces within the finer-grained matrix and that these objects are likely not restricted to lithic fragment environments. Uranium mobility may therefore be curtailed by precipitation of uranyl silicate and phosphate phases, with additional possible influence exerted by capillary barriers. Consequently, equilibrium-based desorption models that predict the concentrations and mobility of U in the subsurface matrix at Hanford are unnecessarily conservative.