Comparison of laboratory uranium sorption data with 'in situ distribution coefficients' at the Koongarra uranium deposit, Northern Australia

Distribution coefficients derived from laboratory sorption experiments are commonly used to model the migration of long-lived radionuclides in the env ironment. However, it has been suggested that field measurements in natural systems ('in situ distribution coefficients') may provide a more accurate indication of 'true' partitioning coefficients than laboratory experiments. In this paper, the relationship between field and laboratory sorption data for uranium is evaluated, using data from the Koongarra uranium deposit in Northern Australia. An extensive suite of laboratory sorption measurements and in situ partitioning data for U has been obtained at this site. A vali d comparison can only be made when the calculation of field partitioning is based on U in 'accessible' phases (rather than total U in the solid) and U species in true solution (i.e. excluding particles). In this study, access ible U was estimated using a chemical extraction and the results were verif ied using an isotope exchange technique. A satisfactory correspondence betw een field and laboratory partitioning data was obtained when the pH values and partial pressures of CO2 in laboratory sorption experiments were simila r to those found in the field. Under these conditions, the measured laborat ory sorption ratios (Rd) and in-field partitioning values (P-acc) for U at Koongarra were in the range between approximately 1 x 10(3) and 2 x 10(4) m l/g. However, the distribution of U in solid and groundwater phases at Koon garra is extremely heterogeneous. This variability must be taken into accou nt when modelling radionuclide migration at this site. (C) 2001 Elsevier Sc ience Ltd. All rights reserved.