URANIUM SERIES ISOTOPIC STUDY OF FRACTURE INFILL MATERIALS FROM EL BERROCAL SITE, SPAIN

In order to establish the natural radionuclide distribution in the min eral phases of the fracture infill materials at El Berrocal Site a num ber of water-bearing horizons have been selected. The infill material from these horizons has been characterised mineralogically and radiome trically in terms of U/Th isotopes. Corresponding fracture waters were sampled and also analysed for U/Th isotopes with the aim of setting u p an input radiometric data base for a rock/water interaction model. T he fracture infill materials are U-rich with concentrations ranging fr om 20 to 400 ppm. Thorium concentrations are typically 5 to 15 ppm. Th e corresponding fracture waters have high U contents (5 to 107 ppb), l ow to moderate U-234/U-238 activity ratios (1.2 to 3.4) and very low T h content (<0.005) ppb). The fracture infill material 'surfaces' have U-234/U-238 activity ratios greater than unity indicating 'recent' U u ptake from the fracture water. A sequential leaching procedure has rev ealed that in some cases the U-234/U-238 activity ratios in the solid 'surfaces' are identical to those in the fracture water indicating iso topic equilibrium between the two phases with respect to U. A mass-bal ance model has been applied to the U/Th isotope data to estimate times cales for the observed rock/water exchange processes, and retardation factors for U and Th assuming dynamic equilibrium. Thus, the Th sorpti on rate constants are all greater than or equal to 10(6) yr(-1), the d esorption rate constants are low (<10(2) yr(-1)), and the correspondin g precipitation rate constants are all small. Uranium retardation fact ors range from 10(2) to 10(4) (comparable to some laboratory measured values and model predictions in other geochemical contexts). The rock/ water interaction times are of the order of several to hundreds of yea rs consistent with the conceptual model timescales for the El Berrocal site fracture waters. Five carbonate samples were analysed for U and Th isotopes. They were subjected to a mild leaching procedure to separ ate a 'pure' carbonate phase from the residue. A binary mixture pseudo -isochron dating technique was then applied to the data to estimate ca lcite 'ages' assuming a geochemically closed system. The estimated age s, corrected for the possible leaching of Th from the residue, range < 64 ka to 190 ka. Other likely geochemical scenarios such as open-syst em conditions allowing continuous carbonate accumulation or intermitte nt episodes of carbonate growth and dissolution should yield much shor ter timescales consistent with the timescales predicted by the mass-ba lance rock/water interaction model.