Boom clay pore water geochemistry at Mol site - State of the art

Document available in extended abstract form only. Boom Clay and Ypresian clays are investigated in the framework of the Belgian research program on the long term management of high-level and/or long-lived radioactive waste coordinated by ONDRAF/NIRAS for their potential to host a deep geological disposal repository for radioactive waste. The Boom Clay pore water geochemistry has been studied for more than two decades with the main objective to better understand the mechanisms controlling its composition. The earlier attempts to determine the pore water composition employed ex-situ batch leaching, mechanical squeezing and in-situ piezometric water sampling techniques. These studies have demonstrated that (1) the ex-situ pore water sampling techniques are prone to artifacts, so do not provide representative pore water samples, (2) in-situ pore waters from piezometers are the least affected by sampling artifacts and are therefore considered as the most representative of the real Boom Clay conditions, (3) numerous pore water samples from piezometers over large scale are needed to study the variability of the Boom Clay geochemistry, (4) to unambiguously interpret a pore water composition, it is of paramount importance to determine reliably in-situ pH and pCO2 in the same pore water sample series. To provide a sufficient amount of high quality pore water samples, a piezometric network was installed around the HADES-URF. The architecture of the network allows to evaluate the variations in the Boom Clay pore water composition in both vertical and horizontal direction. The results show that the Boom Clay pore water composition is not constant at the formation scale. The concentrations of the two most abundant species (Na and HCO3-) decrease in the Boeretang Member (top Boom Clay) in contrast to relatively stable values in the Putte Member. Small variations of these concentrations were also detected at the boundary between the Putte and Boeretang Member. An abrupt increase of most elements at the base of the Putte Member reflects the presence of the 'double band', i.e. two coarser (silty) layers at this location. A coherent geochemical model was proposed by De Craen et al. (2004) to account for the pore water chemistry observed in the MORPHEUS piezometer (installed vertically downward from HADES URF). The model, assuming mineral dissolution/precipitation and cation exchange reactions as the principal processes, could reasonably well reproduce the observed data ranges. However, due to technical difficulties to measure reliably in situ pH and pCO2 at the time, these parameters were fitted by the model. In addition, the observed data ranges covered only a limited part of the Boom Clay and thus were not necessarily representative of the whole Boom Clay Formation. Recent progress in the optimisation of the sampling and measurement procedures led to the installation of a new circulation system around the PRACLAY gallery and an in-house routine application of gas analyses. In addition, a systematic screen for microbial activity became an integral part of the sampling campaigns in the piezometers around PRACLAY to verify the absence of microbial interference with the observed geochemical data. Today, we are able to probe in situ pH, pCO2 and pore water chemistry simultaneously. In this context, we can test, improve and further develop the present Boom Clay pore water geochemical model. Also, the use of additional piezometers for pore water sampling provides complementary data on larger scale thus better representing the overall natural variations in the Boom Clay pore water at Mol site. The modelling effort focuses on explaining the variations in the pore water chemistry observed in the Boom Clay at the formation scale. In-situ measured pH and pCO2 will serve as independent controllers to verify the robustness of the updated model. (authors)