UNDERSTANDING THE THMC EVOLUTION OF BENTONITE BARRIER — MODELING AN IN SITU TEST FOR BENTONITE BACKFILLED ENGINEERED BARRIER SYSTEM

The most common buffer material for engineered barrier system (EBS) is compacted bentonite, which features low permeability and high retardation of radionuclide transport. The safety functions of EBS bentonite include limiting transport in the near field; damping the shear movement of the host rock; preventing the sinking of canisters, limiting pressure on the canister and rock, and reducing microbial activity. To assess whether EBS bentonite can maintain these favorable features when undergoing heating from the waste package and hydration from the host rock, we need a thorough understanding of the thermal, hydrological, mechanical, and chemical evolution of bentonite under disposal conditions. The FEBEX (Full-scale Engineered Barrier EXperiment) in situ test was dismantled after 18 years’ heating and hydration. The comprehensive THMC data obtained in the test provide a unique opportunity to validate coupled THMC models and deepen our understanding of the THMC evolution in bentonite. In this presentation, coupled THMC models were developed for the in situ test. Water content data obtained after dismantling and relative humidity data measured real time showed that the hydration of bentonite is slower than predicted by the typical Darcy flow model. Including Non-Darcian flow into the model however leads a significant underestimation of the relative humidity data. The reason could be that the calibration of relative permeability (and retention curve) already encompasses the nonlinear relationship between gradient and flux for bentonite, which would obviate the consideration of Non-Darcian flow in the model. THMC models that take into account the porosity and permeability changes due to mechanical processes match reasonably well all the THM data. However, they did not provide a desirable fit of the measured Cl concentration profile, further calibration of porosity/permeability changes over the course of hydration and swelling or considering other coupled process such as thermal osmosis are needed for the model to sufficiently explain all the THMC data. Model results also showed that transport processes, i.e. advection and diffusion, control the concentration profile of conservative species (Cl for example) and play a major role in shaping the profile of most reactive species except pH and bicarbonate.