Lawrence Berkeley National Laboratory

Salt formations are a promising host rock for the disposal of heat-emitting nuclear waste due to their ability to heal fractures, high thermal conductivity, and near-zero permeability. This study investigates the concept of waste emplacement in salt formations by placing waste packages in drifts and backfilling them with crushed salt. The aim is to minimize fracturing and restore the initial properties of the host rock. One of the key issues in assessing the safety and performance of nuclear waste repositories in salt is determining the reconsolidation rate of the crushed salt backfill, which has been traditionally predicted using the C-WIPP model. However, this model was calibrated using data from uniaxial oedometer lab tests, where specimens with high initial porosity values have been compacted to medium porosity values. The observed compaction behavior has then been extrapolated to the low porosity range, leading to a prediction of rapid full reconsolidation of the backfill in less than 20 years. In contrast, the new C-WIPP/TUC model used in this study was calibrated for the medium porosity range, between 17% and 8%, based on more suitable triaxial lab tests. Using coupled Thermo-Hydro-Mechanical (THM) simulations with the TOUGH-FLAC simulator, it was found that the C-WIPP/TUC model predicts a slower reconsolidation rate of the backfill compared to the C-WIPP model, with an average porosity of the backfill remaining roughly 10 times that of natural salt after 10,000 years. The effects of the relatively slow reconsolidation of the crushed salt backfill on the THM behavior of the repository are also examined, highlighting the importance of accurately capturing the behavior of crushed salt for the study of the long-term integrity of a nuclear waste repository in salt.