ABSTRACT
The Meuse/Haute-Marne underground research laboratory (URL) operated by the French national radioactive waste management agency (ANDRA) have conducted drift experiments in Callovo-Oxfordian claystone (COx), with extensive characterization of the excavation damaged zone (EDZ). Hybrid finite-discrete element method (FDEM) models were simulated as a proof-of-concept to understand if it can accurately predict the formation of the EDZ, and in doing so would provide insight into its evolution. Two cases were compared in this study, which are differentiated by the orientation of the tunnel relative to the horizontal principal stresses. The first case considered was when the drift was excavated parallel to the maximum horizontal principal stress (σH), and the latter was when the drift was excavated parallel to minimum horizontal principal stress (σh). The preliminary simulations were able to fairly predict the dominance of shear failure in the rock mass and overall fracture geometry around the excavation. However, due to simplifications made to the model leading to the absence of several inherent mechanisms in the experiment (i.e., rock bolts, intermediate principal stress, and confinement near the advancing face), the simulation generally overpredicts the extent of the EDZ.
Increasing global energy demand is a growing issue in modern society that is being addressed by different means of electricity production. In 2017, a quarter of the total energy production in Europe was generated using nuclear power, with France accounting for almost half of the production (Eurostat, 2020). The by-product of spent nuclear fuel is radioactive waste that will need to be properly contained and stored for long periods of time. Deep geological repositories (DGR) have been proposed as a solution for long-term nuclear waste storage in geological clay formations with favourable properties such as low permeability, low molecular diffusion, and excellent radioisotope retention capability (Armand et al., 2013). However, understanding the long-term structural integrity of the DGR must be verified through experiments and numerical simulation for it to be a feasible solution.
