Gas and water flow in an excavation-induced fracture network around an underground drift: A case study for a radioactive waste repository in clay rock

The Excavation Damaged Zone (EDZ) surrounding a drift, and in particular its evolution, is being studied for the performance assessment of a radioactive waste underground repository. A specific experiment (called CDZ) was designed and implemented in the Meuse/Haute-Marne Underground Research Laboratory (URL) in France to investigate the EDZ. This experiment is dedicated to study the evolution of the EDZ hydrogeological properties (conductivity and specific storage) of the Callovo-Oxfordian claystone under mechanical compression and artificial hydration. Firstly, a loading cycle applied on a drift wall was performed to simulate the compression effect from bentonite swelling in a repository drift (bentonite is a clay material to be used to seal drifts and shafts for repository closure purpose). Gas tests (permeability tests with nitrogen and tracer tests with helium) were conducted during the first phase of the experiment. The results showed that the fracture network within the EDZ was initially interconnected and opened for gas flow (particularly along the drift) and then progressively closed with the increasing mechanical stress applied on the drift wall. Moreover, the evolution of the EDZ after unloading indicated a self-sealing process. Secondly, the remaining fracture network was resaturated to demonstrate the ability to self-seal of the COx claystone without mechanical loading by conducting from 11 to 15 repetitive hydraulic tests with monitoring of the hydraulic parameters. During this hydration process, the EDZ effective transmissivity dropped due to the swelling of the clay materials near the fracture network. The hydraulic conductivity evolution was relatively fast during the first few days. Low conductivities ranging at 10(-10) m/s were observed after four months. Conversely, the specific storage showed an erratic evolution during the first phase of hydration (up to 60 days). Some uncertainty remains on this parameter due to volumetric strain during the sealing of the fractures. The hydration was stopped after one year and cross-hole hydraulic tests were performed to determine more accurately the specific storage as well as the hydraulic conductivity at a meter-scale. All hydraulic conductivity values measured at the injection interval and at the observation intervals were all below 10(-10) m/s. Moreover, the preferential interconnectivity along the drift disappeared. Specific storage values at the observation and injection intervals were similar. Furthermore they were in agreement with the value obtained at the injection interval within the second hydration phase (60 days after starting hydration). The graphical abstract synthesizes the evolution of the hydraulic/gas conductivity for 8 intervals since the beginning of the CDZ experiment. The conductivity limit of 10(-10) m/s corresponds to the lower bound hydraulic definition of the EDZ and it is demonstrated that EDZ can be sealed. This is a significant result in the demonstration of the long-term safety of a repository. (C) 2014 Elsevier B.V. All rights reserved.