The current concept for a deep geological disposal of intermediate-level long-lived (ILW-LL) and high level (HLW) radioactive waste in France, called Cigéo, is based on the emplacement of exothermic waste packages in long parallel micro-tunnels drilled from access tunnels. The heat emitted by the HLW provokes the temperature rise within the host formation and its surrounding layers. In a water saturated porous medium with low permeability such as Callovo-Oxfordian claystone (COx), this temperature rise causes a pore pressure increase essentially due to the difference between the thermal expansion coefficient of pore water and of solid skeleton. The design of the HLW disposal sections for Cigéo is therefore intended to ensure that there is no additional diffuse fracture or damage at the scale of the COx formation due to transient thermo-hydro-mechanical (THM) loading.
Understanding the migration of gas produced by corrosion of metals, microbial degradation and radiolysis of water within the COx is also of importance for assessing the performance and long-term evolution of the repository. If the rate of gas production exceeds the rate of diffusion of dissolved gas in the pore-water of the host rock and the engineered barriers, the pressure build-up related to the gas phase may overcome the hydraulic pressure and the capillary resistance of the surrounding rock. This may lead to the damage of the rock mass and could affect the safety function of the host rock. The evacuation capacity of the gas by the processes described above and the kinetic of the gas pressure rise control both the creation of tensile fractures and their extension.
Task A addresses both areas related to the fluid pressurization within the COx and its resulting fracturing. Based on in-situ experiments conducted by Andra at the Meuse/Haute-Marne Underground Research Laboratory (MHM URL) in France, the understanding of fundamental processes and the improvement of capabilities of numerical models will be enhanced and will help the design and optimization of the repository. It will also contribute to a robustness demonstration that these processes will not occur at the repository scale.
Experimental data at the lab and at the field scale are available for the characterization and the numerical reproduction of the thermo-hydro-fracturing process in the COx induced by thermal heating. The lab tests consisted in cylindrical samples subjected to different thermal loads. A novel triaxial apparatus was developed for these thermal extension tests (Braun, 2019); pore pressure, total stresses, and axial strains were recorded until the specimens reached their tensile strength and failed. The lab tests complement an in-situ heating experiment conducted at the MHM URL, the so-called CRQ experiment. The experimental set-up consists of ten 20 m long heating boreholes, drilled horizontally at the GCS drift wall: eight peripheral boreholes arranged in a square of 3 m and two central ones. Heating devices were installed at the last 10 m. Given the in-situ stress field at the MHM URL level and the heating design, the fracture location was expected to be between the two central boreholes. To limit drainage zones, there are only few measurement boreholes around to monitor the heated area, two pore-pressure and temperature measurement boreholes inside the square. Located outside the square, there are two pore pressure and temperature measurement boreholes and four boreholes for acoustic measurements and seismic velocities to identify the occurrence of fracture and its location. The heating strategy consisted of two phases separated by a cooling phase reaching temperature values of about 90 °C in the COx. Further analyses to confirm the fracture are planned to be carried out through horizontal boreholes drilled in the heated zone.
Several gas injection tests (PGZ) were performed at the MHM URL through packed off sections of two 28 m long boreholes in order to study the gas migration mechanisms into the COx (De La Vaissière et al., 2019). The gas injection tests (nitrogen) were performed at various flow rates from ~1 mLn/min to ~500 mLn/min in parallel inclined boreholes (PGZ1201 and PGZ1202) drilled at the GMR drift wall. These boreholes were equipped with a multiple packer system to monitor water/gas pressure in three isolated intervals. A third borehole (PGZ1031) was drilled at the GEX drift wall and was equipped with a multiple magnetic extensometer probe to measure potential axial deformation. New tests are planned to be performed in two new boreholes (PGZ1002 and PGZ1003) drilled at the GEX drift. These two boreholes are 35 m long, horizontal and oriented parallel to the in-situ principal stresses at the MHM URL level.
The task is divided into two steps that will be undertaken in parallel. Step 1 addresses the thermo-hydro-fracturing process and Step 2 addresses gas-hydro-fracturing. Both steps have similar structure: (1) definition of conceptual models, (2) numerical reproduction of an in-situ experiment and (3) application at the repository scale.
Step 1.1: Defining THM models for tensile failure and modelling of thermal extension tests.
Step 2.2: Modelling of the CRQ experiment: (a) Interpretative modelling of the 1st heating and cooling phases, (b) blind prediction of the 2nd heating phase, and (c) complete reproduction.
Step 2.3: Application at the repository scale.
Step 2.1: Developing conceptual models of gas flow injection through benchmarking exercises in plane strain, axisymmetric and three-dimensional domains.
Step 2.2: Modelling of the PGZ experiment: (a) Interpretative modelling of gas injection test at high rate in PGZ1002, (b) blind prediction of gas injection test at low rate in PGZ1002 and at low and high rates in PGZ1003, and (c) complete reproduction.
Step 2.3: Application at the repository scale.
For further information, please contact the task leader, Carlos Plúa.