Description

In France, a deep geological repository is considered the final solution for high-level and intermediate-level long-lived nuclear waste (HLW and ILW-LL). The French National Radioactive Waste Management Agency (Andra) is leading the development of the Cigéo project (the French geological disposal facility for HLW and ILW-LL) which will be emplaced in the Callovo-Oxfordian claystone (COx) at a depth of about 500 m.

Long-term safety in deep geological repositories for radioactive waste disposal requires isolating the disposal zones from all access points leading to the biosphere at the closure of the repository. This isolation limits groundwater flow and the migration of radionuclides through the various repository components. Engineered barrier systems ensure this isolation by constructing seals at several locations within the repository.

Another important issue related to the long-term safety of a repository is the impact of the generation and accumulation of gases, mainly due to the anaerobic corrosion of ferrous components within low-permeability host rocks such as claystone. Generated gas may significantly affect the long-term containment functions of engineered barriers through over-pressurization of the repository and alteration of the hydraulic and mechanical properties of the engineered barriers and host rock (Rodwell et al., 1999), with a risk of fracturing if gas pressure build-up is too high.

Seal properties, such as gas permeability and gas entry pressure, play a significant role in gas pressure build-up. This is why the French sealing concept for Cigéo must have two opposing properties: a very low water permeability but allow gas to pass through as easily as possible. To meet this challenge, bentonite-based mixtures are preferred as sealing materials because they combine low hydraulic permeability in a saturated state with enhanced gas permeability. Additionally, bentonite-based mixtures exhibit adequate swelling capacity, allowing them to close gaps at interfaces with other components of the sealing system and recompact the excavation-damaged zone surrounding the emplacement drifts (Sellin and Leupin 2013).

Understanding and modelling the activation of the seal through natural hydration from the surrounding rock mass and gas flow through the seal is paramount to demonstrating the robustness of engineered barrier systems. Task BaSiSS (Bentonite and Sand in Sealing Systems) aims to enhance the knowledge of the processes and mechanisms governing the hydromechanical performance of sealing systems. In particular, BaSiSS focuses on bentonite-sand mixtures with high sand content for the swelling core, which will be emplaced in the Callovo-Oxfordian claystone (COx) formation.

Andra is conducting an experimental program on engineered barrier systems to demonstrate their technical feasibility and performance. Part of this program is being carried out at the main level of the Meuse/Haute-Marne Underground Research Laboratory (MHM URL), located at a depth of 490 m within the COx formation. The GES drift hosts the NSC experiment (an acronym for Noyau de SCellement in French), an ongoing large-scale seal experiment with a swelling core comprising 40% MX-80 bentonite and 60% sand. The NSC experiment has a twofold objective (de La Vaissière and Talandier): to verify the overall performance and to assess the equivalent hydraulic permeability of the seal, as well as the near-field behavior around the seal. A second objective is to characterize the gas performance of the seal.

The insights gained from the acquired NSC experimental data provide a direct description of the processes and mechanisms governing the advective movement of water and gas in bentonite-sand mixtures with high sand content. Experimental data at the sample scale will be used to characterize the hydromechanical behaviour of the compacted bentonite-sand material used at the field scale.

Experimental Data

The NSC’s experimental zone is surrounded by boreholes dedicated to monitoring the hydromechanical behaviour of the rock mass (Figure 1). These boreholes measure pore pressure using multi-packer systems and radial displacements using extensometers. Additionally, some boreholes facilitate the passage of cables from sensors and hydraulic lines between the swelling core and the NRM niche.

The experiment comprises four zones: the injection chamber (Zone 1), the swelling core (Zone 2), the concrete plug (Zone 3), and the watertight drift (Zone 4). Additionally, the experimental zones contain eleven instrumented cross-sections to monitor the evolution of the degree of saturation, pore pressure, total pressure, displacement, deformation, and temperature (Figure 2).

The construction of the swelling core began with the emplacement of a base layer of pellets and pure bentonite powder on the drift’s counter-vault to start construction from a flat floor. Then, five walls, each 1 m thick, were constructed on the base layer using bricks of the bentonite-sand mixture. Six hydration membranes (Sections B, S1 to S4, and G) were placed between the walls of the core, spaced 1 m apart. The voids between the bricks and the rock were filled with pure bentonite powder, pellets, and pieces of bricks.

The hydration of the swelling core started on January 27, 2014. The hydration was carried out sequentially on the six membranes, beginning with the injection into the membrane at Section B. Currently, the swelling core’s degree of saturation has reached an average of 99%. Water injection tests are ongoing, and new hydration phases are planned before conducting gas injection tests to assess the core’s performance.

Approach

The task follows an incremental approach from sample to field scale and is divided into three steps:

Step 1 – Modelling hydration and gas injection on a bentonite-sand mixture sample: This step includes interpretative modelling and calibration of the models adopted by the research teams to characterize the hydromechanical behavior of the core in NSC.

Step 2 – Modelling of a simplified case of NSC: This step is based on the real geometry of NSC but neglects the role of uncertainties inherent to the in-situ experiment, such as interfaces (core/rock, and plug/rock) and hydration membranes. The objective is to test the hydromechanical response of the core using the constitutive models developed in Step 1. Therefore, this step consists of a well-defined benchmark in terms of elements, initial conditions, and boundary conditions to allow intercomparison between the constitutive models adopted by the research teams for the bentonite-sand mixture. This step is divided into two sub-steps:

• Step 2a – Hydration (including drift excavation and seal emplacement)
• Step 2b – Gas injection test

Step 3 – Modelling of NSC: This step represents a realistic case of the NSC, taking into account uncertainties handled by the research teams. The research teams are required to propose approaches to numerically address elements such as interfaces and hydration membranes. Step 3 follows the experiment’s history, including the numerical reproduction of the water injection test. Step 2 begins with interpretative modelling of hydration, followed by the modelling of the performance tests, consisting of blind prediction exercises and subsequent interpretative modelling once data is released.

• Step 3a – Hydration (including drift excavation and seal emplacement)
• Step 3b – Water injection test
• Step 3c – Gas injection test

Participating Groups

• France: Andra (Amphos 21, CIMNE-UPC, EDF, and Lamcube)
• Germany: BGR – GRS/UFZ
• Republic of Korea: KAERI (KAERI and University of Ulsan)
• UK: NWS (Quintessa Ltd)

Further Information

For further information, please contact the task leader, Carlos Plúa.

References

1. de La Vaissière R, Conil N, Morel J, et al (2014) Design and construction of a large-scale sand-bentonite seal in the Meuse/Haute Marne underground research laboratory: NSC experiment. Bundesanstalt fur Geowissenschaften und Rohstoffe (BGR), p 133
2. de La Vaissière R, Talandier J Ten years of hydration of an in situ large scale sealing experiment. In preparation
3. Rodwell W, Horseman S, Harris A (1999) Gas migration and two-phase flow through engineered and geological barriers for a deep repository for radioactive waste. EC/NEA Status Report, European Commission
4. Sellin P, Leupin OX (2013) The Use of Clay as an Engineered Barrier in Radioactive-Waste Management a Review. Clays Clay Miner 61:477–498. https://doi.org/10.1346/CCMN.2013.0610601