Autor: Sebastián González Fuentes (Doctoral Researcher, IDAEA-CSIC)

Abstract:

Deep geological disposal is extensively studied for the long-term containment of high-level nuclear waste. A disposal system should guarantee that the waste is not a hazard for the environment and remains isolated over long periods of time (millions or hundreds of thousands of years). Mined repositories are suitable to be implemented, as they rely on the multi-barrier concept to achieve the required degree of isolation. This study focuses on assessing the hydro-chemical-mechanical (HCM) behavior of cementitious materials, mainly used as liners or barrier systems, under repository conditions through the development of predictive models. The objective is to analyze the interaction between a support concrete structure and the surrounding clay rock in a deep waste-disposal facility. The elaborated methodology includes models for the hydro-mechanical (HM) response of materials and the hydro-chemical (HC) interaction at the concrete-rock interface, which will be later integrated into a coupled HCM model. The HM model is developed with the finite element method software CODE_BRIGHT and simulates the response of the concrete structure and surrounding rock. The two-dimensional plane-strain model consists of four stages: equilibration, simulating the excavation and alterations caused by drilling, constructing the support structure, and evaluating the long-term behavior of the system. Displacements, porosity variations and stress distributions are analyzed. The HC modeling focuses on the reactive transport modeling of the concrete-rock system. Retraso-CODE_BRIGHT is used to simulate the geochemical processes. The initial modeling stages considers a simplified one-dimensional system with a restricted number of mineral species involved in the concrete-surrounding rock interaction. Data from earlier studies conducted at the Mont Terri Underground Rock Laboratory are used as the basis for the model, regarding the composition of concrete, the rock materials and pore-water compositions. The variation in pH, mineral precipitation/dissolution, volume fraction of minerals and concentrations of elements as a function of time and space are analyzed. The simulation time is 25 years. Preliminary HM results show a generalized compression state of stress, vertical y-displacements of 2 mm and horizontal x-displacements of 0.8 mm in the concrete structure. It is observed that porosity remains invariable because of the geomechanical effect. In the inner region of the lining, the failure limit is exceeded vertically. The excavation damage zone of the host rock extends 10 cm from the concrete-rock interface in the y-direction and 2 m away to the flanks in the x-direction. Preliminary HC results show significant dissolution of portlandite and calcium-silicate-hydrate (C-S-H) phases in the concrete-rock interface region. This results in an increase in the concrete porosity from 0.15 to a maximum value of 0.28 at the concrete-rock interface, and affects the concrete up to a distance of 1 cm from the interface. Future work will focus on determining an appropriate coupling method that integrates both models into an HCM model. Upscaling of concrete degradation with a correlation between porosity variations and changes in the Young’s modulus will be developed, ranging from the aggregate scale to continuum scale. In addition, model results will be compared with the experimental data obtained from the EURAD-MAGIC project.

Webinar