Coupled THC model of heating and hydration corrosion experiments with compacted bentonite

Document available in extended abstract form only. Unsaturated compacted bentonite is foreseen as a backfill and sealing material in radioactive waste repositories. The strong interplays among thermal (T), hydrodynamic (H), mechanical (M) and chemical (C) processes during the hydration and thermal stages of a repository call for coupled THMC models. Most of the THC modeling work performed in recent years was carried out within the context of performance assessment purposes. On the other hand, many laboratory experiments have been performed on canister corrosion, corrosion-bentonite interactions and concrete-bentonite interactions. There is a clear need to test the models used in performance assessment analyses with laboratory data. PEBS (Long-term Performance of the Engineered Barrier System, EBS) is a European Research Euratom Project which aims at evaluating the sealing and barrier performance of the EBS with time, through the development of a comprehensive approach involving experiments, model development and consideration of the potential impacts on long-term safety functions. The work presented here was performed within the framework of PEBS and deals with multiple-continua THC(m) models for the clay barriers of the EBS. These models were developed by incorporating into existing models: (1) Different types of waters such as free, adsorbed and interlayer waters in clays and different types of pores such as macro-pores, inter-aggregate and intra-aggregate pores in multiple continua models; (2) Mechanical and geochemical couplings to account for the changes in the porosity caused by swelling phenomena by using a state-surface approach; and (3) The role of reactive gases such as O2(g), CO2(g) and H2(g). These multiple-continua THC(m) models have been implemented in INVERSE-FADES-CORE and tested with heating and hydration corrosion experiments performed at the Ciemat facilities to study the corrosion products generated at the canister/bentonite interface under repository conditions and analyse how the corrosion products affect the properties of the bentonite. Some of these experiments were performed already within the NF-PRO project while others are performed within the PEBS project. The experiments have been modeled with 1D finite element grids. A uniform liquid pressure of 500 KPa is adopted at the injection boundary. Hydrodynamic, thermal and solute transport parameters were taken from those calibrated previously from a heating and hydration experiment. A zero vertical displacement is imposed at the hydration boundary. A Cauchy condition was used for the energy equation at the hydration boundary of the cell with a thermal coefficient calibrated equal to 4016 W/deg. C and an external temperature of 25 deg. C. The temperature at the bottom of the cell is fixed at 100 deg. C. A Neuman condition was used for the solute transport according to which solute flux is equal to the product of water flux times solute concentration of inflow water. In the experiment granitic water was used to hydrate the bentonite. Bentonite has an initial porosity of 0.40 and a gravimetric water content of 14% which correspond to a saturation degree of 56.8% and a suction of 1.29.108 Pa. The initial gas pressure is equal to the atmospheric pressure. The diffusion coefficient is assumed the same for the all the species equal to 2.10-10 m2/s and for the Cl- is 9.10-11 m2/s. The initial chemical composition of the bentonite was taken from Fernandez et al. (2001). The chemical system is defined in terms of the following primary species: H2O, H+, Ca2+, Mg2+, Na+, K+, Cl-, SO42-, HCO3- and SiO2(aq). The model accounts for homogeneous reactions (acid-base, aqueous complexation and redox reactions) and heterogeneous reactions such as mineral dissolution/precipitation of calcite, gypsum/anhydrite, quartz and Fe minerals, cation exchange of Ca2+, Mg2+, Na+, K+ and Fe2+ using the Gaines-Thomas convention and surface complexation using three types of proto-lysis sites, SSOH, SW1OH and SW2OH. Iron corrosion is modeled with a constant corrosion rate. The experimental data show a sequence of corrosion products from the interface towards the heater. Several hypotheses have been tested with the model to explain the observed mineral phases. Model results indicate that: 1) There is a sequence of corrosion products, magnetite being the end member; 2) Fe is sorbed by surface complexation; 3) Fe cation exchange is less relevant than Fe sorption; 4) No alteration of bentonite takes place. The THM model was calibrated with transient temperature data and relative humidity data collected at two locations within the bentonite column and water content and dry density data measured at the end of the experiment. The reactive transport model was calibrated with porewater chemical data derived from aqueous extract data. For the most part, simulations agree well with experimental data