PROSPECTS FOR COUPLED MODELLING - A REGULATORY PERSPECTIVE

Clay-based buffer and tunnel backfill materials are important barriers in the KBS-3 repository concept for disposal of spent nuclear fuel in Finland. Significant changes can be expected to occur to the properties of clays, especially during re-saturation and the thermal period. Reactions will occur in response to thermal, hydraulic and chemical gradients, induced by the heat output of the waste and at interfaces between different barriers. Desired properties should be achieved both in the ‘initial’ and ‘target’ states. As Posiva point out (Posiva, 2010, p295): “the performance targets, and requirements and the target properties of the host rock consider those properties of the EBS and host rock that are essential for the safety functions, but their fulfilment can usually be examined only indirectly, by modelling”. This places modelling (conceptual and numerical) in a key role in repository design and safety assessment. Consequently, the status and future prospects for modelling of the following technical areas have been reviewed (Savage, in press): � Non-isothermal (T-H-M-C-B) modelling and the potential for cementation. � The consistency of models. � Swelling pressure. � Cement-bentonite interactions. � Iron-bentonite interactions. � Mechanical (shear) behaviour. � Bentonite erosion. A key issue for all topics is the issue of up-scaling of models and data from laboratory and in situ experiments to the physical and temporal scales of interest to safety assessment (e.g. Figure 1). Models of T-H-M-C-B processes have been developed to address T-H-M or T-H-C problems, but modelling the interaction of M (mechanical) and C (chemical) processes has been very limited, in part due to a lack of understanding of process couplings. Chemical processes have been mainly restricted to cation exchange, surface complexation and the (equilibrium) dissolution-precipitation of minerals other than montmorillonite in the bentonite. Clay dissolution-precipitation reactions have been omitted from most pore fluid evolution and TH-M-C-B modelling studies in the past, mainly due to the limitation of computer codes, but also due to a perception of the potential minor impact of such processes. In future, it is likely that these reactions will be included, bringing greater consistency between the modelling of pore fluid evolution and clay transformation (e.g. illitisation).