Bentonite barriers in integrated performance assessment (Benipa)
Most repository concepts for high-Ievel wastes and spent fuel in both crystalline and clay host rocks call for the use of significant amounts of bentonite barriers as a buffer between the waste packages and the host rock. The behaviour of bentonite barriers in the repository environment is influenced by mechanical, hydrological, thermal, chemical and radiological conditions, which are variable in both time and space. Integrated performance assessment needs ta identify and assess all relevant aspects for the performance of this component of the repository system. Benipa (Bentonite barriers in... integrated performance assessment) was a research project, within the fifth framework programme of the European Union, which started in September 2000 and was planned for 36 months. Participants in the project were: two national agencies, which are responsible for national HLW management (Enresa as project coordinator, Spain; NAGRA, Switzerland) and six research centres (GRS, Germany; IRSN, France; NRG, the Nederlands; SCK-CEN, Belgium; VTT, Filand; ZAG, Slovenia). The overall objective of Benipa was to assess the state of the art in the treatment of bentonite barriers in integrated performance assessment, evaluating the capacity and consistency of methods and data available to convincingly justify the capacity of bentonite to perform their assigned safety functions. The project analysed and integrated the numerous developments achieved in the study of the bentonite barriers. Current repository concepts were reviewed and referent cases were selected for modelling. The necessity of using predictive models to evaluate '- potential consequences in performance assessment is well recognised. To this aim, available information on FEPs was compiled and processed in order to evaluate the bases for decisions in integrated performance assessment. The evaluation of the influences of many significant FEPs requires the development and application of conceptual models, comparing results between models of different degrees of abstraction and simplification. Assessment codes at process and integrated level must be analysed and upgrated where necessary. New, more advanced calculations at process and integrated level need to be carried out, in order to verify model performance and to evaluate the sufficiency of available assessment methods. Also, sensitivity and optimisation analyses must be perfomed.