Review of the properties and uses of bentonite as a buffer and backfill material

Research carried out by SKB on the use and behaviour of bentonite as a buffer and backfill material in a radioactive waste repository has been reviewed. The following research areas have been evaluated: mechanical properties; hydraulic and other transport properties; geochemical properties; thermal properties and resaturation; gas migration; manufacturing and emplacement procedures. This review has shown that SKB has carried out much pioneering and world-leading research on bentonite, particularly with regard to analogue studies, microtextural work and practical manufacturing and emplacement procedures. However, there are a number of subject areas which appear less well addressed than others which require further attention: The extrapolation of experimental results of the mechanical properties of bentonite to repository timescales and repository conditions should be investigated further. There is a need for detailed microstructural analysis of materials as part of experimental programmes. This would enable SKB to build confidence in the interpretations of results and reveal whether the mechanical processes occurring during experimentation truly reflect expectations of the performance of the repository. The large amount of experimental, theoretical, empirical datasets and computer models of the mechanical properties of bentonite need to be collated to form a database which is assessable and relevant to those involved in performance assessment calculations. At present, the valuable results of many excellent research projects on mechanical properties of bentonite buffer are not readily available. There seems to be a relatively poor understanding of the mechanisms of radionuclide diffusion through compacted bentonite. Other international work suggests that diffusion coefficients are much lower than those applied by SKB in its PA work. The importance of surface diffusion to describe diffusion in bentonite for certain chemical species ascribed by SKB is not reflected in similar studies elsewhere and is therefore contentious and open to debate. There seems to be a general lack of integration of analogue studies with modelling and experimental work in terms of a model of the evolution of the chemistry of bentonite pore fluids with time (i.e. analogue evidence demonstrates the importance of mineral dissolution and precipitation, but this is not incorporated into the chemical modelling approach). Some rationalisation of approaches is required for a credible model of bentonite pore water evolution to be created. There is little apparent use of modelled bentonite pore water chemistry for the calculation of radionuclide solubility, sorption and speciation in PA (c.f. approaches used by Nagra, JNC, TVO). Although recent work in part redresses this imbalance, SKB is alone amongst disposal agencies in this approach. There is little published work by SKB on the interaction of bentonite with cement. There is now a growing literature elsewhere on this subject which does not seem to be reflected in studies carried out by SKB. Collaboration with other waste disposal authorities interested in this topic is recommended. It may be necessary to incorporate chemical processes into the current SKB conceptual model of gas transport in buffer and backfill materials. Dissolved or gaseous H2 is generated by Fe corrosion under anaerobic conditions, and this gas may be reactive with several minerals in bentonite, including smectite. Although field tests carried out by SKB suggest that the emplacement of compacted bentonite will not be a problem from the perspective of buffer performance, borehole, shaft and vault sealing, there are apparent problems for block manufacture. The introduction of oil as a lubricant in block manufacture, for example may pose problems for long-term behaviour of the near-field due to the presence of these organic materials. Better manufacturing methods are therefore required