Oxidative Dissolution of Spent Fuel and Release of Nuclides from a Copper/Iron Canister : Model Developments and Applications

Three models have been developed and applied in the performance assessment of a final repository. They are based on accepted theories and experimental results for known and possible mechanisms that may dominate in the oxidative dissolution of spent fuel and the release of nuclides from a canister. Assuming that the canister is breached at an early stage after disposal, the three models describe three sub-systems in the near field of the repository, in which the governing processes and mechanisms are quite different. In the model for the oxidative dissolution of the fuel matrix, a set of kinetic descriptions is provided that describes the oxidative dissolution of the fuel matrix and the release of the embedded nuclides. In particular, the effect of autocatalytic reduction of hexavalent uranium by dissolved H2, using UO2 (s) on the fuel pellets as a catalyst, is taken into account. The simulation results suggest that most of the radiolytic oxidants will be consumed by the oxidation of the fuel matrix, and that much less will be depleted by dissolved ferrous iron. Most of the radiolytically produced hexavalent uranium will be reduced by the autocatalytic reaction with H2 on the fuel surface. It will reprecipitate as UO2 (s) on the fuel surface, and thus very little net oxidation of the fuel will take place. In the reactive transport model, the interactions of multiple processes within a defective canister are described, in which numerous redox reactions take place as multiple species diffuse. The effect of corrosion of the cast iron insert of the canister and the reduction of dissolved hexavalent uranium by ferrous iron sorbed onto iron corrosion products and by dissolved H2 are particularly included. Scoping calculations suggest that corrosion of the iron insert will occur primarily under anaerobic conditions. The escaping oxidants from the fuel rods will migrate toward the iron insert. Much of these oxidants will, however, be consumed by ferrous iron that comes from the corrosion of iron. The nonscavenged hexavalent uranium will be reduced by ferrous iron sorbed onto the iron corrosion products and by dissolved hydrogen. In the transport resistance network model, the transport of reactive actinides in the near field is simulated. The model describes the transport resistance in terms of coupled resistors by a coarse compartmentalisation of the repository, based on the concept that various ligands first come into the canister and then diffuse out to the surroundings in the form of nuclide complexes. The simulation results suggest that carbonate accelerates the oxidative dissolution of the fuel matrix by stabilizing uranyl ions, and that phosphate and silicate tend to limit the dissolution by the formation of insoluble secondary phases. The three models provide powerful tools to evaluate "what if" situations and alternative scenarios involving various interpretations of the repository system. They can be used to predict the rate of release of actinides from the fuel, to test alternative hypotheses and to study the response of the system to various parameters and conditions imposed upon it.QC 20100521