After nuclear waste is buried in a repository, hydrogeological processes can dissolve, transport, separate, and rearrange radionuclides inside or outside the repository. If fissile material becomes separated from neutron absorbers and precipitates in a far-field geologic formation, a critical mass may be formed. Far-field criticality could greatly increase the dose to the biosphere by releasing highly mobile and radioactive fission products outside all engineered barriers.

The scope of this study is to assess the impact of the spent fuel composition and host rock type on the risk of criticality in the far field. In particular, this study performs neutronics analysis in order to determine the minimum theoretical mass of fissile material needed to achieve criticality in a water-saturated far-field deposition under conservative conditions. Light water reactor spent fuel compositions are determined using ORIGEN-ARP as a function of initial enrichment and burnup for various fuel and reactor types. Different compositions of potential host rocks are considered. For each combination of spent fuel type and host rock, the deposition minimum critical mass is obtained at 200,050 years after fuel discharge by iterative MCNP calculation in the space of initial enrichment and discharge burnup. The results are compared for different types of spent fuels to demonstrate the effects of reactor, fuel, and fuel assembly types on the minimum critical mass. Fissile material from MOX and PWR spent fuels is shown to have a smaller minimum critical mass than that of UO2 or BWR spent fuels due to increased reactivity whereas the assembly type is insignificant. For a fixed fissile content in the deposition, the critical mass increases linearly with spent fuel burnup. For representative compositions of fissile material, perturbation calculations are carried out on the composition of each host rock to identify minerals with a large impact on neutronics. Overall, this work yields insights into how the fuel cycle can be controlled to mitigate or eliminate the risk of far-field criticality. In addition, the results provide a scientific basis for future criticality safety assessment studies and the engineeringinformed decisions that will be required when repository sites are selected.