Advanced BWR criticality safety part II: Cask criticality, burnup credit, sensitivity, and uncertainty analyses

Abstract

In this study, an analysis on burnup credit for cask criticality safety in BWR spent fuel is conducted. Accurate burnup credit can be used to reduce overly conservative safety margins to increase shipping and storage efficiency while maintaining criticality level within regulatory limits. This analysis is based on advanced lattice depletion models that capture various complexities associated with BWR operation. This paper describes the second part of the two-part study which performs an out-of-core analysis of spent fuel in a transportation/storage cask. The first part of the study (Radaideh et al., 2019) developed the set of depletion models used here. In this paper, the spent fuel compositions resulting from these depletion models are used for cask criticality calculations. Uncertainty quantification of cask keff is performed by combining the uncertainty in isotope inventory, nuclear data, and the statistical sampling in KENO-V.a. The uncertainty in isotopic inventory is quantified by performing a validation analysis by comparing spent fuel compositions calculated by 2D TRITON to experimentally determined spent-fuel assay data for three reactors: Fukushima Daini-2, Cooper-1, and Gundermmingen-A. The validation results demonstrate good agreement for the uranium isotopes as compared to the plutonium isotopes. Also, it was found that the uncertainty in cask keff is dominated by the isotopic uncertainty and can reach about 2500 pcm, and as low as about 1700 pcm. Final results show that axial power profile, axial coolant density, control rod modeling, and the presence of gadolinium in 3D simulations have the largest effects on BWR burnup credit. This implies the need for detailed 3D modeling for accurate BWR burnup credit analysis. In addition, based on the UQ analysis considering both actinide only and actinide and fission products sets, the cask remains subcritical within 2σ for all depletion cases analyzed (C0-C9), even though the cask is assumed to be flooded with water and the lattices are discharged at their peak reactivity.