Impact of Irradiation History on Boltzmann/Bateman Coupled Sensitivities: Application to Sodium-Cooled Fast Reactor Subassembly

Uncertainty quantification of neutronics quantities of interest during irradiation needs to be based on reliable sensitivities that are able to correctly describe the fuel depletion and the impact of the input data. The classical approach with the Standard Perturbation Theory (SPT) is not sufficient to obtain sensitivity for the entire set of nuclear data involved in the reactivity loss phenomenon. Recent developments in the APOLLO3® code package allow computing Boltzmann/Bateman coupling at the sensitivity scale with the Depletion Perturbation Theory. This improved functionality helps in quantifying contributions from all nuclear data involved in the depletion process: cross-sections, fission yields, and KERMA, among others. However, its applicability in the case of “real” reactor context often requires restricting the calculation to a single cycle, thus forgetting information from the previous irradiation cycles due to complex reloading patterns that takes place in between. To address this challenge, approximations based on a restricted irradiation history can be employed. This paper demonstrates that the corresponding “partial” sensitivities effectively replicate the global behavior of the reference sensitivities for the majority of nuclear reactions when compared to SPT sensitivities. However, they also result in a significant overestimation of the sensitivity norm for the main heavy-nuclei cross-sections as the unconsidered irradiation time increases. The impact of this bias on reactivity loss uncertainty is quantified, and the primary affected contributors are highlighted.