Uncertainty quantification in fluoride-based molten salt reactors with open sourced tools

In this study, nuclear data uncertainties originating from the JEFF-3.3 evaluations of fluorine-19 are systematically examined, with a focus on their impact on steady-state neutronics calculations for a thermal fluoride salt-based Molten Salt Reactor (MSR). The two primary objectives of this article are the quantification of uncertainties in reactivity and power distribution, and the establishment of an entirely open-sourced code suite for educational purposes. The reactor under consideration is based on the TAP design from Transatomic Power, representing a fluoride salt MSR. The uncertainty propagation is investigated with a pseudo Total Monte Carlo approach: utilizing SANDY, perturbed PENDF files are generated from the evaluated data, and are further processed with PyNJOY2016. Lattice-level calculations employ the DRAGON5 code to produce few-group cross sections, which, in turn, are used in full-core simulations performed with the KOMODO nodal diffusion code. The data transfer between the codes is implemented in Python. While the primary focus in this work is on demonstrating the functionality of the suite, nuances such as assembly discontinuity factors and optimal group structures are intentionally kept limited. The results demonstrate that uncertainties linked to fluorine-19 nuclear data remain modest for thermal systems, with their significance becoming more notable in epithermal systems.