Nuclear Data Uncertainty Quantification for Reactor Physics Parameters in Fluorine-19-based Molten Salt Reactors

¹ Division of Applied Nuclear Physics, Department of Physics and Astronomy, Uppsala University, Uppsala, 751 20, Sweden
² Seaborg ApS, Titangade 11, 2200 Copenhagen N, Denmark

^* e-mail: wonder@stjarnholm.com

Published online: 17 April 2024

Abstract

The use of the F-19 isotope in the nuclear fuel cycle is already well established for fuel enrichment, but future plans for Gen-IV reactors, such as Molten Salt Reactors, could utilize a fluorine-based salt as a basis for the fuel. It is therefore imperative that an understanding of the characteristics of F-19 is instituted, and one component of key interest is the quantification of reactor parameter uncertainties that arise from the uncertainties in the nuclear data. The results from such analyses can shed light on where experimentalists need to further improve nuclear data for F-19, as well as yielding critical information for developing and optimizing reactor designs thanks to greater knowledge of the uncertainties that result from nuclear data.

In this work, we analysed a molten salt reactor based on the designs made by Transatomic Power. We conducted uncertainty quantification on three reactor operating modes: thermal, semi-epithermal, and epithermal. In the epithermal mode, the neutron spectrum is faster than in the thermal mode because fewer moderator rods are used. We generated nuclear data that was sampled from the covariance matrices in the JEFF-3.3 nuclear data library using SANDY[1] and NJOY. By utilising the Total Monte Carlo approach, we propagated the uncertainties from the samples to uncertainties in the neutron multiplication by simulating the reactor in OpenMC, a Monte Carlo-based neutron transport code. By perturbing individual reaction channels while keeping others constant, it was possible to quantify the amount of contribution each single reaction channel has to the overall uncertainty.

For the thermal reactor, the F-19 data sampling resulted in an uncertainty in reactivity of 62 pcm. The main contributors to the reactivity uncertainty for the thermal reactor are elastic scattering, neutron capture and alpha production. The epithermal reactor, with a reactivity uncertainty of 213 pcm, is mostly affected by elastic scattering, inelastic scattering, and alpha production. The alpha production channel had an unexpectedly large contribution, and it should be investigated further. The results should be considered preliminary. Quantitatively, we observe that scattering plays a bigger role for the uncertainty in the epithermal system, a phenomenon which could be explained by the fact that with less moderation in the form of moderator rods, the role of F-19 in slowing down neutrons is greater, and hence its contribution to the uncertainty is greater.