Verification of nuclear codes is an important step in licensing nuclear reactors. For molten salt reactors, the involved physics phenomena are strongly coupled and include those introduced by the movement of liquid fuel that are not present at nominal conditions in solid fuel reactors. This movement of fuel inside and outside the core poses new simulation challenges. In this paper, a benchmark for a graphite-moderated molten salt reactor with a simplified out-of-core model is proposed and studied. The benchmark addresses both neutronics and thermal-hydraulics phenomena, including the delayed neutron precursor drift inside and outside of the active core region, as well as the temperature feedback. As for the thermal-hydraulics, a laminar flow field with conjugate heat transfer, delayed neutron precursor movement, and a simplified heat exchanger is modeled. The benchmark is investigated with the MOOSE tools Griffin and Squirrel, coupled with the MOOSE internal thermal-hydraulics abilities, the Monte Carlo code iMC coupled with OpenFOAM, Nek5000 with a custom point kinetics solver, the coupled neutronics and fluid dynamics code SIMMER with capabilities for severe accident simulations, and the Modelica-based library TRANSFORM. By employing a variety of high- and low-fidelity modeling approaches, a robust comparison across different codes is ensured. OpenMC and Serpent are employed as reference codes to verify the correct implementation of the neutronics. This paper provides a comprehensive comparison of the strengths and weaknesses of the codes and their underlying modeling assumptions. It examines how modeling assumptions affect the steady-state solution and how they propagate into the transient analysis. • Proposes simplified graphite-moderated multy-physics MSR benchmark with out-of-core model. • Couples neutronics, thermal-hydraulics and delayed neutron precursor drift. • Compares 6 multi-physics codes in steady state and during transients.
Pfahl et al. (Wed,) studied this question.