Transient Neutronic behavior in Molten Salt Reactors

Molten salt reactors are emerging as a promising alternative to conventional nuclear reactor technologies due to their potential for improved fuel efficiency, enhanced safety features, and operational flexibility. Key advantages include higher fuel utilization, strong negative temperature feedback, higher operation temperatures, and lower operating pressure. Despite these advantages, the development of molten salt reactors is still in the early stages, and significant challenges remain. To accurately model molten salt reactors, simulation tools must account for the complex interactions between neutron transport and fluid dynamics. This includes modeling the movement of delayed neutron precursors and temperature effects on reactivity, which arise from the thermal expansion of the liquid fuel and the Doppler effect. The Point Kinetics equations allow for a fast and reliable way to simulate transients. For flowing-fuel systems, these equations have to be modified to account for the movement of delayed neutron precursors. This work investigates the limitations of Point Kinetics to describe transients of molten salt reactors within multiphysics simulations. To accurately model the impact of the moving delayed neutron precursors on the reactor kinetics, the adjoint flux is used to weigh their importance, allowing for the determination of their spatially weighted contribution to the kinetics at each time step. The temperature effect on the reactivity is calculated by weighting the spatial temperature change with the adjoint flux. To allow for the coupling to a high-fidelity thermal-hydraulics solver and to enable the direct comparison to codes with full spatial resolution, a custom solver is implemented in the MOOSE (Multiphysics Object-Oriented Simulation Environment) framework. This solver expands the open-source capabilities of the MOOSE framework to investigate transients in molten salt reactors under consideration of flowing fuel. The implemented solver is tested on the frequency response in the Centre National de la Recherche Scientifique benchmark, a 2D model of a pool-type fast molten salt reactor, and an unprotected loss of flow in a novel benchmark for thermal molten salt reactors with an outer core model. It is thereby verified against spatially resolved codes included in the Centre National de la Recherche Scientifique benchmark, Griffin, the severe accident code SIMMER, the transient Monte Carlo code iMC, and coarser system codes based on Matlab and Modelica. This verification ensures that the spatial Point Kinetics solver can reproduce transient behavior with comparable accuracy to spatially resolved neutronics solvers. The validation against experimental data obtained from the molten salt reactor experiment was done by reproducing a reactivity insertion in a 1D single-channel model, and the pump start, coast-down transients, and the power evolution of the natural circulation transient of the molten salt reactor experiment in a more detailed 2D model. In particular, it is shown that the effect of delayed neutron movement during startup and the temperature feedback effects during the natural circulation transient are accurately modeled. These comparisons further strengthen the confidence in the spatial Point Kinetics solver’s ability to handle complex transient scenarios in molten salt reactors. Within the same 2D model and the same fluid dynamics model of the Molten Salt Reactor Experiment, the presented solver was able to reproduce the loss of reactivity due to delayed neutron precursor movement calculated with Griffin up to 2.75 pcm and calculate the power evolution in the natural circulation transient accurately with an average error of 0.3%. This research expands the applicability of Point Kinetics to molten salt reactors in 1D, 2D, and 3D and shows that spatially dependent Point Kinetics can accurately model reactor transients with runtime improvements in the order of two orders of magnitude. This contribution paves the way for faster design iterations of molten salt reactors and more efficient safety analyses. Showing that spatial Point Kinetics is a valuable tool for the future of molten salt reactor development. Within this thesis, it was not possible to determine limits to the Point Kinetics formulation that are specific to molten salt reactors, but it was not possible to rule out that such transient scenarios might exist. This work demonstrates that the spatial Point Kinetics solver can accurately reproduce the transients of molten salt reactors, achieving results comparable to those obtained from more computationally expensive reference codes.