Conference Agenda

This paper presents the development and validation of a high-fidelity thermal-hydraulic model of a molten salt natural circulation flow loop, designed for integration into a digital twin framework. This paper compares OpenFOAM and Idaho National Laboratory’s code Pronghorn against experimental data from Texas A&M University’s (TAMU) Molten Salt Flow Loop (MSFL).

Validation includes three natural circulation test cases: two-dimensional single-phase, two-dimensional with bubble injection, and three-dimensional single-phase flows. Key figures of merit include qualitative flow profile, accuracy of steady state temperature prediction, and computational efficiency for assessing the codes’ performance.

Preliminary OpenFOAM and Pronghorn results for two-dimensional single-phase agree qualitatively with flow profile. Properly accounting for the experiment’s unaccounted heat losses and the high computational cost have been the biggest obstacle to full validation. Additionally, Pronghorn’s PIMPLE algorithm is under rapid development, with heat-flux boundary conditions to be added in the near future. Initial three-dimensional single-phase models currently exhibit prohibitively-high runtimes and computational costs.

Before the final submission, we will improve model performance to adequately simulate the steady-state single phase models in two and three dimensions, as well as endeavor to implement the proper bubble boundaries. Future work will explore complexity reduction techniques for implementation of a fast-running model within a digital twin framework.

The modeling and simulation of Molten Salt Reactors (MSRs) require a comprehensive multiphysics approach to capture the complex interactions between thermal-hydraulics, neutronics, structural performance, and salt chemistry. This paper introduces an integrated multiphysics modeling framework to support MSRs development using Idaho National Laboratory (INL)’s Pronghorn.

At the core of this framework, thermal-hydraulics is coupled with neutronics, enabling accurate predictions of the dynamic behavior of the MSR core and fuel salt under varying operational conditions. The framework includes detailed neutron transport models combined with weakly compressible thermal-hydraulics models for fuel salt with void transport. Additionally, corrosion modeling, informed by thermochemistry, simulates material degradation and its long-term impact on reactor performance. Furthermore, the integration of redox potential control provides a crucial mechanism for regulating corrosion rates and maintaining fuel salt chemistry stability. This comprehensive approach facilitates the evaluation of safety margins, optimization of reactor designs, and development of strategies to minimize corrosion and ensure long-term reactor reliability.

This integrated approach is unique and novel and is being applied to the Molten Chloride Reactor Experiment (MCRE) design and analysis which demonstrates its practical utility. The results are significant for advancing the safety, performance, and sustainability of MSR technology, reinforcing its potential role in the future of clean energy production.

The Molten Salt Fast Reactor (MSFR) design has the particularity that the fuel is the coolant itself. This produces a tight coupling between neutronics and thermal-hydraulics as the fuel circulates through the primary system. Therefore, developing computational models for the analysis of the MSFR requires a multi-physics approach.

The fission process generates fission products, some of them which decay releasing both decay heat and delayed neutrons. These are known as delayed neutron precursors and decay heat precursors (DNPs), respectively. In the MSFR, these precursors originate and are carried by the liquid fuel throughout the primary circuit. The generation, transport, and decay of the DNPs affect the neutron flux, heat source, and temperature distributions in the MSFR.

In the research, we propose to develop a neutronic – thermal-hydraulics computational model of the MSFR that considers the transport of the delayed neutron and heat precursors along the primary circuit. The principal computational tool chosen for this purpose is the high-fidelity code Cardinal, a wrapping within the MOOSE framework that integrates the Computational Fluid Dynamics code NekRS and the Monte Carlo particle transport code OpenMC.

Flow distribution through the core of a nuclear reactor is a key consideration when predicting full field temperatures and pressure drops. While these metrics are also dependent on the neutron flux distribution in any reactor, a liquid fueled molten salt reactor presents added complexity due to the fact that the heat is generated in the flowing fluid itself. The Natura Resources’ MSR-1 design calls for support grids in the upper and lower plenum of the reactor, which in turn can significantly impact the flow distribution throughout the core. Numerical modeling is performed on a one quarter core with five different grid cases using ANSYS FLUENT with an inlet pipe Reynolds number equal to 1.7E4. The baseline case considers the geometry with no support grids. Each grid is represented as a radially weighted porous media with greater porosity at the extremities to facilitate uniform flow distribution. The deviation of the results from the baseline case are determined and a relationship for predicting a given channel’s mass flow as a function of the grid porosity is proposed.

The liquid-fuel molten salt reactor (MSR), uniquely employing molten salt as both nuclear fuel and coolant, exhibits distinct thermal-hydraulic characteristics due to internal heat generation during flow. This study investigates the flow and heat transfer behavior of molten salt with internal heat sources using CFD simulations. Results reveal significant deviations (up to 52%) in the Nusselt number predicted by traditional correlations (e.g., Gnielinski) for transition flow regimes (Re = 5×10³–1×10⁴), while the Di Marcello model reduces errors to 15%. Friction pressure drops align with classical models (Blasius, Guo and Julien，McAdams) within 17% deviation. In addition, microwave heating is proposed as a new internal heat source experimental method to verify the influence of heterogeneous power distribution on Nu number (less than 33% deviation). The results provide a basis for the thermal design and experimental method optimization of molten salt reactor.

Molten Salt Reactors (MSR) make for a promising technology for nuclear reactor design, due to their flexibility, inherent safety features and waste-burning capabilities. For unmoderated MSRs, the core consists of a large vessel without internal structure guiding the fluid, characterized by very high Reynolds numbers and a highly turbulent salt flow. Moreover, in those reactors, neutronics and thermal-hydraulics are strongly coupled physics due to the significant thermal feedback coefficients and need to be considered together. In previous studies on the Molten Salt Fast Reactor (MSFR) concept, the flow used to be computed with Reynolds Average Navier-Stokes models, which are unable to capture the temporal fluctutations. More recent studies applied a Detached Eddy Simulations (DES) calculation to address this problem and optimized the power stability. However, those studies were using wall functions and high aspect ratio cells in order to reduce computational cost. This led to low precision and prevented eddy computations in this region, resulting in an apparent viscous layer damping all temperature variations. Consequently the wall temperature remains an open question.
In this work, we aim to investigate the impact of resolving near wall region with LES models in specific regions to assess if the temperature fluctuations are potentially inducing thermal fatigue in the vessel of the MSFR. These studies are conducted with coupled neutronic-thermal-hydrauylics simulations on the full core of the MSFR, using Ansys Fluent for the thermal-hydraulics part and the Transient Fission Matrix approach for the neutronics.