Effect of Uncertainties in the MSRE Model Salt Properties Using SPECTRA/SUE Analysis

The reliable prediction of heat transfer in molten salt reactors (MSRs) critically depends on the accurate characterization of the molten salt thermophysical properties. However, significant uncertainties remain in key thermophysical properties, such as thermal conductivity, dynamic viscosity, density, and specific heat capacity, largely due to limitations in historical measurement techniques and the limited availability of high-quality experimental data. These uncertainties can directly influence the assessment of convective heat transfer behavior in MSR systems.In this work, sensitivity analyses were performed using the existing Molten Salt Reactor Experiment model implemented in the SPECTRA code, with a specific focus on heat transfer during normal operation. The analyses were carried out using SUE, an uncertainty quantification and sensitivity analysis tool developed at NRG PALLAS to investigate the influence of uncertainties in thermal conductivity, viscosity, specific heat capacity, and density on the heat transfer coefficient (HTC). Both deterministic sensitivity analyses and stochastic sampling approaches were employed, and a global sensitivity assessment based on the Plischke method was applied to quantify and rank the relative importance of these thermophysical parameters.The results demonstrated that thermal conductivity has the strongest positive impact on the HTC, identifying it as the dominant parameter governing heat transfer predictions. Viscosity showed a smaller but negative effect, while specific heat capacity exhibited a comparatively weak positive influence. The effect of density was found to be less significant than that of thermal conductivity.The sensitivity indices obtained using the Plischke method confirmed these findings, with viscosity and specific heat capacity exhibiting similar secondary effects. Very similar results obtained from deterministic and random sampling approaches indicate the robustness of the sensitivity analysis and highlight the importance of accurately characterizing molten salt thermophysical properties for reliable heat transfer modeling in MSRs.