Study of thermal-hydraulics of a sinusoidal layered heat exchanger for MSR

•Verification of turbulence model selected for FLUENT using the heat transfer and pressure drop measurement data in zigzag channels.•Verification of turbulence model by comparing linear duct computation result with conventional empirical correlation.•Two types of heat exchangers with sinusoidal channels for molten salt reactor are computed.•Evaluation of required channels of an actual heat exchanger using FLUENT and RELAP5-3D.•Sensitivity analysis of channel geometry in terms of height. Plate-type heat exchangers have been investigated as promising heat exchangers for molten salt reactors. It has been found that the heat transfer coefficient of complex channel geometries such that with S-shaped fins studied as printed circuit heat exchangers (PCHE) is generally very high. However, because there is a risk of blockage in the heat exchanger of the extremely small flow area for a molten salt reactor, a sinusoidal-shaped heat exchanger with a large flow area and a simpler shape has been selected. Thermal-hydraulic analyses of several types of flow channels have been performed using the FLUENT code. The turbulence model and the wall function selected in the present calculation were verified in advance by the experimental data. The calculated results were compared with the experimental results using the zigzag flow channels and experimental correlations. The computation results using FLUENT reproduce the empirical correlations previously obtained with smooth ducts for heat transfer and pressure drop. These comparisons also demonstrate that the turbulence model and the mesh of the computation in this study are appropriate. The proposed heat exchanger with a sinusoidal flow geometry is promising because the heat transfer coefficient is larger than that of the empirical correlation for a smooth duct and is almost the same as that of zigzag geometry. The heat transfer coefficient of the sinusoidal geometry is approximately twice as large as the Dittus and Boelter correlation. The friction factor is far below that of the zigzag geometry. In the case of a heat exchanger of this shape, it has been found from the extrapolation of the results of FLUENT that the number of channels required to remove heat of 175 MW is approximately 6000 sets of hot and cold channels. This result is also supported by a real-scale analysis using the RELAP5-3D code.