The curious temperature dependence of fluoride molten salt thermal conductivity

Abstract

To optimize the efficiency and safety of molten salt-based energy applications, accurate molten salt thermophysical property data are required. For molten fluorides, existing thermal conductivity results have large uncertainties and contradict the current theory by eliciting a positive temperature coefficient. Transient grating spectroscopy (TGS), a technique previously deemed reliable by the theoretical community, has been used to measure the thermal conductivity of fluorides (FLiNaK) for the first time. Results show a fairly flat but slightly increasing thermal conductivity as a function of temperature. The technique has been shown to not suffer from contributions from convection and radiation, an explanation used to discount the results of alternative experimental techniques. In addition to thermal conductivity, sound speed data as a function of temperature have also been obtained for the first time in FLiNaK. The use of accurate sound speed data in theoretical models of thermal conductivity provides better but not complete agreement with the results from TGS. The continued existence of a positive temperature coefficient highlights the need for new mechanistic proposals for why TGS, or current theoretical models, are unable to capture the correct temperature dependence for fluoride molten salt thermal conductivity.