Current Status of Knowledge of the Fluoride Salt (FLiNaK) Heat Transfer

A compilation and reevaluation of data from the 1950s and 1970s from three forced convective heat transfer experiments using the ternary fluoride salt FLiNaK (46.5 LiF-11.5 NaF-42 KF mol%) using presently known thermophysical properties of this salt has been performed. The previous experiments each analyzed their data using different values for the properties of the liquid salt, thus leading to differences in the reported heat transfer coefficients. For turbulent flow in experiments conducted in chambers constructed of Inconel® alloys (as used in these three previous experiments), it was determined that FLiNaK salt behaves as a "normal" fluid and can be modeled using the Dittus-Boelter (DB) correlation within ±15% accuracy. The DB correlation can thus be used for preliminary calculations of salt heat transfer. Despite the success of the DB correlation for tests conducted in Inconel® chambers, forced convective data on heat transfer in nickel and Type 316 stainless steel produced different results. The physical effects contributing to the difference in the data measured in different container materials are not understood. The concentration of Cr (the main corrosion product added to the salt during a test) is similar between Inconel® alloys and Type 316 stainless steel and should affect each test similarly. Nickel is a relatively inert container material to fluoride salts and should not affect the heat transfer. To reconcile the experiments, a simplified approach was undertaken to determine if the radiant heat transfer to the FLiNaK salt could account for the differences. It was found that under the experimental conditions used by previous investigators, the radiant heat transfer from container to salt was <2%. However, the amount of energy transferred by radiation can be significant in applications involving high temperatures (T = 1123 K) and laminar flow conditions (Re < 500) in pipes with a diameter of 1 cm or greater.