The Comparison between a Second-Sound Thermometer and a Melting-Curve Thermometer from 0.8 K Down to 20 mK

Abstract

BNM‐INM has realised the Provisional Low Temperature Scale of 2000 (PLTS‐2000). Additionally, for several years, the low temperature team of BNM‐INM has studied the possibility of using the properties of dilute mixtures of 3He in 4He in order to develop a local temperature scale. The team made the choice to develop a new type of thermometer based on the propagation of sound in dilute solutions of 3He in superfluid 4He, a second‐sound thermometer. For low temperatures, the properties of low concentrations of 3He in superfluid 4He are those of a nearly ideal Fermi gas. The experiments of Greywall and Owers‐Bradley et al. have shown that the velocity of second sound in 3He‐4He mixtures is very sensitive to temperature, especially below 0.6 K. In the second‐sound thermometer developed by BNM‐INM, the speed of sound is determined from the resonance spectra of an acoustic cavity. The temperature is deduced from the measurement of the resonance frequencies, by using a physical model describing the relation between the speed of sound, the acoustic length of the cavity and the temperature. This paper presents the experimental results obtained for the temperature determinations using the second‐sound thermometer and for the comparison between this thermometer and the melting‐curve thermometer in the temperature range from 0.8 K down to 20 mK. A detailed uncertainty budget on the temperature comparison is given.