2022 Update for the Differences Between Thermodynamic Temperature and ITS-90 Below 335 K

Abstract

In 2011, a working group of the Consultative Committee for Thermometry published their best estimates of the differences between the thermodynamic temperature T and its approximation ( T90), the temperature according to the International Temperature Scale of 1990, ITS-90. These consensus estimates, in combination with measurements made in accordance with ITS-90, are an important alternative to primary thermometry for those requiring accurate measurements of thermodynamic temperature. Since 2011, there has been a change in the definition of the kelvin and significant improvements in primary thermometry. This paper updates the ( T − T90) estimates by combining and analyzing the data used for the 2011 estimates and data from more recent primary thermometry. The results of the analysis are presented as a 12th-order polynomial representing the updated consensus values for the differences and a sixth-order polynomial for their uncertainty estimates.

Binary metal-carbon phase-transition temperatures

“…[8] We note that the differences between thermodynamic temperatures and those on ITS-90 below 335 K have recently been updated by Gaiser et al (2022). [9] Companion to the ITS-90 temperature scale are recommended secondary fixed points -these are not part of the official ITS-90 scale, but are nevertheless important for realization and dissemination of temperatures, see Bedford et al (1996). [10] The secondary fixed points include: the boiling points of water and Na at (373.124 and 1156.09) K with very low uncertainties of (2 and 10) mK, respectively; at high temperatures above the melting point of copper, the freezing points of Pd and Pt at (1828.0 and 2041.3) K with modest uncertainties of (0.2 and 0.8) K, respectively; the freezing points of Ni, Co, Fe, and Rh, and the melting points of Ti and Al 2 O 3 in the range (1828 to 2326) K have yet larger uncertainties of (1 to 3) K; and at very high temperatures the melting points of Ir, Mo, and W in the range (2719 to 3687) K have very large uncertainties of (4 to 7) K.…”

Speed of sound measurements and derived third and fourth acoustic virial coefficients of supercritical neon

“…Before the experimental results can be compared to the theoretical predictions, temperature measurements on the ITS-90 must be converted into thermodynamic temperatures. Below 300 K, equation (1) in [42] was used, which applies for temperatures T 90 between 4 K and 288.418 K. At 300 K and higher temperatures, equation (2) in [43] was used as recommended by Gaiser et al [42], who also pointed out that both equations have a smooth transition at T 90 = 288.418 K. The thermodynamic temperatures for each measured state point are included in tables 2 and 3. The influence of the uncertainty in the temperature conversion on the uncertainty in the speed of sound measurement is negligible.…”

Small corrections to 1989 NIST constant-volume gas thermometry data

“…The ability to determine thermodynamic temperature continues to improve due to advances in gas-based metrology based on acoustic, dielectric, and refractivity techniques, enabled in part by the ability to calculate thermophysical properties of helium gas from first principles [7]. Improved values of (T − T 90 ) were recently presented by a CCT working group [8], but this recommendation was limited to temperatures below 335 K where most of the new data have been measured. Any future recommendation for (T − T 90 ) at higher temperatures will likely rely in part on the data of Edsinger and Schooley [1], because there are still very few measurements of absolute temperature above approximately 550 K.…”