We describe measurements of thermodynamic temperature in the range 5 K to 24.5561 K (the triple point of neon) using single-pressure refractive-index gas thermometry (SPRIGT) with 4He. In the wake of the May 2019 re-definition of the kelvin and its associated mise en pratique, the main purpose of the work is to provide values of T–T90, the discrepancy between thermodynamic temperature and that of the International Temperature Scale of 1990 (ITS-90). The link to ITS-90 is made via calibrated rhodium-iron resistance thermometers. Innovations required to reach the level of accuracy required for meaningful measurements (uncertainty in T–T90 less than the expected deviation) include the suppression of temperature oscillations in a cryogen-free cryostat, a pressure stabilization scheme based on a non-rotating piston balance, modelling of the hydrostatic head correction and refinements of the measurement of microwave resonances in a quasi-spherical copper resonator. The accuracy of measurements varies from 0.05 mK to 0.17 mK and is competitive with that of all previous ones in this temperature range using other techniques. The improvement stems partly from the new techniques used for the new definition of the kelvin as well as ab initio calculations of the thermophysical properties ofgaseous 4He. In addition to confirming the validity of SPRIGT as an accurate, easier-to-implement alternative to other low-temperature primary thermometry techniques (e.g. acoustic gas thermometry) yet with scope for improvement, the results should provide important input data for any future revision of ITS-90.
We used absolute primary acoustic gas thermometry (AGT) to calibrate a Pt–Co resistance thermometer on the thermodynamic temperature scale by measuring the speed of sound in helium at a temperature T* chosen to be near the temperature of the triple point of neon, TNe. Prior to the present AGT, the Pt–Co thermometer was used with a neon triple-point cell as part of an interlaboratory comparison. Taken together, the results of the interlaboratory comparison and the present AGT redetermined the thermodynamic temperature TNe = (24.555 15 ± 0.000 24) K. This new value of TNe is consistent with other recent determinations obtained with various primary methods. After completing the AGT thermodynamic calibration, we used the Pt–Co thermometer to link T* to the temperature ratios measured by single-pressure refractive-index gas thermometry (SPRIGT) in a different laboratory. (Gao et al 2020 Metrologia57 065006) Now, the T*-linked SPRIGT system can calibrate other thermometers on the thermodynamic temperature scale T in the range 5 K ⩽ T ⩽ TNe without using the international temperature scale ITS-90. At most temperatures in this range, the uncertainties of the T*-linked SPRIGT system are smaller than those of the ITS-90 systems used by National Metrology Institutes to calibrate resistance thermometers.
In the temperature range between 0.65 K and 1 K, the international temperature scale of 1990 (ITS-90) is based on 3He vapour–pressure thermometers and overlaps with the provisional low temperature scale of 2000 (PLTS-2000) defined by the melting pressure of 3He. An indirect comparison at PTB revealed differences between the two scales of up to 1.5 mK at 0.65 K (Engert et al 2007 Metrologia 44 40–52). Stimulated by the PTB results, we have performed a direct comparison T90–T2000 from 0.65 K to 1 K at LNE-CNAM. To test repeatability, the experiment was conducted twice: in 2019 and 2020. We find differences T90–T2000 of 0.28 mK at 1 K, increasing to 1.58 mK at 0.65 K. The direct comparison, eliminates the uncertainty component due to the transfer resistance thermometer and its calibration. Except for a point near 1 K, the new results are in accordance with those obtained at PTB (differences of less than 0.22 mK), which makes it possible to improve the accuracy of the equation specified in ITS-90.
We describe a primary gas pressure standard based on the measurement of the refractive index of helium gas using a microwave resonant cavity in the range between 500 Pa and 20 kPa. To operate in this range, the sensitivity of the microwave refractive gas manometer (MRGM) to low-pressure variations is substantially enhanced by a niobium coating of the resonator surface, which becomes superconducting at temperatures below 9 K, allowing one to achieve a frequency resolution of about 0.3 Hz at 5.2 GHz, corresponding to a pressure resolution below 3 mPa at 20 Pa. The determination of helium pressure requires precise thermometry but is favored by the remarkable accuracy achieved by ab initio calculations of the thermodynamic and electromagnetic properties of the gas. The overall standard uncertainty of the MRGM is estimated to be of the order of 0.04%, corresponding to 0.2 Pa at 500 and 8.1 Pa at 20 kPa, with major contributions from thermometry and the repeatability of microwave frequency measurements. A direct comparison of the pressures realized by the MRGM with the reference provided by a traceable quartz transducer shows relative pressure differences between 0.025% at 20 kPa and −1.4% at 500 Pa.
This paper presents a comparison between the provisional low-temperature scale of 2000 (PLTS-2000) and the magnetic field fluctuation thermometer (MFFT) scales, as realized at LNE-Cnam. The comparison was carried out using three different realizations (i.e. different melting pressure sensors) of PLTS-2000 and a magnetic field fluctuation thermometer operated with a long measurement time (about 3 h) and a more precise power spectral density model of the MFFT response that included an offset term to account for the background noise in the MFFT electronics. The comparison over the range between 8 mK and 1 K shows that the PLTS- 2000 and MFFT scales agree within about 0.2 mK. The results confirm the thermodynamic accuracy of PLTS-2000 at the same level of uncertainty, and show that, under suitable conditions, MFFT measurements can achieve comparable performance. This highlights its potential use in certain practical applications where direct realization of PLTS-2000 is not feasible.