International Temperature Scale Research Articles

The non-uniqueness of ITS-90 above the silver point and its impact on values of T − T90

Non-uniqueness is a fundamental limit to the reproducibility of a defined temperature scale. In this paper, the non-uniqueness of temperatures, T90, on the International Temperature Scale of 1990 (ITS-90) above the silver point is fully investigated for the first time. The non-uniqueness arises from three causes: inconsistency in extrapolation with the choice of the three allowable ITS-90 reference fixed points; a wavelength dependence of T90 due to the difference between thermodynamic temperature, T, and T90 at these reference fixed points; and a further wavelength dependence arising from the discrepancy between the value of the second radiation constant defined on ITS-90 and its value derived from fundamental constants. The non-uniqueness is quantified by deriving an exact expression for T − T90 above the silver point and calculating the differences in this expression as a function of reference fixed point and wavelength. In the worst case, the non-uniqueness is about 100 mK, and T − T90 is as much as 400 mK, at 3000 °C.

Challenges for In-Flight Calibration of Thermal Infrared Instruments for Earth Observation

Satellite instruments operating in the thermal infrared wavelength range >3 µm provide information for applications such as land surface temperature (LST), sea surface temperatures (SST), land surface emissivity, land classification, soil composition, volcanology, fire radiative power, cloud masking, aerosols, and trace gases. All these instruments are dependent on blackbody (BB) calibration sources to provide the traceability of the radiometric calibration to SI (Système International d’Unités). A key issue for flight BB sources is to maintain the traceability of the radiometric calibration from ground to orbit. For example, the temperature of the BB is measured by a number of precision thermometers that are calibrated against a reference Standard Platinum Resistance Thermometer (SPRT) to provide the traceability to the International Temperature Scale of 1990 (ITS-90). However, once calibrated the thermometer system is subject to drifts caused by on-ground testing, the launch and space environments. At best the uncertainties due to thermometer ageing can only be estimated as there is no direct method for recalibrating. Comparisons with other satellite sensors are useful for placing an upper limit on calibration drifts but do not themselves provide a traceable link to the SI. In this paper, we describe we describe some of the technology developments, including phase change cells for use as reference standards, thermometer readout electronics and implementation of novel coatings, that are in progress to enhance the traceability of flight calibration systems in the thermal infrared.

Interpolating Equation of Precision Platinum Resistance Thermometers in the Temperature Range between -80°C and 300 °C

This paper compares two interpolation equations for the calibration of precision platinum resistance thermometers (PPRTs) in the temperature range between -80°C and 300°C. They are the deviation function specified in the International Temperature Scale of 1990 (ITS-90), and the quadratic Callendar–van Dusen (CVD) equation. It was found that when the calibration range was between -80°C and 300 °C, ITS-90 deviation function resulted in a measurement error-of-fit of ±23mK, and the CVD equation resulted in ±24mK. In the temperature range of (-80∼0)°C, ITS-90 deviation function showed distinctly better performance than the CVD equation. When the temperature range was between 0°C and 100°C, the difference among the two equations was insignificant. In the temperature range of (100∼300)°C, the measurement error-of-fit of ITS-90 deviation function increased with temperature, and the maximum error-of fit under the CVD equation appeared at 200°C. If the temperature range is wider, the ITS-90 deviation function had better calibration effect. While the CVD equation worked better in a narrow temperature range.

Standard platinum resistance thermometer interpolations in a revised temperature scale

The thermal metrology community is considering revising the International Temperature Scale of 1990 (ITS-90), motivated by the opportunity to improve the thermodynamic accuracy, reproducibility, and ease of use of the scale, as well as health and safety concerns with the use of mercury in the current scale. This paper considers the mathematical structure of the standard platinum resistance thermometer (SPRT) interpolations of the ITS-90 and identifies (i) mathematical features that are advantageous and should be retained, (ii) opportunities for improvements, and (iii) the research required to maximise the benefits from such improvements. The improvements considered include minor adjustments that leave ITS-90 intact, numerical adjustments to reference resistance ratios that preserve the structure of ITS-90, fixed-point replacements, new subranges, and large-scale changes in the mathematical structure of the interpolations. A significant research effort will be required to implement some of the changes. Overall, an improvement in the thermodynamic accuracy by a factor of about ten is relatively easily realised, but improvements in reproducibility of the scale of even a couple of tens of percent will be hard won. For most users, the costs and inconvenience of a substantial scale revision may outweigh the benefits.

Deviation of Temperature Determined by ITS-90 Temperature Scale from Thermodynamic Temperature Measured by Acoustic Gas Thermometry at 79.0000 K and at 83.8058 K

A difference was determined between temperature defined by International Temperature Scale (ITS-90) and thermodynamic temperature measured by relative acoustic gas thermometry at 79.0000 K and at 83.8058 K. Measurement were carried out using misaligned spherical acoustic resonator filled with helium at different pressures. The isotherms were fitted for four different acoustic modes simultaneously assuming the same thermal accommodation coefficient and third acoustic virial coefficient for all modes. Our measurements yield the following differences between the thermodynamic temperature T and ITS-90 temperature $$T_{90}$$: $$T-T_{90}=-4.81 \pm 1.02$$ mK at 83.9058 K and $$T-T_{90}=-4.47 \pm 0.97$$ mK at 79.0000 K.

Investigation of the thermal stability of PtO2 and dislocations in Pt wire

Abstract The thermal stability of PtO2 and dislocation density within Pt wire were investigated using differential scanning calorimetry (DSC), Thermogravimetry (TG), X-ray photoelectron spectroscopy (XPS) and X-ray diffraction (XRD) to understand and evaluate the stability of standard platinum resistance thermometer (SPRT), which is the interpolation instrument in the International Temperature Scale of 1990 (ITS-90). Through analyzing the thermal stability of PtO2, the results showed that PtO2 decomposition was closely related to the thermal process. PtO2 did not decompose below 500 °C in air but began to decompose into PtO and Pt above 550 °C. The decomposition rate of PtO2 was shown to be the fastest at 620 °C, but the decomposition was not necessarily complete at this temperature. The decomposition process of PtO and PtO2 further occurred from 800 °C to 850 °C. Through analyzing the thermal stability analysis of Pt wire dislocation, the results showed that the total dislocation density of platinum wire decreased exponentially with an increase in annealing time. After 100 h annealing, the dislocation density decreased from 1012 cm−2 to 1011 cm−2, and after 300 h annealing, the dislocation density tended to be stable. Twelve SPRTs recently made in NIM were annealed at 660 °C, the resistance changes of SPRTs at the triple point of water were similar to the trend of dislocation density with annealing time. Therefore, to reduce long-term downward drifts, it is recommended to anneal more than 300 h for new SPRTs before calibration.

Determination of T–T90 from 234 K to 303 K by acoustic thermometry with a cylindrical resonator

The determination of the differences (T–T90) between the thermodynamic temperature T and the international temperature scale of 1990 (ITS-90) T90, is important for an evaluation of the approximation to T by T90. Such evaluations are necessary for the potential revision of ITS-90. A number of efforts have been devoted to the determination of (T–T90) by the acoustic gas thermometry (AGT) using spherical or quasi-spherical resonators. We report in this paper a new study in the temperature range from 234 K to 303 K using AGT in argon with a cylindrical resonator. Piezo-electric acoustic transducers were used to measure the acoustic resonant frequencies. The resonant frequencies of the transverse magnetic microwave modes of the cavity were measured using straight probe antennas. This work further illustrates the high performance of the microwave resonant procedure to measure the thermal expansion of a cylindrical cavity at different temperatures and pressures. The (T–T90) measurements, with standard uncertainties in the range from 0.5 mK to 0.8 mK, agree well with the existing data from AGT with spherical or quasi-spherical resonators. We believe that cylindrical acoustic gas thermometry, with an uncertainty comparable with that using spherical or quasi-spherical resonators but simpler mechanical assembly, has the potential for accurate measurements of (T–T90) at higher temperatures.

Non-contact temperature measurement at the Physikalisch-Technische Bundesanstalt (PTB)

ABSTRACT At the Physikalisch-Technische Bundesanstalt, the National Metrology Institute of the Federal Republic of Germany, the traceability of non-contact temperature measurement to the International Temperature Scale (ITS-90) is realised without gaps over a temperature range from −170°C to 3000°C. In this paper the main calibration facilities for the calibration of radiation thermometers at temperatures above 1000°C and measurement of spectral radiance, the calibration of radiation thermometers at temperatures below 1000°C, the calibration of infrared thermographic cameras together with a method developed at PTB to correct the non-uniform response of sensor arrays, the calibration of infrared hyperspectral imagers for remote sensing, and emissivity measurement for industry and remote sensing are described. These facilities are tailored for different temperature ranges and different spectral ranges to meet the demand of industry and research for precise, practical non-contact temperature measurement.

Thermodynamic Temperature Measurements from the Triple Point of Water up to the Melting Point of Gallium

A new acoustic gas thermometry (AGT) system, which introduces a 1-L quasi-spherical resonator (QSR) made of oxygen-free copper, was built at the National Metrology Institute of Japan (NMIJ/AIST). The inner surface of the new QSR was machined using a diamond-turn tool. Improvement in the pressure measurement was introduced to allow direct measurement of the gas pressure in the resonator. The new AGT system was evaluated based on the measurement of the speed of sound in argon at the triple point of water and the melting point of gallium under the pressure range from 700 kPa down to 50 kPa. Based on these speed of sound measurements, the thermodynamic temperatures at the melting point of gallium were determined. The speed of sound measurements at isotherms of 283.15 K and 293.15 K were also conducted, and the related thermodynamic temperatures were determined. Based on the measured thermodynamic temperature T, the values of difference between T and the temperature T90 based on the International Temperature Scale of 1990 (ITS-90), (T − T90), along with the associated uncertainties, were calculated. The (T − T90) obtained in the present work were 1.3 ± 0.7 mK, 2.7 ± 0.8 mK, and 4.1 ± 0.8 mK for T90 of 283.15 K, 293.15 K, and 302.9146 K, respectively. These values were found to be in agreement within the estimated uncertainty with the currently reported values that exist in temperature range overlapping with the present work.

Evaluation of High-Temperature Platinum Resistance Thermometers Based on ITS-90

High-temperature standard platinum resistance thermometers (HTSPRTs) are used to establish the International Temperature Scale of 1990 (ITS-90) for temperatures up to 960 °C. Currently available thermometers, which may differ in the design of the platinum winding of the sensor, the shape and finish of the sensor frame, the frame material, the protective sheaths, and sealing materials, need some treatments to exhibit the stabilities expected of them and some of the causes of the instability are not well understood. This paper demonstrates a novel model of commercial HTSPRT that reportedly employs a novel sensor winding construction designed to eliminate instability arising from strain on the sensor winding due to differential thermal expansion between the sensor and the sensor frame. The stability of eight different HTSPRTs, including the novel model, are compared at the triple point of water (TPW), the melting point of gallium, and the freezing points of tin, zinc, aluminum and silver. Heat treatments included annealing at 500 °C, a preheating temperature for use at the freezing points of aluminum and silver, and at 800 °C, which was selected for checking the stability of the HTSPRTs in the middle of the temperature range between the freezing points of aluminum and silver. The consistency of the resistance values at the different fixed points was also confirmed by calculating the S parameter. We found three HTSPRTs stable during the described thermal conditions within ± 0.5 mK at the TPW. The novel model of commercial HTSPRT was the best one with a stability in the resistance at the TPW of within ± 0.3 mK over the entire temperature range. These indicate that the stable three HTSPRTs are promising instruments to provide accurate temperature scale at temperatures from the freezing point of aluminum up to the freezing point of silver based on the ITS-90.

Thermodynamic temperature of the triple point of xenon measured by refractive index gas thermometry

Measurements of the refractive indices of helium and argon using a quasi-spherical microwave resonator and upgraded experimental protocol are reported at the temperatures of the triple points of water and xenon. The results at the triple point of water are used to determine the compressibility of the resonating cavity, which is bounded by the copper shell of the resonator. This experimentally-determined compressibility is consistent with the literature value for copper, but with much smaller measurement uncertainty. The measured compressibility is extrapolated to the triple point of xenon, and combined with the refractive index results at that temperature to determine the thermodynamic accuracy of the International Temperature Scale of 1990 (ITS-90): mK at ITS-90 temperature K, corresponding to a xenon triple point thermodynamic temperature of K. The experimental compressibility is further extrapolated to the triple points of argon, oxygen and neon, and used to re-analyze earlier refractive index gas thermometry measurements made using the same resonator, yielding updated values with reduced uncertainties: mK at T90 = 83.8058 K, mK at T90 = 54.3584 K, and mK at T90 = 24.5561 K. The ITS-90 thermodynamic accuracy results of the present refractive index gas thermometry study agree with those previously reported by acoustic gas thermometry and dielectric constant gas thermometry.

T − T90 for Radiation Thermometry Realization Above the Copper Point

The differences, T − T90, between thermodynamic temperature, T, and temperature, T90, on the International Temperature Scale of 1990 (ITS-90) above the Cu fixed-point temperature, 1357.77 K, were investigated using relative primary radiometric thermometry. The Ag and Cu fixed points were used as reference points for temperature realization by the ITS-90 (n = 1) scheme. The values of T − T90 for the Ag and Cu fixed points have been previously determined as 46.2 mK and 52.1 mK, respectively. Extrapolating these differences based on the sensitivity coefficient used to propagate the uncertainty in the fixed-point temperature to other temperatures results in values of T − T90 of 325 mK or 303 mK at 3000 °C when using the Ag or Cu point T − T90 difference, respectively. These values were confirmed by realizing n = 1 temperature scales using either the thermodynamic values or the ITS-90 values of these fixed points as the reference temperatures and comparing the differences on the two scales. Measurements at the Co–C eutectic point indicated the consistency of thermodynamic temperature realization using the n = 1 scheme, and also demonstrated the equivalence of absolute and relative radiometric thermometry.

Evaluation of the triple point temperature of sulfur hexafluoride and the associated uncertainty at NMIJ/AIST

This study determined the triple point temperature of sulfur hexafluoride (SF6), as well as the associated uncertainty, to evaluate the potential use of this value as an alternative to the triple point of mercury as a fixed point defined in the International Temperature Scale of 1990 (ITS-90). The triple point temperature was obtained using an adiabatic calorimeter incorporating a sealed cell made from oxygen-free copper. The cell was filled with 99.9992% SF6 and included three standard platinum resistance thermometers that had been calibrated at eight fixed points defined by the ITS-90 protocol. Applying a thermal treatment to the solidified sample prior to determining the triple point temperature was found to decrease the width of the melting curve. Based on melting curves acquired following this thermal treatment, the triple point temperature of SF6 was found to be 223.556 47(53) K. The temperature scale constructed using SF6 was found to be consistent with that generated using mercury within a maximum deviation of 70 K from 13.8033 K to 273.16 K.

Realization of the triple point of carbon dioxide evaluated by the ITS–90

We performed a precise temperature measurement of the triple point of carbon dioxide (CO2), which is one of the secondary reference points of the International Temperature Scale of 1990 (ITS–90), using an adiabatic calorimeter with five capsule-type standard platinum resistance thermometers (SPRTs). Impurity analyses showed that the purity of the CO2 sample was 99.999 37%. A thermal treatment process before the realization of the triple point reduced the width of the melting curve of the triple point of CO2 and increased the linearity of the melting curves with respect to the inverse of the melting fraction F. We determined the triple-point temperature of CO2 at the liquidus point by extrapolating the six melting curves obtained after the thermal treatment process. The triple-point temperature of CO2 measured in this work is 216.590 90 K 0.36 mK (k = 1). This result indicates that the triple point of CO2 can potentially be used as a fixed point based on the ITS–90 to replace the triple point of mercury.

Temperaturskalan och Boltzmanns konstant

Temperature scale and the Boltzmann constant: The newest system of units is based on a compatible set of natural constants with fixed values. An example is the Boltzmann constant k which defines the thermal energy content kT. To express the base unit T, the absolute temperature in kelvin, an international agreement for the temperature scale is needed. The scale is defined using fixed points, which are temperatures of various phase transitions. Especially important has been the triple point of water at 273.1600 K. These fixed point temperatures determine the international temperature scale ITS within the Si system. Temperature measurement itself is based on physical laws and on the properties of appropriate thermometric materials selected to determine the temperature scale. For determining the Boltzmann constant, new precision techniques have been developed during the last two decades. Examples are different types of gas thermometry, which ultimately are based on the ideal gas law, and thermal noise of electric charge carriers in conductors. With these means it has become possible to fix the value of the Boltzmann constant with a relative uncertainty of < 1 ppm. As of 2019, the value of k has been agreed to be fixed at 1.380649x10^(-23) J/K. This agreement replaces the earlier definition of a Kelvin degree.

Type 3 Non-uniqueness in Interpolations Using Standard Platinum Resistance Thermometers Between − 196 °C and 100 °C

This paper reports on precise measurements of the non-uniqueness (interpolation errors) associated with using Standard Platinum Resistance Thermometers (SPRTs) in the range from − 196 °C to 100 °C. Six long-stem SPRTs were compared with unprecedented (< 0.1 m°C) precision in a stirred oil bath from 100 °C down to − 95 °C, with an additional comparison measurement in a bath of liquid nitrogen at − 196 °C. The measurements resulted in a highly self-consistent set of resistance ratios W(t), where W(t) = R(t)/R(0 °C). Selecting the first SPRT as the reference, the differences (Wx − Wref), x = 2 to 6, were calculated, in the manner of the International Temperature Scale of 1990, ITS-90, and then used to investigate the consistency of interpolations between the SPRTs in various temperature ranges. With some minor exceptions, the data suggest that the interpolation consistency (taken to be an upper limit for the Type 3 non-uniqueness between the SPRTs) was ± 0.03 m°C from − 38 °C to 30 °C (representing the mercury-to-gallium subrange in the ITS-90), ± 0.1 m°C from − 95 °C to 80 °C, and ± 0.2 m°C at − 95 °C for interpolations extending from − 196 °C to 0 °C (no data could be taken between − 196 °C and − 95 °C). Various interpolation schemes are investigated, especially to examine ways of avoiding the use of the mercury triple point. One scheme investigated is simply to omit the mercury point and instead to extend the subrange from the argon point up to the gallium point or further, up to 80 °C. Similarly, the ITS-90 subranges to the indium, tin, zinc, or aluminum points could usefully be extended down to the argon point, with suitably adapted interpolation equations. These possibilities remain to be investigated.

Effects of Quartz Glass Insulation on Platinum Gold Thermocouples

Abstract Au/Pt thermocouples are considered as an alternative to High Temperature Platinum Resistance Thermometers and are one of the prime candidates to replace them as the interpolating instrument of the International Temperature Scale of 1990 (ITS-90) in the temperature range between about 660 °C and 962 °C. This work presents the results of investigation of two Au/Pt thermocouples that used exclusively quartz glass (SiO2) as insulation material. Measurements in fixed points of Zn, Al, and Ag were realized on these thermocouples as well with interchanged inner insulation made of high purity aluminium oxide (Al2O3). The conducted experiments tested the performance of Au/Pt thermocouples with the use of different insulation materials. The measured electromotive forces were found to be sensitive to the replacement of the quartz glass by aluminium oxide as an insulation material of the Au/Pt thermocouples. This change of insulation has resulted in a temperature increase up to about 0.5 K measured at the freezing point of silver. The decreasing insulation resistance of quartz glass at higher temperatures is believed to be the source of thermoelectric instability.

Immersion effects in zinc ITS-90 fixed-point cells

The freezing point of zinc (419.527 °C) is an important defining fixed point of the International Temperature Scale of 1990 (ITS-90). To realize it, a crucible of high purity graphite containing an ingot of high purity zinc is placed in a glass container containing an inert gas, with a central re-entrant tube, also glass, to allow insertion of a standard platinum resistance thermometer (SPRT). A key problem with these cells is difficulty in obtaining good immersion of SPRTs; it is commonly thought that the glass construction causes ‘light piping’ whereby the glass could facilitate heat transfer away from the region close to the SPRT sensing element by radiation, both directly and via reflections. In this paper it is shown that at the zinc freezing temperature light piping via reflections between the surfaces of the glass components is not an important heat transfer mechanism. It is also shown that light piping via transmission within the glass components is insignificant because of attenuation at wavelengths where the magnitude of the Planck function is important. In the case of the zinc fixed point, direct line-of-sight radiative heat transfer and thermal effects within the ingot appear to be the most important contributors to spurious immersion effects. It is shown that where the immersion profile is poor, a short platinum coil wrapped around the SPRT in the vicinity of the sensing element to improve radial conduction and to block vertical radiative heat transfer substantially improves the performance.

Measurement of the Melting Point of Gallium Using a Johnson Noise Thermometer

We measured the thermodynamic temperature of the melting point of gallium $T_{\mathrm {Ga}}$ using a Johnson noise thermometer (JNT) with an integrated quantum voltage noise source (IQVNS) as a reference. $T_{\mathrm {Ga}}$ was calculated using the fundamental constants recommended by the Committee on Data for Science and Technology (CODATA) for the revision of the International System of Units (SI). The measured $T_{\mathrm {Ga}}$ is consistent with the value defined in the International Temperature Scale of 1990.

Realization of the Triple Point of Hg and Observation of a Large Supercooling Using Small Glass Cell

We have developed an adiabatic calorimeter with a small glass Hg cell to calibrate capsule-type standard resistance thermometers (SPRTs) at the triple point of Hg (TPHg), which is one of the defining fixed point of the International Temperature Scale of 1990 (ITS-90). Using this system with high-purity Hg (99.999 99 %), we calibrated SPRTs at the TPHg and evaluated the uncertainty of the calibration. Compared with a calibration system at the National Metrology Institute of Japan (NMIJ/AIST), which uses a large stainless steel Hg cell, for a small cell in an adiabatic calorimeter the temperature distribution from hydrostatic pressure can be suppressed. In addition, the influence of heat flux can also suppressed by using the adiabatic calorimeter. Consequently, the standard combined uncertainty for the realization of TPHg was estimated to be 0.12 mK, much less than the uncertainty of the conventional system using the large stainless Hg cells. Furthermore, by measuring the cooling curve of Hg at different cooling rates, we confirmed that the degree of supercooling of high-purity Hg in the small glass cell reached approximately 18 K, which is much larger than that achievable in not only large Hg cells but also small cells. This large supercooling may be related to the high purity of the Hg employed, the small hydrostatic height of the Hg, and/or the surface contact between Hg and cell.