In order to ensure the reliability of research results obtained in very low temperature fields, it is important that very low temperature measurement is traceable to the internationally agreed the Provisional Low Temperature Scale of 2000 (PLTS 2000). The PLTS-2000 is defined by the relationship of 3He melting pressure pm to temperature T2000 in the range of 0.9 mK to 1 K and provides four intrinsic fixed points on the melting curve. In order to realize the PLTS-2000, it is necessary to cool 3He at melting pressure down to the possible lowest temperature using a nuclear demagnetization cryostat. This cryostat must be capable of working reliably and continuously over long periods of time during ultra-low temperature experiments. In this article, we give information about low-temperature techniques for realizing the PLTS-2000 based on our experiences of designing and constructing in-house ultra-low temperature apparatus.
Accurately calibrated thermometers are an indispensable tool for reliable and precise temperature measurements. The International Temperature Scale of 1990 (ITS-90) was adopted by the Comité International des Poids et Mesures, CIPM in 1989 to enable accurate and reliable temperature measurements around the world. We realized a temperature scale for the lowtemperature region in accordance with the ITS-90 and calibrated standard resistance thermometers. We also developed a resistance thermometer calibration apparatus based on a pulse tube / 3He JT cryocooler. The calibration apparatus is intended to operate between 24.5561 K (temperature of the triple point of Ne) and 0.65 K (lowest defied temperature of the ITS-90), and is used for comparison and calibration of standard resistance thermometers. The apparatus can reach 0.5 K in continuous operation mode. The temperature control stability at the thermometer comparison block of the apparatus over the operation temperature is better than 0.07 mK, expressed as standard deviation of the temperature variation measured for 15 min. The vibration at the comparison block was also measured and found to be approximately 0.01 m/s2 at maximum. This level of vibration does not affect the calibration of the standard resistance thermometer.