The design and performance of a compact and easy-to-build thermometer for the temperature range of 10 mK-4.2 K, which is wide enough to cover the entire temperature range of a typical dilution refrigerator, are described here. A commercial low-T(c) dc superconducting quantum interference device (SQUID) is used to measure the thermal current noise of a resistor, i.e., an insulated copper wire, that scales linearly with the temperature, as expected from the Nyquist formula. Although, in principle, this thermometer is a primary thermometer, its performance was evaluated as a secondary thermometer. The relative precision of the temperature, which is independent of the temperature, was estimated at 0.4% in a measuring time of 130 s, in good agreement with the expectations. The accuracy evaluated from the comparison with other calibrated thermometers (i.e., superconducting reference point (SRP), cerium magnesium nitrate, and germanium) on different temperature ranges is better than 2% between 10 mK and 4.2 K. If only the SRP thermometer is considered for the comparison in the temperature range of 20-207 mK, the accuracy improves to 1%. The calibrations performed with a single fixed point have a typical accuracy value of about 1%.
A Simple Wide-Range Noise Thermometer With DC SQUID Readout for Operation Down to 10 mK
Falferi Paolo;
2011
Abstract
The design and performance of a compact and easy-to-build thermometer for the temperature range of 10 mK-4.2 K, which is wide enough to cover the entire temperature range of a typical dilution refrigerator, are described here. A commercial low-T(c) dc superconducting quantum interference device (SQUID) is used to measure the thermal current noise of a resistor, i.e., an insulated copper wire, that scales linearly with the temperature, as expected from the Nyquist formula. Although, in principle, this thermometer is a primary thermometer, its performance was evaluated as a secondary thermometer. The relative precision of the temperature, which is independent of the temperature, was estimated at 0.4% in a measuring time of 130 s, in good agreement with the expectations. The accuracy evaluated from the comparison with other calibrated thermometers (i.e., superconducting reference point (SRP), cerium magnesium nitrate, and germanium) on different temperature ranges is better than 2% between 10 mK and 4.2 K. If only the SRP thermometer is considered for the comparison in the temperature range of 20-207 mK, the accuracy improves to 1%. The calibrations performed with a single fixed point have a typical accuracy value of about 1%.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
