We describe a method for precisely controlling temperature using a Gifford–McMahon (GM) cryocooler that involves inserting fiber-reinforced-plastic dampers into a conventional cryosystem. Temperature fluctuations in a GM cryocooler without a large heat bath or a stainless-steel damper at 4.2 K are typically of the order of 200 mK. It is particularly difficult to control the temperature of a GM cryocooler at low temperatures. The fiber-reinforced-plastic dampers enabled us to dramatically reduce temperature fluctuations at low temperatures. A standard deviation of the temperature fluctuations of 0.21 mK could be achieved when the temperature was controlled at 4.200 0 K using a feedback temperature control system with two heaters. Adding the dampers increased the minimum achievable temperature from 3.2 to 3.3 K. Precise temperature control between 4.200 0 and 300.000 K was attained using the GM cryocooler, and the standard deviation of the temperature fluctuations was less than 1.2 mK even at 300 K. This technique makes it possible to control and stabilize the temperature using a GM cryocooler.
A limiting mean free path was considered in order to better understand the temperature and wire diameter dependence of the resistivity and Seebeck coefficient of bismuth microwire and nanowire samples. The mean free path limited mobility was numerically calculated from experimentally measured mobility in a bulk bismuth sample, and the electron and hole mobilities were dramatically decreased to a mean free path. Therefore, the temperature dependence of resistivity in very thin wire was quite different from that of a bulk sample, which had a positive temperature coefficient. The calculations showed that the temperature coefficient decreased gradually with decreasing mean free path, and the coefficient became negative for a mean free path of less than at about 150 K. The Seebeck coefficient was also calculated, but showed only a weak dependence on mean free path compared with the resistivity. Experimental comparisons were made to previous measurements of bismuth microwire or nanowire samples, and the temperature and wire diameter dependencies of the resistivity and Seebeck coefficient were qualitatively and quantitatively in very good agreement. Therefore, the temperature dependencies of nanowire samples over 850 nm in diameter were well described using the mean free path limitation.
Modeling and experiments are performed to decrease temperature fluctuation generated by the periodic motion of the displacer in a Gifford–McMahon (GM) type cryocooler within the low-temperature region. The one-dimensional heat equation allows us to show that thermal diffusivity is an essential factor to achieve much smaller temperature fluctuation, and fiber-reinforced plastic (FRP) with low thermal diffusivity makes it possible to reduce the temperature fluctuation dramatically. Based on the model, experiments are performed to vary the thickness of two FRP dampers, on the cryohead of the cryocooler and on the sample stage. As a result, the FRP dampers enable us to achieve the temperature fluctuations of only 0.7 mK, corresponding to a standard deviation of 0.25 mK, when the sample stage is maintained at 4.2000 K, even if a GM cryocooler is utilized for cooling the temperature, which introduces an initial temperature fluctuation of 282 mK at the cryohead.
The Seebeck and Peltier effects have been widely studied and used in various thermoelectric technologies, including thermal energy harvesting and solid-state heat pumps. However, basic and applied studies on the Thomson effect, another fundamental thermoelectric effect in conductors, are limited despite the fact that the Thomson effect allows electronic cooling through the application of a temperature gradient bias rather than the construction of junction structures. In this article, we report the observation of a giant Thomson effect that appears owing to magnetic phase transitions. The Thomson coefficient of FeRh-based alloys reaches large values approaching –1000 μV K−1 around room temperature because of the steep temperature dependence of the Seebeck coefficient associated with the antiferromagnetic–ferromagnetic phase transition. The Thomson coefficient is several orders of magnitude larger than the Seebeck coefficient of the alloys. Using the active thermography technique, we demonstrate that the Thomson cooling can be much larger than Joule heating in the same material even in a nearly steady state. The operation temperature of the giant Thomson effect in the FeRh-based alloys can be tuned over a wide range by applying an external magnetic field or by slightly changing the composition. Our findings provide a new direction in the materials science of thermoelectrics and pave the way for thermal management applications using the Thomson effect.