For some twenty years the rhodium-iron resistance thermometer has been regarded as one of the most reliable for low temperature measurement, and in various forms has been used from millikelvin regions up to room temperature. Although it is known that the annealing treatment is one of the most important factors controlling the thermometric properties of such thermometers, only limited information is available about their stability when the rhodium-iron (mole fraction 0.5 per cent) is in the form of wire.

A recent communication from the National Research Laboratory of Metrology, Japan, reports on the effects of annealing on a new type of rhodium-iron thermometer, in which the 50 µ m diameter wire is wound bifilarly around a cross-shaped frame machined from fused silica. (O. Tamura and H. Sakurai, “Rhodium-Iron Resistance Thermometer with Fused-Silica Coil Frame”, Cryogenics, 1991, 31, 10, 869-873). The use of fused silica enables the sensing element to be annealed at temperatures above 600°C. The four lead wires and the protective sheath are made of platinum. The influence of annealing temperatures between 700 and 900°C upon the resistance of the thermometer has been investigated, and a calibration method proposed for cryogenic use of the thermometers.

It is concluded that an annealing temperature of 800°C is required to remove the strain produced in the wire by coiling; thermometers annealed at or above this temperature have similar temperature-resistance characteristics and, after calibrating the deviation from a reference function at only three calibration points, can be used with an accuracy better than 0.5 mK over the range 4.2 to 25 K. Self-heating effects were found to be of a reasonable magnitude.