SMALL SAMPLE CALORIMETRY AT LOW TEMPERATURES

This thesis describes techniques for making specific heat measurements on small samples at low temperatures, primarily superconducting materials of technological interest (wires, tapes, and films) in the 1 (--->) 40 Kelvin temperature range. A relaxation technique is used and it is shown that samples may be measured with this technique even if they have low thermal conductance or unusual geometries. For samples of good thermal conductivity which must be attached to a thermometer using thermally conductive grease, it is shown that the heat capacity of the sample as well as its thermal conductance can be measured. For wires or tapes with poor thermal conductance because of their shape, it is shown that their heat capacity and thermal conductance can be measured as well as the thermal conductance of any grease used. Finally, techniques are developed so that these three quantities can be measured for free standing thin films.; The characteristics of a very small silicon on sapphire thermometer which was developed specifically for these measurements is discussed. The silicon is doped with phosphorous by ion implantation to produce thermometers sensitive from .25 to 40 Kelvin. A computer controlled data acquisition system used for making the measurements is described. The influence of thermometry errors, errors due to nonlinearities in the analog electronics, and self-heating corrections is discussed in detail. A low noise alternating current bridge is discussed and its performance evaluated. Data acquisition procedures which are necessary for eliminating systematic errors are also discussed.; It is shown that measurements on a thin film superconductor weighing a few hundred micrograms have an absolute accuracy as low as (+OR-) 2%; data is reported on samples weighing as little as 100 micrograms. A measurement is made on two samples of radically different geometries, one a small cube and the other a thin wire. The specific heat measured for the two samples is found to be identical even though the raw data is quite different due to the large differences in thermal conductance. Measurements of the thermal conductance of the sample and the grease are also discussed.