Search for a command to run...
The introduction of the concept of spin-temperature into the literature has not been adequately supported by theory and experiment, and it has not been clear when the concept is essential, merely convenient, or possibly inapplicable. In fields high compared to a defined "local field" and where the individual energy levels are equidistant, it is argued that spin temperature is a convenient but unnecessary notion. Where the levels are not equidistant a spin temperature cannot generally be defined. Experiments are cited or described which demonstrate the rapid establishment of a Boltzmann population ratio when three or more levels are equidistant.It is for low fields, those which are comparable with the local field, that a theoretical examination making use of thermodynamic principles yields new quantitative predictions. In particular, a spin temperature is analytically defined, and its identity with thermodynamic temperature is experimentally established. Studies in "spin calorimetry," or the thermal mixing of two or more spin systems, strengthen the validity of the concept.