Comment on pulsed NMR thermometry at very low temperatures

At temperatures below 1 mK the rf power in pulsed NMR experiments is normally kept as low as possible in order to minimize eddy current heating in the metallic samples used for thermometry. However, for metallic samples with short Korringa constants a temperature dependence enters via the spin-lattice relaxation mechanism in addition to the paramagnetic dependence. It may cause systematic errors if the temperature of the conduction electrons is raised by eddy current heating due to the action of the rf pulse to a temperatureT _e which is not very small compared to ₁, with being the Korringa constant and ₁ the rotation frequency of the nuclear moments in the rf field.