Nuclear Magnetic Resonance Measurements in SolidHe3

Abstract

Measurements of the magnetic susceptibility of solid ${\mathrm{He}}^3$ are complicated by the nature of the coupling of the spin system to the lattice. This coupling is examined at temperatures down to 0.1°K, magnetic fields up to about 2 kG, and molar volumes between 18.5 and 24 ${\mathrm{cm}}^3$ using a continuous-wave nmr apparatus. For pure ${\mathrm{He}}^3$, the three-bath model of Garwin and Landesman gives a good description of the way in which the spins achieve equilibrium with the lattice. By measuring the various relaxation times, values of the exchange integral are derived by several different methods with good internal consistency. Values of the exchange-bath specific heat very close to the calculated values are observed. We have measured values of the exchange-lattice relaxation time over the same wide range of temperature and molar volume and find values very different from the results of Garwin and Reich. This and their large discrepancy between observed and calculated exchange-bath specific heats we believe to be due to the presence of about 1% ${\mathrm{He}}^4$ in their samples. In the bcc phase, over certain temperature regions, diffusion appears to be the dominant coupling process, and values of the self-diffusion coefficient are calculated which suggest that the diffusion process is thermally activated at values down to at least ${10}^{-16\mathrm{}}$ ${\mathrm{cm}}^2$ ${\sec }^{-1\mathrm{}}$. Experiments on samples containing 0.5% ${\mathrm{He}}^4$ show that the presence of small amounts of ${\mathrm{He}}^4$ impurity affects very strongly certain parameters in the three-bath model, and concentrations as low as 0.05% are detectable by their effect on the exchange-bath specific heat. Also in the impure samples a new long relaxation time appears below 0.7°K, which we believe was incorrectly identified as the exchange-lattice relaxation time by Garwin and Reich and which was the cause of nonequilibrium between spins and lattice in the magnetic susceptibility experiments of Adams, Meyer, and Fairbank.