Spin-Lattice Relaxation of Liquid ${\mathrm{He}}^3$ and ${\mathrm{He}}^3$-${\mathrm{He}}^4$ Mixtures

The relaxation times $T_1$ and $T_2$ of ${\mathrm{He}}^3$ nuclei in pure liquid ${\mathrm{He}}^3$ and in liquid ${\mathrm{He}}^3$ - ${\mathrm{He}}^4$ mixtures have been measured in the temperature range $0.9°\mathrm{K}<T<\mathrm{}2.6\mathrm{}°\mathrm{K}$. Reproducible relaxation times longer than those previously reported have been obtained by employing an extensive cleaning procedure. The Pyrex sample chamber and filling capillary were repeatedly cleaned in the presence of several mm of pure ${\mathrm{He}}^4$ by heating the glass to almost melting and, at the same time, touching the glass surface with a Tesla coil. This coil creates a high-frequency discharge which drives absorbed gases off the wall. $T_1$ for pure ${\mathrm{He}}^3$ was found to have the value and the temperature dependence predicted by the modified Bloembergen-Purcell-Pound theory to within ±3%. $T_1$ is proportional to the diffusion coefficient, and increases from 350 sec at 1.0°K to 450 sec at 1.8°K. $T_2$ was found to be approximately equal to $T_1$. These long relaxation times were significantly shortened by contamination of the clean chamber with air or by inadequate purification of the ${\mathrm{He}}^3$ sample. Since the work on pure ${\mathrm{He}}^3$ showed that wall effects could be minimized by cleaning, it was considered of interest to measure $T_1$ in ${\mathrm{He}}^3$ - ${\mathrm{He}}^4$ mixtures. The relaxation times of a 33% ${\mathrm{He}}^3$ solution were determined in an extensively cleaned system. Above the $\lambda$ point, $T_1$ is about 50% greater than for pure ${\mathrm{He}}^3$, while just below this temperature it increases sharply by about 15%. Measurements of $T_1$ of other mixtures were made in partially cleaned systems. The results were affected by wall relaxation and bulk impurities. Some conclusions about $T_1$ are drawn from these data.