DEUTERON NMR OF METHYL-GROUPS IN THE TUNNELING REGIME - A SINGLE-CRYSTAL STUDY OF ASPIRIN-CD3

We report the first single crystal deuteron NMR spectra of CD3 groups which display the so-called +/-beta, +/-(\alpha\+/-beta) and +/-(2\alp ha\+/-beta) lines characteristic of rotational tunneling in a sufficie ntly clear manner to allow a quantitative comparison with the respecti ve theory developed in 1988 by the group of W. Muller-Warmuth. The mol ecular system we study is aspirin-CD3. We recorded spectra for differe ntly oriented single crystals and measured spin-lattice relaxation tim es T-1 in a wide temperature range. At 12.5 K we exploit the dependenc e of the +/-(\alpha\+/-beta) and +/-(2\alpha\+/-beta) lines on the ori entation of the applied field B-0 for determining the equilibrium orie ntation of the CD3 group in the crystal lattice. The spectra display f eatures which allow, by comparison with simulated spectra, a measureme nt of the tunnel frequency nu(t). Its low temperature limit is (2.7+/- 0.1) MHz. It allows to infer the height V-3 Of the potential V(phi) in which the CD3 group moves, provided that this potential is purely thr eefold. We get V-3 = (47.2+/-0.5) meV. The transition from the tunneli ng to the classical, fast reorienting regime occurs in the 15 K less t han or similar to T less than or similar to 35 K temperature range. In this range we observe a broadening, merging and eventually narrowing of the +/-\alpha\ and +/-2\alpha\ lines in very much the way predicted by Heuer. His theory, however, must be extended by taking into accoun t all librational levels. The behaviour of the +/-beta lines in the tr ansition temperature range signalizes a reduction of the observable tu nnel frequency with increasing temperature. This reduction allows an i ndependent measurement of the potential height and represents a test o f the assumption of a purely threefold potential. From the T-1-data we derive the temperature dependence of the correlation time tau(c) of t he reorientational jumps. The plot of log tau(c) vs. 1/T follows a str aight line for more than five decades. From its slope we get yet anoth er independent number for the potential height. It agrees well with th e other ones, which confirms the assumption of the essentially threefo ld potential V(phi) in aspirin-CD3.