The spin-lattice relaxation rates of the aromatic, alkene, hydroxyl, methine and methyl protons of 19 steroid derivatives have been measured using the null point method. A simple procedure is described whereby the Itl values of molecules which have different motional tumbling rates can be directly inter-compared, and it is shown that such 'normalized' relaxation data can provide novel insight concerning both the geometry and the local molecular motion of these substances in solution.With the development' of rapid, routine procedures for determining the spin-lattice relaxation rates (R, values) of the protons of complex organic molecules, it is appropriate to begin to survey the diagnostic potential of these parameters in various areas of natural products chemistry. In this paper we report data for the directly assignable protons (aromatic, alkene, hydroxyl, methine and methyl) of several families of steroid derivatives.Under conditions which are readily realizable in routine studies2 (typically using c O . 1 M solute concentrations in a deuteriated solvent), the spin-lattice relaxation of most protons is dominated3 by the intramolecular dipole-dipole m e~h a n i s m ,~ which is summarized in Eqn (1) in terms of the relaxation contribution between two non-equivalent nuclei i, j :where yc and y1 are the magnetogyric ratios of the i and j nuclei, respectively, ri,j is their internuclear separation and -rc(i, j ) is the motional correlation time of the vector between them. As a result, the R , value of each proton reflects the composite influence of (a) the total spatial separation of that proton from the other protons of the same molecule, together with (b) its motional correlation time; the latter represents the summation of the overall tumbling rate of the molecule as a unit, plus any additional motion of the substituent group which bears that particular proton.Separate identification of the two contributions (a) and (b) can lead to two distinct types of structural information-solution geometry and molecular mobility-and one of the questions posed here is how best to effect this separation in a simple fashion. Since, for chemical purposes, change in the overall tumbling rate is not as important as knowledge of local motional freedom, our strategy has been to 'normalize' the experimental data in such a manner that the effect of * Author to whom correspondence should be addressed. Permanent address: Department of Chemistry, Concordia University, 1455 de Maisonneuve Blvd., West, Montreal, Quebec, Canada H3G 1M8.the overall tumbling represents a constant background (see Ref. 4 for a similar expression of this approach), and, as we shall show, the subsequent distinction between effects associated with changes in 'distance' and those involving 'mobility' can then be made because of the rapid decrease of the former with distance.
RESULTS AND DIXUSSIONThe general features of the relaxation experiment itself and of the chemical significance of the R , values are concisely illustrated by reference to l-dehydrotestosterone ...