Optimal sampling strategies for the measurement of spin-spin relaxation times

It is shown how Cramer-Rao theory may be used to determine the optimal sampling pattern for measuring the NMR spin-spin relaxation time, T-2. The results may also be applied to the measurement of any other exponential decay, including some pulse sequences for measuring the spin-lattice relaxation time, T-1. The optimal sampling pattern involves placing 22% of the sample points at zero time and the remaining 78% at 1.28 T-2, or, more practically and almost as accurately, one point at zero and four at 1.30 T-2, These sampling patterns are very different from those commonly used, The Cramer-Rao results are compared with experimental measurements and computer simulations, Some limitations of the method are described, and its extension to the simultaneous measurement of a range of T-2 values is addressed. (C) 1996 Academic Press, Inc.