JOURNALOF MAGNETIC RESONANCE@ 523-527 (1988) Phase Coherence and Solvent Suppression in Rotating-Frame Correlation Experiments in Liquids GENNARO ESPOSITO, * WILLIAM A. GIBBON&~ AND RENZO BAZZO$ *Eniricerche SpA 00015 Monterotondo, Rome, Italy; t The School of Pharmacy, University of London, 29/39 Brunswick Sq.. WC1 London, United Kingdom: and $ Biochemistry Department, University of Oxford, South Parks Road, OXl3QU Oxford, United Kingdom Received January 25,198s Proton 1 D and 2D rotating-frame experiments have proven very useful in elucidat- ing scalar and dipolar connectivities in both small and large molecules. Rotating- frame homonuclear correlation spectra, like TOCSY ( 1) and HOHAHA (2-4)) ex- hibit increased sensitivity due to the in-phase character of the coherence transfer. Intra- and inter-cross peak cancellation are therefore prevented. In addition, particu- larly for large molecules, a further sensitivity gain is expected when using MLEV- 17 or WALTZ- 16 pulse cycles, instead of fixed or phase-alternated spin-lock fields, due to T, lengthening of the decay rate of the magnetization ( 5, 6). On the other hand, rotating-frame Overhauser spectroscopy, like CAMELSPIN ( 7) or ROESY (8)) overcomes the adverse effect of unfavorable correlation time r,, which leads to vanishingly small net cross relaxation when 7, N ~0’. These experi- ments are also valuable for distinguishing between cross-relaxation and chemical- exchange processes. Beyond their attractive advantages, the routine application of rotating-frame experiments encounters a number of theoretical and instrumental problems that demand careful consideration, Several theoretical aspects have been discussed in recent NMR literature (1, 5, 7, 9-11, 12), after the seminal papers by Hartmann and Hahn (13), Hubbard (14), and Jones (15). As far as instrumental problems are concerned, an essential prerequisite for these experiments is the phase coherence among the different RF sources and the receiver. In most commercial spectrometers, the receiver reference phase is given by the main transmitter RF, the decoupler RF being generated by a separate, independent, device. Therefore it is not possible to run these experiments using a decoupler-gener- ated mixing RF field and transmitter-generated hard pulses tout court, without under- mining the experimental performances. A practical solution (in the absence of a dedicated device for phase locking different RF sources) is to employ a single RF source, both for the pulses and for the mixing RF field. The main transmitter cannot, usually, be used for this purpose, due to duty- cycle limitations preventing the execution of long pulses or long pulse trains. The power delivered by observe transmitter amplifiers is normally too high to avoid sam- ple overheating and damages to the probe. When a suitable RF amplifier is available for the main transmitter, the previous problems can be overcome. 523 0022-2364/88 $3.00 Copyri&t 0 1988 by Academic Press, Inc. All rights of reproduction in any form revved.