Increased HMBC Sensitivity for Correlating Poorly Resolved Proton Multiplets to Carbon-13 Using Selective or Semi-selective Pulses

Abstract

Correlations based on scalar coupling between protons ity, in much the same way as the use of selective H pulses in the refocused INEPT (7, 8) experiment can be used to and carbons separated by two or three chemical bonds are very useful for structure determination of organic molecules. transfer magnetization more effectively through long-range H– C couplings (9, 10) . In particular, the heteronuclear multiple-bond correlation experiment (HMBC) has become widely used for this purpose Figure 1 shows three different implementations of the Hselective HMBC experiment. Selection of the appropriate (1) . This experiment enjoys the increased sensitivity resulting from detection of the NMR-sensitive H spin, and, scheme depends on the particular application. Several of the experimental details will be discussed later, but first we in contrast to most of the C-detected analogs, long-range correlation intensities are not affected by the size of the briefly discuss when to use which pulse sequence. The scheme of Fig. 1A is the simplest and is identical to the one-bond JCH couplings. The HMBC experiment requires a relatively long delay for obtaining H magnetization in regular HMBC pulse scheme, except that the H RF power is attenuated and the carrier position is selected such that antiphase with its long-range coupled C, prior to generation of multiple-quantum coherence. Similarly, after the evoluthe proton of interest, but not the protons J-coupled to the proton of interest, experiences the 1807 pulse. If the H of tion period, H magnetization becomes observable only after it is rephased with respect to its long-range C coupling interest is sufficiently separate from its coupling partners in the H spectrum, such that the duration of the 1807 H pulse partner. During these lengthy (30–100 ms) dephasing and rephasing periods, the H magnetization is also subject to does not need to exceed 2–3 ms for achieving the desired selectivity, the scheme of Fig. 1A can be used. If the duration homonuclear H– H J modulation, and the detected H multiplet generally will have significant antiphase contributions of the 1807 pulse needs to be longer than a few milliseconds, the scheme of Fig. 1B is preferred. In contrast to the regular resulting from the homonuclear multiplet structure. Particularly for protons with an unresolved or poorly resolved H– nonselective HMBC experiment, the schemes of Figs. 1A and 1B do not remove the heteronuclear J coupling evolution H multiplet structure, this tends to result in low sensitivity for its long-range correlations to C. Below, we describe of the carbon in the t1 dimension. In practice, this will limit the spectral resolution that can be obtained in the C dimenhow the sensitivity of the HMBC experiment can be increased, sometimes by as much as an order of magnitude, sion. The scheme of Fig. 1C solves this problem and yields a spectrum with full heteronuclear decoupling. In contrast by using semi-selective pulses which suppress homonuclear H– H J modulation. to the schemes of Figs. 1A and B, this sequence permits the spectrum to be recorded in the phase-sensitive mode, further Selective and semi-selective pulses previously have been widely used on the C channel of the HMBC pulse sequence increasing the obtainable spectral resolution. Examples of all three schemes will be shown for the cyclic decapeptide (2–6) . The purpose of using frequency-selective pulses in these experiments was to reduce the spectral window needed gramicidin S. As mentioned above, the scheme of Fig. 1A is essentially in the F1 ( C) dimension of the 2D spectrum, thereby permitting the spectrum to be recorded at higher resolution for identical to the original HMBC experiment, except for the use of pulsed field gradients to achieve coherence-pathway a given number of increments. Note, however, that using selective pulses in this application does not increase the selection, and the use of selective H pulses to ensure refocusing of the homonuclear couplings at time t1 / D after sensitivity of the experiment: The increased number of scans per t1 increment is offset by the lower number of t1 increthe 1807 H pulse. The RF field strength used for the 907 H pulse is arbitrary and may, for convenience, be set to the ments if the total measuring time is kept constant. On the other hand, the purpose of using selective or semi-selective same value as that used for the 1807 H pulse. The 1807 H pulse may be either shaped or rectangular, and its frequency H pulses in the HMBC experiment is to increase its sensitiv-