Gs‐HSQC‐NOESY versus gs‐NOESY‐HSQC experiments: signal attenuation due to diffusion; application to symmetrical molecules

Gradient-selected (gs) HSQC-NOESY type experiments are often applied in order to obtain NOE cross peaks in symmetrical molecules or in complex organic molecules, e.g. carbohydrates. Since the coherence-selecting gradients in these pulse sequences are separated by several delays, including the mixing time, the gs-HSQC-NOESY spectra exhibit severe signal attenuation due to diffusion effects. The respective NOE information unaffected by diffusion effects can be obtained by X-half-filtered gs-NOESY-HSQC spectra. In this paper, an X-half-filtered version of the gs-NOESY-HSQC experiment is presented and the NOE cross peak integrals of its spectra are compared with those of a standard gs-HSQC-NOESY experiment. With phenanthrene as an example of symmetrical molecules, it is shown that the relative signal attenuation of gs-HSQC-NOESY cross peaks versus gs-NOESY-HSQC cross peaks is caused by diffusion effects and follows the Stejskal-Tanner equation. In contrast, the X-half-filtered gs-NOESY-HSQC experiment provides NOE cross peaks with reasonable signal-to-noise ratios even for long-range interactions in the presence of quadrupolar relaxation. This is demonstrated by the spectra of lithium dimethylcuprate.