Since TOCsY cross-peaks are also in-phase, but of the opposite sign as ROESY cross-peaks (TOCSY: same sign as diagonal peaks; ROESY: opposite sign as diagonal), they will partially cancel the ROESY cross-peak intensities, leading to distances that are too long. However, when the cross- peak used for distance reference contains a large TOCSY contribution(as do all CH2 peaks! ) then the other cross-peaks appear too intense, resulting in distances that are too short Luckily, TOCSY contributions can be minimized by choosing an appropriate spin-lock sequence. The TOCSY transfer relies on perfect synchronization of the two spins involved, while the non-coherent ROESY cross-relaxation is much more robust with respect to imperfect alignment of the two spins during Tm. This leads to the following possibilities for suppression of TOCSY contributions choose a weak spin-lock field, typically 2-4 kHz( depending on spectral width in Hz to be covered i.e., spectrometer frequency)! For the ROESY transfer, this will only cause a faster decrease of cross- eak intensities from the center of the spectrum towards its edges (cf. equation for 0). For the TOCSY transfer, the performance drops dramatically(usually spin-lock fields of 8-10 kHz are used) choose a"bad"spin-lock sequence instead of the optimized sequences used for a TOCSY(MLEV, WALTZ, DIPSI). Good results can be achieved, e.g., with just continuous(CW) irradiation,or thousands of evenly spaced short pulses(<< 90, equivalent to a Cw spin-lock). A spin-lock consisting of just a train of 180° pulses with inverted phase(-180°x-180°×-) also works well Even with these"poor"spin-lock sequences, there are still certain regions in the spectrum where cross-peaks might contain significant TOCSY contributions close to the diagonal(again affecting especially CH2 groups) with all spin-lock sequences, and also close to the counter-diagonal crossing the diagonal in the center under a 90 angle The spin-lock sequence of alternating 180 pulses does not produce toCsY peaks along the counter-diagonal, but pretty severe ones along the diagonal) If an important cross-peak (involving two protons with a scalar coupling! would fall onto the counter-diagonal, the spectral window can be slightly shifted, so that the transmitter frequency is moved away from the center between the two protons, and the potential roEsY cross-peak does not fall onto the counter-diagonal anymore For two peaks with very close resonance frequencies(cross-peak close to the diagonal), TOCSY contributions cannot be avoided!106 Since TOCSY cross-peaks are also in-phase, but of the opposite sign as ROESY cross-peaks (TOCSY: same sign as diagonal peaks; ROESY: opposite sign as diagonal), they will partially cancel the ROESY cross-peak intensities, leading to distances that are too long. However, when the cross￾peak used for distance reference contains a large TOCSY contribution (as do all CH2 peaks!), then the other cross-peaks appear too intense, resulting in distances that are too short. Luckily, TOCSY contributions can be minimized by choosing an appropriate spin-lock sequence. The TOCSY transfer relies on perfect synchronization of the two spins involved, while the non-coherent ROESY cross-relaxation is much more robust with respect to imperfect alignment of the two spins during tm . This leads to the following possibilities for suppression of TOCSY contributions: choose a weak spin-lock field, typically 2-4 kHz (depending on spectral width in Hz to be covered, i.e., spectrometer frequency)! For the ROESY transfer, this will only cause a faster decrease of cross￾peak intensities from the center of the spectrum towards its edges (cf. equation for q). For the TOCSY transfer, the performance drops dramatically (usually spin-lock fields of 8-10 kHz are used). choose a "bad" spin-lock sequence instead of the optimized sequences used for a TOCSY (MLEV, WALTZ, DIPSI). Good results can be achieved, e.g., with just continuous (CW) irradiation, or thousands of evenly spaced short pulses (<< 90°, equivalent to a CW spin-lock). A spin-lock consisting of just a train of 180° pulses with inverted phase ( - 180°x - 180°-x - ) also works well. Even with these "poor" spin-lock sequences, there are still certain regions in the spectrum where cross-peaks might contain significant TOCSY contributions: close to the diagonal (again affecting especially CH2 groups) with all spin-lock sequences, and also close to the counter-diagonal crossing the diagonal in the center under a 90° angle. (The spin-lock sequence of alternating 180° pulses does not produce TOCSY peaks along the counter-diagonal, but pretty severe ones along the diagonal). If an important cross-peak (involving two protons with a scalar coupling!) would fall onto the counter-diagonal, the spectral window can be slightly shifted, so that the transmitter frequency is moved away from the center between the two protons, and the potential ROESY cross-peak does not fall onto the counter-diagonal anymore. For two peaks with very close resonance frequencies (cross-peak close to the diagonal), TOCSY contributions cannot be avoided!