Rotor-driven spin diffusion

Proton and r.f.-driven spin diffusion are nonselective and ideally influence all carbon spin pairs in the same way. Rotor-driven spin diffusion, however, is highly selective and enhances the spin-diffusion rate constant only if the spinning speed is matched to the isotropic chemical-shift difference (Equation... 

Figure 46 (Top) Experimental 31P NMR spectra of three component sample crystallised from toluene (A) powdered sample (PS) recorded with CP/MAS sequence and spinning rate 8 kHz (B) three-component single crystal (TCSC) A held in rotor filled with silica gel recorded with CP/MAS sequence and spinning rate of 8 kHz. Note the much better NMR resolution of resonance lines and small distinction of 31P chemical shifts for the monocrystal compared to powdered sample. (Bottom) 31P-31P proton-driven spin diffusion 2D correlation recorded with mixing times of (A) 0.2 and (B) 10 s. Taken from Ref. [229].

The dynamics of the rotor speed can be conveniently analysed and adjusted by feeding a signal from the rotor motion monitor (via an optical fibre) to a console ADC (Fig. 7). The corresponding spectra are shown in Fig. 8. The faster sweep clearly reduces K12, and to a lesser extent K13. Both direct complementary peaks (1-Kmn) and relayed peaks depend on the product of complementary transfer coefficients, and correlate well with the expected influence of the sweep rate variation. The observed relay process can be shorted by proton-driven spin diffusion. Consequently, efficient rotation-speed independent (or carefully synchronized) decoupling is required during the entire mixing period. 

---