High speed spinning of samples

Slow sample spinning (like few tens of Hz) has been used for decades to average out small magnetic field inhomogeneities seen by different parts of the sample. It has also been known for a long time that spinning the solid sample rapidly about an axis tilted with respect to the laboratory 

3cos 0 -1=0) eliminates those parts of interactions like the m 

Since the spinning rate required to average an interaction is approximately equal to the interaction itself, a spinning rate of a few to tens of kHz is required for the usual solid with an abundant NMR nucleus having a dipolar interaction of several kHz. As Zilm, et al. (1978) have pointed out, the centripetal acceleration at the rim of a 1 cm diameter sample spinning at 10 kHz is two million g s which gives an indication of why this technique has remained a difficult one for so long. 

At the time of this writing, a rapidly growing area of research within NMR is the high resolution study of solids brought about by the development of certain multiple pulse sequences (such as the WAHUHA and MREV-8) on the one hand and an adaptation of the Hartmann and Hahn cross polarization experiment to obtain spectra of dilute spins on 

A novel suggestion by these authors is to make the spin-axis be 45 degrees with respect to the support stem. In that way, the adjustment of the magic angle is easier in an iron magnet because of the relatively small dependence of the angle on the orientation of the support stem about its symmetry axis. 

---