ScienceSoft, LLC
www.ScienceSoft.net
TOC PREV NEXT INDEX


CHAPTER 2: The Need For Magnet Shimming

Magnet shimming remains one of the few areas of NMR where automated procedures have failed to replace the tedious manual optimization of experimental conditions. Even experienced spectroscopists spend a significant amount of spectrometer time on "touching up" shims before each data acquisition. In cases where a good starting set of shims is not available, such as when bringing up a magnet or installing a new probe, the magnet shimming can take hours or even days to complete. Shimming is often regarded as a "black art"1 requiring a good knowledge of the particular magnet and extensive experience with magnet shimming in general.

The field distortions of an NMR magnet over the sample volume should be no greater than a fraction of the natural linewidth of the slowest relaxing nucleus to be observed. For a 750 MHz magnet and a 0.1 Hertz proton resonance linewidth this calls for a field accuracy of better than 1:7,500,000,000. Raw magnetic fields are typically several orders of magnitude less homogeneous. Hence shim coils of different shape are placed inside the raw magnet to produce additional magnetic fields to compensate for remaining inhomogeneities. State-of-the-art NMR super-conducting magnets have typically between 10 and 40 room temperature shim coils.

The best shim setting could be determined analytically if the shape of the raw magnetic field, the shape of all shim fields, and the field distortions caused by probe and sample were known. In NMR imaging (MRI) the shape of the magnetic field can be approximately determined through the use of field gradients. In the absence of field gradients, fields would have to be mapped mechanically using specially designed probes. In either case, shimming through field determination is both expensive and of limited accuracy.

Most shim coil sets are build to produce fields which resemble spherical harmonics. The shape of a magnetic field produced by a shim coil should hence be known and the effect of different shim fields should be independent. In practice however, every coil produces higher order effects and, due to design compromises and winding imperfections, lower order effects as well. So the design of the shim set might be used as a guiding principle during the shimming process but the final shim result should not depend on any assumed functional form of the shim coils.

1Chmurny, G.N.; Hoult, D.I.; "The Ancient and Honourable Art of Shimming"; Concepts in Magnetic Resonance 1990, 2, 131-149.



ScienceSoft, LLC
www.ScienceSoft.net
TOC PREV NEXT INDEX