Hello Everyone;
Two weeks ago I wrote concerning negative glitches that we sometimes see
on the upfield side of strong peaks, always in the directly-detected
dimension, but only with experiments written in-house, not canned Varian
sequences. I would like to thank Paul Keifer (University of Nebraska
Medical Center), Krish Krishnamurthy (Lilly), Jan Runsink (Institut f.
Organische Chemie der RWTH Aachen), and Dimitris Argyropoulos (Varian
Deutschland GmbH) for their responses.
All suggested that this could be due to mis-setting of the lock phase.
(See also Varian NMR News 2003-10-25.) I briefly checked this out, and
found that although there is an effect, it seemed to be relatively small
compared to what I was seeing. However, as Krish implied, this may depend
on the details of the particular pulse sequence and/or parameter set.
Krish also mentioned the issue of the gradient recovery time. I have
found that this can indeed be a very large effect, even though our system
handily passes the gradient recovery test as outlined in the Acceptance
Test Procedures manual. For example, in a diffusion experiment I found
that if I greatly increased the gradient recovery time (from hundreds of
microseconds to ~6 milliseconds) I could eliminate the glitch. I suspect
that the length and the strength of the gradient is also a factor.
However, I still think that there may be other factors at work here. I
ran an in-house Watergate NOESY and a comparable one from Varian's
ProteinPack, using the sample and parameters from a student who had
observed the glitches. (For simplicity I used Z-gradients only; the
gradients were very strong--30000.0--and relatively long (either 4 or 1
msec). In both spectra I could detect the effect, although it was greatly
reduced using the Varian pulse program. What is especially interesting is
that the Varian sequence employed a shorter gradient recovery delay prior
to acquistion than we did. We are still pondering the matter; if we learn
any more I will let you know.
Thanks,
Karl Koshlap
--------------------------------------------------------
Karl M. Koshlap, Ph.D.
Manager, Biomolecular NMR Laboratory
Department of Biochemistry & Biophysics
Texas A&M University
College Station, TX 77843-2128
Tel: (979) 458-2272
Fax: (979) 845-9274
email: koshlap_at_tamu.edu
--------------------------------------------------------
Received on Thu Nov 20 2003 - 16:53:32 MST