Summary of suggestions received on 'Resonance distortions when using wet
for solvent suppression'
On Wed, 6 Sep 2000, I wrote:
>We have been using the wet solvent suppression sequence for
>suppressing the water resonance in 1D proton spectra. When we do
>this on our pfg id probe on the Varian Inova 500 we observe a
>distortion on resonances removed from the suppression point. The
>distortion is most noticeable on sharp intense singlets. Such a
>distorted singlet seems unusually sharp on one side and has a marked
>negative spike, whilst the other side is normal or maybe somewhat
>broadened. Its like I've applied resolution enhancement to one side
>of the resonance and line-broadening to the other side! The effect
>is present even using only one transient and no steady state pulses.
>
>Has anyone seen this and if so an explanation? It would also be nice
>with a solution as well!
The following replies were received:
1) This distortion is a result of "lock pull". Make sure you are right on
resonance with Z0 and your lock phase is properly adjusted.
2) This sounds like a gradient-recovery / lock-stability problem. It is
possible that your setup is not optimized or that your gradient coil
is not performing correctly (broken). The WET sequence works
beautifully in our hands.
Things to consider:
(a) Be sure the lock-phase is adjusted properly for pure dispersion
(maximum lock signal intensity). Even a small error, few percent,
can induce a frequency shift in the lock channel and
"field-wandering".
(b) Check the gradient recovery delays. You need many tens of
microseconds on the latest *Varian* probes and hundreds of
microseconds on the earlier *Varian* probes (or perhaps even longer).
Obviously, other probes (Nalorac, modified Bruker, homebuilt) may
need something longer or shorter.
(c) Check that the gradient probe is responding properly to gradient
pulses. The easiest way to do this is with the Varian test routines
or craft your own. Also, be aware that on older (usually
retrofitted) magnets, the entire magnet assembly, shim coils, etc.
can be giving you problems.
(d) Check that your probe is centered in the magnet/shim-coil.
If all this checks out then you may have something wrong with your RF
system? in the way that the pulse train is generated...purely a guess
there.
3) You probably need more time between the last gradient pulse and the start of
acquisition (for the WET sequence, between the last gradient pulse and the 90
degree pulse). What you desribe sounds like the gradients have not settled
prior to acquisition. When I was doing bio-NMR, I used my own version of the
WET sequence so I don't know the Varian parameters that you can change. But if
you make that delay longer, you should solve your problem. These negative-type
peaks are generally the result of an unstable field (due to eddy currents,
etc.). I guess the delay can't be too long or the solvent protons will have too
much time to recover.
Also, I have found that if you are not right on resonance with the lock, you can
observe these problems. You might just try a 1-pulse experiment without lock to
see if the distortions go away. In the version of WET that I used, I blanked
the lock during the gradient pulses. I don't think that the Varian version does
this.
4) I have had a similiar problem on my Bruker instrument. The affect is
called "leed pickup". I have been unsuccessful to completely eliminate
this problem from my instrument . What I have done is decrease the
gradient power slightly, adjust the power of the last shape pulse, and
readjust the 1H 90 saturation power all seemed to decrease the problem.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
Thank you very much for your responses. I have checked out the lock-phase/frequency
suggestion and this scored bullseye. The lock frequency did not seem to be very
important, but setting the lock-phase correctly gave a perfect spectrum and
missetting it reproduced exactly what we have been observing. Joy!
Thank you all once again - you have all earned your coffee today or
whatever the politically correct drink for NMR spectroscopists is at the
moment.
best wishes,
David Grace
Nycomed Amersham
Manager NMR group, Oslo
voice +47 23185006
fax +47 23186014
Received on Thu Sep 07 2000 - 10:06:51 MST