I received a request to summarize the suggestions on how to do Silicon-29 on our
AC-200. After teaching both summer sessions and some other lame excuses for not
doing this sooner, here goes.
Everyone agreed on how to set O1 and SW to avoid folding the spectrum. That is,
start with SW at 100000, find an O1 for which a signal can be obtained in one
scan, and decrease SW without changing the frequency of the peak.
Also, everyone agreed on how to find O2, i.e., get a proton spectrum and set O2
in the middle.
Not everyone agreed one how to decouple. The list of suggestions follows:
1. Use CPD
2. Don't use CPD because of the negative spin and resulting negative NOE.
3. Use CPD in combination with INVGATE.AU
4. Decouple during acquisition only. (also INEPT with decoupling during acq.)
5. Use BB decoupling.
For us, CPD did not result in a decoupled spectrum. Obeying the KISS principle,
I tried the next easiest thing to do. That was BB decoupling. That
successfully decoupled our standard TMS sample (NS = 1 is sufficient). I have
not tried any of the other choices. I suspect that for a dilute sample
requiring many pulses, the negative NOE will be a problem if simple BB
decoupling is used. Items 3 and 4 above have been reported to be successful.
Which one works probably will depend on the particular sample and its T1.
Finally, the question regarding the relationship amongst SF0, O1 and O2 had two
differing answers. The more common answer was that there is no relationship.
The other answer is "SF0 is the 'base-level' offset. It is the spectrometer
frequency in the absence of any offset correction, O1=0. Changing SF0 by 100
kHz means you need 100 kHz less O1 (or more, depending on the sign) to maintain
the same spectrometer frequency." Nothing was said about how that affects O2.
Bottom line: Empirical observation is easiest.
Thanks again for all that responded to my request.
Jim
James C. Howard, Ph.D.
Associate Professor
Department of Chemistry
Middle Tennessee State University
Murfreesboro, TN 37132 USA
JHOWARD@A1.MTSU.EDU