Dear BUMs,
Many thanks goes to Phil, Sara, Nick, Mark Cool, Donna, Kees Erkelens and Dan
for their helpful replies.
There are two ways to get reasonable 1D 13C spectra on the cryoprobe nmr
machine. Both of them work well on our NMR machine.
1) increase DE from 6usec to 20-25usec and re-run rga. The only way to get good
13C baseline is to throw away the first few (16 or 32 or 64 or ??) data points
and then do a back linear prediction. Phil wrote an AU program to calculate rg
using 100usec DE and automatedly put the DE back to the original one such as
6usec and he also wrote a macro program to run the back linear prediction
automatedly.
2) avoid the problem by using a C13 spin_echo pulse program.
With best regards,
Flora
Phd students,
The department of chemistry,
The University of Hong Kong.
-------------
Here are some replies
-------------------------------------------------------------------------
Dear Flora,
I have a TCI cryoprobe, and we always get a bog spike at the
beginning of 13C FIDs. My solution is to temporarily increase DE when
setting the receiver gain, and then use reverse linear prediction to
correct the FID when processing. I use the attached au program 'rgac'
to set RG, and then the macro 'roll' to tidy up the data.
Hope this is of help,
Phil.
-- Dr Phil Dennison NMR Facility Director (949)824-6010 (office) Department of Chemistry (949)824-5649 (lab) University of California, Irvine (949)824-8571 (fax) Irvine, CA 92697-2025 dennison@uci.edu USA -------------------------------------------------------------------------------Flora, If this is a TCI probe like ours? We avoid the problem by using a C13 spin_echo pulse prog to avoid the large initial spike from solvent (esp DMSO or C6D6). Then rg can be set at a reasonable 8K or higher. This pp also removes the broad lump at @110ppm picked up presumably from the glue in the PFG coil of our inverse probe. I would be interested in others comments. I can send the pp to you if you wish. regards, Nick
Nick Burlinson PhD Director NMR Facility Chemistry Dept. Univ. British Columbia Vancouver BC V6T 1Z1 604-822-6787(nmr lab) 604-822-6956(office) ------------------------------------------------------------------------------- Hi Flora,
The rolling baseline comes from the spike at the beginning of your fid. To avoid this, try increasing your DE from 6usec to 20usec and re-running rga. You can eliminate the spike in your current dataset by setting in edp the following: ME_mod to LPbc, NCOEF to 32 and TDoff to say about 32-64 to replace the first few points in your fid that make up the spike. Then type efp. If your data had been acquired with AQ_mod = DQD (as opposed to qsim) you would have to first type convdta before doing the efp. This would eliminate the decimated points that are at the beginning of oversampled or dqd data.
Good luck.
Regards, Donna
Donna Baldisseri, Ph.D. Senior Applications Scientist 978-667-9580 x5381 ------------------------------------------------------------------------------- Dear Flora, We have the same problem with our 13C cryoprobe. The carbon channel of the probe has a very long deadtime. The only thing that helps is using a long DE ( 25 usec),acquirr the data in the analog mode ( DIGMOD analog) and do some backwords Lineair prediction ( ME_mod?LPbc with TDoff 16 to 32 and NCOEF 1000) In this way you get a accepable 13C spectrum. APT, DEPT and Inept are all spin-echo experment so you do not have any problems with long deadtimes. With kind regards, Kees Erkelens ------------------------------------------------------------------------------- Dear Flora, Did you get a response from anyone? Here is my rexperience... rga on a cryoprobe is useless, especially for 13C on our system (700 & 800 with TCI cryoprobes, dru's). It should be near max gain, although that is not critical with a CryoProbe (because the S/N does not improve much with gain once you are above a certain level. You can increase DE from the nominal 6 microseconds to try to avoid the spike at T=0. The only way to get good 13C baseline is to throw away the first few (32 or 64 or ??) data points and then do a back linear prediction. You convert the data to 'analog' with convdta command (you must specify an new expno to hold the converted data... I usually justadd 100 to the current experiment number). Then you set the backprojection parameters abd then transform. I hope this helps; if you need more detail, just let me know.
Dan
--------------------------------------------------------------------- original information -----------------------------------------
Dear Bruker Users,
Happy new year:)
I recently run a 1D small organic molecular NMR spectrum on our AV600 with cryoprobe, when I input "rga" and the automatedly calculated rg was 7. After 3 hours even overnight, we can not get any signal. After increase the rg to 4K, I can get normal signals but with sinuose baseline. There is an unusual large spike at the beginning of the fid spectrum. There are the 1D spectrum http://147.8.148.20/public/yyh/1D-C13.JPG and fid pectrum http://147.8.148.20/public/yyh/13c-fid.JPG
I have tried other 1D 13C pulse program and got the same result at all times. I also run another sample on the machine, the result was just the same. However, all 1H proton and dept spectra behave well on the same machine.
Thanks in advance!
Flora PhD student The department of chemistry, The University of Hong Kong.
This archive was generated by hypermail 2b29 : Wed Dec 26 2007 - 09:47:03 PST
