[asa029]

Hi,

Many of my compounds contain both aromatic and aliphatic carbons. These vary from primary to quaternary. Characterization with 13C-NMR sometimes gives spectra where the aromatic carbons have positive intensities and the aliphatic have negative intensities (or the other way around) (see attached image).

It's been too long since I had any formal NMR-training, but I seem to remember that this has to do with the carbon relaxation times. Is this correct? I usually go for a delaytime (d1) of 10 seconds. I tried to muck around with variations on d1, but that didn't really help. How do I progress to find the correct d1-times? Is it just trial and error?

[azmanam]

On first glance, it seems like a post-FT phasing problem.  Immediately after running the Fourier transform, try manually phasing to a) flatten out the baseline of your CDCl₃ peaks, and b) perhaps swing your negative peaks to positive.

If that doesn't fix it...then it may be a relaxation issue - and I don't have any experience correcting those issues.

[OC pro]

Definitely a problem with phase correction which is only a software problem. The measurement is just fine. Try to use a proper NMR software, e.g. Mestrec.

[nox]

3rd'd for phase correction issue.

Although I guess we should mention it is possible to obtain "legitimate" negative peaks when you run a DEPT-135 experiment.

[OC pro]

But the spectrum above is not a DEPT. Simply the phase correction is not properly fixed.

[nox]

Yes I'm aware of that. My point was OP should be informed that DEPT-135 does give negative peaks, so they're not mistaken for phase correction problems should such an experiment be run in the future.

[asa029]

Doing software phase-correction does not result in aliphatic and aromatic peaks being on the same intensities (like ++ or --) but always results in +- or -+. I therefore don't think this is a phase-correction issue. I normally always phase-correct all my spectra and the spectrum shown in my first post is allready phase-corrected, both using software automatic and manual phase-correction.

I think this is problem has to do with relaxation times.

[enahs]

Doing software phase-correction does not result in aliphatic and aromatic peaks being on the same intensities (like ++ or --) but always results in +- or -+. I therefore don't think this is a phase-correction issue. I normally always phase-correct all my spectra and the spectrum shown in my first post is allready phase-corrected, both using software automatic and manual phase-correction.

I think this is problem has to do with relaxation times.

You phased poorly, there is no question about it. It is not an insult, it is sometime hard. You can not really go much further until you phase it correctly. Ingratiating on poorly phased NMRs is not a good idea.  You can attach the raw XY points and I can help re-phase it.

Not really sure what you are trying to say by intensities. Intensities are irrelevant.


Also, sometimes when you have signals so far out of phase it is a sign of the probe not been tuned properly, or the pulse-width way off, or a combination of both.

Note, it would be really helpful if you would attach what you think the molecule is, or what you are hoping for.

[asa029]

You phased poorly, there is no question about it. It is not an insult, it is sometime hard. You can not really go much further until you phase it correctly. Ingratiating on poorly phased NMRs is not a good idea.  You can attach the raw XY points and I can help re-phase it.

Not really sure what you are trying to say by intensities. Intensities are irrelevant.

I've taken _many_ 13C-spectra in my life, and I have phased all of them (its not that hard!), so I know how to do it. In this case (for the spectrum shown) there are no way of phasing so that all signals "face the same way" (have the same intensities).

I would never integrate on a 13-C spectra, by the way.

[ATMyller]

Manual phase correction should fix most of that, but 10 seconds is rather small delay for 13C spectra. If you are planning on integrating those peaks I suggest you use inverse gated pulse program and 60+ seconds D1 times to get results that can be considered even remotely quantative.

[arit]

I would also say it's just a matter of phasing.

I'd also like to hijack this thread for a bit, and ask whether people generally acquire their 13C-spectra using small flip-angles
and short repetition delays, or do you go for quantitative spectra?

[zolarpwr]

To the OP, looks like a phasing issue to me too. Do a first order phase correction on the largest peak in the spectra (solvent probably), then do a second order to get the peaks farther away to line up properly. In unusual circumstances this still isn't enough but then I'd blame it on an instrument issue.

To the last poster, for routine C-13 spectra small flip angles and fast pulses with short delays is the way to go. This is, of course, if you don't plan on trying to integrate the peaks for quantitative work.

[asa029]

Normally I use Topspin for spectra analysis and manipulation, but after the problems with phasing I switched to Meternova. Using the phasing options for this program, the spectrum was perfect (ie, all the peaks were "facing the same way"). The problem seems to be with the software. Weird, though! Has anyone else had this problem? According to people at my faculty, this happens from time to time.

[nox]

I love MNova! It's easy to use and the keyboard shortcuts are awesome.

I do sometimes notice "phasing" issues with MNova as well, but 99% of the time it's because the sample wasn't shimmed properly, rather than a software issue.