Jean-Claude Bradley Says:
October 11th, 2007 at 1:32 pm e
http://www.chemspider.com/spectra.aspx
and for example benzaldehyde:
http://www.chemspider.com/Chemical-Structure.235.html
I think a lot of the errors in the literature are due to poor quality spectra and improper referencing. It is really useful to verify that the TMS peak is indeed exactly at zero and that phasing was done properly. JSpecView used by us and ChemSpider allows people to expand any part of the spectrum and integrate at will, which is very handy for measuring fine splitting patterns, including J constants in H NMR spectra. The problems with JSpecView and the new Java version have been resolved by Robert Lancashire and fixed on ChemSpider.
PMR: I think there are the following subtasks:
- identify the peaks in the spectrum. This could be done automatically or by a human
- identify the solvent
- identify the atom labels in the connection table
- create a peakList where each peak is linked to one or more atoms.
I haven’t looked at JSpecView for a bit and maybe it does (1). If not I have some code somewhere that could be revitalised that uses second derivatives. (3) should be automatic in the generation of CML. Initially (4) must be done by humans, but there could come a phase where the system calculated the shifts and proposed the result to the humans and they then agreed or disagreed. The humans *may* have implicit knowledge which helps the assignment (e.g. relaxation times).
I am really hoping we can find some errors that the method can help with, especially if they are in clodes access publications (open access is junk science anyway).
Just a comment from an NMR spectroscopist…re. TMS. It’s about a decade since I actually ran an NMR spectrum…but in my career I have run 2 NMR labs. NMR technology has come a long way..the systems of today are so stable, so well set up in automation re. solvent referencing that the majority of industrial labs probably don’t even use TMS anymore. I MAYBE used it 2-4 times a year in the last 5 years in industry and only because I couldn’t see the solvent peak to reference. If you ask the deuterated solvent providers I have to believe that their sales of TMS spiked solvents has decreased in recent years. It’s interesting to read that TMS persists in academic labs. There’s nothing wrong with using it …except when too much is spiked in and you get dynamic range issues if you are not using pre-spiked solvents.
(1) Thanks. I can see this might be a problem if some people report relative to actual TMS and some use a machine calibration.
Since the spectrum is relative to “TMS” we have to know that value – ca 200 – so we can subtract the calculated shifts (isotropic tensors). It may be that we have to fit to y = x + c and use c as the TMS offset.
In my lab, especially with new students using our NMRs, we have often had to re-zero the JCAMP file because it was not done properly when the spectrum was saved. Sometimes H NMR are off by a few ppm, where the error is obvious but other times by a fraction of a ppm, which would not be detectable without TMS. My guess is this is typical for lots of academic labs. It makes me wary of reported peaks in articles when I can’t access the original raw data.