cheminfo / nmrium

React component to display and process nuclear magnetic resonance (NMR) spectra.
https://docs.nmrium.org
MIT License
55 stars 25 forks source link

'massif' versus 'multiplet' #1215

Open lpatiny opened 3 years ago

lpatiny commented 3 years ago

I tend to make the difference between a multiplet, one kind of protons for which the multiplicity is too complex to be determined and a 'massif', a range in which we have couple of different protons.

In a publication I would write:

However I think the term 'massif' does not really exists in english. Is there a better terminology to make the difference between those 2 cases ?

nes-cgn commented 3 years ago

Hello Luc, could you describe this a bit clearer or with an example? Do you mean, there is a range of chemical shifts which correspond to signals of several protons, maybe? In general, describing such kind of a 'massif' (i.e. a range of chemical shifts that covers overlapping signals of several atoms) is tricky, because the 'range' depends on the strength of the magnet. I am attaching an example (Org Lett) where I marked with two blue circles your two cases

Lucexample.pdf

If that is what you mean, I would describe it as I did (range of overlapping signals)

lpatiny commented 3 years ago

@nes-cgn Consider only the blue and green H in the following molecule:

HT H

If in a specific place there are only NMR equivalent atoms and the multiplicity is too complex to be described (blue H) then I would put 2.0 (m, 1H). While if we have in a range non NMR equivalent atoms I would but a range 1.4-1.6 (m, 6H)

jliermann commented 3 years ago

@lpatiny I understand your point but I am not aware that there is any common standard terminology for that differentation. I have also come across the term "mc" meaning "centered multiplet" which would correspond to the multiplet or the blue proton in your example.

In terms of software design, my suggestion would simply be to let the users decide whether they want to define a multiplet by a center frequency or a range.

nes-cgn commented 3 years ago

Well, the charming thing of the ranges picking is that there is a center frequency for each defined range. If this is connected with the HSQC picking along the 1H dimension (I think Alejandro and Michael are working on that), it would be the better alternative. As I said, the range (1.4 ... 1.6 ppm) may be different depending on the spectrometer

djeanner commented 3 years ago

What is unknowned when people use "m" in spectral description may be:

In any case, I think that whatever the name NMium distinguishes regions (that are just a part of a spectrum - therefore having boundaries) from the properties of a proton signal (which normally has a (single value) chemical shift). In a spectral anaylsis workflow, you start from regions (of the spectrum), and try to assign and reach NMR parameters of nuclei of the compound. In case of strong coupling you may not know the chemical shift or chemical shifts if the strongly coupled spins overlap. In this case, you have a problem to give chemical shifts to the nuclei and it is tempting to provide the range. For single spin regions it would be OK, even if I would probably prefer to give a single chemical shift in the middle of the region (midle defined according to the integral) but if you have multiple strongly coupling nuclei, you have a problem and may want to propose the range.

In the NMReDATA, properties of spectral regions can be a range or a single value. But for chemical shifts (of a nucleus), we do not accept ranges, we want a single chemical shifts. In the problematic case mentionned above, you would have no chemical shift for the nucleus, but the region of the spectrum could have the (multiple) nuclei assigned to it. (Indeed, a spectral regions may have one or more label assigned to it - a label corresponds to one or more isochronus nuclei - aka chemically equivalent).

I hope that this is helping.