Sciencemadness Discussion Board

1H NMR predictor

budullewraagh - 14-11-2006 at 15:41

I have a problem set due tomorrow and I need some help assigning a few peaks on some of the spectra. Is there a reliable proton NMR predictor available for Mac OSX that anyone here knows of?

Drunkguy - 15-11-2006 at 03:25

I think programs such as chemdraw and hyperchem have this facility, although I have never used them.

budullewraagh - 16-11-2006 at 20:17

chemdraw ultra has it. even the nmr guru at my former lab doesn't have access to it.

Phel - 17-11-2006 at 12:12

I believe there's a copy on one of the FTPs, and I got a copy from bittorrent. But IIRC they are both for Windows.

lakersno834 - 3-9-2007 at 16:08

I belive ACD makes NMR predictor softwares.. does anyone have it.. If yes. Can someone please post it here?

Phel - 4-9-2007 at 08:47

lakersno, did you try searching for chemdraw on isohunt as I suggested in the PM?

vulture - 4-9-2007 at 09:36

Chemdraw H-NMR prediction is pretty shitty to say the least. ACD is better, but there is still a significant error margin which one must consider.

lakersno834 - 4-9-2007 at 13:51

Originally posted by Phel
lakersno, did you try searching for chemdraw on isohunt as I suggested in the PM?

I did try isohunt but apparently they don't have the latest chemdraw ultra (v. 10) .. they have biodraw.. but not chemdraw..

I looked at forums.. and i found back in the days someone posted the ACD 1H NMR predictor.. however the link is broken and i cannot get it. ACD NMR predictor is better than chemdraw.. so I was wondering if i could get the ACD.. if not chemdraw will work fine for me.. but preferably version 10 (sorry for being too picky).

Thank you!

Sauron - 4-9-2007 at 14:16

On two or three occasions I downloaded ACDLabs from various links in References, a laborious process, but found that the supplied SN does not work.

It is a waste of time. So in this instance, encountering a dead link is better.

If I were in your shoes I'd concentrate on ChemDraw.

lakersno834 - 4-9-2007 at 20:58

If one could provide a link to download the latest chemdraw or ACD, it would be great.. A reliable third party link will also be good. Thanks!

trilobite - 5-11-2007 at 16:59

One should never believe there was a truly reliable NMR predictor. Chemdraw seems to give the delta value of 5 for every phenolic hydroxyl, even for o-nitrophenol whose phenolic hydrogen gives a sharp peak (due to hydrogen bonding) at 10.57 in CDCl3. Fortunately it does remind you that the estimate is rough.

Then if you go and draw the enol form of acetylacetone, the programs are going to miss the fact that hydrogen exchange between the two oxygens is so fast at the temperatures you are most likely using for your measurement, that the molecule will appear symmetrical. In the actual measurement the methyl hydrogen peaks are overlap perfectly and one will see only three different C13 signals for the enol form. The hydroxyl hydrogen signal is again a sharp one because of hydrogen bonding. And with increasing molecular complexity the situation can very soon get pretty hairy for several reasons. The reality is far too complicated to be properly expressed with chemical drawing programs.

That said, the predictions these programs give are useful if you understand when they are likely to fail but are too lazy or busy to do the same calculations by hand.

Nicodem - 4-4-2008 at 07:28

Just stumbled upon this online 1H NMR predictor and remembered to post here:

I only checked for a couple of compounds and it does seem quite good, at least for those who do not have ChemDraw installed or other software supporting NMR prediction. The only drawback I noticed is that it refuses to predict exchangeable protons - which on the other hand is not so bad since every NMR predictor generally misses these proton shifts for a long shot anyway. The best part is in that you can find the multiplicities written out in the output (which for some reason ChemDraw doesn’t give).

It even supports SMILE codes being incorporated into the URL so that one can link a structure to be processed. For example, the URL for prediction of vanillin 1H NMR spectra:

(Here is the real NMR spectra of vanillin from Sigma in case someone wants to compare)

Interestingly they also provide an NMR spectral database. However I could not find any information about the number of spectra they have (I could not find vanillin there :P):

Yet, quite a nice addition to the SDBS and Sigma-Aldrich NMR databases.

And some other tools…

[Edited on 4/4/2008 by Nicodem]