I'm trying to find a program/algorithm/API that can find the equivalent atoms in a molecule when given a molecule (knowing the complete geometry with x, y and z coordinates, and bonds between the atoms). Equivalent in the sense that they show up as one peak in the NMR spectrum.

Example: I supply ethanol and the program needs to return that the three H's on one C are equivalent and that the two H's on the other C are also equivalent. So that my program knows that it should only have 3 (sets of) peaks in my NMR spectrum and not 6.

  • 1
    $\begingroup$ Without 3D structure, this will be a heuristic at best. Also, equivalency is a much stronger condition that show up as a single peak.There is also chemical equivalence and magnetic equivalence. What are your actually asking for?. $\endgroup$
    – Zhe
    Sep 20 '17 at 13:04
  • $\begingroup$ As Zhe said. The two hydrogens on the oxy-carbon are actually enantiotopic, not homotopic, so in a chiral environment they will give two peaks. $\endgroup$
    – Jan
    Sep 20 '17 at 13:27
  • $\begingroup$ @Zhe I do have the complete geometry of the molecule with all the x,y and z coordinates of the nuclei plus the bonds that are present (and if they are single, double or triple). I am personally more interested in the magnetic equivalence since this creates a different spectrum than just chemical equivalence, but if an algorithm for magnetic equivalence is too expensive then I may settle for an algorithm that just calculates the chemical equivalence (if it is cheaper). $\endgroup$
    – LinG
    Sep 21 '17 at 7:58
  • $\begingroup$ Some drawing programs come with a heuristic prediction feature for NMR. Try ChemDraw and /or ChemSketch. $\endgroup$
    – TAR86
    Sep 21 '17 at 9:54
  • $\begingroup$ For chemical equivalence, you just need to check the environment of the atom, i.e., the spatial arrangement of all other atoms, and compare that to the environment of the other atom. In principle, chirality matters, though in general, NMR is achiral. The big caveat is watching out for conformational changes that are fast on the time scale of NMR. That would lead to an averaging that could merge some peaks. $\endgroup$
    – Zhe
    Sep 21 '17 at 13:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.