14
$\begingroup$

Said software is useful when you need to assign the peaks in vibrational spectra to different vibrational modes found in a hessian calculation for a given molecule, as described in Jamróz, Michał H. “Vibrational Energy Distribution Analysis (VEDA): Scopes and Limitations”. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, vol. 114, outubro de 2013, p. 220–30. ScienceDirect, doi:10.1016/j.saa.2013.05.096. It's becoming popular, the paper has nearly 450 citations as of april 2020, but it has some downsides for me:

  • Despite being a freeware that can be downloaded from their site by anyone, it's not free software, or at least open source, and it's windows-only;
  • Doesn't support people using free tools like Gamess-US or Psi4, only accepting Gaussian output files as its input, a paid and very expensive quantum chemistry software;

So I did some research for alternatives that could avoid these downsides. Among the citations to the VEDA paper, I've found what seems to be a alternative approach to solve the same problem, as described in Teixeira, Filipe, e M. Natália D. S. Cordeiro. “Improving Vibrational Mode Interpretation Using Bayesian Regression”. Journal of Chemical Theory and Computation, vol. 15, no 1, janeiro de 2019, p. 456–70. ACS Publications, doi:10.1021/acs.jctc.8b00439. Their software, called vibAnalysis, is open source, and available from a GitHub repository. It doesn't support Gamess-US or Psi4, but at least supports ORCA, which is free for academics, besides Gaussian.

I wonder if there are other alternatives I have missed in my research, specially if they run on linux and work with Gamess-US or Psi4. If somebody here could provide some guidance, I'd be grateful. I posted this question in chemistry.stackexchange at first, before materials.stackexchange entered beta. As it went unanswered, I'm giving it a try here.

$\endgroup$
  • 3
    $\begingroup$ PGOPHER is free and can be used to view vibrational peaks, but probably doesn't support GAMESS-US or Psi4. Would it be difficult to convert the GAMESS-US or Psi4 output into ORCA format? $\endgroup$ – Nike Dattani May 5 at 16:12
  • 4
    $\begingroup$ Hi Ksousa / Nike, You may be able to use the Molden software to read the Gamess-US software output and save the matrices into the Molden format. You then may be able to import the Molden format into PGOPHER; which can simulate and fit vibrational spectra. cheminf.cmbi.ru.nl/molden/molden_format.html pgopher.chm.bris.ac.uk/Help/index.htm pgopher.chm.bris.ac.uk/Help/importlmatrix.htm $\endgroup$ – Su Stephens May 5 at 16:41
  • 1
    $\begingroup$ 10 days later, still no progress? It's been posted on twitter, and bumped up to the top on May 20th and then again on June 15th. There's not much more we can do now! Maybe you can put a bounty on it after all that rep you earned from the O(N logN) question! $\endgroup$ – Nike Dattani Jun 25 at 17:19
  • 1
    $\begingroup$ @NikeDattani, I still could not complete the full sequence proposed. I'm used to Avogadro as a GUI, and the Molden interface is a bit odd. Still figuring out how to use it. $\endgroup$ – ksousa Jun 26 at 18:38
  • 2
    $\begingroup$ I'm so sorry it didn't work out. She saw my message on Messenger, but was unable to reply (perhaps too busy). I wish the bounty was more successful, but I hope this doesn't turn you away from issuing one on a different question in the future. This question might have just been to specialized for the very small and narrow user-base we currently have. Today we reached 1000 registered users, but hopefully 1 year from now we will have closer to 10,000, and situations like this will turn out happier. $\endgroup$ – Nike Dattani Jul 4 at 1:42
6
$\begingroup$

LocVib (part of MoViPac)

I have never used VEDA before, but according to the paper, it decomposes the normal modes into vibrations of atomic groups, bond stretching, bending or something else. Something similar is LocVib, which localizes normal modes to localized vibrational modes by maximizing the "distances" between vibrational modes and "atomic contribution" to a mode, like localization of molecular orbitals.

LocVib is not same as VEDA, and the operation is somehow more complex. The user need to select a subset of all normal modes on their own and the localization would be done by the software, and finally localized modes, typically related to several atoms will be printed out, with derived frequencies from the normal modes.

The software is in Python 2 and the github page is https://github.com/chjacob-tubs/LocVib. It is also part of MoViPac.

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ +1. Welcome back!! $\endgroup$ – Nike Dattani Jul 11 at 3:21

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.