Viewing Vibrational Spectra of van der Waals Clusters

General Discussion Support
1

Hi there,

I am a student researcher, and am currently using the earlier 1.2.0 version of Avogadro for my academic research. I have been successful in generating an IR spectra for several gaseous molecules that I am hoping to collect data on, such as H2O and CO2; however, would like to get some assistance on how to accurately perform the same procedure with multiple molecules in one file?

For example, I have a file with multiple H2O molecules and have run the auto-optimization with the MMFF94s force field, but generating a spectra of that produces a long list of absorption frequencies (which may be accurate due to the changed molecular dynamics of the file). However, one notable point is that the Avogadro CML file views the separate H2O molecules as one molecule with the chemical formula of “H12O6,” for example.

I would greatly appreciate information regarding how to properly create Avogadro representations of clusters of the same molecule in order to collect accurate data. Thank you for all of the help in advance, and Happy New Year!

Regards,
Zaid Contractor.

P.S. I am utilizing the MOPAC2016 package along with Avogadro to view vibrational spectra.

2

I’m not 100% sure I understand. I think your question is something like this:

  1. You want to create a cluster of H2O molecules
  2. You calculate the IR spectra with MOPAC2016
  3. You’re see many frequencies, but you’re concerned the auto-generated formula is H12O6 instead of, say, (H2O)6.

If so, you’re doing basically the right thing.

  1. The formula as I mentioned is autogenerated. It counts up the atoms and applies the Hill order to sort the elements.
  2. For N atoms, you’ll see 3N-6 frequencies because you don’t have a linear molecule. So for H12O6, that’s 18 atoms and you’d expect to see 48 vibrational modes. Many will be very similar (e.g., you have 6 water molecules, and so there will be six modes generated from the symmetric stretch of H2O).

You don’t mention, but at the very least, you should make sure to optimize the geometry with MOPAC2016 before generating the vibrational spectra. Particularly with multiple molecules, it can be hard to get a fully optimized system, so take a look at the MOPAC output file (as text) to make sure it doesn’t complain or yield negative frequencies.

1 Like
3

Thank you for the response, sir. And yes, my question is exactly as what you have mentioned in the reply.

As for the MOPAC output file, there indeed was a negative frequency value present, and I will check the file again and optimize the geometry with MOPAC2016.

Thanks again,
Zaid.