Chalcogen interaction not picked up with AIM

corey.taylor · 2025-02-14 08:28:12

Hi all,

I was reading a paper on intramolecular interactions involving Sulfur and Nitrogen (https://pubs.acs.org/doi/epdf/10.1021/j … le_openPDF) and the Fig 2 on page 3 (see attached).

I wanted to see if I could verify that an interaction is present and ran a DFT opt (B3LYP/def2-TZVPPD) and then the AIM modules in mwfn but it wasn't able to find a BCP between those two atoms.

Their hypothesis for how it occurs seems clear enough (basic donation into the σ* C-S orbital) so shouldn't this be visible with AIM? If not, how else could one try to verify the presence of an interaction there? Is it usual that AIM doesn't always find CPs for these sorts of interactions?

Things I've tried in mwfn:

- Setting the scale factor to 0.5
- option with 1000 starting points around each atom (ran this twice)

Other things I'm considering;

- Heavier hybrid (M06) or range-separate functional (e.g. B97)
- basin analysis to calculate DIs.
- heavier method (e.g. MP2).

I don't think just using a heavier method necessarily solves the issue, though. Any insights or hints?

sobereva · 2025-02-14 23:05:30

There is unlikely a BCP. Topology analysis in AIM theory is unable to exhibit all kinds of interaction, as carefully discussed in my paper Chemistry—Methods 2021, 1, 231–239 https://doi.org/10.1002/cmtd.202100007
Even for a more evident intramolecular interaction, including nonnegligible intramolecular H-bonds, sometimes there is no corresponding BCP. The IRI method introduced in the above paper is able to exhibit much more interactions, and IRI analysis can be easily realized by Multiwfn.

In addition, the hyperconjugation interaction can be unveiled by NBO second-order perturbation analysis.

corey.taylor · 2025-02-15 08:50:04

Ah I didn't realise this was implemented in mwfn. Good tip, thank you. I did run the RDG analyses too and found similar very weak interactions.

If you do a torsion scan, there is a net energetic stability with this configuration. After discussion with others, a colleague believes it's mostly electrostatic. He did an IQA calculation, which I don't have access to and mwfn doesn't support. One could get similar from a SAPT calculation but the only method similar in mwfn from what I can see is sobEDA, which requires Gaussian. Are there other mwfn options?

Last edited by corey.taylor (2025-02-15 08:56:22)

sobereva · 2025-02-15 16:19:38

You can consider to perform NAdO analysis between N and S at this configuration, NAdO is able to convert interatomic or interfragment orbital interactions into graphical representation to provide chemical insights. See Section 3.200.20 of Multiwfn manual for introduction and 4.200.20 for examples. In addition, the ETS-NOCV in Multiwfn may also be useful (you need to divide the molecule as two fragments, left and right, perhaps you can find NOCV pair(s) that directly characterize the N...S interaction), see Section 3.26 of manual for introduction and 4.23 for examples. More introduction of these methods in Multiwfn can be found in Section "F. Bond order and relevant analyses" of recent Multiwfn introductory paper J. Chem. Phys., 161, 082503 (2024) DOI: 10.1063/5.0216272.

corey.taylor · 2025-02-17 08:21:20

Fantastic, I'll have a look into those. Thanks again, Tian.

Multiwfn forum

#1 2025-02-14 08:28:12

Chalcogen interaction not picked up with AIM

#2 2025-02-14 23:05:30

Re: Chalcogen interaction not picked up with AIM

#3 2025-02-15 08:50:04

Re: Chalcogen interaction not picked up with AIM

#4 2025-02-15 16:19:38

Re: Chalcogen interaction not picked up with AIM

#5 2025-02-17 08:21:20

Re: Chalcogen interaction not picked up with AIM

Board footer