I'm trying to calculate an accurate electrostatic repulsion (and possibly attraction) energy term for various small molecules, using Quantum Mechanics (QM) methods. These may include molecules such as bifuran, bipyridine, etc... I'm mainly interested in, for example, the electrostatic interaction energy of the two oxygens in a bifuran.

I've already tried using Molecular Mechanics-based methods, which use point charges (ie. GAFF charges) but they seem to be unable to account for the nuance orbital interactions, especially when molecules are in close contact. I was wondering if there exists a robust QM method, which is able to give accurate electrostatic interaction energies for molecules?

  • $\begingroup$ With QM, be it HF, CC, MP2 etc, it will be hard to disentangle different contributions - You could compare fragments though. E.g. for 2,2'-Bipyridine: compare energy of conformer by dihedral angle as a proxy for N distance as well as comparison to energy of two seperate pyridine fragments. But would also include dispersion and effects of aromaticity/conjugation in the molecule. $\endgroup$ – user213305 Apr 4 '19 at 22:32
  • $\begingroup$ I agree that it is difficult to disentangle the different interactions. I was hoping that there would be a robust method (ie. NBO equivalent interaction energy for full orbitals), which may or may not be possible. $\endgroup$ – Astronomer Apr 4 '19 at 22:57
  • $\begingroup$ Although many localisation schemes exist (NBO, Mulliken population analysis, atoms in molecules) they each have various failing and I would be wary to call them robust for this purpose. Though partitioning space/electron density does suggest ideas for DFT based schemes for calculating the interaction energy of two densities - but I'm not aware of the literature/previous work on that type of approximation. $\endgroup$ – user213305 Apr 4 '19 at 23:09

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.