Multiwfn forum

saeed_E

Member

Registered: 2019-12-21

Posts: 244

IQA and Many-body contribution

Dear Tian,

Please suppose a trimer composed of three interacting monomers as, for instance, A...B...C. Also, assume the total electronic energy of this fully optimized trimer is calculated using Gaussian to be X.

You know much better than me based on the Interacting Quantum Atom (IQA), the total electronic energy of this fully optimized trimer is expressed as the sum of electronic energy of any monomer (namely, self atomic energy) plus the sum of pairwise (two-atomic) interaction energies. Using AIMAll, one can easily show that the total electronic energy of this trimer is identical to that obtained by Gaussian (X).

On the other hand, in terms of many-body contribution concept, the total electronic energy of this fully optimized trimer is the sum of electronic energy of monomer plus the sum of pairwise interaction energies (called two-body contribution) and plus an additional term namely three-body contribution. For many cases, three-body contribution is considerable and cannot safely be ignored.

The question is that:
Within IQA and only considering monomeric energies plus pairwise interaction energies, we can reach the value of X but an additional term namely three-body contribution needs to be included if one wants to reach value of X through many-body theory. Could you please let me know why these two approaches function quite different while both are quite reasonable?

Sincerely,
Saeed

Last edited by saeed_E (2024-09-03 22:48:02)

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,835

Website

Re: IQA and Many-body contribution

Dear Saeed,

This mainly comes from different choices of reference states.

The reference states of IQA are fragments in complex, namely charge polarization and transfer have take place between the fragments. However, in the case of so-called many-body theory, the references states corresponds to the fragments in their isolated states, the charge polarization and transfer to be occur among the fragments cannot be decomposed exactly in a pairwise manner, so three-body effect sometimes is not negligible.

Best regards,

Tian

saeed_E

Member

Registered: 2019-12-21

Posts: 244

Re: IQA and Many-body contribution

Dear Tian,
Your highly valuable and informative comments are so appreciated.
Please let me state that it seems the reference states in these two approaches should be interchanged. Indeed, the reference state in IQA is fragments in their isolated state while in many-body, this reference state is fragments in the fully optimized complex geometry.
Also, please let me ask you if possible provide much evident explanations. Frankly speaking, I cannot well understand your mean or, better, I cannot accept your rationalization while it can be of my wrong understanding.

Sincerely,
Saeed

sobereva

Tian Lu (Multiwfn developer)

From: Beijing

Registered: 2017-09-11

Posts: 1,835

Website

Re: IQA and Many-body contribution

The reference state I mentioned refers to the electronic structure of fragments rather than geometric structure of fragments. And when I referred to many-body theory, I assume you are calculating interaction energy rather than binding energy, the former doesn't consider geometry variation owing to interfragment interaction.

saeed_E

Member

Registered: 2019-12-21

Posts: 244

Re: IQA and Many-body contribution

Thank you very much.

Saeed