| Full text | |
| Author(s): |
Duarte, Leonardo J.
;
Bruns, Roy. E.
Total Authors: 2
|
| Document type: | Journal article |
| Source: | Journal of Physical Chemistry A; v. 126, n. 48, p. 10-pg., 2022-12-08. |
| Abstract | |
The Interacting Quantum Atoms (IQA) energy decomposition scheme divides the total energy of a molecule into intra-and interatomic contributions. While the former relates to the kinetic and potential energies of electrons inside a unique individual atomic basin, the latter contains the Coulomb and exchange-correlation potentials between electrons from two atomic basins. Considering that the molecular energy is a sum of IQA contributions, the Hessian matrix can also be written as a sum of "IQA Hessian" matrices, whose elements are second derivatives of IQA terms. Herein, we present a mathematical formalism for the IQA decomposition of force constants revealing their energetic origins. The method consists of adding a new dimension to the Hessian matrix, which becomes 3N x 3N x N2, with N being the number of atoms in the molecule and N2 the number of IQA terms. Since there is no analytical method that produces the IQA second derivatives, the three-dimensional IQA Hessian is numerically calculated. When studying molecular vibrations, force constants, providing information about the nature of chemical bond and related to infrared frequencies, can be obtained by Wilson's FG method, which involves detailed manipulations of the Hessian matrix. In this paper, the methodology is reported and validated for a set of 30 molecules and more than 200 force constants and their interactions. Energetic origins of force constants are presented for diatomics and small molecules containing carbon-carbon, oxygen-oxygen, and carbon-oxygen bonds with different bond orders. It is found that bond stability and stiffness can have strikingly different energetic origins. (AU) | |
| FAPESP's process: | 18/08861-9 - Application of the QTAIM / CCTDP model and machine learning for the forecast of chemical reactivities |
| Grantee: | Roy Edward Bruns |
| Support Opportunities: | Regular Research Grants |
| FAPESP's process: | 17/22741-3 - Using atomic multipoles and developing machine learning models to investigate transition states |
| Grantee: | Leonardo José Duarte |
| Support Opportunities: | Scholarships in Brazil - Doctorate (Direct) |