| Full text | |
| Author(s): |
Silva, Thiago Soares
;
Cruz, Allefe Barbosa
;
Oliveira Rodrigues, Karinna Gomes
;
Pereira, Douglas Henrique
Total Authors: 4
|
| Document type: | Journal article |
| Source: | Journal of Molecular Modeling; v. 28, n. 9, p. 10-pg., 2022-09-01. |
| Abstract | |
Composite methods are the combination of ab initio calculations used to achieve high precision in the face of a computational reduction. Weizmann-n theories (n = 1, 2, 3, and 4) stand out for presenting a high precision, and a version of the W1 theory is the W1BD theory that uses ab initio Brueckner Doubles (BD) methods. One way to reduce the computational cost of composite methods and maintain accuracy is to use pseudopotentials in the calculation steps; in this context, W1BDCEP composite method was developed from the respective W1BD all-electron version by considering the implementation of compact effective pseudopotential (CEP). The test set used to evaluate the theory were 8 proton affinities (PA(0)), 46 electron affinities (EA(0)), 54 ionization energies (IE0), 80 enthalpies of formation (Delta H-f(0)), and 10 bond dissociation energies (BDE). The mean absolute deviation values (MADs) for W1BD and for the version adapted to the pseudopotential, W1BDCEP, were similar, with values of 0.97 kcal mol(-1) and 1.03 kcal mol(-1), respectively, when the properties PA(0), EA(0), IE0, and Delta H-f(0) were evaluated together. Comparing the versions of the theories that employ ab initio Brueckner Doubles calculations with the W1 and W1CEP theories, it is possible to observe that the W1BD and W1BDCEP theories are more accurate than the W1 theory (MAD(W1) = 1.25 kcal mol(-1)) and W1CEP (MAD(W1CEP) = 1.44 kcal mol(-1)), proving the accuracy of using the BD method. Pseudopotential reduces computational time by up to 30% and thus enables more accurate calculations with less computational time. (AU) | |
| FAPESP's process: | 13/08293-7 - CCES - Center for Computational Engineering and Sciences |
| Grantee: | Munir Salomao Skaf |
| Support Opportunities: | Research Grants - Research, Innovation and Dissemination Centers - RIDC |
| FAPESP's process: | 17/11485-6 - Computational and theoretical developments based on ab initio methods and the Density Functional Theory |
| Grantee: | Rogério Custodio |
| Support Opportunities: | Regular Research Grants |