Full text
| |
| Author(s): |
Mariana Mieko Odashima
Total Authors: 1
|
| Document type: | Doctoral Thesis |
| Press: | São Carlos. |
| Institution: | Universidade de São Paulo (USP). Instituto de Física de São Carlos (IFSC/BT) |
| Defense date: | 2010-06-08 |
| Examining board members: |
Klaus Werner Capelle;
Marilia Junqueira Caldas;
Mauricio Domingues Coutinho Neto;
Antonio Carlos Pedroza;
Albérico Borges Ferreira da Silva
|
| Advisor: | Klaus Werner Capelle |
| Abstract | |
Electronic-structure calculations play a fundamental role in solid-state physics and quantum chemistry. Density-functional theory (DFT) is today the most-widely used electronic-structure method, from atomic and nanoscopic scales to biomolecular aggregates. The accuracy of DFT depends essentially on approximations to the exchange and correlation energy, which are controlled by exact constraints. This is a very important issue, since the improvement of functionals is the key to a better description of many-body effects. In the present work, we investigate the exchange-correlation energy and approximate functionals from the viewpoint of an universal constraint on interacting Coulomb systems: the Lieb-Oxford lower bound. Initially we present evidence that for several classes of systems (atoms, ions, molecules and solids), the actual exchange-correlation energies are far from the Lieb-Oxford lower bound. A tighter form of this bound was conjectured; implemented in the Perdew-Burke-Erzenhof (PBE) functionals, and tested for atoms, molecules and solids. Finally, we propose to use the Lieb-Oxford bound not just to fix the value of a parameter as in PBE, but as a starting point for a new family of hyper-GGA functionals. For these, we explored a non-empirical construction, investigating its performance for atoms and small molecules post-selfconsistently. The particular HGGA proposed benefited from the tightening of the Lieb-Oxford bound and exhibited satisfactory correlation energies. (AU) | |