WHY THE GENERALIZED GRADIENT APPROXIMATION WORKS AND HOW TO GO BEYONDIT

Citation
K. Burke et al., WHY THE GENERALIZED GRADIENT APPROXIMATION WORKS AND HOW TO GO BEYONDIT, International journal of quantum chemistry, 61(2), 1997, pp. 287-293
Citations number
49
Categorie Soggetti
Chemistry Physical
ISSN journal
00207608
Volume
61
Issue
2
Year of publication
1997
Pages
287 - 293
Database
ISI
SICI code
0020-7608(1997)61:2<287:WTGGAW>2.0.ZU;2-Y
Abstract
The local spin density (LSD) approximation, while of only moderate acc uracy, has proven extremely reliable over three decades of use. We arg ue that any gradient-corrected functional should preserve the correct features of LSD, even if the system under study contains no regions of small density gradient. The Perdew-Wang 1991 (PW91) functional respec ts this condition, while, e.g., the Lee-Yang-Parr (LYP) correlation fu nctional violates it. We extend this idea to the next generation of de nsity functionals, those which incorporate exact exchange via the opti mized effective potential (OEP), with a model in which the correlation hole is constructed from the exact exchange hole. The resulting excha nge-correlation hole is deeper and less diffuse than the exact exchang e hole. We denote such a functional as ''locally correlated Hartree-Fo ck'' and list a variety of conditions such a functional should satisfy . We demonstrate the promise of this approach with a crude simple mode l. (C) 1997 John Wiley & Sons, Inc.