There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.
Abstract
The neglect of non-local electron correlation effects is a serious drawback of common
DFT methods. To remedy this, we have recently developed double-hybrid density functionals
(X2PLYP family) [1,2], which add a second order perturbation correction for correlation
to a standard hybrid functional in an empirical way.
Here we give an overview of the extensions of our previous work. We discuss the analytical
gradient for structure optimisations [3], the combination with an empirical dispersion
correction (DFT-D) [4], and the computation of excitation energies in a time-dependent
framework [5]. We present results for several benchmark sets and for some challenging
applications. In all cases very accurate results are obtained at a reasonable computational
expense. These show, that our method outperforms common (TD)DFT approaches and is
even competitive to more sophisticated approaches like CCSD(T).