Introduction to the Fifth-rung Density Functional Approximations: Concept, Formulation, and Applications

Abstract

The widespread use of (generalized) Kohn-Sham density functional theory (KS-DFT) lies in the fact that hierarchical sets of approximations of the exchange-correlation (XC) energy functional can be designed, offering versatile choices to satisfy different levels of accuracy needs. The XC functionals standing on the fifth (top) rung of the Jacob's ladder incorporate the information of unoccupied Kohn-Sham orbitals, and by doing so can describe seamlessly non-local electron correlations that the lower-rung functionals fail to capture. The doubly hybrid approximations (DHAs) and random phase approximation (RPA) based methods are two representative classes of fifth-rung functionals that have been under active development over the past two decades. In this review, we recapitulate the basic concepts of DHAs and RPA, derive their underlying theoretical formulation from the perspective of adiabatic-connection fluctuation-dissipation theory, and describe the implementation algorithms based on the resolution-of-identity technique within an atomic-orbital basis-set framework. Illustrating examples of practical applications of DHAs and RPA are presented, highlighting the usefulness of these functionals in resolving challenging problems in computational materials science. The most recent advances in the realms of these two types of functionals are briefly discussed.