Combining Wave Function Methods with Density Functional Theory for Excited States

Ghosh, S.; Verma, P.; Cramer, C. J.; Gagliardi, L.; Truhlar, D. G.
Chem. Rev. 2018, 118, 7249 (doi:10.1021/acs.chemrev.8b00193).

[Chemical Reviews does not have abstracts--the following is the opening paragraph of the review.]

The Born-Oppenheimer approximation simplifies the solution to the Schrödinger equation by separating the nuclear and electronic motions, but even after this simplification, the electronic Schrödinger equation remains unsolvable for many- electron systems. Because electronic structure is the key to understanding many chemical problems, a variety of approximation techniques have been proposed. For a long time, these approximations were mainly based directly on the Schrödinger equation, and this may be called wave mechanics or wave function theory (WFT). Slower to emerge as a powerful electronic structure method (although it has roots in the early days of quantum theory) was another approach, this one based on electron density rather than electronic wave functions. The second approach is called density functional theory (DFT), the most popular version of which is based on the Kohn-Sham (KS) equation. Owing to having more favorable scaling of demand on computational resources with respect to system size, DFT eventually became the most common method for large and complex systems with accuracy for many cases comparable to or better than (affordable) WFT approximations. However, approximate DFT suffers from some generic problems, many of which can be traced to so-called delocalization error (or the closely related problem of self-interaction error). This has motivated many workers to seek hybridized theories that combine the best of both approaches. The present review is about methods that combine WFT and DFT to treat electronic excitation.