In the formative years of Density-Functional Theory (DFT), late 1970s to late 1980s, local and semi-local exchange-correlation approximations were discovered whose efficacy rivalled that of non-DFT wavefunction methods. Our own work on semi-local exchange approximations culminated in the 'B88' functional [Phys. Rev. A 38, 3098 (1988)] which has garnered enormous popularity in computational chemistry. Indeed DFT now underlies most computations in quantum chemistry and physics. As users apply existing semi-local DFT approximations to ever more challenging electronic-structure problems, however, the catalogue of known failures is growing. A new paradigm is needed.The failures of semi-local DFT approximations reside in the exchange part of the exchange-correlation energy. If exchange is computed exactly, as in Hartree-Fock theory, then a clean path forward is possible. We have been exploring this path, post-Hartree-Fock, correlation-only DFT, for the last ten years. Its present status will be discussed.Research on Hartree-Fock-based strategies, thirty years after semi-local DFT largely replaced them, brings us full circle! One might ask: has thirty years of DFT development been wasted?

Award Address  (ACS Award in Theoretical Chemistry sponsored by Dell Incorporated). Full circle: A career in density-functional theory

247th National Meeting (2014)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.