Skip to main content

Research Repository

Advanced Search

Assessment of tuning methods for enfacing approximate energy linearity in range-separated hybrid functionals

Gledhill, Jonathan D.; Peach, Michael J.G.; Tozer, David J.

Assessment of tuning methods for enfacing approximate energy linearity in range-separated hybrid functionals Thumbnail


Authors

Jonathan D. Gledhill

Michael J.G. Peach



Abstract

A range of tuning methods, for enforcing approximate energy linearity through a system-by-system optimization of a range-separated hybrid functional, are assessed. For a series of atoms, the accuracy of the frontier orbital energies, ionization potentials, electron affinities, and orbital energy gaps is quantified, and particular attention is paid to the extent to which approximate energy linearity is actually achieved. The tuning methods can yield significantly improved orbital energies and orbital energy gaps, compared to those from conventional functionals. For systems with integer M electrons, optimal results are obtained using a tuning norm based on the highest occupied orbital energy of the M and M + 1 electron systems, with deviations of just 0.1–0.2 eV in these quantities, compared to exact values. However, detailed examination for the carbon atom illustrates a subtle cancellation between errors arising from nonlinearity and errors in the computed ionization potentials and electron affinities used in the tuning.

Citation

Gledhill, J. D., Peach, M. J., & Tozer, D. J. (2013). Assessment of tuning methods for enfacing approximate energy linearity in range-separated hybrid functionals. Journal of Chemical Theory and Computation, 9(10), 4414-4420. https://doi.org/10.1021/ct400592a

Journal Article Type Article
Publication Date Oct 1, 2013
Deposit Date Feb 21, 2014
Publicly Available Date Mar 11, 2014
Journal Journal of Chemical Theory and Computation
Print ISSN 1549-9618
Electronic ISSN 1549-9626
Publisher American Chemical Society
Peer Reviewed Peer Reviewed
Volume 9
Issue 10
Pages 4414-4420
DOI https://doi.org/10.1021/ct400592a

Files

Accepted Journal Article (260 Kb)
PDF

Copyright Statement
This document is the Accepted Manuscript version of a Published Work that appeared in final form in Journal of chemical theory and computation, copyright © American Chemical Society after peer review and technical editing by the publisher. To access the final edited and published work see http://dx.doi.org/10.1021/ct400592a





You might also like



Downloadable Citations