Journal
COMPUTATIONAL AND THEORETICAL CHEMISTRY
Volume 1090, Issue -, Pages 147-152Publisher
ELSEVIER
DOI: 10.1016/j.comptc.2016.06.018
Keywords
Isomerization energies; Density functional theory; Benchmarking; ISOL24/11
Categories
Funding
- Western Canada Research Grid (WestGrid) [100185]
- Shared Hierarchical Academic Research Computing Network (SHARCNET) [sn4612]
Ask authors/readers for more resources
The isomerization energy (Delta E-isom) performance of a wide range of density functionals was tested against the set of 24 individual intramolecular reactions collectively comprising the ISOL24/11 benchmark database. Among 50 non-dispersion corrected functionals examined, the M06-2X single-component density functional offered the best estimation capacity, yielding a mean unsigned error (MUE) of 2.4 kcal/mol. The addition of molecular mechanics dispersion-like terms to the M06-2X functional offers minimal improvement in its capacity to reliably estimate isomerization energies, whereas the choice of basis set played a significant role. Triple zeta Dunning, Ahlrichs, and Pople basis sets offer effectively equivalent Delta E-isom performance, modestly better than double zeta equivalent basis sets, with Delta E-isom performance subsequently rapidly declining at basis set sizes below this level. Among a reduced set of dispersion corrected functionals considered, the APFD functional displayed Delta E-isom performance (MUE = 1.8 kcal/mol) superior to both the non-dispersion corrected M06-2X functional and the wB97X-D and PBE0-D3 dispersion corrected functionals (1.9 kcal/mol). At this accuracy, the APFD functional is competitive with computationally expensive high-accuracy methods such as MC3MPWB and MC3MPW and nearing that of composite methods and dispersion corrected double-hybrid density functionals. (C) 2016 Elsevier B.V. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available