Correction to “Can B3LYP Be Improved by Optimization of the Proportions of Exchange and Correlation Functionals?”

Abstract

L. Lu, “ Can B3LYP be Improved by Optimization of the Proportions of Exchange and Correlation Functionals?,” International Journal of Quantum Chemistry 115 (2015): 502–509, https://doi.org/10.1002/QUA.24876.

An improved version of B3LYP obtained by refining empirical parameters in B3LYP functional against accurate experimental, which was denoted as OpB3LYP in this article, is the same functional reported as revB3LYP in the “An improved B3LYP method in the calculation of organic thermochemistry and reactivity” published on Computational and Theoretical Chemistry in volume 1015 (2013), 64–71. In this article, we have presented more details on parameter optimization and more extensive tests on the general good performance of revB3LYP. Unfortunately, the previous CTC paper has not been cited correctly in IJQC paper. To correct this oversight, we add the erratum to this article, and a statement should be added in the introduction section in our IJQC paper as follows:

In this work, empirical parameters in B3LYP functional, a₀, a_X, and a_C, are refined against accurate experimental data, the detailed procedure that we have performed to improve B3LYP functional is presented, and the performance of a better version of B3LYP denoted as OpB3LYP, which is identical to the revB3LYP functional in the previous work [R1], has been systematically characterized and statistically analyzed.

[R1] L. Lu, H. Hu, H. Hou, B. Wang, Computational and Theoretical Chemistry 1015 (2013): 64.

B3LYP is the most widely used hybrid DFT functional. The relative weight of each component in B3LYP is determined by three empirical parameters. A global optimization of these parameters against accurate experimental reference is performed to obtain an improved version of the functional called OpB3LYP. OpB3LYP significantly improves the accuracy and overcomes the size dependence problems of B3LYP.