The development of practical density functional theory (DFT) methods has provided the science community with a very important tool for modeling variety of systems such as materials, molecular and bio–molecular systems. Nonetheless, most practitioners of the method did not give enough attention to the deficiencies in modeling the dispersion interactions with the commonly used density functionals until a few years ago. Since then there have been many methods proposed to solve this problem and it is still a very active research area. I have tested a number of these dispersion–corrected DFT schemes for various systems that are of interest to our research group such as a water molecule interacting with a series of acenes and isomers of the water hexamer to see which of these methods give accurate results. Based on the tests, DFT–D3 of Grimme et al. and dispersion–corrected atom–centered pseudopotentials (DCACPs) attracted on our attention. DCACP procedure provided accurate interaction energies for the test cases, but the interaction energies fall too quickly as the distance between the molecules increases. I further investigated the effects of DCACPs on the employed density functionals with a detailed study of the interaction energies of isomers of the water hexamers and determined that with the original implementation it corrects for limitations of the BLYP functional in describing exchange-repulsion interaction as well as for dispersion interactions. We propose two different methods, namely DCACP+D and DCACP2, for improving the problems associated with the DCACP approach. These methods both provide improvements in the accuracy of the original DCACPs and also correct the quick fall-off problem of the interaction energies at long–range.
展开▼
