Organic crystal modeling: On the importance of accurate London dispersion interactions

CCMMP Seminars
Dr Jan Gerit Brandenburg
October 11th, 2016 at 14:00
GO Jones Room 610

"Organic crystal modeling: On the importance of accurate London dispersion interactions"

The interest in molecular crystals and their property prediction has grown in the past decade. This is due to their relevance in pharmaceuticals, organic semiconductor materials, and other applications.[1,2,3]

I will present a hierarchy of quantum chemical methods designed for this purpose. The methods range from London dispersion inclusive density functional approximations to semiempirical tight-binding methods and thus cover about four orders of magnitudes in computational efficiency. Comprehensive benchmark results show the overall accuracy of the different approaches, where the impact of dispersion interactions is highlighted. The test systems focus on geometrical properties as this is a most important property that is typically not feasible via high level wavefunction methods.[4,5,6,7]
I will demonstrate the application of some discussed methods to the 6th blind test for organic crystal structure prediction. Comparisons to other state-of-the-art methods indicate both problems and success in the recent method developments.[8,9,10]

Key references:
[1] G. Beran, Chem. Rev., 116, 5567 (2016)
[2] S. Grimme, A. Hansen, JGB, C. Bannwarth, Chem. Rev., 116, 5105 (2016)
[3] JGB, S. Grimme, Top Curr Chem, 345, 1 (2014)
[4] JGB, E. Caldeweyher, S. Grimme, Phys. Chem. Chem. Phys., 18, 15519 (2016)
[5] E. Johnson, A. Otero-de-la-Roza, J. Chem. Phys. 137, 054103 (2012)
[6] A. Reilly, A. Tkatchenko, J. Chem. Phys. 139, 024705 (2013)
[7] JGB, T. Maas, S. Grimme, J. Chem. Phys. 142, 124104 (2015)
[8] S. Price, Chem. Soc. Rev. 43, 2098 (2014)
[9] A. Reilly, et al. Acta Cryst. B, 72, 439 (2016)
[10] JGB, S. Grimme, Acta Cryst. B, 72, 502 (2016)


Dr Brandenburg is a Humboldt Fellow currently working with Prof Price and Prof Michaelides at University College London