Various flavors of dispersion corrected DFT have been extensively tested in the last 2-3 years on benchmark sets consisting of rather small molecular complexes.In general good accuracy compared to wavefunction based ab initio data (mostlyCCSD(T) level) with errors on the order of 5-10 % of the interaction energyhas been obtained. However, less is known how these methods perform for largercomplexes with about 100-200 atoms that are important in supramolecular aswell as bio-chemistry. One of the reasons is that CCSD(T) or similar referencedata are difficult to obtain for complexes of this size and that experimentallymostly free enthalpies of association are available that can not be directly comparedwith theoretical data. We suggest a non-empirical theoretical procedure to back-correct these experimental free enthalpies in solution in orderto obtain estimated experimental gas phase interaction energies (i.e. De values) for a set of 12 typical supramolecular complexes[1]. The De values are in a rangeof 20-130 kcal/mol which is about one order of magnitude larger than in the standardS22 or S66 test sets. A few common dispersion corrected DFT methods and M06-type densityfunctionals are applied to this new set (termed S12L). The importance of  asymptotically correct long-range dispersion as well as three-body effects is discussed.[1] S. Grimme, Chem. Eur. J., 2012, 18, 9955-9964

Supramolecular interactions by dispersion corrected density functional theory

245th National Meeting (2013)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.