Using the DFT-D method to describe dispersion interactions in systems of weakly-bonded Xe-aromatic molecules

The optimum version of the DFT-D class of methods (BHHLYP-D2, 6-31G*) is chosen to describe binding in a Xe-phenol system with the aim of subsequent KM/MM calculations for complex Xe-containing protein systems. It is shown that the stability of the Xe-phenol system is due to weak dispersion interactions not described in conventional approaches using the density functional. The MP2 approach using the (aug)-cc-pVTZ basis and Stuttgart pseudopotential, which yield the best reproduction of the characteristics of a Xe2 xenon dimer, is chosen as the reference standard. It is noted that the 2010 DFT-D3 methods underestimate the binding energy by a factor of nearly three, while DFT methods without dispersion corrections do not reproduce the stability of Xe2 and Xe-phenol systems. It is found that in the best version of calculations, BHHLYP-D2, the binding energy in Xe-phenol complex is estimated to be 2.7 kcal/mol versus the 3.1 kcal/mol found using the comparative approach. It is concluded that BHHLYP-D2 adequately reproduces the difference between the two conformers of the Xe-phenol complex and trend toward an increase in binding energy in the series of aromatic amino acids (phenylalanine, tyrosine, and tryptophan). DFT-D can also indicate the existence of excess conformers that are missing in systems according to more precise descriptions (MP2/(aug)-cc-pVTZ).