Benzene, pyridine and dispersion interactions
by tzvicky
Some time ago a paper dealing with adsorption and reactivity of aromatic hydrocarbons on the surface of water was published. My part of work was to perform ab initio molecular dynamics simulations of benzene and pyridine adsorbed on water. I was using a moderately small slab (72 water molecules + hydrocarbon) and, as usual, cp2k/Quickstep code with DFT method for electronic structure. Pretty soon it occured tha pure DFT is giving wrong results, namely, benzene ring was not adsorbed on water surface at all. The dispersion interactions which are ommited by DFT are crucial for ‘gluing’ benzene and water molecules. With assistance of Jan Rezac from prof. Hobza’s group, we employed DFT+D scheme in Quickstep and that really helped. Pyridine behaved the same way – only DFT+D scheme was able to describe its adsorption on water. These ab initio MD simulations were only a support for classical MD were the time scale was large enough to get a proper statistics for localization of both benzene and pyridine rings on water. In addition to these classical and ab initio MD simulations the calculations using SAPT method were used. On the top of that a lot of experimantal work of our colleagues was done so the final paper looks very interesting.
(DOI: 10.1021/jp711813p)
