Evaluation of the Hartree-Fock dispersion (HFD) model as a practical tool for probing intermolecular potentials of small aromatic clusters: Comparison of the HFD and MP2 intermolecular Potentials

The performance of the Hartree-Fock dispersion (HFD) model for aromatic clusters has been evaluated by comparing the HFD/6-31G intermolecular potentials with the MP2/6-31G potentials for dimers of four aromatic hydrocarbons (benzene, naphthalene, anthracene, and pyrene) and the trimer of naphthalene. The computationally efficient HFD model yields equilibrium geometries and binding energies that are essentially identical to those from the MP2 calculations for all aromatic clusters. For the T-shaped dimer of benzene and the cyclic trimer of naphthalene for which experimental geometries are known, the computed geometry and intermolecular separations are in excellent agreement with the experimental data. Although the MP2/6-31G (not corrected for basis set superposition errors) and HFD/6-31G binding energies (D-e) of the dimers of benzene and naphthalene, and the trimer of naphthalene, are almost a factor of 2 greater than the experimental values (D-0), they are considerably in better agreement with experiment than the MP2 interaction energies computed by using larger and diffuse basis sets, 6-31G* (0.25) and aug-cc-pVDZ. The calculated minimum-energy structures of the four aromatic hydrocarbons of differing sizes support the notion that electrostatic interaction favors edge-on (T-shaped) structures, whereas dispersion interaction favors stacked structures. The computed dimer binding energy is approximately a linear function of the number of hexagons in the monomer.