Item

Abstract

A non-empirical quantum mechanical method for the efficient and accurate quantification and analysis of intermolecular interactions is presented and tested on existing benchmark sets. The leading idea here is to focus on the intermolecular part of the correlation energy that contains the all-important London dispersion (LD) interaction. In order to keep the cost of the method low, essentially at the level of a Hartree-Fock (HF) calculation, the intramolecular part of the correlation energy is neglected. We also neglect the non-dispersive parts of the intermolecular correlation energy. This scheme that we denote as HFLD (Hartree-Fock plus London dispersion) can be readily realized on the basis of the recently reported multilevel implementation of the domain-based local pair natural orbital coupled cluster (DLPNO-CC) theory in conjunction with the well-established Local Energy Decomposition (LED) analysis. The accuracy and efficiency of the HFLD method is evaluated on rare gas dimers, on the S66 and L7 benchmark sets of noncovalent interactions and on an additional set consisting of bulky Lewis pairs held together by intermolecular interactions of various strengths, with interaction energies ranging from –8 to –107 kcal/mol. It is first shown that the LD energy calculated with this approach is essentially identical to that obtained from the full DLPNO-CCSD(T)/LED calculation, with a mean absolute error of 0.2 kcal/mol on the S66 benchmark set. Moreover, in terms of the overall interaction energies, the HFLD method shows an efficiency that is comparable to that of the HF method, while retaining an accuracy between that of the DLPNO-CCSD and DLPNO-CCSD(T) schemes. Since the underlying DLPNO-CCSD method is linear scaling with respect to system size, the HFLD approach also does not lead to new bottlenecks for large systems. As an illustrative example of its efficiency, the HFLD scheme was applied to the interaction between the substrate and the residues in the active site of the cyclohexanone monooxygenase enzyme. The excellent cost/performance ratio indicates that the HFLD method opens new avenues for the accurate calculation and analysis of noncovalent interaction energies in large molecular systems.