An effective methodology to predict infrared spectra of van der Waals complexes: A case of Ar–CO complex

Highlights: • An effective methodology is developed to predict the infrared spectra of van der Waals complexes and successfully applied to the complex Ar–CO. • The full-dimensional Hamiltonian of a complex can be strictly divided into three parts which are solved with higher accuracy and less computing resources. • The vibrational shifts are accurately predicted with the errors of 0.16% and 1.68% for two infrared bands. In this work, we developed an effective methodology to predict infrared spectra of van der Waals complexes and successfully applied it to the prototype system of Ar–CO complex. The basic idea is to divide strictly the full-dimensional Hamiltonian of a complex into three parts, named as rigid-rotor-approximation Hamiltonian, monomer Hamiltonian, and intramolecular and intermolecular coupling Hamiltonian, respectively. Following the three parts, intermolecular potential energy surfaces (IPESs) were firstly constructed at the rigid rotor approximation for ground and vibrationally excited states of Ar–CO complex. Then, potential curves were calculated for ground and vibrationally excited states of CO monomer with the intermolecular equilibrium structural parameters fixed. Based on these PESs, the bound state calculations were performed to obtain the rovibrational levels and average structural parameters. Finally, the intermolecular and intramolecular coupling Hamiltonian was calculated using the average structural parameters obtained from the previous steps. Our predicted vibrational shifts and spectroscopic parameters for Ar–CO complex are all in good agreement with available experimental data. The vibrational shift was calculated to be –0.4384 cm^–1 for the fundamental and –0.8925 cm^–1 for the overtone bands, which reproduce experimental results with the error of 0.16% (–0.4377 cm^–1) and 1.68% (–0.8778 cm^–1), respectively. In brief, the advantages of this methodology are its higher accuracy for predicting infrared spectra and are less computing resources for constructing IPESs than those of the full-dimensional quantum calculations.