Highly accurate HF dimer ab initio potential energy surface

TL;DR

This paper presents a highly accurate ab initio potential energy surface for the HF dimer, enabling precise calculations of vibrational and rotational energy levels that closely match experimental data.

Contribution

The authors developed a new analytical PES for HF dimer using extensive high-level ab initio calculations, achieving unprecedented accuracy in energy level predictions.

Findings

Predicted ground state rotational constants within 2 MHz of experimental values

Reproduced dissociation energy within 1 \,cm^{-1} of experimental data

Achieved accurate energy levels for higher vibrational and rotational states

Abstract

A very accurate, (HF)$_2$ potential energy surface (PES) is constructed based on \ai\ calculations performed using the MOLPRO package at the CCSD(T) level of theory with an aug-cc-pvQz-F12 basis set at about 161~000 points. a higher correlation correction is computed at CCSDT(Q) level for 2000 points and is considered alongside other more minor corrections due to relativity, core-valence correlation and Born-Oppenheimer failure. The analytical surface constructed uses 500 constants to reproduce the \ai\ points with a standard deviation of 0.3 \cm. Vibration-rotation-inversion energy levels of the HF dimer are computed for this PES by variational solution of the nuclear-motion Schr\"{o}dinger using program WAVR4. Calculations over an extended range of rotationally excited states show very good agreement with the experimental data. In particular the known empirical rotational constants $B$ for the ground vibrational states are predicted to better than about 2 MHz. $B$ constants for excited vibrational states are reproduced several times more accurately than by previous calculations. The experimental dissociation energy of the HF dimer is reproduced \ai\ within the experimental accuracy of about 1 \cm\ for the first time. This level of accuracy is shown to extend to higher excited inter-molecular vibrational states $v$ and higher excited rotational quantum numbers $(J,K_a)$.