Lattice energies of molecular solids from the random phase approximation with singles corrections

TL;DR

This study evaluates the effectiveness of the RPA method with singles corrections in calculating lattice energies of molecular solids, showing significant improvements over standard RPA, especially for hydrogen-bonded systems.

Contribution

It demonstrates that adding singles corrections at the GWSE level substantially enhances RPA's accuracy for molecular solid lattice energies.

Findings

RPA underestimates lattice energies by 13.7% on average.

Including singles corrections reduces the error to 3.7%.

The method performs well for hydrogen-bonded systems but less so for dispersion-dominated systems.

Abstract

We use the random phase approximation (RPA) method with the singles correlation energy contributions to calculate lattice energies of ten molecular solids. While RPA gives too weak binding, underestimating the reference data by $13.7$\% on average, much improved results are obtained when the singles are included at the GW singles excitations (GWSE) level, with average absolute difference to the reference data of only $3.7$\%. Consistently with previous results, we find a very good agreement with the reference data for hydrogen bonded systems, while the binding is too weak for systems where dispersion forces dominate. In fact, the overall accuracy of the RPA+GWSE method is similar to an estimated accuracy of the reference data.