Abstract
Accurate calculation of binding energies of molecular solids is challenging due to the need to reliably model electron correlations in extended systems. Moreover, the energy differences between different phases or polymorphs are often small, leading to high requirements not only on the accuracy but also on the precision of the numerical set-up.
To date, two main approaches are in use to obtain binding energies: direct evaluation using periodic boundary conditions (PBC) and many-body expansion (MBE). Here we compare their efficiency on calculations of four molecular solids: ethane, ethylene, and two forms of acetylene using the random phase approximation (RPA) and second-order Moller-Plesset perturbation theory (MP2).
We assess how difficult it is to reach highly precise values (converged with numerical parameters) in the MBE and PBC approaches. The RPA and MP2 binding energies are also compared with reference values obtained using the coupled cluster scheme (CCSD(T)).
We find that RPA with singles corrections is more accurate than MP2 for all considered systems.