Abstract
While many-body wavefunction theory has long been established as a powerful framework for highly accurate molecular quantum chemistry, these methods have only fairly recently been applied to extended systems in a significant scale. This is due to the high computational cost of such calculations, requiring efficient implementations and ample computing resources. To further aggravate this, second-order Møller-Plesset perturbation theory (MP2) (the most cost effective wavefuntion method) is known to diverge or fail for some prototypical condensed matter systems like the homogeneous electron gas (HEG). In this paper, we explore how the issues of MP2 for metallic and strongly correlated systems can be ameliorated through regularization. To this end, two regularized second-order methods (including a new, size-extensive Brillioun-Wigner approach) are applied to the HEG, the one-dimensional Hubbard model and the graphene-water interaction energy. We find that regularization consistently leads to improvements over the MP2 baseline and that different regularizers are appropriate for metallic and strongly correlated systems, respectively.