Efficiently Capturing Substrate Screening Effects on Level Alignments at Molecule-Metal Interfaces with GW Calculations

Quasiparticle energies and level alignments at molecule-metal interfaces can be accurately captured by the ab initio GW approach. However, the computational cost for such GW calculations for typical interfaces is relatively high, given the large system size and chemical complexity of typical interface systems. Approximate self-energy corrections constructed from image-charge models have been used to compute level alignments with good accuracy but require the definition of an image plane. In this work, we study a primary bottleneck of GW calculations at interfaces, namely the calculation of the polarizability. We explore models and approximations to compute this quantity more efficiently and accurately, with the aim of computing molecular resonance energies at interfaces and capturing substrate screening effects without the need of defining an image plane.