Physical origin of satellite in photoemission of doped graphene: An ab-initio GW plus cumulant study

TL;DR

This study uses advanced ab initio GW plus cumulant methods to accurately model photoemission spectra of doped graphene, revealing the importance of substrate screening and challenging previous plasmaron interpretations.

Contribution

It introduces an ab initio GW plus cumulant approach to better match experimental satellite features in doped graphene photoemission spectra.

Findings

Standard GW fails to reproduce satellite features

GW plus cumulant with substrate screening matches experiments

No evidence of plasmarons in the spectra

Abstract

We calculate the photoemission spectra of suspended and epitaxial doped graphene using an ab initio cumulant expansion of the Green's function based on the GW self-energy. Our results are compared to experiment and to standard GW calculations. For doped graphene on a silicon carbide substrate, we find, in contrast to earlier calculations, that the spectral function from GW only does not reproduce experimental satellite properties. However, ab initio GW plus cumulant theory combined with an accurate description of the substrate screening results in good agreement with experiment, but gives no plasmaron (i.e., no extra well-defined excitation satisfying Dyson's equation).