Fingerprints of dynamical correlation in electron spectroscopies

Invited

Abstract

One of the great challenges of condensed-matter physics is the description, understanding, and prediction of the effects of the Coulomb interaction on materials properties. In electronic spectra, the Coulomb interaction causes a renormalization of excitation energies and a transfer of spectral weight. Most importantly, it can lead to qualitatively new structures, such as satellites in photoemission or double-plasmon resonances in energy-loss spectra. Being a genuine signature of dynamical correlation, they are absent in a non-interacting picture but can be understood in terms of the coupling between different elementary excitations.
In this framework, a key physical ingredient is the dynamical screening of the Coulomb interaction, containing charge excitations such as plasmons and excitons. It can be accurately calculated within time-dependent density-functional theory (including double plasmons [1]) or by solving the Bethe-Salpeter equation. Its microscopic picture can be obtained from the mixed dynamic structure factor that is measured by coherent inelastic X-ray scattering spectroscopy [2].
Building upon a detailed knowledge of dynamical screening, the cumulant expansion of the Green's function can efficiently explain plasmon and exciton satellites in the photoemission spectra of several materials, ranging from simple metals to correlated transition-metal oxides [3].
Finally, the combined effect of many-body interactions and experimental conditions can lead to novel signatures in the measured spectra [4], underlining the need to bridge the gap between theory and experiments.
[1] M. Panholzer et al. Phys. Rev. Lett. 120, 166402 (2018). [2] I. Reshetnyak et al. Phys. Rev. Research, in press (2019). [3] See e.g. J. S. Zhou et al. Phys. Rev. B 97, 035137 (2018). [4] J. S. Zhou et al. https://arxiv.org/abs/1811.12217

Presenters

  • Matteo Gatti

    LSI, CNRS, Ecole Polytechnique, LSI, CNRS-Ecole Polytechnique, Palaiseau

Authors

  • Matteo Gatti

    LSI, CNRS, Ecole Polytechnique, LSI, CNRS-Ecole Polytechnique, Palaiseau