Electron energy-loss spectrum and exciton band structure of ${\mathrm{WSe}}_{2}$ monolayer studied by ab initio Bethe-Salpeter equation calculations

TL;DR

This study uses ab initio calculations to analyze the exciton band structure and electron energy-loss spectra of monolayer WSe2, revealing how EEL spectroscopy selectively probes bright excitons and matches experimental results.

Contribution

It provides a detailed theoretical framework for interpreting EEL spectra in 2D materials and clarifies the connection between exciton band structure and EEL measurements.

Findings

EEL spectra follow the non-parabolic upper band of the A exciton.

Only bright excitons with in-plane dipole moments are excited in EEL spectroscopy.

The calculated spectra agree well with previous experimental data.

Abstract

Bounded excitons in transition metal dichalcogenides monolayers lead to numerous opto-electronic applications, which require a detailed understanding of the exciton dynamics. The dynamical properties of excitons with finite momentum transfer $\textbf{Q}$ can be investigated experimentally using electron energy-loss (EEL) spectroscopy. The EEL spectrum depends on the response function of the material which in turn is determined by the exciton energies and eigenvectors in the exciton band structure. In this work, we utilize ab initio density-functional theory plus Bethe-Salpeter equation (DFT+BSE) approach to explore the exciton band structure and also $\textbf{Q}$-resolved EEL spectrum in monolayer ${\mathrm{WSe}}_{2}$. In particular, we carefully examine the discrepancies and connections among the existing EEL spectrum formulas for quasi-two-dimensional (2D) systems, and establish a proper definition of the EEL spectrum, which is then used to calculate the EEL spectra of monolayer ${\mathrm{WSe}}_{2}$. We find that remarkably, the dispersion of the calculated lowest-energy EELS peaks for the in-plane momentum transfer follows almost precisely the non-parabolic upper band of the lowest bright A exciton, and also agrees well with the previous experiment. Furthermore, we show that only the bright exciton with its electric dipole being parallel to the direction of the transfered momentum is excited, i.e., EEL spectroscopy selectively probes bright exciton bands. This explains why only the upper band of the A exciton, which is a longitudinal exciton with an in-plane dipole moment, was observed in the previous experiment. Our findings will stimulate further EEL experiments to measure other branches of the exciton band structure, such as the parabolic lower band of the A exciton, and hence will lead to a better understanding of the exciton dynamics in quasi-2D materials.