Electronic and vibrational properties of $\textup{TiSe}_2$ in the charge-density wave phase from first principles

TL;DR

This study uses first-principles calculations to analyze the electronic and vibrational properties of TiSe₂ in its charge-density wave phase, successfully reproducing experimental structural and electronic features and exploring the effects of electron correlation.

Contribution

It demonstrates that density-functional calculations can accurately model the CDW phase of TiSe₂ and investigates the impact of electron correlations using LDA+U, revealing complex effects on phonon stability.

Findings

DFT reproduces structural instability and experimental distortion.

Distorted band structure aligns with ARPES data at low temperature.

LDA+U improves band agreement but removes phonon instability.

Abstract

We study the charge-density wave phase in $\textup{TiSe}_2$ by using first principle calculations. We show that, regardless of the local functional used and as long as the cell parameters are in agreement with the experiment, density-functional calculations are able to reproduce not only the structural instability of $\textup{TiSe}_2$, but also the effective distortion observed in the experiments. We study the electronic structure evolution of the system under the charge-density wave deformation. In particular, we show that the energy bands for the distorted superstructure, unfolded into the original Brillouin zone, are in reasonable agreement with angle-resolved photoemission spectroscopy (ARPES) data taken at low temperature. On the contrary, the energy bands for the undistorted structure are not in good agreement with ARPES at high temperature. Motivated by these results, we investigate the effect of the correlation on the electrons of the localized Ti-$d$ orbitals by using the LDA+$U$ method. We show that within this approximation the electronic bands for both the undistorted and distorted structure are in very good agreement with ARPES. On the other hand, the $U$ eliminates the phonon instability of the system. Some possible explanations for this counter intuitive result are proposed. Particularly, the possibility of taking into account the dependence of the parameter $U$ from the atomic positions is suggested.