Self-energy and excitonic effects in the electronic and optical properties of<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mml:mrow><mml:msub><mml:mrow><mml:mtext>TiO</mml:mtext></mml:mrow><mml:mn>2</mml:mn></mml:msub></mml:mrow></mml:math>crystalline phases

Letizia Chiodo, J. M. García‐Lastra, Amilcare Iacomino, Stefano Ossicini, Jin Zhao, Hrvoje Petek, Ángel RubioPhysical 2010 · DOI

TL;DR

This paper presents a comprehensive theoretical study of the electronic and optical properties of TiO₂'s rutile and anatase phases. By combining density functional theory (DFT) with many-body perturbation theory (MBPT), the authors provide a highly accurate description of these properties, offering new insights into the role of excitonic effects and their spatial distribution, which is crucial for understanding TiO₂'s applications in green energy technologies.

What Problem Does This Paper Address?

Titanium dioxide (TiO₂) is a cornerstone material for many green energy applications, including photocatalysis for hydrogen production and solar cells. However, a complete and consistent understanding of its fundamental electronic and optical properties, particularly the interplay between self-energy and excitonic effects, has been lacking. Previous theoretical and experimental studies often focused on individual phases or specific properties, leading to inconsistencies and uncertainties, especially regarding the precise nature and behavior of excitons. This paper aims to bridge this gap by providing a unified, high-quality theoretical description of both rutile and anatase TiO₂'s electronic and optical properties, clarifying the role of many-body effects and exciton localization.

Methods

The authors employ a sophisticated theoretical framework that combines Density Functional Theory (DFT) with Many-Body Perturbation Theory (MBPT).

  • DFT (specifically PBE functional): This is used for calculating the ground-state properties, including the equilibrium structure, density of states, and electronic band structure. DFT is a standard tool for this purpose, but it's known to underestimate electronic band gaps.
  • G₀W₀ calculations: This is a key MBPT technique used to correct the DFT eigenvalues and eigenfunctions. G₀W₀ calculations account for the self-energy effects, which significantly improve the description of quasiparticle energies and thus the electronic band gap, bringing it into better agreement with experimental values.
  • Bethe-Salpeter Equation (BSE) calculations: This is the most advanced technique used, which explicitly includes the electron-hole interaction (excitonic effects). By solving the BSE, the authors can accurately describe the optical absorption spectra, including the shape, intensity, and energy of optical transitions, and characterize the excitonic states.

The paper also explores different levels of approximation for screening within the BSE calculations to understand their impact on the results.

Datasets

Not discussed in the paper. The study relies on theoretical calculations and compares its results to existing experimental data from the literature.

Key Insights and Ideas

  • The critical role of many-body effects: The study highlights that accurately describing TiO₂'s electronic and optical properties requires going beyond standard DFT. Both self-energy (G₀W₀) and electron-hole interactions (BSE) are essential for achieving quantitative agreement with experimental optical spectra.
  • Exciton localization is phase-dependent: The authors found that excitons in anatase tend to be more localized, particularly within a single atomic layer in the (001) plane, compared to rutile where they are more delocalized in the (110) plane. This difference is linked to the crystal structure and anisotropic screening.
  • Anisotropic screening matters: The significant anisotropy in the electronic dielectric constant of rutile plays a crucial role in the optical response and the screening of electron-hole interactions, leading to different spectral behaviors for in-plane and out-of-plane polarizations.
  • Dark vs. Bright excitons: The analysis reveals the presence of "dark" excitons (optically forbidden transitions) at lower energies, which contribute to the overall optical spectrum but are not directly observed in absorption experiments. The first "bright" excitons are identified and characterized.
  • Bridging theory and experiment: The work provides a consistent theoretical framework that reconciles many previously disparate experimental observations and theoretical calculations, offering a more unified understanding of TiO₂'s optical properties.

Key Results

  • Improved electronic band gaps: G₀W₀ calculations yielded electronic band gaps of 3.59 eV for rutile and 3.96 eV (indirect) / 4.24 eV (direct) for anatase, showing good agreement with experimental photoemission data.
  • Accurate optical spectra: BSE calculations, incorporating electron-hole interactions, produced optical absorption spectra that closely match experimental data for both rutile and anatase, across different polarization directions.
  • Characterization of excitonic transitions: The paper provides detailed information on the energies, k-points, and spatial distributions of the first few excitonic transitions for both phases, including the identification of dark and bright excitons.
  • Exciton binding energies: The study estimates exciton binding energies, finding them to be small for direct excitons in rutile (around 0.65 eV for a low-intensity exciton) and more influenced by electron-hole interactions in anatase.

Limitations and Future Work

  • Exclusion of phonon effects: The current theoretical framework does not explicitly include electron-phonon coupling or polaronic effects. These are acknowledged as important contributors to the experimental optical edge, particularly for indirect transitions, and explain some of the energy differences between theoretical and experimental results.
  • Defects not considered: The calculations were performed for defect-free bulk TiO₂. The authors note that defects, such as oxygen vacancies, can significantly alter optical properties and are a crucial area for future investigation.
  • Experimental comparison nuances: Direct comparison with experimental data for optical transitions is challenging due to the co-presence of various physical effects (phonons, defects, etc.) in experiments that are not fully captured in the theoretical model.
  • Future directions: The authors suggest that this work serves as a crucial starting point for investigating exciton-phonon interactions, polarons, and the effects of defects on TiO₂'s optical properties. The characterization of excitons in bulk phases is also expected to aid in understanding the photophysical response of TiO₂ nanostructures and surfaces.

Summary updated

Abstract

We present a unified ab initio study of electronic and optical properties of ${\text{TiO}}_{2}$ rutile and anatase phases with a combination of density-functional theory and many-body perturbation-theory techniques. The consistent treatment of exchange and correlation, with the inclusion of many-body one-particle and two-particles effects in self-energy and electron-hole interaction, produces a high-quality description of electronic and optical properties, giving, for some quantities, the first available estimation for this compound. In particular, we give a quantitative estimate of the electronic and direct optical gaps, clarifying their role with respect to previous measurements obtained by various experimental techniques. We obtain a description for both electronic gap and optical spectra that is consistent with experiments by analyzing the role of different contributions to the experimental optical gap and relating them to the level of theory used in our calculations. We also show the spatial properties of excitons in the two crystalline phases, highlighting the localization character of different optical transitions. This paper aims at understanding and firmly establishing electro-optical bulk properties, yet to be clarified, of this material of fundamental and technological interest for green energy applications.
BibTeX
@inproceedings{chiodo2010self,
  author = {Letizia Chiodo and J. M. García‐Lastra and Amilcare Iacomino and Stefano Ossicini and Jin Zhao and Hrvoje Petek and Ángel Rubio},
  title = {{Self-energy and excitonic effects in the electronic and optical properties of<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mml:mrow><mml:msub><mml:mrow><mml:mtext>TiO</mml:mtext></mml:mrow><mml:mn>2</mml:mn></mml:msub></mml:mrow></mml:math>crystalline phases}},
  booktitle = {Physical},
  year = {2010},
  doi = {10.1103/physrevb.82.045207},
}
Figure