[Paper Review] Optical properties of anatase and rutile TiO2 studied by GGA+U

Jinping Li, Meng Song-he+2 more|arXiv (Cornell University)|Nov 25, 2016

Gas Sensing Nanomaterials and Sensors5 citations

TL;DR

This study investigates the optical and electronic properties of anatase and rutile TiO2 polymorphs using the first-principles GGA+U method, which includes Hubbard U corrections for Ti 3d and O 2p orbitals to accurately reproduce the experimentally observed band gaps. The results show excellent agreement with experimental optical data and reveal distinct anisotropy and birefringence differences between the two phases, providing a reliable theoretical framework for designing TiO2-based optoelectronic devices.

ABSTRACT

The optical properties of thermally annealed TiO2 samples depend on their preparation process, and the TiO2 thin films usually exist in the form of anatase or rutile or the mixture of the two phases. The electronic structures and optical properties of anatase and rutile TiO2 are calculated by means of First-principles generalized gradient approximation (GGA) +U approach. By Introducing the Coulomb interactions on 3d orbitals of Ti atom (Ud) and 2p orbitals of O atom (Up), we can reproduce the experimental values of the band gap. The optical properties of anatase and rutile TiO2 are obtained by means of GGA+U method, well agreeing with experimental results and other theoretical data. Further we present the comparison of the electronic structure, birefringence and anisotropy between the two phases of TiO2.

Motivation & Objective

• To accurately model the electronic structure and optical properties of anatase and rutile TiO2, which are critical for optoelectronic applications.
• To address the known band gap underestimation in standard DFT by applying the GGA+U method with Hubbard U corrections on Ti 3d and O 2p orbitals.
• To compare the electronic structure, birefringence, and optical anisotropy between anatase and rutile phases of TiO2.
• To validate theoretical predictions against experimental optical data and other theoretical results.

Proposed method

• Employed first-principles density functional theory (DFT) within the generalized gradient approximation (GGA) framework.
• Applied Hubbard U corrections (Ud for Ti 3d, Up for O 2p) to improve band gap prediction accuracy.
• Calculated optical dielectric function and absorption spectra using the GGA+U approach.
• Performed electronic structure analysis including band dispersion and orbital contributions.
• Compared results between anatase and rutile phases to assess anisotropy and birefringence.
• Validated results against experimental optical measurements and other theoretical studies.

Experimental results

Research questions

• RQ1 How does the GGA+U method improve the prediction of the band gap in anatase and rutile TiO2 compared to standard GGA?
• RQ2 What are the differences in optical dielectric response between anatase and rutile TiO2 phases?
• RQ3 How do the electronic structures of anatase and rutile TiO2 differ in terms of band dispersion and orbital character?
• RQ4 What is the degree of optical anisotropy and birefringence in each TiO2 polymorph?
• RQ5 To what extent do GGA+U calculations reproduce experimental optical spectra for both phases?

Key findings

• The inclusion of Hubbard U corrections (Ud for Ti 3d and Up for O 2p) successfully reproduces the experimental band gap values for both anatase and rutile TiO2.
• The calculated optical dielectric function and absorption spectra using GGA+U show strong agreement with experimental data.
• Anatase and rutile phases exhibit distinct electronic structures, with rutile showing more pronounced band dispersion and lower effective mass characteristics.
• Significant differences in optical anisotropy and birefringence are observed between the two phases, with rutile displaying stronger anisotropic optical response.
• The theoretical results are consistent with other ab initio studies, confirming the reliability of the GGA+U approach for TiO2 systems.
• The study provides a validated theoretical model for predicting optical and electronic behavior in mixed-phase TiO2 thin films.