Predicting the suitability of photocatalysts for water splitting using Koopmans spectral functionals: The case of TiO$_2$ polymorphs

TL;DR

This work demonstrates that Koopmans spectral functionals provide accurate and efficient predictions of band gaps and ionization/electron affinity for TiO$_2$ polymorphs, enabling reliable band-edge alignment relative to water redox levels. By combining a DFT slab calculation (for vacuum reference) with a bulk Koopmans correction (KI/pKIPZ), the authors obtain band structures and band offsets that show anatase as the most promising photocatalyst among rutile, anatase, and brookite. The approach matches or surpasses many-body methods in accuracy while significantly reducing computational cost, making it suitable for high-throughput screening of new photocatalyst candidates. The results underscore the potential of Koopmans functionals to streamline the identification of materials with favorable band gaps and band-edge positions for photocatalytic water splitting, with practical implications for designing efficient hydrogen production catalysts.

Abstract

Photocatalytic water splitting has attracted considerable attention for renewable energy production. Since the first reported photocatalytic water splitting by titanium dioxide, this material remains one of the most promising photocatalysts, due to its suitable band gap and band-edge positions. However, predicting both of these properties is a challenging task for existing computational methods. Here we show how Koopmans spectral functionals can accurately predict the band structure and level alignment of rutile, anatase, and brookite TiO$_2$ using a computationally efficient workflow that only requires (a) a DFT calculation of the photocatalyst/vacuum interface and (b) a Koopmans spectral functional calculation of the bulk photocatalyst. The success of this approach for TiO$_2$ suggests that this strategy could be deployed for assessing the suitability of novel photocatalyst candidates.

Figures (5)

• Figure 1: Schematic illustration of photocatalytic water splitting. The conduction-band and valence-band regions are shown in green and red, respectively. When a photon (yellow line) with energy equal to or greater than the semiconductor band gap is absorbed, an electron is excited from the valence band to the conduction band, leaving a hole behind. The electron and hole subsequently participate in the reduction of hydrogen (top right) and oxidation of water (bottom right).
• Figure 2: Crystal structures of three $\mathrm{TiO_2}$ polymorphs
• Figure 3: Cartoon of the band alignment procedure. The black line represents the macroscopic average potential $\Delta V$ calculated across the slab for most stable surface facet. The IP and EA correspond to the VBM and CBM relative to the vacuum reference level, which is obtained via alignment of the average potential $V_\mathrm{bulk}$ for the slab and bulk systems.
• Figure 4: KI@PBE band structures (blue) with respect to the PBE bands (red); the zero reference energy is set to $\varepsilon_\mathrm{VBM}^\mathrm{DFT}$.
• Figure 5: Band alignment of TiO$_2$ polymorphs using KI and pKIPZ. Blue rectangles represent the IPs of three polymorphs, while red rectangles the EAs. Experimental values are given as red solid lines (from Refs. thomas2003resonantkashiwaya2018worknguyen2018koopmansthomas2007comparison).