Water splitting using a semiconductor photocatalyst with sunlight has long been viewed as a potential means of large-scale H production from renewable resources. Different from anatase TiO , rutile enables preferential water oxidation, which is useful for the construction of a Z-scheme water-splitting system. The combination of rutile TiO with a suitable H -evolution photocatalyst such as a Pt-loaded BaZrO -BaTaO N solid solution enables solar-driven water splitting into H and O . While rutile TiO is a wide-gap semiconductor with a bandgap of 3.0 eV, co-doping of rutile TiO with certain metal ions and/or nitrogen produces visible-light-driven photocatalysts, which are also useful as a component for water oxidation in visible-light-driven Z-scheme water splitting. The key to achieving highly efficient water oxidation is to maintain a charge balance of dopants in the rutile, because single doping typically produces trap states that capture photogenerated electrons and/or holes. Here we provide a concise summary of rutile TiO -based photocatalysts for water-splitting systems.

Nitrogen/fluorine-codoped rutile TiO2 photocatalyst was developed toward solar-driven Z-scheme water splitting.

Mixed‐anion materials, consisting of metal cation(s) and more than two anionic species in a single phase, are promising as photocatalysts for water splitting and CO2 fixation using visible light. Oxynitrides, oxysulfides, and oxyhalides that contain d0 transition‐metal or d10 typical‐metal cations have absorption bands in the visible‐light region and band‐edge potentials suitable for these reactions. In general, visible‐light absorption by these materials arises from the p orbitals of anions that are less electronegative than oxygen, forming a valence band above the level of oxygen 2p orbitals. Less electronegative anions in mixed‐anion compounds are essential for visible‐light absorption but are inherently less stable than oxygen, potentially causing oxidative decomposition of the materials during photoreaction. More importantly, mixed‐anion compounds are more difficult to synthesize than single‐anion compounds like oxides, and reducing the density of defects that act as recombination centers for photogenerated charge carriers in the material remains challenging. There is therefore much room for improving the activities of mixed‐anion photocatalysts. Herein, the development of mixed‐anion materials for solar‐to‐fuel conversion over the past 10 years is highlighted with a focus on key milestones in research developments.

Substitution of oxide anions (O2–) in a metal oxide for nitrogen (N3–) results in reduction of the band gap, which is attractive in heterogeneous photocatalysis; however, only a handful of two-dimensional layered perovskite oxynitrides have been reported, and thus, the structural effects of layered oxynitrides on photocatalytic activity have not been sufficiently examined. This study reports the synthesis of a Ruddlesden–Popper phase three-layer oxynitride perovskite of K2Ca2Ta3O9N·2H2O, and the photocatalytic activity is compared with an analogous two-layer perovskite, K2LaTa2O6N·1.6H2O. Topochemical ammonolysis reaction of a Dion–Jacobson phase oxide KCa2Ta3O10 at 1173 K in the presence of K2CO3 resulted in a single-phase layered perovskite, K2Ca2Ta3O9N·2H2O, which belongs to the tetragonal P4/mmm space group, as demonstrated by synchrotron X-ray diffraction, scanning transmission electron microscopy measurements, and elemental analysis. The synthesized K2Ca2Ta3O9N·2H2O has an absorption edge at around 460 nm, with an estimated band gap of ca. 2.7 eV. K2Ca2Ta3O9N·2H2O modified with a Pt cocatalyst generated H2 from an aqueous solution containing a dissolved NaI as a reversible electron donor under visible light (λ > 400 nm) with no noticeable change in the crystal structure and light absorption properties. However, the H2 evolution activity of K2Ca2Ta3O9N·2H2O was an order of magnitude lower than that of K2LaTa2O6N·1.6H2O. Femtosecond transient absorption spectroscopy revealed that the lifetime of photogenerated mobile electrons in K2Ca2Ta3O9N·2H2O was shorter than that in K2LaTa2O6N·1.6H2O, which could explain the low photocatalytic activity of K2Ca2Ta3O9N·2H2O.