Perovskite and layered perovskite oxynitrides for efficient sunlight-driven artificial synthesis

Abstract

Perovskite and layered perovskite oxynitrides are regarded as promising visible-light-responsive semiconductors for efficient artificial photosynthesis to produce renewable value-added energy resources, including H₂, formic acid (HCOOH), and ammonia (NH₃). This is because of their chemical stability in aqueous electrolytes at various pH values, along with a narrow bandgap and a suitable band position bestowing ideal optical properties. In this review, we discuss recent advances in photocatalysis using perovskite and layered perovskite oxynitrides responsive to a wide wavelength range of the visible-light spectrum. Next, we address in detail how the photoactivity of oxynitrides can be enhanced with respect to their synthesis, including bulk and surface engineering such as doping (or substitution), controlling their morphology, and crystal facet engineering. Finally, we discuss the existing challenges to realizing efficient artificial photosynthesis using these materials.

---

Jeongsuk Seo

Jeongsuk Seo is currently an associate professor in Department of Chemistry at Chonnam National University, Republic of Korea. She received her PhD degree under the supervision of Prof. Kazunari Domen in Chemical System Engineering from The University of Tokyo in 2014. Since then, she joined the Japan Technological Research Association of Artificial Photosynthetic Chemical Process (ARPChem) as a postdoctoral researcher. She moved to Shinshu University as an assistant professor in 2017 and then shifted to Chonnam National University as an assistant professor in 2019. At present, she has been working on materials science and catalysis for (photo)electrochemical energy conversion.

Kazunari Domen

Kazunari Domen is currently a university professor at The University of Tokyo and special contract professor at Shinshu University, Japan. He received his PhD degree in chemistry from the University of Tokyo in 1982. He joined Chemical Resources Laboratory at Tokyo Institute of Technology in 1982 as an assistant professor and was promoted to an associate professor in 1990 and a professor in 1996. He moved to The University of Tokyo as a professor in 2004 and was appointed by Shinshu University as a special contract professor in 2017. Since 1980, he has been working on overall water splitting using heterogeneous photocatalysts to produce clean and renewable hydrogen.

---

1. Perovskite and layered perovskite oxynitrides

Perovskite-type oxynitrides are mixed metallic compounds with the general formula AB(O,N)₃, where A and B represent a large alkali-earth metal or La (i.e., A = Ca, Sr, Ba, or La) and a relatively small transition metal (i.e., B = Ti, Ta, or Nb), respectively.^1–3 The stoichiometry of anions O and N in the perovskite is determined by the combination of cations at the A and B sites. Fig. 1 displays the crystal structure of an ideal cubic perovskite AB(O,N)₃ whereby the B(O,N)₆ octahedron is at the center of the cube while the large A atoms occupy the corners. The relationship between the ionic radii (r) expressed as r_A + r_(O,N) = √2(r_B + r_(O,N)) is valid for the ideal cubic perovskite structure. The ratio of left- to right-hand sides of the expression is called the tolerance factor t, and the ideal cubic perovskites have a t value of unity. Tilting of the octahedron in the cubic structure originates from the subtle differences between the ionic radii of the cations leading to a t value less than unity. However, structural distortion producing orthorhombic or tetragonal shapes is commonly found, except for very few combinations including BaBO₂N (B = Ta or Nb). In addition, the perovskite oxynitrides can adopt the A1_x−1A2_x and/or B1_y−1B2_y configurations at the A and/or B sites. In these cases, the stoichiometry of the anions deviates from the ideal coordination configuration if the cations at the A and B sites undergo a change in their oxidation state.

Fig. 1 Crystal structures of perovskite oxynitride BaTaO₂N and Ruddlesden–Popper (RP) phase layered perovskite oxynitride Ba₂TaO₃N.

Besides the ideal cubic structure, perovskites can also adopt layered structures in which AB(O,N)₃ layers are separated by thin layers of intrusive materials (commonly metal oxides), as shown in Fig. 1. Depending on the intruding layers, the perovskite derivatives designated layered perovskites are generally classified as (i) Aurivillius ((Bi₂O₂)(A_n−1B_nO_3n+1)), (ii) Ruddlesden–Popper (RP) (A_n+1B_nO_3n+1 or A′A_n−1B_nO_3n+1), or (iii) Dion–Jacobson (A′A_n−1B_nO_3n+1) phases (n = 1, 2, or 3 represents the number of stacked octahedra separated by a rock salt AO layer). The oxygen in the layered perovskites can be partially exchanged with another anion such as halogen and nitrogen. Sillen–Aurivillius layered perovskite oxyhalides A₄A′_n−1B_nO_3n+5X (A and A′ = Sr, Ba, Pb, or Bi; B = Ti, Nb, or Ta; X = Cl or Br) were prepared by the substitution of oxygen by halogen.^4,5 Moreover, RP phase oxynitrides A_n+1B_n(O,N)_3n+1 and Dion–Jacobson phase oxynitrides A′A_n−1B_n(O,N)_3n+1 (A = Ca, Sr, Ba, or La; A′ = Li, Na, K, or Cs; B = Ti, Nb, or Ta), composed of mixed anions of O²⁻ and N³⁻, have been studied as a novel layered perovskite oxynitride family active for various artificial photosynthesis approaches.^6–10

Perovskite oxides (e.g., SrTiO₃) and layered perovskite oxides (e.g., Ba₅Nb₄O₁₅) (well-known active catalysts for water splitting to produce H₂ gas) possess large bandgap energy values (E_g) of up to or even above 3.2 eV (λ < 380 nm), which are too high to harvest a wide range of the solar spectrum, including visible light.^11–13 Introducing N into the oxide (i.e., nitridation) shifts the valence band maximum (VBM) to a more negative potential because the potential of the N 2p atomic orbitals is lower than that of the O 2p orbitals.¹⁴ Although nitridation leaves the conduction band (CB) potential of the oxide almost unchanged, hybridization of the N 2p and O 2p orbitals during nitridation narrows the E_g value of the oxide, resulting in a visible-light-responsive perovskite or layered perovskite oxynitride.

The wavelength edges of the visible-light absorption by perovskite oxynitrides vary depending on the combination of cations at the A and B sites, as presented in Fig. 2. For instance, SrNbO₂N can be prepared by the partial exchange of 3O²⁻ for 2N³⁻ during the nitridation of layered perovskite Sr₅Nb₄O₁₅, with the resulting reddish-brown oxynitride SrNbO₂N able to absorb visible light up to a wavelength of approximately 680 nm.¹⁵ In the same manner, in the syntheses of SrTaO₂N, LaTiO₂N, BaTaO₂N, and BaNbO₂N, complete nitridation of the corresponding starting oxide resulted in oxynitrides capable of absorbing photons up to ca. 560, 620, 660, and 740 nm, respectively.^16–19 The apparent colors of the oxynitrides became orange, red, dark red, or brown based on their light absorption edge. As discussed previously, the VBM positions of perovskite oxynitrides are composed of hybridized N 2p and O 2p atomic orbitals. The CB minimum (CBM) of the oxynitrides mainly consists of empty Ti 3d, Nb 4d, or Ta 5d orbitals for the respective B-site cation with d⁰ electronic configuration.^20,21 The E_g value of oxynitrides typically decreases as the effective electronegativity of the B-site cations increases.²¹ The effective electronegativity of Nb⁵⁺ is larger than that of Ta⁵⁺, so Nb-based oxynitrides can absorb longer wavelengths of visible light (Fig. 2). Furthermore, the CBM positions of the oxynitrides are determined by the B–O/N–B bond angles.^21,22 The tilt of the B(O,N)₆ octahedra in a perovskite structure is larger for a smaller A-site cation, resulting in a narrower CB. The reduced dispersion of the CBs shifts the CBM positions to a more negative potential, thereby causing an increase in E_g. Thus, the E_g values of BaBO₂N, SrBO₂N, and CaBO₂N are sequentially larger according to the size of the A-site cation (r_Ba > r_Sr > r_Ca).

Fig. 2 Photographic images and UV-vis diffuse reflectance spectroscopy (DRS) spectra of perovskite oxynitrides AB(O,N)₃ (A = Sr, Ba, or La and B = Ti, Ta, or Nb) prepared via nitridation of layered perovskites A₅B₄O₁₅.

In addition to the narrow bandgap, the band positions of the perovskite and layered perovskite oxynitrides straddle the water redox potential (Fig. 3). This indicates that perovskite oxynitrides acting as a single absorber thermodynamically drive overall water splitting to produce H₂ and O₂ without the need to supply additional energy for the reaction. The favorable optical properties of the oxynitrides are distinguishable from those of other n-type oxide semiconductors exclusively driving water oxidation, such as visible-light-responsive BiVO₄ (λ < 530 nm) and α-Fe₂O₃ (λ < 600 nm).^23–25 Moreover, since the CBM potentials of the oxynitrides are more negative than the reduction potential of CO₂, HCOOH, carbon monoxide (CO), and methanol (CH₃OH) can be produced via photocatalysis. N₂ fixation to synthesize NH₃ gas is theoretically possible via photocatalysis with oxynitrides. Finally, since the VBM potentials of the oxynitrides are located below the potential of water oxidation reactions, they can catalyze the oxidation of various anions in seawater (including mineral salts).

Fig. 3 Bandgap energies (eV) and band edge potentials (V_NHE) of representative n-type perovskite and layered perovskite oxynitrides in contact with aqueous electrolytes at pH 0. The electrochemical potentials for H₂ and O₂ evolution, CO₂ reduction, Cl₂ evolution, and N₂ fixation are also presented for comparison.

Over several decades, perovskite and layered perovskite oxynitrides have been regarded as promising visible-light-responsive semiconductors for efficient artificial photosynthesis to produce value-added eco-friendly and renewable energy resources such as H₂, HCOOH, CH₃OH, and NH₃.^2,3,26–28 This is because the oxynitrides are chemically stable in aqueous electrolytes at various pH values and have ideal optical properties due to a narrow bandgap and a suitable band position, which makes them favorable for sunlight-driven photocatalysis. Therefore, we discuss recent advances in photocatalysis using various perovskite and layered perovskite oxynitrides responsive to a wide wavelength range of the solar spectrum. How the photoactivity of these compounds can be enhanced by their synthesis method for use in photoelectrodes, bulk and surface engineering by doping (or substitution), controlling their morphology, and crystal facet engineering is discussed in detail. We also discuss the challenges associated with using perovskite and layered perovskite oxynitrides for efficient artificial photosynthesis.

2. Visible-light-driven photocatalysis

A representative photocatalytic process using perovskite or layered perovskite oxynitrides is water splitting to produce renewable H₂ energy. The half water oxidation of water splitting typically exhibits very slow kinetics compared with half water reduction so that alternative oxidations using the oxynitrides have been addressed to boost water splitting and then produce H₂ gas with a higher evolution rate.²⁹ Recently, CO₂ reduction using these compounds has also been reported since carbon neutrality is becoming critical globally.^30,31 First, we discuss the various reactions that can be photocatalyzed by perovskite and layered perovskite oxynitrides under sunlight (or only visible-light irradiation), the standard reduction potentials for which are reported in Table 1.

Table 1 The thermodynamic redox potentials of various reactions that can be photocatalyzed by perovskite and layered perovskite oxynitrides in contact with aqueous electrolytes at pH 0 under sunlight

Half redox reactions	Balanced chemical reactions	Potential (VNHE)
CO2 reduction	CO2 + e^− ⇌ CO2^−	−1.90
	CO2 + 2H^+ + 2e^− ⇌ HCOOH	−0.20
	CO2 + 2H^+ + 2e^− ⇌ CO + H2O	−0.11
	CO2 + 4H^+ + 4e^− ⇌ HCHO + H2O	−0.07
	CO2 + 6H^+ + 6e^− ⇌ CH3OH + H2O	0.03
	2CO2 + 12H^+ + 12e^− ⇌ C2H5OH + 3H2O	0.08
	CO2 + 8H^+ + 8e^− ⇌ CH4 + 2H2O	0.17
Water reduction (HER)	2H^+ + 2e^− ⇌ H2	0
N2 reduction	N2 + H^+ + e^− ⇌ N2H	−3.2
	N2 + 2H^+ + 2e^− ⇌ N2H2	−1.1
	N2 + 4H^+ + 4e^− ⇌ N2H4	−0.36
	N2 + 8H^+ + 6e^− ⇌ 2NH4^+	0.27
	N2 + 6H^+ + 6e^− ⇌ 2NH3	0.55
Water oxidation (OER)	O2 + 4H^+ + 4e^− ⇌ 2H2O	1.23
Chloride oxidation (CER)	Cl2 + 2e^− ⇌ 2Cl^−	1.36
	ClO4^− + 8H^+ + 8e^− ⇌ Cl^− + 4H2O	1.39
	ClO3^− + 6H^+ + 6e^− ⇌ Cl^− + 3H2O	1.45
	HClO + H^+ + 2e^− ⇌ Cl^− + H2O	1.48
	HClO2 + 3H^+ + 4e^− ⇌ Cl^− + 2H2O	1.57
Hydrogen peroxide production	H2O2 + 2H^+ + 2e^− ⇌ 2H2O	1.78
Sulfate oxidation	S2O8^2− + 2e^− ⇌ 2SO4^2−	2.01

---

2.1. Overall water splitting via one-step photoexcitation

The most ideal means of producing H₂ is overall water splitting using a single semiconductor capable of absorbing visible light. Water redox reactions, i.e., hydrogen evolution reaction (HER) and oxygen evolution reaction (OER), are simultaneously activated at the separate surfaces of a semiconductor particle via one-step photoexcitation to produce H₂ and O₂, respectively, as follows:

2H₂O → 2H₂ + O₂ ΔG⁰ = 237 kJ mol⁻¹ (E⁰ = −1.23 V) (1)

where E⁰ is the thermodynamic potential difference for overall water splitting and ΔG⁰ is the Gibbs free energy change. Thus, a semiconductor particle to drive water splitting must have an E_g greater than 1.23 eV and its band position must span the water redox potentials of 0 and 1.23 V_NHE. As shown in Fig. 3, the thermodynamic characteristics of almost all perovskite and layered perovskite oxynitrides satisfy these requirements, thus making them promising semiconductor candidates for water splitting via one-step photoexcitation.

The solar-to-hydrogen (STH) conversion efficiency (η), a benchmark figure used to estimate the performance of a semiconductor for solar water splitting, is defined by

where r_H2 is the rate of hydrogen evolution, P_sun is the photon flux of sunlight, and S is the photoreaction area of a panel or electrode. The STH efficiency greatly depends on the E_g of the semiconductor, because E_g determines the photon flux available for water splitting. Fig. 4 presents the relationships between the E_g values of perovskite oxynitrides and their theoretical maximum STH efficiency, photocurrent, and resulting hydrogen evolution rate per unit area (cm²). Visible light in the wavelength range of 400–800 nm provides a large proportion of solar energy, and so the STH efficiency is higher in semiconductors with a wider light absorption edge. It is also known that an STH efficiency of 10% and above is required for practical water-splitting applications.^32,33 To satisfy this requirement, the semiconductor as a single absorber should be capable of absorbing visible light up to ca. 520 nm without performance loss of the water-splitting device. It is thus noteworthy that besides their suitable band potentials, perovskite and layered perovskite oxynitrides have a small enough E_g to achieve a high STH efficiency for water splitting. For instance, perovskite LaTiO₂N (E_g = 2.1 eV) can adsorb visible light up to a wavelength of approximately 600 nm (equivalent to a theoretical STH efficiency of 16%). Moreover, LaTiO₂N on its own or in a water-splitting device produces hydrogen gas at 6.1 mL h⁻¹ cm⁻², with a maximum photocurrent density of 13 mA cm⁻². The total photon energy harvested by BaNbO₂N (with an absorption edge of approximately 740 nm) for solar water splitting is much greater than that by LaTiO₂N, resulting in a maximum STH efficiency of 29% corresponding to hydrogen production of 11 mL h⁻¹ cm⁻². Thus, it is necessary to develop perovskite and layered perovskite oxynitrides with long light absorption edges of above 600 nm to realize commercial water-splitting applications with a high STH value.

Fig. 4 A graphical representation of the theoretical maximum solar-to-hydrogen (STH) conversion efficiency, maximum photocurrent density, and hydrogen evolution rate as a function of wavelength. Representative perovskite oxynitrides are positioned along the theoretical STH efficiency curve at their respective absorption edges. The three numerical values beside each perovskite oxynitride indicate the efficiency, photocurrent density, and hydrogen evolution rate.

Overall water splitting with a quantum efficiency of almost unity using one-step photoexcitation of cuboidal Al-doped SrTiO₃ (SrTiO₃:Al) particles with a perovskite crystal structure has been reported.¹² The remarkable water-splitting activity was attributed to the selective depositions of Rh/Cr₂O₃ (for HER) and a CoOOH cocatalyst (for OER) at the (100) and (110) crystal facets of SrTiO₃:Al, respectively. HER and OER could be separated and strongly promoted by the anisotropic transport of photogenerated electrons and holes in the single crystal. Overall water splitting on a 100 m² panel comprising modified SrTiO₃:Al particles is an example of a safe mass hydrogen production system with a maximum STH of 0.76%.³⁴ Although the system design of the panel for scale-up is simple and cost-effective, the low STH efficiency for mass water splitting is still a significant limitation.

Nevertheless, several semiconductors capable of absorbing visible light, such as graphitic-C₃N₄ (E_g = 2.8 eV),³⁷ (Ga_1−xZn_x)(N_1−xO_x) (E_g = 2.4 eV),³⁸ Ta₃N₅ (E_g = 2.1 eV),³⁹ and Y₂Ti₂O₅S₂ (E_g = 1.9 eV),⁴⁰ have been successfully applied to overall water splitting via one-step photoexcitation. Moreover, several perovskite oxynitrides have also been utilized. Complex LaB(O,N)₃ (B = Ta or Nb) solid solutions were successfully employed in the overall water splitting, resulting from the precise tuning of both E_g and bandgap positions of the oxynitrides via the substitution of B⁵⁺ with Mg²⁺.^41,42 These results are discussed in detail in the following section concerning doping (or substitution). Recently, overall water splitting under visible-light irradiation has also been achieved using ATaO₂N (A = Sr, Ba) particles with few defects.^35,36Fig. 5 represents the first demonstration of overall water splitting by particulate ATaO₂N crystals and the corresponding strategies based on the deposition of active cocatalysts. Under Xe lamp irradiation (λ > 420 nm), CrO_x/Ru/IrO₂/SrTaO₂N particles, which are more photoactive than BaTaO₂N, split water to produce H₂ and O₂ gases with initial evolution rates of 9.1 and 3.0 μmol h⁻¹, respectively.³⁶ The apparent quantum yield of surface-modified SrTaO₂N at the wavelength of 420 ± 30 nm was 0.34% in the initial stage of water splitting, which became decreasingly saturated to 0.005% during the reaction over 48 h.

Fig. 5 Overall water splitting by (A) perovskite CrO_x/Ru/IrO₂/SrTaO₂N and (B) IrO₂/Cr₂O₃/Na–Rh/BaTaO₂N:Mg. (a) Deposition strategies for the HER and OER cocatalysts and (b) the corresponding gas evolution activities over time under visible-light irradiation (λ > 420 nm) using a 300 W Xe lamp in ultrapure water under an Ar atmosphere with an initial background pressure of 5 or 10 kPa. (c) The amounts of H₂ and O₂ gases generated during water splitting for 3 h using SrTaO₂N with various cocatalysts: (i) CrO_x/Ru/IrO₂, (ii) CrO_x/Ru, (iii) CrO_x/Ru/IrO_2(AD) (AD: by adsorption of colloidal IrO₂), and (iv) IrO₂/CrO_x/Ru. Reproduced with permission.³⁵ Copyright 2022, ACS. Reproduced with permission.³⁶ Copyright 2023, ACS.

Although the H₂ evolution rate using ATaO₂N particles is very limited compared with that using SrTiO₃:Al particles,¹² it is remarkable that overall water splitting using the perovskite oxynitrides is achieved under visible-light illumination. Interestingly, a general strategy for designing oxynitride particles was employed for overall water splitting via one-step photoexcitation. Both perovskite oxynitrides were initially prepared with few defects via different synthetic routes. During nitridation, SrCl₂ and NaOH as fluxes, as well as Sr and O as sources, were introduced to obtain SrTaO₂N with few defects, while doping with Mg²⁺ with an Mg/Ta ratio of 0.1 into BaTaO₂N suppressed the Ta₃N₅ phase as an impurity. Subsequently, HER and OER cocatalysts were sequentially deposited on the perovskite oxynitride particles to promote the overall water-splitting ability. Ru or Rh covered with CrO_x or Cr₂O₃ was employed as the HER catalyst. It is well-known that chromium oxides prevent the reverse reaction (i.e., the oxygen reduction reaction (ORR)) at HER sites.^43,44 RuO₂ or IrO₂ catalysts deposited on the oxynitride surface promoted OER under light irradiation. For overall water splitting, although the bare surfaces of perovskite or layered perovskite oxides with a large bandgap (e.g., SrTiO₃) can be employed as OER sites for overall water splitting,^11,45 ATaO₂N particles with a narrow bandgap are still necessary to load the OER cocatalyst. In the solid solution case, a thin coating of amorphous oxyhydroxide on the surface of RhCrO_y/LaMg_1/3Ta_2/3O₂N suppressed N₂ evolution by self-photooxidation of the oxynitride, thereby improving overall water splitting.⁴⁶ Moreover, the deposition of HER and OER cocatalysts on ATaO₂N particles effectively suppresses the decomposition of the oxynitride and separates the reaction sites. Subsequently, the separation strategy of both types of reaction sites via the deposition of cocatalysts is mainly ascribed to the overall water-splitting activity of a single ATaO₂N particle.

2.2. Photoelectrochemical (PEC) water splitting

PEC water splitting using semiconductor electrodes is a potential means of producing H₂ and O₂. The HER and OER sites are divided into a p-type photocathode and an n-type photoanode, respectively. Based on this, the STH efficiency of a PEC water-splitting cell is typically higher than that of the overall water splitting using a photocatalyst.⁴⁷ Various combinations of p- and n-type electrodes have been studied for PEC water splitting without supplying electricity.^48–50 Unbiased PEC water splitting using an inverted and planar lead halide perovskite (LHP) based on the material of a perovskite solar cell (i.e., an NiFe/LHP photoanode and an NiPCoP/LHP photocathode) has recently been reported to generate a high STH conversion efficiency of 10.64% with a photocurrent density of 8.65 mA cm⁻².⁵⁰ A PEC cell using a perovskite BaTaO₂N photoanode combined with an La₅Ti₂Cu_0.9Ag_0.1S₅O₇ photocathode has also shown potential for the unbiased overall water splitting, albeit its activity is very low.⁵¹

Eqn (2) can be rearranged for the PEC water-splitting system to provide the STH efficiency (η) as follows:

where J, η_F, and P_sun are the photocurrent density obtained in a two-electrode configuration, faradaic efficiency, and the power density of sunlight (100 mW cm⁻² for AM 1.5G), respectively. According to the definition, a PEC cell must generate high photocurrent density for water splitting without any side reactions to obtain sufficient STH efficiency. The maximum photocurrent density of the cell is dependent on the E_g of the semiconductor, and notwithstanding, both photoelectrodes must be responsive to long wavelengths of visible light, as portrayed in Fig. 4. Moreover, the onset potentials for HER and OER at the photocathode and photoanode, respectively, are critical because the STH value of a PEC cell is estimated by using the photocurrent density obtained at the potential where the plotted linear sweep voltammetry (LSV) curves for HER and OER overlap, which is otherwise known as the operation voltage.³ Thus, a photoanode (or photocathode) capable of both absorbing long wavelengths of visible light and producing a high photocurrent density at a low potential (or high potential) is required for unassisted overall water splitting with high STH efficiency. For efficient water oxidation, the half-cell STH conversion efficiency (η) of the photoanode (a standard measure) is given bywhere J is the anodic photocurrent density, η_F is the faradaic efficiency, V_app is the applied potential, and P_sun is 100 mW cm⁻² for AM 1.5G. Based on these requirements, n-type perovskite and layered perovskite oxynitrides have been employed as a light absorber for the photoanode in a PEC cell to progress the OER because of their favorable optical property of being responsive to long wavelengths of visible light.^17–19,52

The B-site cations in perovskite AB(O,N)₃ can easily be reduced to a lower oxidation state (e.g., Nb⁵⁺ to Nb⁴⁺ or Nb³⁺) during high-temperature nitridation under a reducing NH₃ atmosphere owing to their high electronegativity. The reduced B-site cations cause the generation of anion defects and impurity traces that compensate for the charge imbalance, which increases the recombination of photogenerated holes and electrons during water splitting and leads to a decrease in photoactivity. In fact, the photoreaction takes place on the surface of the oxynitride, and surface defects therein negatively influence the PEC water-splitting activity.^18,53 The sunlight-driven PEC activities of LaTiO₂N, BaNbO₂N, and BaTaO₂N photoanodes have been significantly advanced compared to the other perovskite AB(O,N)₃ by controlling the surface defect density.^17–19 Prior to deposition of the OER electrocatalyst, Akiyama et al. cleaned the surfaces of synthesized LaTiO₂N particles by using mild poly(4-styrene sulfonic acid) (PSS) (Fig. 6(A)).¹⁷ This resulted in etching of the defective surface layer of the oxynitride; although longer acid treatment increased the number of fine pores in the surface, it decreased the weight of the oxynitride layer. Afterward, the CoO_x/LaTiO₂N photoanode produced a high photocurrent density of 8.9 mA cm⁻² at 1.23 V_RHE in a 1 M NaOH electrolyte at pH 13.5 under AM 1.5G irradiation, which is the highest reported so far. Moreover, its PEC water-splitting activity was more than two times higher than that of the as-prepared CoO_x/LaTiO₂N photoanode. Similarly, eliminating the surface defects in an LaTiO₂N layer via acid treatment with aqua regia doubled the photocatalytic HER and OER activities of the oxynitride particles.⁵⁴ Therefore, the effect of the acid treatment on photoactivity suggests that an oxynitride surface with a low number of defects is necessary to suppress the recombination of charge carriers and thereby boost water-splitting activity.

Fig. 6 PEC water splitting. (A) Scanning electron microscopy (SEM) images (a)–(d) and high-resolution (HR) transmission electron microscopy (TEM) images (e)–(h) of the surface of perovskite LaTiO₂N treated with poly(4-styrene sulfonic acid) (PSS) for (a), (e) 0, (b), (f) 1, (c),(g) 17, and (d), (h) 72 h. Corresponding linear sweep voltammetry (LSV) curves of CoO_x/LaTiO₂N photoanodes not treated (lower line) or treated with PSS for 17 h (upper line) used for PEC water splitting in a 1 M NaOH electrolyte at pH 13.5 under chopped AM 1.5G irradiation. Reproduced with permission.¹⁷ Copyright 2016, Wiley-VCH. (B) HRTEM images of the surface of perovskite BaNbO₂N as-prepared (a) or annealed under an Ar atmosphere at 873 K for 1 h (b). Corresponding LSV curves of Co(OH)_x–FeO_y/BaNbO₂N photoanodes used for PEC water splitting in a 0.5 M KBi electrolyte at pH 13 under AM 1.5G irradiation. Reproduced with permission.¹⁸ Copyright 2018, Wiley-VCH. (C) (a) LSV curves in various scan directions of a Co(OH)_x–FeO_y/BaTaO₂N photoanode annealed under an Ar atmosphere at 1073 K for 1 h after nitridation used for PEC water splitting in a 0.5 M KBi electrolyte at pH 13 under AM 1.5G irradiation. (b) The chronoamperometry curve obtained during long-term water splitting at the applied potential of 1.23 V_RHE for 24 h. (c) The half-cell STH (HC-STH) conversion efficiency (%) estimated from the LSV curve presented in (a). Reproduced with permission.¹⁹ Copyright 2019, ACS.

In another approach, annealing the surface of perovskite oxynitrides was effective in improving the water-splitting activity of Co(OH)_x–FeO_y/BaBO₂N (B = Nb, Ta) photoanodes under sunlight (Fig. 6(B) and (C)).^18,19 The surfaces of the as-prepared oxynitrides were amorphous as a result of mild nitridation of Lewis base Ba-rich Ba₅B₄O₁₅ as the starting oxide. Subsequently, their highly defective surface with a high concentration of oxygen atoms became a single crystal by annealing under an Ar atmosphere at a suitable temperature. For instance, Ar-annealing of BaNbO₂N at 873 K for 1 h resulted in a photocurrent density of 5.2 mA cm⁻² at 1.23 V_RHE for sunlight-driven PEC water splitting in a 0.5 M KBi aqueous electrolyte at pH 13, which is five times higher than that of the as-prepared oxynitride.¹⁸ Although using inert Ar gas has provided the most favorable results, the amorphous surface of oxynitrides can also be crystallized by annealing under other gas atmospheres, such as NH₃. The degree of improvement in photoactivity is inversely proportional to the surface defect concentration of the as-prepared oxynitride. Moreover, a highly defective oxynitride surface can become polycrystalline rather than a single crystal via Ar-annealing, leading to a small increase in photocurrent. The effect of the annealing temperature on the oxynitride is also dependent on the thermal stability of the oxynitride under an Ar atmosphere.⁵³ Prolonged annealing at the decomposition temperature of BaNbO₂N (>873 K) decreases the photoactivity. For instance, Ar-annealing highly enhanced the surface crystallinity of BaTaO₂N, probably because of its high thermal stability nearly up to 1200 K. Thus, the increased crystallinity of as-prepared bulk BaTaO₂N and its surface via Ar-annealing at 1073 K for 1 h induced a high photocurrent density of 6.5 mA cm⁻² at 1.23 V_RHE for water splitting in a 0.5 M KBi aqueous electrolyte at pH 13.¹⁹ The photoactivity of the oxynitride corresponded to a maximum half-cell STH energy conversion efficiency of 1.4% at 0.88 V_RHE, which is still the highest value yet reported using perovskites AB(O,N)₃. Moreover, the improved crystallinity both on the surface and in the bulk of BaTaO₂N led to long-term stability during water splitting over 24 h (79% retention of the initial photocurrent). These remarkable results clearly demonstrate that modifying the surface crystallinity of AB(O,N)₃via annealing or using acid etching to improve the surface texture, both of which lower the defect density, is an excellent way of providing strong separation and fast transfer of photogenerated charges therein, thus leading to highly active and stable PEC water splitting.

2.3. Alternative oxidation reactions to the OER

OER driven by four-electron transfer is relatively sluggish compared with HER activated via a two-electron pathway, the former creating a bottleneck for water splitting leading to a significant overpotential. The sluggish kinetics of OER typically become much slower in neutral electrolytes regardless of the electrocatalysts and/or semiconductors used, which indirectly causes a low production rate of hydrogen. Alternatively, a high evolution rate of hydrogen from water at neutral pH can be achieved via other oxidation reactions with relatively fast kinetics,^28,55,56 such as the chlorine evolution reaction (CER). In particular, the splitting of seawater with a high concentration (ca. 0.55 M) of Cl⁻ involving the CER can be driven under sunlight. In seawater, the thermodynamic potential for the oxidation of Cl⁻ driven via two-electron transfer to produce Cl₂ is 1.36 V_NHE at pH 0 (Table 1). Fig. 7(A) represents a calculated Pourbaix diagram for an artificial chlorine system in a 0.5 M NaCl aqueous electrolyte.⁵⁷ The disproportionation of Cl⁻ in water to produce hypochlorous acid (HClO), which is also activated by the two-electron pathway, is possible in the pH range of approximately 3–7.5 while the dissociation of HClO producing hypochlorite (ClO⁻) occurs at neutral and alkaline pH values because of its pK_a of 7.5. Although CER is thermodynamically less favorable than OER (E⁰ = 1.23 V_NHE at pH 0), the oxidation driven via the two-electron pathway in the former is kinetically catalyzed faster than in the latter. The CER is selectively more catalyzed than the competing OER even in strong acidic electrolytes due to its independence from the concentration of H⁺.⁵⁸ Moreover, using seawater instead of pure water is cost-effective for artificial H₂ production systems because the former is very abundant. Based on the advantages of the CER over the OER, PEC seawater splitting using WO₃ and BiVO₄ semiconductors at neutral pH has recently been reported.^58–60

Fig. 7 Chlorine evolution reaction (CER) in seawater as an alternative oxidation pathway. (A) A computed Pourbaix diagram for an artificial chlorine system in a 0.5 M NaCl aqueous electrolyte without any other mineral salts. Reproduced with permission.⁵⁷ Copyright 2016, Wiley-VCH. (B) The feasible mechanism for PEC seawater splitting using a perovskite oxynitride in a 0.5 M NaCl aqueous electrolyte. (C) (a) A schematic of the bottom-up fabrication of an SrNbO₂N/Nb photoanode involving oxidation and two-step nitridation. (b) LSV curves for the corresponding Co(OH)_x/SrNbO₂N/Nb photoanode during seawater splitting in a 0.2 M NaPi buffer with and without 0.5 M NaCl electrolyte (at pH 6.4 and 6.7, respectively) under AM 1.5G irradiation. (c) Time courses of O₂, ClO⁻, and H₂ generation during seawater splitting using the same photoanode and a Pt wire at an applied potential of 1.23 V_RHE in a 0.5 M NaCl electrolyte buffered with 0.2 M NaPi under AM 1.5G simulated sunlight for 150 min. The dashed lines indicate the amounts of H₂, ClO⁻ (e⁻/2), and O₂ (e⁻/4) estimated for a faradaic efficiency of unity. Reproduced with permission.²⁹ Copyright 2023, ACS.

The photoactivity of perovskite oxynitrides for OER is relatively low in neutral electrolytes compared to strong alkaline electrolytes (e.g., at pH 13).^61–63 This is because the band edge and flat band potential energies of AB(O,N)₃ are dependent on the pH value, and their OER kinetics are very slow under neutral conditions.^14,53 However, the limited activity of AB(O,N)₃ at a neutral pH makes them unfavorable for commercial water-splitting applications. Recently, the PEC activity of SrNbO₂N for water splitting at a neutral pH was largely enhanced via the bottom-up fabrication of the photoanode and the addition of NaCl to an aqueous electrolyte.²⁹ The feasible mechanism to simultaneously drive the HER and CER using the perovskite oxynitride in artificial seawater is illustrated in Fig. 7(B). The band edge potentials of SrNbO₂N thermodynamically span the standard electrode potential for CER, as well as those for OER and HER, indicating that the holes photogenerated at the surface of the oxynitride are consumed to drive the CER and OER. Fig. 7(C) demonstrates sunlight-driven seawater splitting in a 0.5 M NaCl aqueous electrolyte at pH 6.4 using an SrNbO₂N photoanode prepared via bottom-up fabrication including oxidation and flux-assisted nitridation: the oxidation process was necessary for growing the crystalline Nb₂O₅ layer on an Nb substrate while the subsequent two-step nitridation process at different temperatures caused the complete conversion of crystalline Nb₂O₅ to porous cuboidal SrNbO₂N with high crystallinity and a large surface area. Consequently, the Co(OH)_x/SrNbO₂N/Nb photoanode exhibited an mA-level photocurrent in a neutral 0.2 M NaPi aqueous electrolyte, which is a remarkable result for water splitting using a perovskite oxynitride. Moreover, the photoactivity of the oxynitride became three times higher in artificial seawater including 0.5 M NaCl providing the activation of the CER as well as the OER. Quantitative analysis proved that seawater splitting by SrNbO₂N resulted in the generation of HClO, ClO⁻, and O₂ in the oxidation compartment and H₂ evolution in the reduction compartment, corresponding to a faradaic efficiency of almost 90%. These results clearly indicate that the CER is more preferentially driven over the OER during seawater splitting at neutral pH. It also indicates that the CER can be an alternative oxidation reaction to the OER that improves the H₂ production activity of perovskite oxynitrides in strong alkaline to neutral environments.

Hydrogen peroxide (H₂O₂) production from water oxidation is regarded as an alternative oxidation reaction to the OER,^28,64 and H₂O₂ is also generated via the ORR using semiconductor photocathodes.^65,66 In this review, we only discuss H₂O₂ production from water oxidation in terms of improving H₂ production. As reported in Table 1, although H₂O₂ production is driven via a two-electron pathway similar to the CER leading to kinetically fast activation, its thermodynamic potential of 1.78 V_NHE is significantly unfavorable compared with the competing OER. The overall reaction for H₂O₂ production and water reduction via one-step photoexcitation is as follows:

2H₂O → H₂ + H₂O₂, E⁰ = −1.78 V (5)

Despite the thermodynamic disadvantage, the production of H₂O₂ based on the photocatalytic oxidation of water has recently been attempted because H₂O₂ is a much higher value-added product than O₂.^67–69 Moreover, attempts at suppressing the competitive OER and selectively improving the valuable H₂O₂ production have been successful. In particular, it has been reported that the HCO₃⁻ electrolyte in the pH range of 7–8 promotes the PEC activity of a BiVO₄/WO₃ photoanode for H₂O₂ production.⁷⁰ However, the very low faradaic efficiency for the PEC H₂O₂ production (approximately 54%) was improved up to 79% by introducing an Al₂O₃ overlayer on the photoanode.⁶⁷ In the reaction mechanism, the weakly basic HCO₃⁻ as a hole acceptor is adsorbed at the weakly acidic Al₂O₃ surface and oxidized to unstable HCO₄⁻, which then reacts with H₂O to produce H₂O₂. Furthermore, a faradaic efficiency of more than 90% for H₂O₂ production using the Gd-doped BiVO₄ photoanode in the potential range of 1.8–2.5 V_RHE has been reported.⁶⁹ The authors theorized that the Gd doping of BiVO₄ shifted the binding energy of OH⁻ on the active sites of Bi–Bi to a more optimal energy level, thereby boosting H₂O₂ production. Although the catalytic mechanism for H₂O₂ production is still uncertain,⁶⁴ it is remarkable that the selectivity and faradaic efficiency of the H₂O₂ production can be highly improved despite its unfavorable thermodynamic characteristics compared with the OER.

Although semiconductors with a wide bandgap (e.g., TiO₂, WO₃, and BiVO₄) have been employed as photocatalysts or photoelectrodes for H₂O₂ production,^28,64,65 photocatalytic oxidation to produce valuable H₂O₂ using perovskite or layered perovskite oxynitrides has not yet been reported. Based on the electromotive force of H₂O₂ production (1.78 V) being higher than that of water splitting, the band structures of perovskite oxynitrides with a negative VBM potential could make them unsuitable. However, the band structures of layered perovskite oxynitrides with relatively large E_g values possibly make them suitable for H₂O₂ production (Fig. 3). Moreover, the bandgap engineering of perovskite oxynitrides via substitution with alien elements could make them capable of thermodynamically driving H₂O₂ production.^41,46 Therefore, in addition to CER, H₂O₂ production using perovskite or layered perovskite oxynitrides may be an alternative oxidation reaction to OER that efficiently produces not only valuable H₂O₂ but also H₂.

2.4. CO₂ reduction

The photocatalytic reduction of CO₂ using a semiconductor is driven via a multiple-electron transfer process that generates many different products, such as formic acid (HCOOH), carbon monoxide (CO), methanol (CH₃OH), methane (CH₄), and so on, depending on the number of transferred electrons and the subsequent reaction pathway (Table 1). The catalytic mechanism is complex and has been unclear until very recently. Indeed, much research effort toward understanding it and increasing the selectivity for its high-value reaction products has been made because of the current global drive toward carbon neutrality.^71–73 Photocatalytic and PEC systems employing perovskite or layered perovskite oxynitrides for CO₂ reduction under visible-light irradiation have also been reported.^30,74

The band edge potentials of perovskite and layered perovskite oxynitrides straddling the various CO₂ reduction potentials in Fig. 3 suggest their suitability as a semiconductor for sunlight-driven CO₂ reduction. In fact, the overall reaction including the CO₂ reduction and OER in (sea)water containing dissolved CO₂ catalyzed by oxynitrides is thermodynamically feasible via one-step photoexcitation. However, this has still not been realized because of the small driving force of oxynitrides with a narrow bandgap for CO₂ reduction and sluggish OER kinetics driven via the simultaneous transfer of four electrons. Alternatively, a hybrid PEC cell constructed using a CuGaO₂/PRu-Re photocathode and a CoO_x/TaON photoanode exhibited visible-light-driven CO₂ reduction to produce CO and oxidation via the OER to release O₂ in an aqueous electrolyte with no external bias,⁷⁵ albeit the efficiency of the system was very poor.

In the study by Yoshitomi et al., although triethanolamine (TEOA) in an organic solvent instead of water was used as the hole acceptor, a hybrid perovskite CaTaO₂N coupled with a binuclear Ru(II) complex (RuRu′) catalyst activated CO₂ reduction under visible-light illumination to produce HCOOH with high selectivity.⁷⁶Fig. 8(A) illustrates the expected Z-scheme CO₂ reduction mechanism of the hybrid catalyst under visible light. The binuclear RuRu′ complex is composed of a redox photosensitizer unit (Ru(PS)) and a catalytic unit (Ru(Cat)). Both CaTaO₂N and RuRu′ are capable of absorbing visible light up to a wavelength of 500 nm. The authors proposed that the transfer of photoexcited electrons from the CBM potential of CaTaO₂N to Ru(PS) and then to Ru(Cat) provides the Z-scheme for CO₂ reduction. The effect of each component in the hybrid catalyst on the CO₂ reduction activity is summarized in Table 2. The combination of CaTaO₂N with RuRu′ exclusively enabled the detection of formate resulting from the dissociation of HCOOH with a high selectivity of greater than 99%. The oxynitride with only Ru(Cat) adsorbed thereon produced less formate and accompanying byproduct H₂, while its combination with Ru(PS) did not produce either product. These results indicate that both CaTaO₂N and the binuclear RuRu′ complex were necessary to realize HCOOH production. Moreover, the deposition of Ag particles on the CaTaO₂N surface resulted in three times higher CO₂ reduction activity. The metallic Ag particles did not act as a cocatalyst but instead as an intermediate promoter by accumulating photogenerated electrons and then mediating charge transfer from CaTaO₂N to RuRu′. This high activation and selectivity for visible-light-driven CO₂ reduction can be attributed to fast electron transfer caused by the suitable arrangement of the energy levels of CaTaO₂N and RuRu′ and by the deposition of Ag particles.

Fig. 8 CO₂ reduction. (A) The catalytic mechanism for visible-light-driven Z-scheme CO₂ reduction using a hybrid catalyst consisting of perovskite and layered perovskite oxynitrides and a binuclear Ru complex (RuRu′). Reproduced with permission.⁷⁶ Copyright 2015, ACS. (B) Band structure diagrams of layered perovskite Li₂LaTa₂O₆N and perovskites CaTaO₂N and LaTaON₂ estimated in an anhydrous acetonitrile (MeCN) electrolyte containing 0.1 M tetraethylammonium tetrafluoroborate (Et₄NBF₄) at pH 7. Reproduced with permission.²⁶ Copyright 2018, Wiley-VCH.

Table 2 The photocatalytic CO₂ reduction activity of various hybrid catalysts consisting of perovskite or layered perovskite oxynitrides and a binuclear Ru complex (RuRu′). Reproduced with permission.²⁶ Copyright 2018, Wiley-VCH. Reproduced with permission.⁷⁶ Copyright 2015, ACS

Parameter	Photocatalyst^a	Product amount (nmol)		Selectivity for formate (%)
		Formate	H2
a Reaction conditions: CO2 reduction for 15 h using 4.0 mg of photocatalyst in a CO2-purged N,N-dimethylacetamide (DMA)/triethanolamine (TEOA) (4 : 1 v/v) electrolyte of 4 mL under a 400 W high-pressure Hg lamp with a filter comprising sodium nitrite (NaNO2) solution. In each case, the adsorbed amount of RuRu′ and the loading amount of Ag on the semiconductor were 2.5 μmol g^−1 and 1.0 wt%, respectively. b The semiconductor upon which 3 μmol g^−1 of RuRu′ was adsorbed was immersed in mixed anhydrous acetonitrile (MeCN)/TEOA (4 : 1 v/v). RuRu′ is composed of a redox photosensitized unit (Ru(PS)) and a catalytic unit (Ru(Cat)). ND, not detected.
Reaction medium	CaTaO2N	ND	ND	—
	Ag/CaTaO2N	ND	ND	—
	RuRu′	ND	ND	—
	RuRu′/CaTaO2N	93	ND	—
Ag	RuRu′/Ag/CaTaO2N	320	ND	>99
RuRu′	Ru(Cat)/Ag/CaTaO2N	114	2.6	—
	Ru(PS)/Ag/CaTaO2N	ND	3.8	—
Semiconductor^b	Li2LaTa2O6N	ND	ND	—
	RuRu′/LaTaON2	ND	ND	—
	RuRu′/Li2LaTa2O6N	660	16	97
	RuRu′/Ag/Li2LaTa2O6N	1440	16	99

---

It has been reported that visible-light-driven CO₂ reduction can be highly activated by introducing a 2D-layered perovskite Li₂LaTa₂O₆N instead of CaTaO₂N.²⁶ Although the E_g value of layered perovskite Li₂LaTa₂O₆N is the same as that of CaTaO₂N, its band structure is slightly shifted to a more positive potential from that of the perovskite oxynitride due to its different A-site cations and lower concentration of nitrogen (Fig. 8(B)). According to the CO₂ reduction activity reported in Table 2, the CBM potential of Li₂LaTa₂O₆N, even though reduced, is sufficient to transfer photogenerated electrons to RuRu′. The tendency for the photoactivity of Li₂LaTa₂O₆N is identical to that of CaTaO₂N, i.e., the combination with RuRu′ and the deposition of Ag nanoparticles thereon are essential for driving the photoreaction with high selectivity. Interestingly, the CO₂ reduction activity using Li₂LaTa₂O₆N was almost five times higher than that using CaTaO₂N. The largely enhanced photoactivity was mainly ascribed to the high crystallinity of Li₂LaTa₂O₆N with a lower density of defect traps and higher density of reactive electrons, as analyzed by using transient absorption spectroscopy. However, the effect of the 2D structure of Li₂LaTa₂O₆N on the photoactivity has not yet been elucidated.^9,77

Perovskite oxynitrides employed as a single photocatalyst rather than in a hybrid catalyst with organic compounds for visible-light-driven CO₂ reduction have recently been reported.^78–80 The deposition of a core–shell Ni–Ag bicomponent cocatalyst on CaTaO₂N increased interfacial electron transfer and thus enhanced the CO₂ reduction activity, showing a synergistic effect of the bicomponent catalyst.⁷⁸ Moreover, heterojunction structures composed of a metal oxide (e.g., CeO₂) and LaTiO₂N have been shown to improve CO₂ reduction, which is a well-known strategy for effectively suppressing the recombination of photogenerated charges and thereby boosting the photoreaction.^79,80 This approach for CO₂ reduction mainly results in CO production accompanied by CH₄ as a secondary product, thereby indicating relatively low selectivity for them. Therefore, there is still room for improvement in sunlight-driven CO₂ reduction activity and product selectivity using perovskite and layered perovskite oxynitrides. Nevertheless, successful CO₂ reduction using the oxynitrides also demonstrates the suitability of their band structures for various artificial photosynthetic processes and H₂ production via water splitting.

3. Synthesis of perovskite and layered perovskite oxynitrides

3.1. Powder

Perovskite and layered perovskite oxynitrides are typically prepared via high-temperature nitridation of the starting oxides with a stoichiometric A/B ratio identical to that of the resulting oxynitride under an NH₃ atmosphere (e.g., Ca₂Ta₂O₇ for CaTaO₂N, Sr₂Nb₂O₇ for SrNbO₂N, or amorphous Sr₂TaO_x for Sr₂TaO₃N).^81–83 During nitridation, NH₃ thermally decomposes into H₂ and N₂, and N³⁻ from N₂ exchanges with O²⁻ in the starting oxide under the reducing atmosphere exemplified in the following equation for CaTaO₂N synthesis:

Ca₂Ta₂O₇ + 2NH₃ → 2CaTaO₂N + 3H₂O (6)

Otherwise, the oxynitrides are synthesized via calcination of the starting nitride precursor under an inert N₂/Ar atmosphere. For instance, BaTaO₂N has been prepared using TaN as the Ta precursor based on the following reaction:⁸⁴

BaCO₃ + TaN → BaTaO₂N + CO (7)

The ammonium chloride (NH₄Cl) powder instead of toxic NH₃ gas has also been used as the nitrogen source for nitridation because it thermally decomposes into NH₃ and HCl at a temperature higher than 610 K.⁸⁵

3.1.1. Polymerized-complex method. Commonly prepared via a solid-state reaction (SSR), the crystalline starting oxide with a cation ratio stoichiometric to the corresponding oxynitride is completely converted to the latter according to eqn (6). For multi-metal oxynitrides such as LaMg_xTa_1−xO_1+3xN_2−3x and K₂LaTa₂O₆N, it is difficult to prepare the single-crystalline starting oxide via an SSR without traces of impurities. Fortunately, a polymerized-complex method is a facile means of preparing homogeneous starting mixed oxides at the atomic level for nitridation to oxynitrides.^41,86,87 In short, citric acid and/or ethylene glycol are blended in methanol in which the metal precursors for the A- and B-site cations are dissolved, followed by heating to catalyze polymerization. The resulting transparent or yellow resin (proof of polymerization) is carbonized at a low temperature (≤773 K) to remove the organic chain in the resin. Subsequently, the carbonized powder is calcined at a higher temperature (≥1073 K) to finally yield a homogeneous white oxide; if it has a crystalline structure, the particle size can be controlled by varying the calcination temperature.⁸⁸ The homogeneous starting oxides are easily transformed via nitridation to the corresponding oxynitride with a single perovskite or layered perovskite structure that is responsive to visible light. Moreover, the resulting oxynitride is capable of water splitting and other artificial photosynthesis processes.^{26,41,42,83,89} The polymerized-complex method is effective at preparing oxynitrides with small particles and high crystallinity that provide many catalytic sites to drive photoreactions.^41,42

3.1.2. Conversion of A-site cation-rich starting oxides. The high crystallinity of a semiconductor is typically a critical factor for suppressing the recombination of photogenerated charges and promoting their separation and fast transfer. Many attempts to prepare perovskite and layered perovskite oxynitrides with high crystallinity and without impurity phases have been reported.^62,91–93 It is typical that the stoichiometric oxides prepared via SSR form crystalline oxynitrides. If the stoichiometric oxide has a crystal structure different from the corresponding oxynitride, then the conversion of the crystal oxide phase to the perovskite or layered perovskite will definitely go ahead during nitridation. This can require a long nitridation period, which sometimes results in a low-crystalline perovskite oxynitride with a high density of defects.^94,95 Moreover, exceptional oxynitrides such as BaBO₂N (B = Ta, Nb) have no corresponding starting stoichiometric crystalline oxide, so an amorphous mixture of oxides of all the elements within the oxynitride must be used instead.^91,96 Particulate ANbO₂N (A = Sr, Ba) can be prepared from the nitridation of stoichiometric perovskite ANbO₃ (Nb⁴⁺), as shown in Fig. 9(A).⁹⁰ Cubic perovskite SrNbO₃ is a red metallic photocatalyst for photocatalytic H₂ and O₂ evolution under visible light (λ > 420 nm).⁹⁷ Single-phase ANbO₃ can be synthesized in sealed evacuated fused quartz ampoules through the following reaction:

A₅Nb₄O₁₅ + xNb → 5ANb_(4+x)/5O₃ (stoichiometric ANbO₃ when x = 1.0) (8)

Fig. 9 Conversion of perovskite BaNbO₃ and layered perovskite Ba₅Nb₄O₁₅ to BaNbO₂N via nitridation. (A) UV-vis (DRS) spectra of starting BaNbO₃ and Ba₅Nb₄O₁₅ oxides for the synthesis of visible-light-responsive BaNbO₂N. (B) SEM images of the different starting oxides and the corresponding oxynitrides after nitridation. (C) X-ray diffraction (XRD) patterns for perovskite ANbO₂N (A = (a), (b) Ba, (c) Sr) and intermediate derivatives obtained via the nitridation of (a) BaNbO₃ and (b), (c) A₅Nb₄O₁₅ at 1123 and 1173 K, respectively. FWHM, full-width at half-maximum. Reproduced with permission.¹⁵ Copyright 2022, Elsevier. Reproduced with permission.⁹⁰ Copyright 2016, ACS.

The SEM images in Fig. 9(B) and the XRD patterns in Fig. 9(C)-(a) show the transformation of smooth perovskite BaNbO₃ to randomly porous BaNbO₂N via nitridation at 1173 K for 20 h. The perovskite BaNbO₃ was monotonically converted to perovskite oxynitride BaNbO₂N without involving any structural transition, leading to the exchange of oxygen with nitrogen. Thus, the use of ANbO₃ as the starting precursor was effective in the preparation of surface- and bulk-crystalline ANbO₂N. Nevertheless, the high water-splitting activity of ANbO₂N was not achieved from stoichiometric ANbO₃ (as x = 1.0). The high bulk crystallinity of ANbO₂N was accompanied by an impurity phase of NbO_xN_y and a surface concentration ratio of A/Nb lower than unity, resulting in an oxynitride with rather low photoactivity. An A-site-rich oxide (as x < 1.0) was necessary to obtain stoichiometric oxynitrides without the impurity phase because of the volatility of the A-site cation in the presence of alkali and alkali-earth metal groups during high-temperature nitridation. In response, the use of A-site-rich precursors has been applied in various syntheses of perovskite and layered perovskite oxynitrides.^62,94–96 Layered perovskite A₅B₄O₁₅ (A = Sr, Ba; B = Ta, Nb) has been employed in the synthesis of photoactive ABO₂N, as presented in Fig. 9(A).^15,18,19 The starting oxides have a crystal structure similar to that of ABO₂N, with the A-rich concentration being replenished during high-temperature nitridation to compensate for the volatility of the A-site cations. The layered perovskite was converted to perovskite oxynitride ABO₂N via the following reaction:

A₅B₄O₁₅ + 4NH₃ → 4ABO₂N + AO + 6H₂O (9)

The smooth-layered perovskites were completely changed to their porous ABO₂N counterpart via a process analogous to the conversion of perovskite ANbO₃. However, they are distinguishable in that the pores are orderly located in a layered structure, probably resulting from the decomposition of an AO slate layer in A₅B₄O₁₅ and the exchange of three O²⁻ ions in the equatorial plains of octahedral BO₆ with two N³⁻ (Fig. 9(B)).⁹⁸ As shown in Fig. 9(C)-(b), (c), the transformation of A₅Nb₄O₁₅ to ANbO₂N was completed with no AO impurity traces after 15 h of nitridation.¹⁵ The excess of A species, undoubtedly in the form of the amorphous phase and/or nanoparticles, was easily removed with distilled water in the washing step of ANbO₂N due to the A species being a Lewis base. The Lewis base A-rich species positively suppressed the reduction of the B-site cation, leading to ABO₂N with fewer defects and an adjusted stoichiometric A/B ratio in the oxynitride. However, it caused an amorphous surface to form on ABO₂N, which became crystalline via subsequent Ar-annealing treatment at a suitable temperature depending on the thermal property of ABO₂N that enhanced water-splitting photoactivity. In another conversion, employing a (Na_1/4Ba_3/4)(Zn_1/4Ta_3/4)O₃ solid solution with a perovskite structure to synthesize active BaTaO₂N maximized the evaporation of volatile Na and Zn elements during nitridation.⁹⁹

3.1.3. Flux-assisted calcination/nitridation. Flux-assisted nitridation using additive molten salts, as well as the adoption of suitable starting precursors, is an effective way of preparing highly crystalline perovskite and layered perovskite oxynitrides. The molten fluxes produced during high-temperature nitridation are employed as a solvent for crystallization of the oxynitrides. It is well-known that the solubility of the solute (i.e., the cation precursors of oxynitrides), which determines the crystallinity and surface morphology of the resulting product (i.e., the oxynitride), is influenced by the solute-to-flux molar ratio, operating temperature, and nitridation duration.^100,101 Oxynitrides are commonly synthesized via the flux-assisted nitridation of the corresponding precursors in a single step or the crystalline oxide precursors prepared via flux-assisted calcination in two steps. Single-crystalline LaTaON₂ prepared via a one-step flux method using KCl exhibited higher photocatalytic and PEC activity than those prepared via polymerized-complex and SSR methods because of the reduced defect density and enhanced crystallinity of the oxynitride.¹⁰² Furthermore, perovskite SrNbO₂N prepared via one- or two-step flux-assisted nitridation became highly crystalline and the resulting oxynitride prepared from the corresponding oxide through NaI-assisted calcination provided an excellent photocurrent density of 1.5 mA cm⁻² at 1.23 V_RHE under AM 1.5G simulated sunlight.⁹³ Crystalline BaNbO₂N particles prepared via the nitridation of Ba₅Nb₄O₁₅ and subsequent NaCl-assisted calcination provided significantly high photocatalytic OER activity under visible-light irradiation.⁹¹ These results demonstrate that flux synthesis is a worthwhile route for preparing highly crystalline perovskite oxynitrides.

Fig. 10(A) illustrates the anticipated growth mechanism of cubic-type BaTaO₂N particles showing how the flux-assisted nitridation conditions change the surface morphology of the resulting oxynitride. The following reaction steps are based on the XRD results:⁹²

BaCO₃ → BaO + CO₂ (10)

BaO + Ta₂O₅ → BaTa₂O₆ (11)

2BaTa₂O₆ + 3BaO → Ba₅Ta₄O₁₅ (12)

2Ba₅Ta₄O₁₅ + 2BaTa₂O₆ + 12NH₃ → 12BaTaO₂N + 18H₂O (13)

First, BaCO₃ as the Ba precursor is decomposed to BaO, after which BaO and Ta₂O₅ as the Ta source are dissolved in the KCl flux at a high temperature. The diffusion of the reactants through the molten flux leads to nucleation and growth of plate-like BaTa₂O₆ and Ba₅Ta₄O₁₅ (eqn (11) and (12), respectively). Finally, nitridation of the crystalline oxides under NH₃ flow causes the crystallization and growth of BaTaO₂N particles (eqn (13)). The crystalline growth during nitridation at 1123 K for 0 h (i.e., as the temperature was rising) provided vertically aligned plate-like shapes (Fig. 10(B)) comprising a mixture of BaTaO₂N, Ba₅Ta₄O₁₅, BaTa₂O₆, and Ta₂O₅. The plate-like shapes became thicker and irregular with increasing nitridation time up to 4 h. After 6 h of nitridation, the irregular particles had completely turned into the intrinsic crystal shape of cubic BaTaO₂N with clear edges. Subsequently, after 10 h, the cubic-like BaTaO₂N particles displaying specific (100) and (110) facets were more homogeneously dispersed. Interestingly, the surface morphology of BaTaO₂N during nitridation with a different flux was dissimilar to that during the KCl-assisted nitridation (Fig. 10(C)).¹⁰³ As RbCl and CsCl were utilized in the latter process, BaTaO₂N particles maintained their cubic crystal structure, although their edges were slightly truncated and the exposed facets were predominantly along the (100) plane. In particular, the BaTaO₂N crystals prepared using CeCl possessed many small steps at the edges of the cubic structure. Meanwhile, the oxynitride crystals prepared via BaCl₂-assisted nitridation had a tetradecahedral shape with exposed (111) and (100) facets. The various fluxes of cations changed the electrostatic forces between the precursor ions during the growth of BaTaO₂N, which thus determined the surface morphology and predominant crystal facets of the resulting oxynitride.

Fig. 10 Flux-assisted nitridation. (A) A schematic of the growth mechanism of cube-like BaTaO₂N crystals during nitridation using molten KCl. Reproduced with permission.⁹² Copyright 2015, ACS. (B) SEM images of BaTaO₂N particles grown via nitridation using molten KCl at 1223 K for (a) 0, (b) 1, (c) 2, (d) 4, (e) 6, (f) 8, or (g) 10 h. Reproduced with permission.⁹² Copyright 2015, ACS. (C) SEM images, TEM images, and SAED patterns of BaTaO₂N prepared via flux-assisted nitridation using different fluxes of (a)–(c) RbCl, (d)–(f) CsCl, and (g)–(i) BaCl₂. Reproduced with permission.¹⁰³ Copyright 2020, ACS.

3.2. Photoelectrodes

3.2.1. Particulate electrodes prepared via spin coating, electrophoretic deposition (EPD), and particle transfer methods. Particulate powder-type perovskite and layered perovskite oxynitrides are typically synthesized via the nitridation of their starting precursors, as discussed previously. To prepare oxynitride photoelectrodes for PEC artificial synthesis systems, the particulate powder must be deposited onto a conductive substrate such as F-doped tin oxide (FTO). To this end, oxynitride photoelectrodes have been fabricated via spin coating, EPD, and particle transfer methods.³ Spin coating is a very facile and scalable method for loading particulate oxynitrides onto conductive substrates, which has been utilized for the deposition of thin films in various applications, including solar cells.^105–107 A suspension of the oxynitride blended with organic solvents is drop-cast on the substrate via a spin coater while controlling the rotation speed. The thickness of the resulting particulate film is easily controllable via the surface morphology of the particles, the dispersion degree of the suspension, and the rotation speed.^106,108 For instance, 2D SrNbO₂N particles doped with Zr⁴⁺ have been uniformly packed onto an FTO substrate to produce a layer with a thickness of 2.5 μm by spin coating, as shown in Fig. 11(A).¹⁰⁵ The semi-transparent oxynitride film showed a significant photocurrent density of 2.0 mA cm⁻² at 1.23 V_RHE for water splitting under sunlight irradiation, thereby proving the feasibility of using the spin coating method to prepare water-splitting films. Meanwhile, EPD has become the most ubiquitous means of preparing photoanodes from particulate oxynitrides.^9,95,109Fig. 11(B) presents a schematic diagram of the fabrication of an oxynitride photoanode via the EPD method. The oxynitride powder is dispersed in an organic solvent including ionized species (e.g., I⁻), and two conductive electrodes are immersed parallel in the suspension solution. The positively and negatively charged species are shifted to the oppositely charged electrode at an applied DC voltage via electrostatic attraction, and the oxynitride particles are commonly deposited on the negative electrode. The distance between the two electrodes, the concentration of the charged species, the applied voltage, the duration, and so on, can all be varied to tune the thickness and deposition density of the resulting oxynitride film. In fact, heat treatment with conductive binders to promote the transfer of photogenerated charges between multiple layers of particles and/or the substrate (called necking) is an important step in preparing photoanodes via EPD.^61,95 The water-splitting photocurrent density over a TaON photoanode has been notably increased by necking treatment of the photoanode using TiCl₄.¹¹⁰ Nevertheless, the water-splitting activity of the photoanodes prepared via EPD is limited due to many grain boundaries between the oxynitride particles in the multiple layers slowing the migration of photoexcited charges. It has been suggested that the particle transfer method (Fig. 11(C)) can overcome the latter disadvantage in multi-layered particulate oxynitride photoanodes.¹⁰⁴ The ohmic contact layer (typically a metal) with a thickness of a few hundred nanometers is first deposited onto the top layer of the oxynitride particles loaded on a glass substrate by using radio-frequency (RF) magnetron sputtering. A thicker Ti conductor layer (∼5 μm) is then deposited on the ohmic contact layer via RF magnetron sputtering, resulting in high electroconductivity and mechanical strength. Subsequently, several of the upper layers physically connected to the conductor and contact layers are lifted off, which finally leads to an ideal oxynitride monolayer photoanode. An LaTiO₂N/Ta/Ti photoanode composed of a Ta ohmic contact and Ti conductive layers produced via the particle transfer method showed water-splitting photoactivity much higher than that of its counterpart prepared via EPD.¹⁰⁴ Moreover, the water-splitting photocurrent density at 1.2 V_RHE of the BaTaO₂N/Ta/Ti photoanode was 10 times higher than that of previously reported photoanodes of the same material.^61,62 The oxynitride particles on the photoanodes prepared via the particle transfer method are well dispersed in a monolayer, which promotes the efficient separation and vertical transfer of charges photogenerated using a single oxynitride. In fact, the highest PEC activities of perovskite oxynitrides LaTiO₂N, BaTaO₂N, and BaNbO₂N were observed in monolayered photoanode systems prepared via the particle transfer method.^17–19 These findings indicate that the monolayer electrode structure resulting from the particle transfer method is effective at controlling charge separation and transfer, thus enhancing the photoactivity for water splitting.

Fig. 11 Particulate photoelectrodes. Schematic diagrams of the preparation of perovskite oxynitride photoanodes via (A) spin coating, (B) electrophoretic deposition (EPD), and (C) particle transfer methods. Reproduced with permission.¹⁰⁴ Copyright 2013, RSC. Reproduced with permission.¹⁰⁵ Copyright 2023, Elsevier.

3.2.2. Bottom-up fabrication. The bottom-up fabrication of perovskite or layered perovskite oxynitride photoelectrodes from conductive substrates increases the conductivity of photogenerated charges between the oxynitride particles and/or the substrate over that of an electrode fabricated using the oxynitride powder, which probably enhances the photoreaction capability. Based on this, various bottom-up fabrication methodologies have been introduced for preparing oxynitride photoelectrodes. LaTiO_xN_y thin films grown on a TiN conductive layer have been synthesized by using a modified pulsed laser deposition method;¹¹¹ determining the surface crystal orientation of the (001), (011), or (112) facets of the LaTiO_xN_y thin films was achieved by using different facet-oriented TiN conductive layers. However, the water-splitting activity of the resulting oxynitride film was very limited, mainly due to the non-stoichiometric composition of the anions in the oxynitride. SrNbO₂N crystals have been grown on an Nb substrate via flux coating Sr(NO₃)₂ and chloride salts followed by nitridation.¹¹² The Sr-to-flux molar ratios and nitridation conditions determined the surface morphology and bulk crystallinity of the resulting SrNbO₂N and increased interfacial adhesion between SrNbO₂N and the Nb substrate. Despite the high interfacial adhesion of the oxynitride film, the water-splitting photoactivity was low. This was because the nitrogen during high-temperature nitridation managed to diffuse into the conductive substrate through the oxynitride semiconductor layer, which caused the generation of additional crystal and/or metallic interlayers. Thus, preparing photoactive oxynitride photoelectrodes via bottom-up fabrication is not straightforward.

Several studies showing the improved PEC water-splitting activity using perovskite oxynitrides prepared via bottom-up fabrication have recently been reported.^52,113,114Fig. 12 presents top-view and cross-sectional SEM images of various bottom-up-fabricated oxynitride films. Vertical SrNbO₂N nanorod arrays have been grown on an Nb substrate via a hydrothermal method and subsequent nitridation.⁵² SrNbO_x nanorod arrays were grown on the Nb substrate at a low temperature of 473 K, after which the oxide was transformed to SrNbO₂N nanorod arrays through nitridation at 1273 K for 2 h. As shown in Fig. 12(A), the diameter and length of the oxynitride nanorods were approximately 60 and 500 nm, respectively, thereby allowing favorable visible-light absorption and vertical separation of photogenerated charges. Interestingly, perovskite oxynitrides with 1D surface morphology have been reported for the first time. After the loading of CoO_x nanoparticles, SrNbO₂N nanorod arrays produced a notable photocurrent density of 1.3 mA cm⁻² at 1.23 V_RHE in an aqueous electrolyte at pH 13 under AM 1.5G simulated sunlight. Although the synthesis of the starting oxide at low temperature and a short nitridation period caused no additional interlayers detrimental to photoreactions, the light absorption edge of the SrNbO₂N nanorod arrays was approximately 630 nm, which is significantly shorter than 700 nm reported previously.^15,93

Fig. 12 Bottom-up fabrication of perovskite oxynitride photoanodes. (a) Top-view and (b) cross-sectional SEM images of (A) SrNbO₂N nanorod arrays prepared via a hydrothermal method followed by nitridation, (B) SrTaO₂N crystals grown on a LiTaO₃ substrate synthesized by using a reactive inorganic vapor method including nitridation, and (C) BaTaO₂N nanoparticle films deposited on a Nb substrate prepared via co-evaporation followed by nitridation. Reproduced with permission.⁵² Copyright 2020, Wiley-VCH. Reproduced with permission.^113,114 Copyright 2022, Elsevier.

Centimeter-scale perovskite SrTaO₂N crystals have been prepared by using a reactive inorganic vapor method including nitridation.¹¹⁴ The oxynitride crystals were epitaxially grown on a (110)-oriented LiTaO₃ substrate via the evaporation of Sr sources under an NH₃ atmosphere. During the evaporation step, Sr²⁺ was exchanged with Li⁺ in the (110)-oriented LiTaO₃ and nitrogen diffused into the oxide substrate simultaneously, leading to the growth of the SrTaO₂N crystals. Interestingly, different crystal orientations of LiTaO₃ (i.e., the (110) and (001) facets) resulted in nitridation rather than the intercalation of Sr²⁺ because the sizes of the lattice voids in the (110) and (001) facets are not large enough for the insertion of Sr²⁺ in the oxide substrate whereas they are in the (100) orientation. As shown in Fig. 12(B), the epitaxial growth of the perovskite SrTaO₂N crystals produced a dense film several tens of micrometers thick because of lattice expansion resulting from the insertion of Sr²⁺. The dense film layer minimized inter-particle interfaces and grain boundaries, thereby resulting in a high photocurrent density of 1.20 mA cm⁻² at 0.6 V_RHE (with a low onset potential of 0.35 V_RHE) for water splitting in a 1 M NaOH electrolyte at pH 13.6. However, lattice expansion during the conversion to oxynitride caused the limited diffusion of Sr²⁺, so the oxide substrate was not completely transformed to SrNbO₂N. Thus, these results indicate that the suitable selection of both the starting substrate and its crystal orientation is critical for the epitaxial growth of perovskite oxynitrides.

A co-evaporation method followed by nitridation to prepare a BaTaO₂N nanoparticle film directly grown on an Nb substrate has been reported.¹¹³ BaF₂ and Ta₂O₅ precursors were uniformly evaporated onto the substrate via electron beam deposition and then completely transformed to BaTaO₂N via nitridation under an NH₃ atmosphere. The resulting ∼760 nm-thick BaTaO₂N film comprised nanosized cubic crystalline particles, as shown in Fig. 12(C). The high crystallinity of the oxynitride was due to the uniform dispersion of Ba and Ta atoms in the resulting film. As a result, the BaTaO₂N photoanode produced a high photocurrent density of 4.7 mA cm⁻² at 1.23 V_RHE for water splitting in a 1 M KOH electrolyte at pH 13.6 under sunlight irradiation. The advantage of the co-evaporation method is the ability to regulate the reduction of Ta⁵⁺ during high-temperature nitridation by tuning the Ba/Ta ratio, which is similar to powder syntheses. In conclusion, the various bottom-up fabrication methods for perovskite oxynitrides reviewed in this article could enhance water-splitting activity by suppressing the generation of defects and additional interlayers in the resulting oxynitride film.

4. Bulk and surface engineering

4.1. Cation doping and substitution

The doping (or substitution) of foreign elements into a semiconductor is a typical means of modifying its bulk and/or surface properties, which can lead to enhanced photocatalytic or PEC water-splitting activities. In this review, doping and substitution are distinguished from each other according to the amount of the alternative species introduced into the perovskite oxynitride: doping is below 10 at% and substitution is above 10 at%. Moreover, we cover the doping (or substitution) of the A- and/or B-site cations in the perovskite or layer perovskite oxynitrides. Perovskite BaBO₂N (B = Ta, Nb) has an ideal cubic crystal structure and the volume of its unit cells is markedly small.^90,115 By contrast, the cubic lattices of AB(O,N)₃ (A = La, Ca, or Sr; B = Ta, Nb) are distorted because of the A-site cations being smaller than Ba²⁺ and the tilting of B(O,N)₆ octahedra. These findings imply that the isostructural replacement of cations in perovskite oxynitrides with alternative ones could be significantly limited in terms of the size and amount of the dopant (or substituent).

The doping of perovskite SrNbO₂N with 2 at% Zr⁴⁺ reduced the bulk and surface defect densities of the oxynitride by suppressing the reduction of Nb⁵⁺ during nitridation, thereby improving its PEC water-splitting activity.¹⁰⁵ Although the limited replacement of Nb⁵⁺ with Zr⁴⁺ maintained the optical properties of the oxynitride (i.e., the absorption of visible light up to a wavelength of 680 nm), using a larger amount of Zr⁴⁺ caused Zr-related impurity phases, thereby notably decreasing the photoactivity. Moreover, doping the Nb⁵⁺ sites in BaNbO₂N with Ti⁴⁺, Zr⁴⁺, W⁶⁺, or Mo⁶⁺ has been attempted to control the donor density in the oxynitride bulk originating from the generation of reduced species during nitridation.¹¹⁶ Doping with lower-valent cations Ti⁴⁺ and Zr⁴⁺ suppressed the generation of Nb⁴⁺ during nitridation whereas doping with higher-valent cations W⁶⁺ and Mo⁶⁺ rather increased the donor density. Moreover, doping with up to 4 at% Ti⁴⁺ resulted in the synthesis of BaNbO₂N without impurity traces, which thus exhibited the highest water-splitting activity. However, doping of the lower-valent cations at the Ta-sites in BaTaO₂N did not effectively lower the defect density and thus improve photoactivity despite the similarity of the ionic radii of 6-coordinate Ta⁵⁺ (78 pm) and 6-coordinate Nb⁵⁺ (78 pm). Selective replacement at the Ta-site in perovskite BaTaO₂N using 5 at% Al³⁺, Ga³⁺, Sc³⁺, or Zr⁴⁺ for improving photocatalytic HER and OER activity has also been considered.¹¹⁷ In particular, the doping of BaTaO₂N with Mg²⁺ or Zr⁴⁺ enhanced OER photoactivity in a sacrificial reagent by modifying the optoelectronic and surface properties of the oxynitride. Therefore, limited doping of AB(O,N)₃ using alternative cations alters the surface properties of oxynitrides (such as defect and donor densities) while maintaining their bulk properties (such as the crystalline structure and optical properties).

Co-substitution of the Ta⁵⁺ site of Ta₃N₅ with 33 at% 6-coordinate Mg²⁺ (86 pm) and Zr⁴⁺ (86 pm) led to significant improvement in the anodic photocurrent density and onset potential for PEC water splitting.¹¹⁸ This large amount of substitution modified not only the band structure but also the surface morphology of the original Ta₃N₅. The difference in valence between parent Ta⁵⁺ and alternative cations Mg²⁺ and Zr⁴⁺ caused additional N³⁻/O²⁻ replacement to replenish the charge imbalance. Increasing and decreasing the concentration of Mg²⁺/Zr⁴⁺ cations and N³⁻ anions respectively shifted the CBM and VBM potentials negatively from those of parent Ta₃N₅. This strategy was also applied in the synthesis of perovskite LaMg_xB_1−xO_1+3xN_2−3x (B = Ta, Nb; x = 0 to 2/3) solid solutions, leading to overall water-splitting activity via one-step photoexcitation.^41,42Fig. 13(A)-(a) portrays the crystal structures of LaMg_xTa_1−xO_1+3xN_2−3x solid solutions after substituting Ta⁵⁺ with Mg²⁺.⁴¹ The general formula of these perovskite oxynitrides is given by

(1 − 3/2x) LaTaON₂ + 3/2x LaMg_2/3Ta_1/3O₃ → LaMg_xTa_1−xO_1+3xN_2−3x (x = 0 to 2/3) (14)

Fig. 13 Substitution of B⁵⁺ with Mg²⁺ in perovskite LaBON₂ (B = Ta, Nb). (A) (a) Crystalline structure and (b) UV-vis DRS spectra of the LaMg_xTa_1−xO_1+3xN_2−3x (x = 0 to 2/3) powder after replacing some of the Ta⁵⁺ with Mg²⁺. Reproduced with permission.⁴¹ Copyright 2015, Wiley-VCH. (B) (a) UV-vis DRS spectra and (b) band structure diagrams of LaMg_xNb_1−xO_1+3xN_2−3x (x = 0 to 2/3) after replacing some of the Nb⁵⁺ with Mg²⁺. Reproduced with permission.⁴² Copyright 2021, RSC.

The composition of the LaMg_xTa_1−xO_1+3xN_2−3x solid solutions was easily modified by mixing perovskite oxynitride LaTaON₂ and perovskite oxide LaMg_2/3Ta_1/3O₃ at the atomic level. The significant substitution of Mg²⁺ in LaTaON₂ was realized after the nitridation of the starting oxides blended at the atomic level via a polymerized-complex method. The greater the level of substitution of Ta⁵⁺ with Mg²⁺, the greater the increase in the composition ratio of LaMg_2/3Ta_1/3O₃, resulting in a marked change in the optical properties of the resulting oxynitrides (Fig. 13(A)-(b)). The wavelength of the light absorption edge of LaMg_xTa_1−xO_1+3xN_2−3x blue-shifted gradually from 640 nm (pure LaTaON₂) to 270 nm (fully substituted LaMg_2/3Ta_1/3O₃).

This phenomenon was also observed in the substitution of Nb⁵⁺ with Mg²⁺ in counterpart LaNbON₂.⁴²Fig. 13(B)-(a) presents UV-vis DRS spectra of LaMg_xNb_1−xO_1+3xN_2−3x (x = 0 to 2/3) showing a gradual blue-shift of the absorption onset wavelengths from 700 to 320 nm. In addition to the increase in the E_g, the CBM and VBM potentials of the solid solutions also shifted to the negative and positive sides, respectively, as portrayed in Fig. 13(B)-(b). The band structure diagrams of the solid solution series demonstrate that the substitution of Nb⁵⁺ with Mg²⁺ in LaNbON₂ not only increases the concentration of the O 2p atomic orbital (leading to the positive shift of the VBM) but also decreases the concentration of the Nb 4d orbital (causing the negative shift of the CBM). The substitution of Mg²⁺ largely reduced the bulk and surface defect densities of LaMg_xNb_1−xO_1+3xN_2−3x solid solutions, which is similar to the effect of the small-cation doping of oxynitrides. Moreover, the subsequent changes in the optical properties (i.e., the E_g value and band structure) of the oxynitrides were significant. In addition, surface modification using cocatalysts finally activated the overall water-splitting capability of LaMg_xB_1−xO_1+3xN_2−3x, thereby producing stoichiometric H₂ and O₂ evolution. Interestingly, the highest gas evolution rates for water-splitting were achieved using LaMg_1/3B_2/3O₂N (i.e., x = 1/3) irrespective of the cationic occupancy of the B sites in the oxynitride. As a result, the visible-light-driven overall water-splitting activity is mainly attributed to the expanded bandgap enabling a photoresponse of up to 600 and 650 nm for LaMg_1/3Ta_2/3O₂N and LaMg_1/3Nb_2/3O₂N, respectively. Likewise, the bulk modification of oxynitrides via substitution with alternative cations has been subsequently applied: Zn²⁺ for Ta⁵⁺ in perovskite BaTaO₂N,¹¹⁹ Zr⁴⁺ for Ta⁵⁺ in LaTaON₂,¹²⁰ Mg²⁺ for Ti⁴⁺ in LaTiO₂N,¹²¹ La³⁺ for Ti⁴⁺ in layered perovskite Sr₂TiO₄,¹²² and so on. All these outcomes obviously demonstrate that the substitution of the B sites in oxynitrides with alternative cations is a very useful way of modifying the bulk properties of the parent oxynitride and thus improving the latter's photocatalytic water-splitting capability.

4.2. Varying the surface morphology

The electrochemical activity of an electrocatalyst (e.g., Pt nanoparticles) for oxidation and reduction reactions is typically proportional to its electrochemically active surface area containing its reaction sites for accessing the electrolyte.^124–127 The surface morphology (including the particle size and shape) of the electrocatalyst to increase reaction sites is thus critical for progressing electrochemical reactions efficiently. In contrast, the surface morphology of a semiconductor influences visible-light harvesting, the reaction efficiency of photocarriers, and the deposition coverage of cocatalysts (e.g., CoO_x) on which the actual photoreaction takes place. Because the small size of semiconductor particles can lead to a short diffusion distance for the photogenerated charges, it increases the charge separation, transfer, and collection efficiencies, thereby enhancing photoactivity.¹²⁸ BaNbO₂N particles of various sizes prepared via the nitridation of Ba₅Nb₄O₁₅ size-controlled in the range from 0.2 to 50 μm exhibit photocatalytic water oxidation activity depending on the particle size in an aqueous electrolyte including a sacrificial reagent.¹²⁹ Similarly, small scheelite-type LaNbO₄ particles as the starting oxide produced small LaNbON₂ particles with fewer defects after nitridation with significantly high photocatalytic water oxidation activity under visible-light Xe lamp irradiation (λ > 420 nm).^94,95Fig. 14(A) presents the size dependence of perovskite BaNbO₂N particles on PEC water-splitting activity⁸⁸ The size of the oxynitride particles was systematically tuned via the nitridation of Ba₅Nb₄O₁₅ with various average sizes (1.05, 1.93, 4.11, or 6.92 μm). The oxide particles synthesized via a polymerized-complex method were gradually grown by increasing the calcination temperature from 1073 to 1373 K. Their conversions to the corresponding oxynitride exclusively increased the surface porosity without an increase in particle size, except for the nitridation of the oxide calcined at 1073 K. The small, porous oxynitride prepared from the oxide via calcination at 1173 K provided both the largest BET and electrochemical surface areas, resulting in the highest photocurrent density of 2.2 mA cm⁻² at 1.23 V_RHE for sunlight-driven water splitting. The photocurrent decreased upon increasing the oxynitride particle size. Although the smallest non-porous BaNbO₂N particles prepared via nitridation of the oxide calcined at 1073 K were highly crystalline, this did not contribute to increasing the surface reaction sites. This result demonstrates that the size reduction of oxynitride particles accompanied by an increase in the surface reaction area clearly enhances the water-splitting photoactivity. In fact, the photon flux in the solar spectrum (used to determine the number of charges generated during photoreactions) is constant depending on the light wavelength. Therefore, the activity result of perovskite BaNbO₂N depending on its particle size implies that the photoresponse of the oxynitride is still relatively poor, and there is a large room for improvement in the activity up to the theoretical maximum STH efficiency of each oxynitride.

Fig. 14 Varying the surface morphology of Nb-based perovskite oxynitrides. (A) SEM images (a) and the corresponding PEC water-splitting activity (b) of BaNbO₂N particles of various sizes. Reproduced with permission.⁸⁸ Copyright 2022, Elsevier. (B) SEM images of starting LaKNaNbO₅ (a) and the corresponding 2D LaNbON₂ (b) after nitridation along with an HRTEM image (c) and the photocatalytic O₂ evolution activity (d) of the 2D oxynitride particles. Reproduced with permission.¹²³ Copyright 2020, RSC. SEM images of starting Sr₅Nb₄O₁₅ (e) and the resulting 2D SrNbO₂N (f) after nitridation along with an HRTEM image (g) and the PEC water-splitting activity (h) and charge separation efficiency (i) of the 2D oxynitride photoanode. The photoactivities of both 2D oxynitrides were compared with those of typical 3D oxynitrides. Reproduced with permission.¹⁰⁵ Copyright 2023, Elsevier.

Besides the particle size, the surface shape of a perovskite or layered perovskite oxynitride also influences its photoactivity. As discussed previously, fluxes during nitridation can be used to readily change the shape of the oxynitride particles. The employment of different fluxes during nitridation and/or calcination of the starting oxide has been applied to vary the shape and size of 3D SrNbO₂N particles.⁹³ Moreover, the nitridation of La₂Ti₂O₇ using an NaCl–KCl flux produced highly porous brick-like LaTiO₂N particles.¹⁶ The surface shape of the oxynitride with a large surface area improved its photocatalytic water oxidation activity because of the high dispersion of CoO_x or IrO₂ cocatalysts over the large surface of the oxynitride. Meanwhile, rarely reported vertical SrNbO₂N nanorod arrays (as discussed earlier) with a 1D structure of the semiconductor promoted the strong separation of photogenerated charges exclusively in the vertical direction, thereby enhancing PEC activity.

Over the past few years, newly emerged 2D configurations of perovskite and layered perovskite oxynitrides have been found to provide a short diffusion transfer route for the photoexcited charges and thus lower their recombination.^31,130–133 The electrons and holes photogenerated in HCa₂Nb₃O₁₀ nanosheets and selectively transferred to the surface of 2D Ca₂Nb₃O₁₀⁻ plates without their self-recombination significantly improved photocatalytic water reduction activity under UV irradiation (λ > 300 nm).¹³⁴ A 2D layered perovskite K₂LaTa₂O₆N was also stably active during the same photoreaction under visible-light illumination (λ > 400 nm).⁸⁹ The optical and electrical properties and charge transport of 2D RP-type A₂BO₃N (A = Ca, Sr, or Ba; B = Ta or Nb) layered perovskites have been studied based on density functional theory.¹³⁵ Thereby, the mobility of the electrons (or holes) photogenerated in the 2D layered perovskites was found to be much quicker than in the 3D perovskites. These results indicate that the advantages of 2D morphology (i.e., the fast vertical charge transfer to the surface of the oxynitride plates) enhance water-splitting photoactivity.

Because the crystal structure of perovskite AB(O,N)₃ is cubic, the 3D shape of the oxynitride is typically synthesized. However, 2D-type starting oxides with a layered perovskite structure could be transformed into 2D oxynitrides capable of absorbing visible light. Fig. 14(B) shows the surface morphologies of the starting oxides and their corresponding oxynitrides, along with the visible-light-driven photoactivity of the 2D oxynitrides compared with the corresponding 3D type. In the upper row, it can be seen that the smooth layered LaKNaNbO₅ plates were converted to porous 2D LaNbON₂via nitridation accompanied by volatilization of the K and Na species.¹²³ This conversion also worked well in the synthesis of counterpart LaTaON₂ using LaKNaTaO₅.¹³¹ Interestingly, volatilization of the layered oxide along the [001] direction resulted in the shape of the oxynitride with exposed (010) surface facets remaining unchanged. The photocatalytic water oxidation activity of 2D LaNbON₂ with (010) facets was much higher than that of conventional 3D LaNbON₂ with mostly (100) facets. The efficient transfer of charges along the (010) facets of the 2D oxynitride and its low defect density resulting from the fast nitridation and the volatilization of the K and Na species led to its enhanced photoactivity. Meanwhile, layered perovskite Zr-doped Sr₅Nb₄O₁₅ with a 2D-type truncated octahedral structure was completely transformed to porous 2D-type SrNbO₂N particles with an identical morphology, as shown in the bottom row of Fig. 14(B).¹⁰⁵ The conversion process yielded oxynitrides with fewer defects, as discussed in detail in the earlier synthesis section using A-site cation-rich starting oxides. The 2D shape of the SrNbO₂N particles, as well as their high crystallinity, significantly heightened the PEC water-splitting activity of the oxynitride. The efficiency of charge separation by the 2D SrNbO₂N was much higher than that by the conventional 3D oxynitride. Moreover, the charge transfer resistance in the 2D oxynitride was relatively low. These results enable us to draw the same conclusion that oxynitrides with a 2D surface structure facilitate strong charge separation and fast transfer of charges photogenerated during water splitting.

4.3. Controlling the crystallographic facets

Crystal facet engineering for more efficient photoreaction activity by effectively promoting the preferential separation of charges between two adjacent anisotropic facets in a semiconductor during the reaction has been recently reported.^27,136 Representatively, it has been demonstrated that monoclinic BiVO₄ particles with two different facets prepared via a hydrothermal method and the selective photo-deposition of cocatalysts could separate photogenerated holes in the (110) facets while collecting the counterpart electrons in the (010) facets.¹³⁷ Au, Pt, or Ag nanoparticles were deposited on the (010) facets of BiVO₄ under light irradiation while MnO_x or PbO₂ particles were coated on the (110) facets of the oxide in the same manner. Both selective depositions of the cocatalysts, particularly the Pt and MnO_x particles, dramatically enhanced the photocatalytic water oxidation activity of the BiVO₄ particles. Similarly, spatial separation of the photogenerated holes and electrons with CoO_x and RhO_x cocatalysts, respectively, also improved the PEC water oxidation activity.¹³⁸ Moreover, optoelectrical simulation proved that facet engineering increased the efficiencies of not only charge injection but also charge separation during the water oxidation reaction.

The facet engineering of cubic perovskite SrTiO₃:Al was very effective at spatially separating the photogenerated charges to two different facets, thus driving overall water splitting via one-step photoexcitation.¹² Rh/Cr₂O₃ as an HER cocatalyst and CoOOH as an OER catalyst were sequentially photo-deposited onto the electron-attracting (100) facets and the hole-collecting (110) facets, respectively. The resulting anisotropic charge transport achieved high overall water-splitting activity that was almost consistent with the theoretical quantum efficiency of SrTiO₃. Unlike the oxide, it is difficult to prepare perovskite oxynitrides with exposed crystal facets to spatially separate charges because their crystal growth is not easily controlled during typical high-temperature nitridation. Fortunately, flux-assisted nitridation may offer a feasible means of tuning the surface morphology of oxynitrides, as discussed in the synthesis section. Symmetric BaTaO₂N particles with only the (100) and (110) facets exposed have been successfully prepared via flux-assisted nitridation, as displayed in Fig. 15(A).¹³⁹ The starting precursors of Ba and Ta were rearranged and then recrystallized in the molten salt during nitridation, which resulted in well-defined cubic BaTaO₂N particles of approximately 200 nm in size. Interestingly, the BaTaO₂N crystals prepared using an NaCl flux exhibited a cubic structure with six isotropic (100) facets whereas the oxynitride synthesized using a KCl flux presented a cubic-like structure with additional smooth (110) facets at the twelve corners of the (100) facets. The ionic species of fluxes become adsorbed on the crystal facets, thereby enabling faster growth and changes in the electrostatic interactions between the crystals. Because the ionic radius of six-coordinate K⁺ (152 pm) is very similar to that of six-coordinate Ba²⁺ (149 pm), K⁺ can more easily occupy the Ba²⁺ positions at the interface between BaTaO₂N and KCl. This not only suppresses the crystal growth of BaTaO₂N with the most stable (100) facets but also leads to the formation of secondary stable (110) facets. According to density functional theory calculations (Fig. 15(B)), the estimated surface energy of the (110) facets is higher than that of the (100) facets, meaning that the CBM and VBM potentials of the (110) facets are located above those of the (100) facets. Thus, a facet junction is generated between the anisotropic (100) and (110) facets of a BaTaO₂N particle, thereby boosting the selective transfer of photogenerated electrons and holes to the (100) and (110) facets, respectively. Like the previously described BiVO₄ system, Pt and MnO_x particles were separately photo-deposited onto the (100) and (110) facets of BaTaO₂N. As a result, the BaTaO₂N crystals prepared using a KCl flux presented significantly improved water reduction activity compared with those using NaCl. Therefore, crystal facet engineering promoting selective separation and transfer of charges during photoreactions offers an effective strategy for enhancing the photoactivity of perovskite and layered perovskite oxynitrides by adjusting the crystal facets during their synthesis.

Fig. 15 Crystal facet engineering of perovskite and layered perovskite oxynitrides. (A) SEM and TEM images of BaTaO₂N particles prepared via flux-assisted nitridation using (a)–(c) NaCl and (d)–(f) KCl molten salts. (B) The density of states of the different energy levels of the (100) and (110) facets of BaTaO₂N prepared via KCl-assisted nitridation. (C) (a) Photocatalytic H₂ evolution activity using the two different BaTaO₂N particles and (b) cyclic testing using the oxynitride particles prepared via KCl-assisted nitridation. Reproduced with permission.¹³⁹ Copyright 2019, ACS.

5. Conclusions and challenges

We discussed the preparation and use of perovskite AB(O,N)₃ and layered perovskite oxynitrides A_n+1B_n(O,N)_3n+1 as potential semiconductors for visible-light-driven artificial photosynthesis, as shown in Fig. 16. The materials are capable of absorbing visible light wavelengths up to 750 nm at most, and their band structures straddle the various redox potentials of the HER, OER, CER, H₂O₂ production, and CO₂ reduction, among others. Moreover, they are chemically stable in aqueous media over a wide pH range where photoreactions can take place. They fulfill the thermodynamic requirements for various photocatalytic processes to generate renewable, eco-friendly, and/or high value-added products such as H₂ and NH₃ with high theoretical solar-to-fuel energy conversion efficiency. Based on their favorable optical and chemical properties, perovskite and layered perovskite oxynitrides have been employed as both light absorbers and photocatalysts for visible-light-driven water splitting to produce H₂ over several decades. In addition, using them for CO₂ reduction and the CER as an alternative oxidation reaction to the OER to obtain high value-added products such as HCOOH have recently been reported. The adaption of the CER, feasible by using seawater instead of fresh water, not only boosted the oxidation kinetics but also enhanced the H₂ production activity in neutral electrolytes. However, the perovskite and layered perovskite oxynitrides have shown low photoactivity and long-term stability during various photoreactions, which are still far from the requirements for commercialization (corresponding to STH values of above 10%).

Fig. 16 Sunlight-driven catalysis and the challenges of perovskite and layered perovskite oxynitrides to artificially synthesize renewable, high value-added products such as H₂.

Because Ti⁴⁺, Ta⁵⁺, and Nb⁵⁺ in the perovskite and layered perovskite oxynitrides are readily reduced during high-temperature nitridation, the surface and bulk properties of the latter can change due to the presence of defects and/or traces of impurities. Defects in perovskite and layered perovskite oxynitrides boost the recombination of photogenerated holes and electrons during photoreactions, thereby limiting their photoactivity. This phenomenon can be reduced by various synthesis strategies and surface and bulk engineering including doping and substitution. Moreover, photooxidation of the oxynitride surface, which results in the release of N₂ and a positive shift in the surface Fermi level, can be completely suppressed via surface modification to provide separated compartments for HER and OER cocatalysts or passivation layers.

Controlling the surface morphology of the oxynitride particles and/or electrodes can spatially separate the transfer orientation of photogenerated holes and electrons, thus leading to dramatic improvement in the sunlight-driven catalytic activity. Although 1D and 2D perovskite AB(O,N)₃ were rarely studied in terms of the synthesis method, the morphologically modified surfaces of the oxynitrides were found to largely increase the separation efficiency of charges generated during PEC water splitting. It is known that the surface shape of metal oxynitrides such as Ta₃N₅ is relatively easy to be transformed even through high-temperature nitridation. However, it is difficult to prepare AB(O,N)₃ with 1D or 2D surface morphology because two different cations at A- and B-sites are typically sintered at the high temperature probably causing the change in the surface structure. Alternatively, the charges generated during photoreactions can be spatially separated by crystal facet engineering and the employment of layered perovskite oxynitrides. The crystal facet engineering of BaTaO₂N, materialized by flux-assisted nitridation including a recrystallization stage, proved that (100) and (110) facets play roles as electron-attracting and hole-collecting sites, respectively. The layered perovskite oxynitrides with a 2D crystal structure, which allow selective transfer of holes and electrons in crystal lattices intrinsically, can be also utilized for sunlight-driven catalysis. The spatial charge separation in photoelectrodes, as well as on oxynitride particles, is likely to be necessary for efficient, stable photoreactions. In particulate photoelectrodes, the electron transfer from oxynitride particles to a conductive substrate is significantly resistive due to the loose adhesion between them. The bottom-up fabrication to directly grow oxynitride crystals on the conductive substrate can be a solution to obtain a fast electron pathway comparable to the hole transfer rate to surfaces of oxynitrides. Therefore, both the novel synthesis of less-defective oxynitrides to reduce the recombination of photogenerated charges and the surface design of the oxynitrides to make spatially selective separation and transfer of the charges must be challenges to achieving efficient, stable sunlight-driven artificial synthesis using perovskite and layered perovskite oxynitrides.

The overall water splitting of perovskite and layered perovskite oxynitrides by one-step photoexcitation can be the most ideal way to facilitate the low-cost production of H₂ gas on a large scale. The overall water splitting of a ATaO₂N (A = Sr, Ba) particle has been demonstrated; however its activity is still limited. The separation of both HER and OER sites at the oxynitride surface is likely to be necessary to drive water splitting, while HER sites with a cocatalyst were distinguished from the surface of oxides such as SrTiO₃. The selective deposition of the cocatalyst and crystal facet engineering of the oxynitrides are challenging for efficient overall water splitting. In a PEC system, the anodic photoresponse of perovskite and layered perovskite oxynitrides should be below 0 V_RHE based on their flat band potentials (which are negative). However, the actual photocurrent onset is limited to the range of 0.5 to 0.6 V_RHE. Unassisted solar water splitting via PEC cells comprising perovskite and layered perovskite oxynitrides combined with other cathode materials has rarely been reported. Thus, efficient photoanodes must be able to generate significant photocurrent density at a moderate potential (0.4 to 0.6 V_RHE) based on the overlap with the photocathode during overall water splitting. It is necessary to apply various approaches to improve their water-splitting activity by, for instance, optimizing their optical and physical properties. Moreover, a better understanding of the effect of pH (photoanodes are more active at higher pH values while photocathodes work well at lower pH values) must also be achieved to obtain a high rate of H₂ production. Therefore, these challenges are directly related to the photoactivity and long-term stability of these materials for efficient artificial photosynthesis.

Conflicts of interest

The authors have no conflicts of interest relevant to this study to disclose.

Acknowledgements

This research was financially supported by the Basic Science Research Program (no. RS-2023-00243439) and Regional Innovation Strategy (RIS) (no. 2021RIS-002) through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (MOE). This study was also supported by the NRF grant funded by the Korean government (MSIT) (no. 2021M3I3A1084818).

References

1. G. Zhang, G. Liu, L. Wang and J. T. Irvine, Inorganic perovskite photocatalysts for solar energy utilization, Chem. Soc. Rev., 2016, 45, 5951–5984.
2. T. Hisatomi, J. Kubota and K. Domen, Recent advances in semiconductors for photocatalytic and photoelectrochemical water splitting, Chem. Soc. Rev., 2014, 43, 7520–7535.
3. J. Seo, H. Nishiyama, T. Yamada and K. Domen, Visible-Light-Responsive Photoanodes for Highly Active, Stable Water Oxidation, Angew. Chem., Int. Ed., 2018, 57, 8396–8415.
4. H. Fujito, H. Kunioku, D. Kato, H. Suzuki, M. Higashi, H. Kageyama and R. Abe, Layered Perovskite Oxychloride Bi₄NbO₈Cl: A Stable Visible Light Responsive Photocatalyst for Water Splitting, J. Am. Chem. Soc., 2016, 138, 2082–2085.
5. A. Nakada, M. Higashi, T. Kimura, H. Suzuki, D. Kato, H. Okajima, T. Yamamoto, A. Saeki, H. Kageyama and R. Abe, Band Engineering of Double-Layered Sillén–Aurivillius Perovskite Oxychlorides for Visible-Light-Driven Water Splitting, Chem. Mater., 2019, 31, 3419–3429.
6. M. Hojamberdiev, M. F. Bekheet, E. Zahedi, H. Wagata, Y. Kamei, K. Yubuta, A. Gurlo, N. Matsushita, K. Domen and K. Teshima, New Dion–Jacobson Phase Three-Layer Perovskite CsBa₂Ta₃O₁₀ and Its Conversion to Nitrided Ba₂Ta₃O₁₀ Nanosheets via a Nitridation–Protonation–Intercalation–Exfoliation Route for Water Splitting, Cryst. Growth Des., 2016, 16, 2302–2308.
7. Y. Suemoto, Y. Masubuchi, Y. Nagamine, A. Matsutani, T. Shibahara, K. Yamazaki and S. Kikkawa, Intergrowth between the Oxynitride Perovskite SrTaO₂N and the Ruddlesden–Popper Phase Sr₂TaO₃N, Inorg. Chem., 2018, 57, 9086–9095.
8. G. Gou, M. Zhao, J. Shi, J. K. Harada and J. M. Rondinelli, Anion Ordered and Ferroelectric Ruddlesden–Popper Oxynitride Ca₃Nb₂N₂O₅ for Visible-Light-Active Photocatalysis, Chem. Mater., 2020, 32, 2815–2823.
9. Y. Tang, K. Kato, T. Oshima, H. Mogi, A. Miyoshi, K. Fujii, K. I. Yanagisawa, K. Kimoto, A. Yamakata, M. Yashima and K. Maeda, Synthesis of Three-Layer Perovskite Oxynitride K₂Ca₂Ta₃O₉N·2H₂O and Photocatalytic Activity for H₂ Evolution under Visible Light, Inorg. Chem., 2020, 59, 11122–11128.
10. Y. Shiroma, H. Mogi, T. Mashiko, S. Yasuda, S. Nishioka, T. Yokoi, S. Ida, K. Kimoto and K. Maeda, Interlayer modification and single-layer exfoliation of the Ruddlesden–Popper perovskite oxynitride K₂LaTa₂O₆N to improve photocatalytic H₂ evolution activity, J. Mater. Chem. A, 2023, 11, 9485–9492.
11. Y. Ham, T. Hisatomi, Y. Goto, Y. Moriya, Y. Sakata, A. Yamakata, J. Kubota and K. Domen, Flux-mediated doping of SrTiO₃ photocatalysts for efficient overall water splitting, J. Mater. Chem. A, 2016, 4, 3027–3033.
12. T. Takata, J. Jiang, Y. Sakata, M. Nakabayashi, N. Shibata, V. Nandal, K. Seki, T. Hisatomi and K. Domen, Photocatalytic water splitting with a quantum efficiency of almost unity, Nature, 2020, 581, 411–414.
13. Y. Miseki, H. Kato and A. Kudo, Water Splitting into H₂ and O₂ over Ba₅Nb₄O₁₅ Photocatalysts with Layered Perovskite Structure Prepared by Polymerizable Complex Method, Chem. Lett., 2006, 35, 1052–1053.
14. W.-J. Chun, A. Ishikawa, H. Fujisawa, T. Takata, J. N. Kondo, M. Hara, M. Kawai, Y. Matsumoto and K. Domen, Conduction and Valence Band Positions of Ta₂O₅, TaON, and Ta₃N₅ by UPS and Electrochemical Methods, J. Phys. Chem. B, 2003, 107, 1798–1803.
15. S. Ramaraj and J. Seo, Intensive-visible-light-responsive ANbO2N (A = Sr, Ba) synthesized from layered perovskite A₅Nb₄O₁₅ for enhanced photoelectrochemical water splitting, J. Energy Chem., 2022, 68, 529–537.
16. F. Zhang, A. Yamakata, K. Maeda, Y. Moriya, T. Takata, J. Kubota, K. Teshima, S. Oishi and K. Domen, Cobalt-modified porous single-crystalline LaTiO₂N for highly efficient water oxidation under visible light, J. Am. Chem. Soc., 2012, 134, 8348–8351.
17. S. Akiyama, M. Nakabayashi, N. Shibata, T. Minegishi, Y. Asakura, M. Abdulla-Al-Mamun, T. Hisatomi, H. Nishiyama, M. Katayama, T. Yamada and K. Domen, Highly Efficient Water Oxidation Photoanode Made of Surface Modified LaTiO₂ N Particles, Small, 2016, 12, 5468–5476.
18. J. Seo, T. Hisatomi, M. Nakabayashi, N. Shibata, T. Minegishi, M. Katayama and K. Domen, Efficient Solar-Driven Water Oxidation over Perovskite-Type BaNbO₂N Photoanodes Absorbing Visible Light up to 740 nm, Adv. Energy Mater., 2018, 8, 1800094.
19. J. Seo, M. Nakabayashi, T. Hisatomi, N. Shibata, T. Minegishi and K. Domen, Solar-Driven Water Splitting over a BaTaO₂N Photoanode Enhanced by Annealing in Argon, ACS Appl. Energy Mater., 2019, 2, 5777–5784.
20. K. Maeda and K. Domen, New Non-Oxide Photocatalysts Designed for Overall Water Splitting under Visible Light, J. Phys. Chem. C, 2007, 111, 7851–7861.
21. H. W. Eng, P. W. Barnes, B. M. Auer and P. M. Woodward, Investigations of the electronic structure of d₀ transition metal oxides belonging to the perovskite family, J. Solid State Chem., 2003, 175, 94–109.
22. S. Balaz, S. H. Porter, P. M. Woodward and L. J. Brillson, Electronic Structure of Tantalum Oxynitride Perovskite Photocatalysts, Chem. Mater., 2013, 25, 3337–3343.
23. T. W. Kim and K.-S. Choi, Nanoporous BiVO₄ Photoanodes with Dual-Layer Oxygen Evolution Catalysts for Solar Water Splitting, Science, 2014, 343, 990–994.
24. Y. Kuang, Q. Jia, G. Ma, T. Hisatomi, T. Minegishi, H. Nishiyama, M. Nakabayashi, N. Shibata, T. Yamada, A. Kudo and K. Domen, Ultrastable low-bias water splitting photoanodes via photocorrosion inhibition and in situ catalyst regeneration, Nat. Energy, 2017, 2, 16191.
25. P. Sharma, J. W. Jang and J. S. Lee, Key Strategies to Advance the Photoelectrochemical Water Splitting Performance of α-Fe₂O₃ Photoanode, ChemCatChem, 2018, 11, 157–179.
26. T. Oshima, T. Ichibha, K. S. Qin, K. Muraoka, J. J. M. Vequizo, K. Hibino, R. Kuriki, S. Yamashita, K. Hongo, T. Uchiyama, K. Fujii, D. Lu, R. Maezono, A. Yamakata, H. Kato, K. Kimoto, M. Yashima, Y. Uchimoto, M. Kakihana, O. Ishitani, H. Kageyama and K. Maeda, Undoped Layered Perovskite Oxynitride Li₂LaTa₂O₆N for Photocatalytic CO₂ Reduction with Visible Light, Angew. Chem., Int. Ed., 2018, 57, 8154–8158.
27. Q. Wang and K. Domen, Particulate Photocatalysts for Light-Driven Water Splitting: Mechanisms, Challenges, and Design Strategies, Chem. Rev., 2020, 120, 919–985.
28. X. M. C. Ta, R. Daiyan, T. K. A. Nguyen, R. Amal, T. Tran-Phu and A. Tricoli, Alternatives to Water Photooxidation for Photoelectrochemical Solar Energy Conversion and Green H₂ Production, Adv. Energy Mater., 2022, 12, 2201358.
29. V.-H. Trinh and J. Seo, Porous Cuboidal SrNbO₂N Crystals Grown on a Nb Substrate as an Active Photoanode for Neutral Seawater Splitting under Sunlight, ACS Sustainable Chem. Eng., 2023, 11, 1655–1665.
30. R. Shi, G. I. N. Waterhouse and T. Zhang, Recent Progress in Photocatalytic CO₂ Reduction Over Perovskite Oxides, Sol. RRL, 2017, 1, 1700126.
31. K. Maeda and T. E. Mallouk, Two-Dimensional Metal Oxide Nanosheets as Building Blocks for Artificial Photosynthetic Assemblies, Bull. Chem. Soc. Jpn., 2019, 92, 38–54.
32. A. Murphy, P. Barnes, L. Randeniya, I. Plumb, I. Grey, M. Horne and J. Glasscock, Efficiency of solar water splitting using semiconductor electrodes, Int. J. Hydrogen Energy, 2006, 31, 1999–2017.
33. A. G. Tamirat, J. Rick, A. A. Dubale, W.-N. Su and B.-J. Hwang, Using hematite for photoelectrochemical water splitting: a review of current progress and challenges, Nanoscale Horiz., 2016, 1, 243–267.
34. H. Nishiyama, T. Yamada, M. Nakabayashi, Y. Maehara, M. Yamaguchi, Y. Kuromiya, Y. Nagatsuma, H. Tokudome, S. Akiyama, T. Watanabe, R. Narushima, S. Okunaka, N. Shibata, T. Takata, T. Hisatomi and K. Domen, Photocatalytic solar hydrogen production from water on a 100-m(2) scale, Nature, 2021, 598, 304–307.
35. H. Li, J. Xiao, J. J. M. Vequizo, T. Hisatomi, M. Nakabayashi, Z. Pan, N. Shibata, A. Yamakata, T. Takata and K. Domen, One-Step Excitation Overall Water Splitting over a Modified Mg-Doped BaTaO₂N Photocatalyst, ACS Catal., 2022, 12, 10179–10185.
36. K. Chen, J. Xiao, J. J. M. Vequizo, T. Hisatomi, Y. Ma, M. Nakabayashi, T. Takata, A. Yamakata, N. Shibata and K. Domen, Overall Water Splitting by a SrTaO₂N-Based Photocatalyst Decorated with an Ir-Promoted Ru-Based Cocatalyst, J. Am. Chem. Soc., 2023, 145, 3839–3843.
37. X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen and M. Antonietti, A metal-free polymeric photocatalyst for hydrogen production from water under visible light, Nat. Mater., 2009, 8, 76–80.
38. K. Maeda, K. Teramura, D. Lu, T. Takata, N. Saito, Y. Inoue and K. Domen, Photocatalyst releasing hydrogen from water, Nature, 2006, 440, 295.
39. Z. Wang, Y. Inoue, T. Hisatomi, R. Ishikawa, Q. Wang, T. Takata, S. Chen, N. Shibata, Y. Ikuhara and K. Domen, Overall water splitting by Ta₃N₅ nanorod single crystals grown on the edges of KTaO₃ particles, Nat. Catal., 2018, 1, 756–763.
40. Q. Wang, M. Nakabayashi, T. Hisatomi, S. Sun, S. Akiyama, Z. Wang, Z. Pan, X. Xiao, T. Watanabe, T. Yamada, N. Shibata, T. Takata and K. Domen, Oxysulfide photocatalyst for visible-light-driven overall water splitting, Nat. Mater., 2019, 18, 827–832.
41. C. Pan, T. Takata, M. Nakabayashi, T. Matsumoto, N. Shibata, Y. Ikuhara and K. Domen, A complex perovskite-type oxynitride: the first photocatalyst for water splitting operable at up to 600 nm, Angew. Chem., Int. Ed., 2015, 54, 2955–2959.
42. J. Seo, D. Ishizuka, T. Hisatomi, T. Takata and K. Domen, Effect of Mg²⁺ substitution on the photocatalytic water splitting activity of LaMg_xNb_1−xO_1+3xN_2−3x, J. Mater. Chem. A, 2021, 9, 8655–8662.
43. K. Maeda, K. Teramura, D. Lu, N. Saito, Y. Inoue and K. Domen, Roles of Rh/Cr₂O₃ (core/shell) nanoparticles photodeposited on visible light responsive (Ga_1−xZn_x)(N_1−xO_x) solid solutions in photocatalytic overall water splitting, J. Phys. Chem. C, 2007, 111, 7554–7560.
44. M. Yoshida, K. Takanabe, K. Maeda, A. Ishikawa, J. Kubota, Y. Sakata, Y. Ikezawa and K. Domen, Role and Function of Noble-Metal/Cr-Layer Core/Shell Structure Cocatalysts for Photocatalytic Overall Water Splitting Studied by Model Electrodes, J. Phys. Chem. C, 2009, 113, 10151–10157.
45. H. Kato, K. Asakura and A. Kudo, Highly Efficient Water Splitting into H₂ and O₂ over Lanthanum-Doped NaTaO₃ Photocatalysts with High Crystallinity and Surface Nanostructure, J. Am. Chem. Soc., 2003, 125, 3082–3089.
46. C. Pan, T. Takata, K. Kumamoto, S. S. Khine Ma, K. Ueda, T. Minegishi, M. Nakabayashi, T. Matsumoto, N. Shibata, Y. Ikuhara and K. Domen, Band engineering of perovskite-type transition metal oxynitrides for photocatalytic overall water splitting, J. Mater. Chem. A, 2016, 4, 4544–4552.
47. J. H. Kim, D. Hansora, P. Sharma, J. W. Jang and J. S. Lee, Toward practical solar hydrogen production – an artificial photosynthetic leaf-to-farm challenge, Chem. Soc. Rev., 2019, 48, 1908–1971.
48. Y. Zhang, H. Lv, Z. Zhang, L. Wang, X. Wu and H. Xu, Stable Unbiased Photo-Electrochemical Overall Water Splitting Exceeding 3% Efficiency via Covalent Triazine Framework/Metal Oxide Hybrid Photoelectrodes, Adv. Mater., 2021, 33, e2008264.
49. S. Ye, W. Shi, Y. Liu, D. Li, H. Yin, H. Chi, Y. Luo, N. Ta, F. Fan, X. Wang and C. Li, Unassisted Photoelectrochemical Cell with Multimediator Modulation for Solar Water Splitting Exceeding 4% Solar-to-Hydrogen Efficiency, J. Am. Chem. Soc., 2021, 143, 12499–12508.
50. R. Rhee, T. G. Kim, G. Y. Jang, G. Bae, J. H. Lee, S. Lee, S. Kim, S. Jeon and J. H. Park, Unassisted overall water splitting with a solar-to-hydrogen efficiency of over 10% by coupled lead halide perovskite photoelectrodes, Carbon Energy, 2022, 5, 1–10.
51. T. Higashi, Y. Shinohara, A. Ohnishi, J. Liu, K. Ueda, S. Okamura, T. Hisatomi, M. Katayama, H. Nishiyama, T. Yamada, T. Minegishi and K. Domen, Sunlight-Driven Overall Water Splitting by the Combination of Surface-Modified La₅Ti₂Cu_0.9Ag_0.1S₅O₇ and BaTaO₂N Photoelectrodes, ChemPhotoChem, 2016, 1, 1–7.
52. M. Cao, H. Li, K. Liu, J. Hu, H. Pan, J. Fu and M. Liu, Vertical SrNbO₂N Nanorod Arrays for Solar-Driven Photoelectrochemical Water Splitting, Sol. RRL, 2020, 5, 2000448.
53. J. Seo, M. Nakabayashi, T. Hisatomi, N. Shibata, T. Minegishi, M. Katayama and K. Domen, The effects of annealing barium niobium oxynitride in argon on photoelectrochemical water oxidation activity, J. Mater. Chem. A, 2019, 7, 493–502.
54. M. Matsukawa, R. Ishikawa, T. Hisatomi, Y. Moriya, N. Shibata, J. Kubota, Y. Ikuhara and K. Domen, Enhancing photocatalytic activity of LaTiO₂N by removal of surface reconstruction layer, Nano Lett., 2014, 14, 1038–1041.
55. C. R. Lhermitte and K. Sivula, Alternative Oxidation Reactions for Solar-Driven Fuel Production, ACS Catal., 2019, 9, 2007–2017.
56. J. C. Hill and K.-S. Choi, Effect of Electrolytes on the Selectivity and Stability of n-type WO₃ Photoelectrodes for Use in Solar Water Oxidation, J. Phys. Chem. C, 2012, 116, 7612–7620.
57. F. Dionigi, T. Reier, Z. Pawolek, M. Gliech and P. Strasser, Design Criteria, Operating Conditions, and Nickel-Iron Hydroxide Catalyst Materials for Selective Seawater Electrolysis, ChemSusChem, 2016, 9, 962–972.
58. M. Jadwiszczak, K. Jakubow-Piotrowska, P. Kedzierzawski, K. Bienkowski and J. Augustynski, Highly Efficient Sunlight-Driven Seawater Splitting in a Photoelectrochemical Cell with Chlorine Evolved at Nanostructured WO₃ Photoanode and Hydrogen Stored as Hydride within Metallic Cathode, Adv. Energy Mater., 2019, 10, 1903213.
59. S. Iguchi, Y. Miseki and K. Sayama, Efficient hypochlorous acid (HClO) production via photo electrochemical solar energy conversion using a BiVO₄-based photoanode, Sustain, Energy Fuels, 2018, 2, 155–162.
60. R.-T. Gao, X. Guo, S. Liu, X. Zhang, X. Liu, Y. Su and L. Wang, Ultrastable and high-performance seawater-based photoelectrolysis system for solar hydrogen generation, Appl. Catal., B, 2022, 304, 120883.
61. M. Higashi, K. Domen and R. Abe, Fabrication of an efficient BaTaO₂N photoanode harvesting a wide range of visible light for water splitting, J. Am. Chem. Soc., 2013, 135, 10238–10241.
62. K. Ueda, T. Minegishi, J. Clune, M. Nakabayashi, T. Hisatomi, H. Nishiyama, M. Katayama, N. Shibata, J. Kubota, T. Yamada and K. Domen, Photoelectrochemical oxidation of water using BaTaO₂N photoanodes prepared by particle transfer method, J. Am. Chem. Soc., 2015, 137, 2227–2230.
63. K. Takanabe, Photocatalytic Water Splitting: Quantitative Approaches toward Photocatalyst by Design, ACS Catal., 2017, 7, 8006–8022.
64. J. Liu, Y. Zou, B. Jin, K. Zhang and J. H. Park, Hydrogen Peroxide Production from Solar Water Oxidation, ACS Energy Lett., 2019, 4, 3018–3027.
65. H. Hou, X. Zeng and X. Zhang, Production of Hydrogen Peroxide by Photocatalytic Processes, Angew. Chem., Int. Ed., 2020, 59, 17356–17376.
66. K. Mase, M. Yoneda, Y. Yamada and S. Fukuzumi, Seawater usable for production and consumption of hydrogen peroxide as a solar fuel, Nat. Commun., 2016, 7, 11470.
67. Y. Miyase, S. Takasugi, S. Iguchi, Y. Miseki, T. Gunji, K. Sasaki, E. Fujita and K. Sayama, Modification of BiVO₄/WO₃ composite photoelectrodes with Al₂O₃via chemical vapor deposition for highly efficient oxidative H₂O₂ production from H₂O, Sustainable Energy Fuels, 2018, 2, 1621–1629.
68. Y. Miseki and K. Sayama, Photocatalytic Water Splitting for Solar Hydrogen Production Using the Carbonate Effect and the Z-Scheme Reaction, Adv. Energy Mater., 2018, 9, 1801294.
69. J. H. Baek, T. M. Gill, H. Abroshan, S. Park, X. Shi, J. Nørskov, H. S. Jung, S. Siahrostami and X. Zheng, Selective and Efficient Gd-Doped BiVO₄ Photoanode for Two-Electron Water Oxidation to H₂O₂, ACS Energy Lett., 2019, 4, 720–728.
70. R. Saito, Y. Miseki and K. Sayama, Highly efficient photoelectrochemical water splitting using a thin film photoanode of BiVO₄/SnO₂/WO₃ multi-composite in a carbonate electrolyte, Chem. Commun., 2012, 48, 3833.
71. S. N. Habisreutinger, L. Schmidt-Mende and J. K. Stolarczyk, Photocatalytic reduction of CO₂ on TiO₂ and other semiconductors, Angew. Chem., Int. Ed., 2013, 52, 7372–7408.
72. J. Ran, M. Jaroniec and S. Z. Qiao, Cocatalysts in Semiconductor-based Photocatalytic CO(2) Reduction: Achievements, Challenges, and Opportunities, Adv. Mater., 2018, 30, 1704649.
73. E. Kalamaras, M. M. Maroto-Valer, M. Shao, J. Xuan and H. Wang, Solar carbon fuel via photoelectrochemistry, Catal. Today, 2018, 317, 56–75.
74. H. Kumagai, Y. Tamaki and O. Ishitani, Photocatalytic Systems for CO₂ Reduction: Metal-Complex Photocatalysts and Their Hybrids with Photofunctional Solid Materials, Acc. Chem. Res., 2022, 55, 978–990.
75. H. Kumagai, G. Sahara, K. Maeda, M. Higashi, R. Abe and O. Ishitani, Hybrid photocathode consisting of a CuGaO₂ p-type semiconductor and a Ru(II)–Re(I) supramolecular photocatalyst: non-biased visible-light-driven CO₂ reduction with water oxidation, Chem. Sci., 2017, 8, 4242–4249.
76. F. Yoshitomi, K. Sekizawa, K. Maeda and O. Ishitani, Selective Formic Acid Production via CO₂ Reduction with Visible Light Using a Hybrid of a Perovskite Tantalum Oxynitride and a Binuclear Ruthenium(II) Complex, ACS Appl. Mater. Interfaces, 2015, 7, 13092–13097.
77. T. Oshima, T. Ichibha, K. Oqmhula, K. Hibino, H. Mogi, S. Yamashita, K. Fujii, Y. Miseki, K. Hongo, D. Lu, R. Maezono, K. Sayama, M. Yashima, K. Kimoto, H. Kato, M. Kakihana, H. Kageyama and K. Maeda, Two-Dimensional Perovskite Oxynitride K₂LaTa₂O₆N with an H(+)/K(+) Exchangeability in Aqueous Solution Forming a Stable Photocatalyst for Visible-Light H₂ Evolution, Angew. Chem., Int. Ed., 2020, 59, 9736–9743.
78. H. Zhou, Y. Luo, J. Xu, L. Deng, Z. Wang and H. He, Bicomponent Cocatalyst Decoration on Flux-assisted CaTaO₂N Single Crystals for Photocatalytic CO₂ Reduction under Visible Light, Chem. – Eur. J., 2022, 28, e202202044.
79. P. Lin, L. Xiao, Y. Lu, Z. Zhang, X. Wang and W. Su, Anchoring 0D CeO₂ Nanoparticles on 2D LaTiO₂N Nanosheets for Efficient Visible-Light-Driven Photocatalytic CO₂ Reduction, ChemCatChem, 2022, 14, e202201041.
80. L. Lin, P. Lin, J. Song, Z. Zhang, X. Wang and W. Su, Boosting the photocatalytic activity and stability of Cu₂O for CO₂ conversion by LaTiO₂N, J. Colloid Interface Sci., 2023, 630, 352–362.
81. H. Urabe, T. Hisatomi, T. Minegishi, J. Kubota and K. Domen, Photoelectrochemical properties of SrNbO₂N photoanodes for water oxidation fabricated by the particle transfer method, Faraday Discuss., 2014, 176, 213–223.
82. J. Xu, C. Pan, T. Takata and K. Domen, Photocatalytic overall water splitting on the perovskite-type transition metal oxynitride CaTaO₂N under visible light irradiation, Chem. Commun., 2015, 51, 7191–7194.
83. S. Wei and X. Xu, Boosting photocatalytic water oxidation reactions over strontium tantalum oxynitride by structural laminations, Appl. Catal., B, 2018, 228, 10–18.
84. S. Nishimae, Y. Mishima, H. Nishiyama, Y. Sasaki, M. Nakabayashi, Y. Inoue, M. Katayama and K. Domen, Fabrication of BaTaO₂N Thin Films by Interfacial Reactions of BaCO₃/Ta₃N₅ Layers on a Ta Substrate and Resulting High Photoanode Efficiencies During Water Splitting, Sol. RRL, 2020, 4, 1900542.
85. Y. Xiang, B. Zhang, J. Liu, S. Chen, T. Hisatomi, K. Domen and G. Ma, A one-step synthesis of a Ta₃N₅ nanorod photoanode from Ta plates and NH₄Cl powder for photoelectrochemical water oxidation, Chem. Commun., 2020, 56, 11843–11846.
86. B. Siritanaratkul, K. Maeda, T. Hisatomi and K. Domen, Synthesis and photocatalytic activity of perovskite niobium oxynitrides with wide visible-light absorption bands, ChemSusChem, 2011, 4, 74–78.
87. K. Maeda and K. Domen, Water oxidation using a particulate BaZrO₃–BaTaO₂N solid-solution photocatalyst that operates under a wide range of visible light, Angew. Chem., Int. Ed., 2012, 51, 9865–9869.
88. T. T. T. Tran, S. Kim and J. Seo, Size dependence of perovskite-type BaNbO₂N particles on sunlight-driven photoelectrochemical water splitting, J. Catal., 2022, 406, 157–164.
89. T. Oshima, T. Ichibha, K. Oqmhula, K. Hibino, H. Mogi, S. Yamashita, K. Fujii, Y. Miseki, K. Hongo, D. Lu, R. Maezono, K. Sayama, M. Yashima, K. Kimoto, H. Kato, M. Kakihana, H. Kageyama and K. Maeda, Two-Dimensional Perovskite Oxynitride K₂LaTa₂O₆N with an H⁺/K⁺ Exchangeability in Aqueous Solution Forming a Stable Photocatalyst for Visible-Light H₂ Evolution, Angew. Chem., Int. Ed., 2020, 59, 9736–9743.
90. J. Seo, Y. Moriya, M. Kodera, T. Hisatomi, T. Minegishi, M. Katayama and K. Domen, Photoelectrochemical Water Splitting on Particulate ANbO₂N (A = Ba, Sr) Photoanodes Prepared from Perovskite-Type ANbO₃, Chem. Mater., 2016, 28, 6869–6876.
91. T. Hisatomi, C. Katayama, Y. Moriya, T. Minegishi, M. Katayama, H. Nishiyama, T. Yamada and K. Domen, Photocatalytic oxygen evolution using BaNbO₂N modified with cobalt oxide under photoexcitation up to 740 nm, Energy Environ. Sci., 2013, 6, 3595.
92. M. Hojamberdiev, K. Yubuta, J. J. M. Vequizo, A. Yamakata, S. Oishi, K. Domen and K. Teshima, NH₃-Assisted Flux Growth of Cube-like BaTaO₂N Submicron Crystals in a Completely Ionized Nonaqueous High-Temperature Solution and Their Water Splitting Activity, Cryst. Growth Des., 2015, 15, 4663–4671.
93. M. Kodera, H. Urabe, M. Katayama, T. Hisatomi, T. Minegishi and K. Domen, Effects of flux synthesis on SrNbO₂N particles for photoelectrochemical water splitting, J. Mater. Chem. A, 2016, 4, 7658–7664.
94. J. Seo, Size Control of LaNbON₂ Particles for Enhanced Photocatalytic Water Oxidation Under Visible Light Irradiation, Bull. Korean Chem. Soc., 2021, 42, 571–576.
95. J. Seo, S. Jeong and S. Kim, Visible-Light-Driven Water Splitting over Particulate LaNbON₂ Prepared from La-Rich Lanthanum Niobium Oxides, ACS Appl. Energy Mater., 2021, 4, 3141–3150.
96. T. Hisatomi, C. Katayama, K. Teramura, T. Takata, Y. Moriya, T. Minegishi, M. Katayama, H. Nishiyama, T. Yamada and K. Domen, The effects of preparation conditions for a BaNbO₂N photocatalyst on its physical properties, ChemSusChem, 2014, 7, 2016–2021.
97. X. Xu, C. Randorn, P. Efstathiou and J. T. Irvine, A red metallic oxide photocatalyst, Nat. Mater., 2012, 11, 595–598.
98. D. Lu, G. Hitoki, E. Katou, J. N. Kondo, M. Hara and K. Domen, Porous Single-Crystalline TaON and Ta₃N₅ Particles, Chem. Mater., 2004, 16, 1603–1605.
99. S. Jadhav, S. Hasegawa, T. Hisatomi, Z. Wang, J. Seo, T. Higashi, M. Katayama, T. Minegishi, T. Takata, J. M. Peralta-Hernández, O. Serrano Torres and K. Domen, Efficient photocatalytic oxygen evolution using BaTaO₂N obtained from nitridation of perovskite-type oxide, J. Mater. Chem. A, 2020, 8, 1127–1130.
100. D. E. Bugaris and H. C. zur Loye, Materials discovery by flux crystal growth: quaternary and higher order oxides, Angew. Chem., Int. Ed., 2012, 51, 3780–3811.
101. T. Yamada, Y. Murata, H. Wagata, K. Yubuta and K. Teshima, Facile Morphological Modification of Ba₅Nb₄O₁₅ Crystals Using Chloride Flux and in Situ Growth Investigation, Cryst. Growth Des., 2016, 16, 3954–3960.
102. L. Zhang, Y. Song, J. Feng, T. Fang, Y. Zhong, Z. Li and Z. Zou, Photoelectrochemical water oxidation of LaTaON₂ under visible-light irradiation, Int. J. Hydrogen Energy, 2014, 39, 7697–7704.
103. Y. Luo, Z. Wang, S. Suzuki, K. Yubuta, N. Kariya, T. Hisatomi, K. Domen and K. Teshima, Fabrication of Single-Crystalline BaTaO₂N from Chloride Fluxes for Photocatalytic H₂ Evolution under Visible Light, Cryst. Growth Des., 2019, 20, 255–261.
104. T. Minegishi, N. Nishimura, J. Kubota and K. Domen, Photoelectrochemical properties of LaTiO₂N electrodes prepared by particle transfer for sunlight-driven water splitting, Chem. Sci., 2013, 4, 1120–1124.
105. T. T. T. Tran, V.-H. Trinh and J. Seo, Two-dimensional perovskite SrNbO₂N with Zr doping for accelerating photoelectrochemical water splitting, J. Mater. Sci. Technol., 2023, 142, 176–184.
106. M. V. Kelso, N. K. Mahenderkar, Q. Chen, J. Z. Tubbesing and J. A. Switzer, Spin coating epitaxial films, Science, 2019, 364, 166–169.
107. S. G. Ebbinghaus, H.-P. Abicht, R. Dronskowski, T. Müller, A. Reller and A. Weidenkaff, Perovskite-related oxynitrides – Recent developments in synthesis, characterisation and investigations of physical properties, Prog. Solid State Chem., 2009, 37, 173–205.
108. M. D. Tyona, A theoritical study on spin coating technique, Adv. Mater. Res., 2013, 2, 195–208.
109. H. Takeuchi, D. Miyamoto, Y. Masubuchi, M. Higuchi, K. Ishii and T. Uchikoshi, Preparation of thick dense ceramic films of dielectric BaTaO₂N through electrophoretic deposition, Ceram. Int., 2023, 49, 21825–21829.
110. R. Abe, T. Takata, H. Sugihara and K. Domen, The Use of TiCl₄ Treatment to Enhance the Photocurrent in a TaON Photoelectrode under Visible Light Irradiation, Chem. Lett., 2005, 34, 1162–1163.
111. M. Pichler, W. Si, F. Haydous, H. Téllez, J. Druce, E. Fabbri, M. E. Kazzi, M. Döbeli, S. Ninova, U. Aschauer, A. Wokaun, D. Pergolesi and T. Lippert, LaTiO_xN_y Thin Film Model Systems for Photocatalytic Water Splitting: Physicochemical Evolution of the Solid-Liquid Interface and the Role of the Crystallographic Orientation, Adv. Funct. Mater., 2017, 27, 1605690.
112. K. Kawashima, M. Hojamberdiev, O. Mabayoje, B. R. Wygant, K. Yubuta, C. B. Mullins, K. Domen and K. Teshima, NH₃-assisted chloride flux-coating method for direct fabrication of visible-light-responsive SrNbO₂N crystal layers, CrystEngComm, 2017, 19, 5532–5541.
113. X. Luo, Y. Xiao, B. Zhang, C. Feng, Z. Fan and Y. Li, Direct synthesis of BaTaO₂N nanoparticle film on a conductive substrate for photoelectrochemical water splitting, J. Catal., 2022, 411, 109–115.
114. X. Xu, W. Wang, Y. Zhang, Y. Chen, H. Huang, T. Fang, Y. Li, Z. Li and Z. Zou, Centimeter-scale perovskite SrTaO₂N single crystals with enhanced photoelectrochemical performance, Sci. Bull., 2022, 67, 1458–1466.
115. Y.-I. Kim, P. M. Woodward, K. Z. Baba-Kishi and C. W. Tai, Characterization of the Structural, Optical, and Dielectric Properties of Oxynitride Perovskites AMO₂N(A = Ba, Sr, Ca; M = Ta, Nb), Chem. Mater., 2004, 16, 1267–1276.
116. T. Iwai, A. Nakada, M. Higashi, H. Suzuki, O. Tomita and R. Abe, Controlling the carrier density in niobium oxynitride BaNbO₂N via cation doping for efficient photoelectrochemical water splitting under visible light, Sustainable Energy Fuels, 2021, 5, 6181–6188.
117. M. Hojamberdiev, R. Vargas, Z. C. Kadirova, K. Kato, H. Sena, A. G. Krasnov, A. Yamakata, K. Teshima and M. Lerch, Unfolding the Role of B Site-Selective Doping of Aliovalent Cations on Enhancing Sacrificial Visible Light-Induced Photocatalytic H₂ and O₂ Evolution over BaTaO₂N, ACS Catal., 2022, 12, 1403–1414.
118. J. Seo, T. Takata, M. Nakabayashi, T. Hisatomi, N. Shibata, T. Minegishi and K. Domen, Mg–Zr Cosubstituted Ta₃N₅ Photoanode for Lower-Onset-Potential Solar-Driven Photoelectrochemical Water Splitting, J. Am. Chem. Soc., 2015, 137, 12780–12783.
119. Y. Bao, H. Zou, N. Yang, G. Li and F. Zhang, Synthesis of perovskite BaTaO₂N with low defect by Zn doping for boosted photocatalytic water reduction, J. Energy Chem., 2021, 63, 358–363.
120. Y. Wang, S. Jin, G. Pan, Z. Li, L. Chen, G. Liu and X. Xu, Zr doped mesoporous LaTaON₂ for efficient photocatalytic water splitting, J. Mater. Chem. A, 2019, 7, 5702–5711.
121. G. Lin, X. Sun and X. Xu, Mg modified LaTiO₂N with ameliorated photocarrier separation for solar water splitting, Appl. Catal., B, 2023, 324, 122258.
122. X. Sun, Y. Mi, F. Jiao and X. Xu, Activating Layered Perovskite Compound Sr₂TiO₄via La/N Codoping for Visible Light Photocatalytic Water Splitting, ACS Catal., 2018, 8, 3209–3221.
123. X. Wang, T. Hisatomi, J. Liang, Z. Wang, Y. Xiang, Y. Zhao, X. Dai, T. Takata and K. Domen, Facet engineering of LaNbON₂ transformed from LaKNaNbO₅ for enhanced photocatalytic O₂ evolution, J. Mater. Chem. A, 2020, 8, 11743–11751.
124. M. Shao, A. Peles and K. Shoemaker, Electrocatalysis on platinum nanoparticles: particle size effect on oxygen reduction reaction activity, Nano Lett., 2011, 11, 3714–3719.
125. E. J. Coleman, M. H. Chowdhury and A. C. Co, Insights Into the Oxygen Reduction Reaction Activity of Pt/C and PtCu/C Catalysts, ACS Catal., 2015, 5, 1245–1253.
126. J.-W. Park and J. Seo, Ultrafine TaOx/CB Oxygen Reduction Electrocatalyst Operating in Both Acidic and Alkaline Media, Catalysts, 2021, 12, 35.
127. J. Seo, W.-J. Moon, W.-G. Jung and J.-W. Park, A N-doped NbOx nanoparticle electrocatalyst deposited on carbon black for oxygen reduction and evolution reactions in alkaline media, Mater. Adv., 2022, 3, 5315–5324.
128. J. Liu, T. Hisatomi, M. Katayama, T. Minegishi, J. Kubota and K. Domen, Effect of particle size of La₅Ti₂CuS₅O₇ on photoelectrochemical properties in solar hydrogen evolution, J. Mater. Chem. A, 2016, 4, 4848–4854.
129. T. Yamada, Y. Murata, S. Suzuki, H. Wagata, S. Oishi and K. Teshima, Template-Assisted Size Control of Polycrystalline BaNbO₂N Particles and Effects of Their Characteristics on Photocatalytic Water Oxidation Performances, J. Phys. Chem. C, 2018, 122, 8037–8044.
130. H. Tsai, W. Nie, J. C. Blancon, C. C. Stoumpos, R. Asadpour, B. Harutyunyan, A. J. Neukirch, R. Verduzco, J. J. Crochet, S. Tretiak, L. Pedesseau, J. Even, M. A. Alam, G. Gupta, J. Lou, P. M. Ajayan, M. J. Bedzyk and M. G. Kanatzidis, High-efficiency two-dimensional Ruddlesden–Popper perovskite solar cells, Nature, 2016, 536, 312–316.
131. X. Wang, T. Hisatomi, Z. Wang, J. Song, J. Qu, T. Takata and K. Domen, Core-Shell-Structured LaTaON₂ Transformed from LaKNaTaO₅ Plates for Enhanced Photocatalytic H₂ Evolution, Angew. Chem., Int. Ed., 2019, 58, 10666–10670.
132. C. Lan, Z. Zhou, R. Wei and J. C. Ho, Two-dimensional perovskite materials: From synthesis to energy-related applications, Mater. Today Energy, 2019, 11, 61–82.
133. C. Xu, Y. Wang, Q. Guo and X. Wang, Defect engineering of two-dimensional Nb-based oxynitrides for visible-light-driven water splitting to produce H₂ and O₂, Nanoscale Adv., 2023, 5, 3260–3266.
134. K. Maeda, M. Eguchi, W. J. Youngblood and T. E. Mallouk, Calcium Niobate Nanosheets Prepared by the Polymerized Complex Method as Catalytic Materials for Photochemical Hydrogen Evolution, Chem. Mater., 2009, 21, 3611–3617.
135. Y. L. Cen, J. J. Shi, M. Zhang, M. Wu, J. Du, W. H. Guo and Y. H. Zhu, Optimized band gap and fast interlayer charge transfer in two-dimensional perovskite oxynitride Ba₂NbO₃N and Sr₂NbO₃/Ba₂NbO₃N bonded heterostructure visible-light photocatalysts for overall water splitting, J. Colloid Interface Sci., 2019, 546, 20–31.
136. S. Wang, G. Liu and L. Wang, Crystal Facet Engineering of Photoelectrodes for Photoelectrochemical Water Splitting, Chem. Rev., 2019, 119, 5192–5247.
137. R. Li, F. Zhang, D. Wang, J. Yang, M. Li, J. Zhu, X. Zhou, H. Han and C. Li, Spatial separation of photogenerated electrons and holes among (010) and (110) crystal facets of BiVO₄, Nat. Commun., 2013, 4, 1432.
138. Z. Pan, V. Nandal, Y. Pihosh, T. Higashi, T. Liu, J. A. Röhr, K. Seki, C. Chu, K. Domen and K. Katayama, Elucidating the Role of Surface Energetics on Charge Separation during Photoelectrochemical Water Splitting, ACS Catal., 2022, 12, 14727–14734.
139. Y. Luo, S. Suzuki, Z. Wang, K. Yubuta, J. J. M. Vequizo, A. Yamakata, H. Shiiba, T. Hisatomi, K. Domen and K. Teshima, Construction of Spatial Charge Separation Facets on BaTaO₂N Crystals by Flux Growth Approach for Visible-Light-Driven H₂ Production, ACS Appl. Mater. Interfaces, 2019, 11, 22264–22271.

---