A Sillén oxyhalide SrBi₃O₄Cl₃ as a promising photocatalyst for water splitting: impact of the asymmetric structure on light absorption and charge carrier dynamics

Abstract

Bismuth-based oxyhalides with layered Sillén(–Aurivillius) structures have attracted significant attention as photocatalysts. Recent studies have unveiled a part of the structure–property relationship of the materials; however, it has not been fully understood. In the present study, we investigated a Sillén-type oxyhalide SrBi₃O₄Cl₃ with single and double halogen layers. Interestingly, SrBi₃O₄Cl₃ showed a visible light response up to ∼460 nm, whereas SrBiO₂Cl and BiOCl with single and double halogen layers, respectively, did not. Rietveld refinement and STEM-EDX mapping determined the asymmetric Bi occupation in the fluorite [Sr_0.5Bi_1.5O₂] layer of SrBi₃O₄Cl₃, which was derived from the coexistence of the halogen layers. DFT calculations and Madelung potential calculations showed that the asymmetric Bi occupation affords both the Bi–Bi interaction across the single halogen layer and the electrostatic destabilization of Cl in the double halogen layer, probably leading to the narrow bandgap of SrBi₃O₄Cl₃. Another merit of possessing the two different halogen layers was revealed by time-resolved microwave conductivity measurements as well as DFT calculations; the spatial separation of the conduction band minimum and valence band maximum based on the coexistence of the halogen layers would promote charge carrier separation. Visible-light-driven Z-scheme water splitting was accomplished using a RuO₂-loaded SrBi₃O₄Cl₃ sample as an O₂-evolving photocatalyst. This study provides another option for engineering band structures and promoting the charge carrier separation of layered oxyhalides for efficient water splitting under visible light.

---

Introduction

Sillén-type oxyhalides possess a layered structure consisting of intergrowth of alternating fluorite [M₂O₂] and halogen [X] layers. Since the first report by Sillén,¹ various compounds that belong to Sillén phases have been reported^2–7 and studied in terms of battery,^8–13 catalysis,^14–18 optical properties,^19–22 photovoltaics,^23–25 photocatalysis,^13–17etc.^18,19 Among them, oxychloride BiOCl with fluorite [Bi₂O₂] and double halogen [Cl] layers is one of the most actively researched materials, along with other analogous BiOBr and BiOI, particularly as a promising photocatalyst for dye degradation,^26–28 selective oxidation reactions,^29–32 and solar energy conversions.^33–36 Although the activity of the BiOCl photocatalyst has been improved via various strategies, such as forming heterojunctions^37–39 and engineering defects and facets,^26,27,33 the wide bandgap (∼3.5 eV) has restricted its applications under natural sunlight. In addition, it has intrinsic stability problems when employed in harsh reactions, such as water splitting (i.e., water oxidation). Because its valence band maximum (VBM) predominantly consists of Cl-3p orbitals rather than O-2p orbitals, photogenerated holes tend to accumulate and oxidize Cl anions, resulting in irreversible and oxidative deactivation.⁴⁰

In contrast, we have demonstrated that some Sillén oxychlorides, including Pb cations, such as PbBiO₂Cl and PbBi₃O₄Cl₃, not only possess narrow bandgaps for visible-light-absorption but also function as stable photocatalysts for water splitting.^41–43 For example, PbBiO₂Cl with fluorite [PbBiO₂] and single halogen [Cl] layers absorbs visible light up to 500 nm, and can oxidize water to O₂ under visible light without experiencing oxidative deactivation.^41,44,45 Density functional theory (DFT) calculations confirmed that the strong interaction between O-2p and Pb-6s orbitals, with the help of empty Pb-6p, provides an up-shifted VBM consisting of O-2p orbitals dominantly, instead of Cl-3p orbitals, thus endowing PbBiO₂Cl with both a narrow bandgap (∼2.5 eV) and high durability against self-oxidative deactivation. In stark contrast, the analogous Sr-counterpart SrBiO₂Cl possesses a much larger bandgap (∼3.5 eV). This implies that the presence of Pb cations is vital for the visible-light absorption of such Sillén-type oxychlorides.

Contrary to this expectation, we herein show that Pb-free Sillén oxychloride SrBi₃O₄Cl₃ possesses a narrow bandgap (∼2.7 eV) and functions as a stable and efficient photocatalyst for water oxidation under visible light. Although this material was first reported by Sillén in 1940,⁴⁶ where it has been studied in terms of crystal structure and catalysis,^16,17 there has been no study focusing on photofunctionality, including photocatalysis. The combination of various characterization studies and theoretical approaches in the present study revealed that both the Bi–Bi interactions across the single halogen layer and the electrostatic destabilization of Cl in the double halogen layer arise from the asymmetric Bi occupation in the fluorite [Sr_0.5Bi_1.5O₂] layer, as well as the coexistence of the halogen layers. These two factors most probably afford a much narrower bandgap than analogous oxychlorides, such as BiOCl and SrBiO₂Cl. Furthermore, it was also suggested that the coexistence of the halogen layers in SrBi₃O₄Cl₃ promoted spatial charge carrier separation.

Experimental

Materials

Bismuth oxychloride (BiOCl, 95.0%), hydrochloric acid (HCl, 1.0 M), hydrogen hexachloroplatinate hexahydrate (H₂PtCl₆·6H₂O, 99.9%), iron(III) chloride hexahydrate (FeCl₃·6H₂O, 99%), iron(III) nitrate enneahydrate (Fe(NO₃)₃·9H₂O, 99%), iron(III) perchlorate n-hydrate (Fe(ClO₄)₃, 70% as anhydrous), iron(III) sulfate n-hydrate (Fe₂(SO₄)₃, 60–80% as anhydrous), methanol (CH₃OH, 99.8%), nitric acid (HNO₃, 1 M), perchloric acid (HClO₄, 70% as assay), rhodium oxide (Rh₂O₃, 98.0%), ruthenium chloride n-hydrate (RuCl₃·nH₂O, 99.9%), silver nitrate (AgNO₃, 99.8%), sodium sulfate (Na₂SO₄, 99.0%), strontium carbonate (SrCO₃, 99.99%), sulfuric acid (H₂SO₄, 0.5 M), and titanium oxide (TiO₂, 99.0%) were purchased from FUJIFILM Wako Chemicals. All chemicals were used as received, except for SrCO₃ and TiO₂ which were dried at 300 °C for 2 h before use.

Sample synthesis

All samples were synthesized by solid-state reactions. SrBi₃O₄Cl₃ was synthesized from a stoichiometric mixture of BiOCl and SrBiO₂Cl. First, SrBiO₂Cl was synthesized from a stoichiometric mixture of SrCO₃ and BiOCl. The mixture was loaded into an alumina crucible and calcined at 700 °C for 12 h in air. SrBi₃O₄Cl₃ can be prepared by heating a stoichiometric mixture of SrBiO₂Cl and BiOCl in an alumina crucible at 600–800 °C for 12 h in air (Fig. S1†). In this study, the SrBi₃O₄Cl₃ sample prepared at 700 °C was used. As seen from scanning electron microscopy (SEM) images, the sample has aggregated particles with a plate-like shape with a few hundred nanometers in size (Fig. S2†). The RuO₂ cocatalyst was deposited on SrBi₃O₄Cl₃ by an impregnation method. SrBi₃O₄Cl₃ was dispersed in a small amount of RuCl₃ aqueous solution (0.7 wt% Ru metal) and calcined at 500 °C for 1 h in air.

As an H₂-evolving photocatalyst in Z-scheme water splitting, rhodium-doped strontium titanate (SrTiO₃ : Rh) was synthesized by a solid-state reaction.⁴⁷ A mixture of TiO₂, SrCO₃, and Rh₂O₃, with a molar ratio of Ti : Sr : Rh = 1 : 1.07 of 0.01, was calcined in an alumina crucible at 600 °C for 2 h and subsequently at 1000 °C for 10 h with an intermediate ground. A Ru-based cocatalyst was deposited on SrTiO₃ : Rh by a photodeposition method.⁴⁸ SrTiO₃ : Rh was suspended in a methanol aqueous solution (10 vol%, 120 mL) containing RuCl₃ (0.7 wt% Ru metal) and irradiated by visible light (400 < λ < 800 nm) for 2 h.

Characterization

Powder X-ray diffraction (XRD; MiniFlex II, Rigaku, Cu Kα), SEM (NVision 40, Carl Zeiss-SIINT), scanning transmission electron microscopy (STEM; JEOL) with energy-dispersive X-ray spectroscopy (EDX), UV-vis diffuse reflectance spectroscopy (V-650, Jasco), and X-ray photoelectron spectroscopy (XPS; ESCA 5500MT, ULVAC-PHI, Mg Kα) were used to characterize the samples. The XPS binding energies were corrected with respect to the Au 4f peak (84.0 eV). Synchrotron X-ray diffraction (SXRD; BL02B2, SPring-8, Japan, λ = 0.419432 Å) patterns were collected at room temperature. Indexing of the SXRD patterns and Rietveld refinement were conducted using EXPO2014 and RIETAN-FP programs.^49,50 The VESTA program was used for drawing crystal structures.⁵¹ The maximum entropy method (MEM) analysis was conducted using the Dysnomia program.⁵² Time-resolved microwave conductivity (TRMC) measurements were performed using the fourth harmonic generation (FHG at 266 nm, I₀ = 1.4 × 10¹⁵ photons per cm² per pulse) from an Nd:YAG laser (Spectra Physics, GCR-200, 5–8 ns pulse duration, 10 Hz) as the excitation source. The resonant frequency and microwave power were set to ∼9.1 GHz (X-band) and 3 mW.

Electrochemical measurement

Mott–Schottky plots were recorded in a three-electrode cell equipped with a Pt wire counter-electrode and Ag/AgCl reference electrode in a phosphate-buffered solution (0.1 M, pH 6.0) using an electrochemical analyzer (VersaSTAT 4, Princeton Applied Research), with an amplitude of 10 mV and a frequency of 1000 Hz. The electrode was prepared using the squeegee method. A particulate sample with a small amount of water was coated on a fluorine-doped tin oxide (FTO) conductive substrate and dried overnight at room temperature.

Calculation

The partial density of states (PDOS) of SrBi₃O₄Cl₃ was calculated using the Cambridge Serial Total Energy Package (CASTEP).⁵³ The energy was calculated using the generalized gradient approximation (GGA) of DFT proposed by Perdew, Brurke, and Ernzerhof (PBE). The electronic states were expanded using a plane-wave basis set with a cut-off of 700 eV. The k-point meshes were set as 3 × 3 × 1. Before the PDOS calculations, geometry optimization was performed using the Broyden–Fletcher–Goldfarb–Shanno (BFGS) algorithm. The crystal orbital Hamilton populations (COHP) were calculated using the Quantum Espresso package⁵⁴ and the Local-Orbital Basis Suite Toward Electronic-Structure Reconstruction (LOBSTER) package.⁵⁵ The Perdew–Burke–Ernzerhof functional for solids (PBEsol) generalized gradient approximation (GGA) was used as the exchange-correlation functional. A cutoff energy of 100 Ry (∼1361 eV) and 6 × 6 × 2 k-points were used. We confirmed that the calculated PDOS (Fig. S3†) was similar to that obtained using the calculation described above using the CASTEP (Fig. 5).

Photocatalytic reactions

Photocatalytic reactions were performed in an overhead irradiation-type Pyrex glass vessel connected to a closed gas circulation system. The system was degassed and purged with Ar gas before each photocatalytic reaction. A Xe lamp (LAMP HOUSE R300-3J, Eagle engineering, 300 W) equipped with a cold mirror (CM-1, Kenko) and a cut-off filter (L-42, HOYA) was used as a visible-light (400 < λ < 800 nm) source. As for UV-vis light (300 < λ < 500 nm) irradiation, the Xe lamp equipped with only a cold mirror (CM-2, Kenko) was used. The evolved gases were analyzed using online gas chromatography (GC-8A, Shimadzu, thermal conductivity detector, 5 Å molecular sieve column packing) that uses Ar carrier gas. For O₂ evolution reactions, the photocatalyst (100 mg) was suspended in an aqueous solution of FeCl₃, Fe(NO₃)₃, Fe₂(SO₄)₃ or Fe(ClO₄)₃ (5 mM Fe³⁺, 120 mL, pH 2.4, adjusted by using HCl, HNO₃, H₂SO₄ or HClO₄) and irradiated with visible light. For the H₂ evolution reactions, the photocatalyst (100 mg) was suspended in an aqueous methanol solution (10 vol%, 120 mL) and irradiated with UV-vis light. A Pt cocatalyst (1 wt%) was loaded onto the photocatalyst by an in situ photodeposition method using H₂PtCl₆ as a precursor. Z-scheme overall water splitting, employing RuO₂/SrBi₃O₄Cl₃ and Ru/SrTiO₃ : Rh as the O₂-evolving and H₂-evolving photocatalysts, respectively, was conducted. The two photocatalysts (50 mg each) were suspended in an aqueous solution of FeCl₃ or Fe₂(SO₄)₃ (2 mM Fe³⁺, 120 mL, pH 2.4, adjusted by HCl or H₂SO₄) and irradiated with visible light. The apparent quantum efficiencies (AQE) for Z-scheme water splitting were measured using a similar experimental setting. A Xe lamp equipped with both a CM-1 cold mirror and bandpass filter (λ = 420 ± 5 nm, ASAHI) was used as the light source. The AQE was calculated using the following equation:

AQE (%) = 100 × A × R/I

where A is the number of photons required to generate one molecule of the product (i.e., eight for O₂ and four for H₂ evolution), R is the rate of gas evolution, and I is the number of incident photons (101.6 mW).

Results and discussion

Synthesis and crystal structure analysis

Rietveld refinement of the SrBi₃O₄Cl₃ sample synthesized was performed using the SXRD pattern collected at room temperature. All diffraction peaks within 2θ = 2–24° were indexed to the tetragonal system. Based on previous reports,¹⁷ space group I4/mmm (#139) was adopted for Rietveld refinement. Reliability factors converged to R_p = 0.0795, R_wp = 0.1301, R_B = 0.0337, R_F = 0.0251, and χ² = 3.240. The final Rietveld plot and the refined crystal structure are shown in Fig. 1 and S4.† The refined parameters (Table S1†) were similar to those in a previous study,¹⁷ whereas in the present study, the positions of the cations were individually refined for Bi and Sr to obtained a more accurate structure. There are two Bi/Sr sites in the unit cell, derived from the asymmetric layered structure, that is, the fluorite [Sr_0.5Bi_1.5O₂] layer sandwiched between the single-halogen [Cl] and double-halogen [Cl₂] layers. Rietveld refinement indicated that the Bi/Sr1 site adjacent to the [Cl] layer is almost fully occupied by Bi³⁺, whereas Bi/Sr2 next to the [Cl₂] layer is shared almost equally by Bi³⁺ and Sr²⁺ (see Table 1). The electron density distribution obtained by MEM analysis (Fig. S5†) shows that the Bi/Sr1–O1 bond has a more covalent bonding character than the Bi/Sr2–O1 bond, which corresponds to the difference in occupations. Although the preferential occupancy of Bi³⁺ in the Bi/Sr1 site is also seen in the previous reports on XRD-based structural analysis,¹⁷ the detailed STEM observation in the present study provides direct evidence. As seen in Fig. 2, a periodic atomic arrangement agreeing with the refined crystal structure was observed in the high-angle annular dark field STEM (HAADF-STEM) and annular bright-field STEM (ABF-STEM) images in the direction of [100]. The elemental mapping (Fig. 2d) further confirmed the validity of the refined crystal structure. Importantly, the occupancies of the Bi/Sr1 and Bi/Sr2 sites are in good agreement with those obtained by Rietveld refinement (Table 1). The preferential occupancy of Bi³⁺ in the Bi/Sr1 site can be explained by Pauling's second rule. The bond valence sum of Bi and Sr cations in the two sites was calculated (Table S2†), where Sr cations in Bi/Sr1 were significantly overbonded. Such a large overbonding indicates that it is difficult for Sr cations to occupy the Bi/Sr1 site. As will be discussed later, the asymmetric distribution of Bi³⁺ cations in the fluorite layer of SrBi₃O₄Cl₃ is most likely the key to its narrower bandgap than that of analogous materials, such as SrBiO₂Cl.

Fig. 1 Final Rietveld plot for SrBi₃O₄Cl₃ using the SXRD pattern collected at room temperature; R_p = 0.0795, R_wp = 0.1301 R_B = 0.0337, R_F = 0.0251, and χ² = 3.240.

Table 1 Site occupancies in SrBi₃O₄Cl₃ obtained by Rietveld refinement (Fig. 1) and STEM-EDX (Fig. 2)

Atom	Rietveld refinementi	EDX
Bi/Sr1	Bi: 0.972(2)	Bi: 0.96(4)
	Sr: 0.028(2)	Sr: 0.04(2)
Bi/Sr2	Bi: 0.528(2)	Bi: 0.53(3)
	Sr: 0.472(2)	Sr: 0.47(4)

---

Fig. 2 STEM images of SrBi₃O₄Cl₃ in the direction of [100]. (a) HAADF, (b) ABF, (c) their enlarged images with the crystal structure, and (d) EDX elemental maps of SrBi₃O₄Cl₃.

Band structure of SrBi₃O₄Cl₃ and other analogues

Fig. 3a and b show the diffused reflectance spectra and the estimated band levels, respectively, of SrBi₃O₄Cl₃, SrBiO₂Cl, and BiOCl, whose refined crystal structures are shown in Fig. 4. The band levels were estimated based on a combination of bandgaps and flat-band potentials determined from the reflectance spectra (Fig. S6†) and Mott–Schottky (MS) plots (Fig. S7†), respectively. Despite their similar compositions and structures, only SrBi₃O₄Cl₃ possesses a narrow bandgap (2.68 eV), enabling visible absorption, which is derived from both the positively shifted CBM and the negatively shifted VBM compared to others.

Fig. 3 (a) Diffuse reflectance spectra of SrBi₃O₄Cl₃, SrBiO₂Cl, and BiOCl. (b) The band levels estimated from the diffuse reflectance spectra and MS plots (Fig. S7†). The band levels of BiOCl were estimated from the reported MS plot.⁵⁶ Flat-band potentials are assumed to be located just below the CBMs due to the n-type nature of these materials.

Fig. 4 Refined crystal structure of SrBi₃O₄Cl₃, along with BiOCl⁵⁷ and SrBiO₂Cl.⁵⁸ For clarity, the Bi/Sr1 site being next to the single [Cl] layer is fully occupied by Bi³⁺, whereas the Bi/Sr2 next to the double [Cl] layer is shared equally by Bi³⁺ and Sr²⁺.

Fig. 5 shows the density of states (DOS) and partial DOS (PDOS) of SrBi₃O₄Cl₃, calculated based on the assumption that Bi³⁺ and Sr²⁺ are ordered in the Bi/Sr2 site, whereas the Bi/Sr1 site is fully occupied by Bi³⁺ in a √2 × √2 ×1 supercell (Fig. S8†). The orbital distributions of the CBM and VBM of SrBi₃O₄Cl₃ are shown in Fig. 6. Given the refined crystal structures (Fig. 4), the significantly lowered CBM of SrBi₃O₄Cl₃ can probably be explained by Bi–Bi interactions across the single halogen layer. As shown in Fig. 5 and S9,† the DOS around the CBM is formed exclusively by Bi 6p orbitals at the Bi/Sr1 site, which Bi³⁺ almost fully occupies. In addition, an obvious interaction between the two Bi³⁺ ions across the single Cl anion layer can be observed in the visualized CBM (Fig. 6). The interaction was evaluated using the crystal orbital Hamilton population (COHP), in which positive and negative values indicate the bonding and antibonding nature, respectively (Fig. S10†). The COHP for the interaction of Bi 6pz–Bi 6pz reveals the bonding nature at the CBM; the Bi-6pz orbitals interact with each other across the single halogen layer, forming bonding and antibonding states. The bonding states, which are more positive than the nonbonding states of the Bi-6pz orbitals, significantly lowered the CBM of SrBi₃O₄Cl₃. Such Bi–Bi interactions can be rationalized by the coordination environment of the Bi/Sr1 site in SrBi₃O₄Cl₃, as previously discussed. In the fluorite [Sr_0.5Bi_1.5O₂] layer, the Bi/Sr1 site adjacent to the [Cl] layer is almost fully occupied by Bi³⁺, whereas Bi/Sr2 next to the [Cl₂] layer is shared almost equally by Bi³⁺ and Sr²⁺. No Cl anion is coaxially arranged with Bi³⁺ in Bi/Sr1, whereas coaxially arranged Bi³⁺ exists across the [Cl] layer. Although the coordination environment of Bi/Sr in SrBiO₂Cl (having single Cl layers only) is similar to that in Bi/Sr1 of SrBi₃O₄Cl₃, the occupancies of the cations are different from each other. Rietveld refinement of SrBiO₂Cl in a previous report⁵⁸ indicated that Bi³⁺ and Sr²⁺ in the Bi/Sr site are ordered, as shown in Fig. 4. In the fluorite [SrBiO₂] layer, Bi³⁺ and Sr²⁺ are distributed in a balanced manner toward the halogen layer. This cation distribution causes the Bi³⁺ cation to face the coaxially arranged Sr²⁺ cation across the [Cl] layer, preventing Bi–Bi interaction across the Cl layer (Fig. S11†). To evaluate the impact of Bi–Bi interactions on the band structure, additional DFT calculations were conducted using a hypothetical model of SrBi₃O₄Cl₃ where the Bi/Sr1 site is orderly and equally shared by Bi³⁺ and Sr²⁺ whereas the Bi/Sr2 site is solely occupied by Bi³⁺ (Fig. S12†), where the presence of Sr²⁺ in the Bi/Sr1 site drastically increases the bandgap from 1.4 to 2.6 eV. As for BiOCl, the double [Cl] layers increased the distance between the vertically arranged Bi³⁺, weakening their interaction (Fig. S13†).

Fig. 5 (a) DOS and PDOS of SrBi₃O₄Cl₃. (b) and (c) their close-ups at around the VBM and CBM, respectively.

Fig. 6 Orbital distributions of the CBM ((a) orange) and VBM ((b) blue) of SrBi₃O₄Cl₃ estimated by DFT calculations (isosurface value: 0.01 e Å⁻³), which are viewed from two different directions corresponding to the (110) (left) and (100) (right) directions for the refined structure (Fig. 1).

As shown in Fig. 5, the VBM of SrBi₃O₄Cl₃ is mainly composed of O-2p orbitals along with a contribution of the Cl-3p orbitals of Cl2 in the double layer. This sizable contribution of Cl-3p in SrBi₃O₄Cl₃ is in stark contrast to SrBiO₂Cl having single halogen layers only, where the contribution of Cl-3p to the VBM is negligible (Fig. S14†). In previous studies, we have revealed that the ionic orbital energy levels of each anion, which can be calculated from the sum of the electron affinity and the Madelung potential at each crystallographic site, can appropriately reproduce the features of the valence band PDOS obtained from DFT calculations.⁵⁹Fig. 7 shows the ionic energy levels of each Cl anion in these compounds, where the ionic energy level of Cl2 in the double layer of SrBi₃O₄Cl₃ exhibits the highest value, while those of Cl anions in SrBiO₂Cl and BiOCl exhibit approximately the same and low values. The highest value indicates that the Cl2 anion in SrBi₃O₄Cl₃ was electrostatically destabilized. Cl anions are more destabilized in SrBi₃O₄Cl₃ than in SrBiO₂Cl due to the strong repulsion between Cl anions in the double layer. Although such destabilization is also expected for Cl anions in the Cl double layer of BiOCl, their ionic energy level is significantly lower than that of SrBi₃O₄Cl₃. This difference in the ionic energies of Cl anions in Cl double layers can be understood by considering the valence state of the surrounding cations. While the Cl anions in the double layer of SrBi₃O₄Cl₃ are surrounded by four Sr²⁺/Bi³⁺ cations (i.e., a positive charge of 2.5) owing to the asymmetric Bi occupation in the fluorite [Sr_0.5Bi_1.5O₂] layer, those in BiOCl are surrounded by four Bi³⁺ cations, resulting in a higher positive charge to stabilize Cl anions. The COHP for the Bi2–O and Bi2–Cl2 interactions (Fig. S15†) shows an antibonding nature at the VBM; the Cl and O orbitals probably interact with each other through the interaction with Bi orbitals, forming the elevated VBM of SrBi₃O₄Cl₃. In conclusion, the most negative VBM level of SrBi₃O₄Cl₃ can probably be rationalized by the destabilization of Cl2 anions in SrBi₃O₄Cl₃ and the accompanying sizable contribution of the Cl-3p of Cl2 to the VBM rather than that of Cl1. In future studies, a more systematic investigation of a series of related layered oxyhalides needs to be conducted to confirm the origin of the elevated VBM of SrBi₃O₄Cl₃ and, thereby, provide a guiding principle for controlling band levels by designing halogen layers (e.g., change in the proportion of single and double halogen layers).

Fig. 7 Ionic energy levels of each halogen site in SrBi₃O₄Cl₃, calculated from the sum of the Madelung potentials at the Cl site and electron affinity, along with those of BiOCl and SrBiO₂Cl. The Madelung potentials for SrBi₃O₄Cl₃ were calculated from the refined structure (Fig. 1 and Table S1†).

Carrier dynamics and photocatalytic reactions

Before discussing the photocatalytic reactions, we evaluated the carrier dynamics of SrBi₃O₄Cl₃ by TRMC, which gives the signal of the product of the charge carrier generation efficiency φ and the sum of the charge carrier mobilities Σμ. As shown in Fig. 8, SrBi₃O₄Cl₃ shows a significantly higher TRMC signal than the others after excitation at 266 nm, despite a similar, high photoabsorption at this wavelength among all the samples. The product of the maximum signal intensity and lifetime, which has been previously reported to well correlate with the photocatalytic activities of some oxyhalide photocatalysts,^44,45 is more than an order of magnitude larger than that of BiOCl and SrBiO₂Cl (Fig. S16 and Table S3†). This superior carrier dynamics of SrBi₃O₄Cl₃ is probably derived from the spatial separation of the CBM and VBM (Fig. 6), which is regarded as an effective strategy to promote charge separation,^60–62 as also suggested in analogous PbBi₃O₄Cl₃.⁵⁶ The present result shows that the coexistence of single and double halogen layers in SrBi₃O₄Cl₃ not only provides visible light response but also improved carrier separation and suppression of charge recombination.

Fig. 8 TRMC transients (φΣμ) of SrBi₃O₄Cl₃, SrBiO₂Cl, and BiOCl after excitation at 266 nm.

SrBi₃O₄Cl₃ showed a negligibly low H₂ evolution activity (Fig. S17†), as expected from the slightly more negative CBM compared to the water reduction potential (Fig. 3b). In contrast, it exhibited a reasonable activity for O₂ evolution from water in the presence of an appropriate electron acceptor, such as Fe³⁺. Fig. 9a shows the time course of photocatalytic O₂ evolution in the presence of an Fe³⁺ electron acceptor with various counter anions; the O₂ evolution rates strongly depend on the type of counter anions. This influence of anions has also been reported for other photocatalysts.^41,63,64 Although FeCl₃ provided the highest rate, we noticed slight but obvious impurity peaks derived from BiOCl after the reaction, except for Fe₂(SO₄)₃ (Fig. 9b). The emergence of such a small amount of BiOCl was also observed under dark conditions, suggesting that the change mainly occurred chemically and not photochemically. The following evaluations were conducted using mainly Fe₂(SO₄)₃, which does not provide a BiOCl phase at all, to avoid the possible influence of the BiOCl phase.

Fig. 9 (a) Time courses of photocatalytic O₂ evolution on SrBi₃O₄Cl₃ under visible light (400 < λ < 800 nm) from an aqueous solution of FeCl₃, Fe(NO₃)₃, Fe₂(SO₄)₃ or Fe(ClO₄)₃ (5 mM as Fe³⁺, 120 mL, pH 2.4). (b) XRD patterns of SrBi₃O₄Cl₃ after the reactions.

The loading of RuO_x cocatalysts on SrBi₃O₄Cl₃ increased the O₂ evolution rate from aqueous Fe³⁺ solution in a similar fashion to other oxyhalides.^41,56,65 The formation of RuO₂ was confirmed by XAFS and XPS measurements (Fig. 10a and S18†). RuO₂-loaded SrBi₃O₄Cl₃ (RuO₂/SrBi₃O₄Cl₃) exhibited steady O₂ evolution repeatedly at least three times from aqueous solutions of Fe₂(SO₄)₃ (Fig. 10b). Importantly, an almost stoichiometric amount of O₂ (150 μmol) estimated from the amount of introduced Fe³⁺ was evolved, strongly suggesting that O₂ evolution in this system was not significantly suppressed by the backward oxidation of Fe²⁺ to Fe³⁺ by holes.⁶⁶ A Z-scheme overall water splitting system was constructed using a combination of RuO₂/SrBi₃O₄Cl₃ and Ru/SrTiO₃ : Rh,^47,48 respectively, as O₂- and H₂-evolving photocatalysts. As shown in Fig. 11, an excess amount of O₂ was generated at the initial stage because of the higher concentration of Fe³⁺ (vs. Fe²⁺), which functioned as the electron acceptor for RuO₂/SrBi₃O₄Cl₃. With increasing time, the rate of H₂ evolution gradually increased as a result of the increased concentration of Fe²⁺ and finally approached an almost stoichiometric value (i.e., H₂ : O₂ = 2 : 1). At 420 nm, the AQE was 0.3% and 0.9% with Fe₂(SO₄)₃ and FeCl₃ (see Fig. S19†), respectively. Although the value is not so high compared with the reported ones,⁶⁷ we expect that further optimizations will improve the performance of photocatalytic water splitting.

Fig. 10 (a) Ru K-edge XANES spectra for the SrBi₃O₄Cl₃ sample loaded with Ru species. (b) Third-cycle test of photocatalytic O₂ evolution on RuO₂/SrBi₃O₄Cl₃ under visible light (400 < λ < 800 nm) from an aqueous solution of Fe₂(SO₄)₃ (5 mM as Fe³⁺, 120 mL, pH 2.4).

Fig. 11 Z-scheme overall water splitting using RuO₂/SrBi₃O₄Cl₃ as an O₂-evolving photocatalyst and Ru/SrTiO₃ : Rh as a H₂-evolving photocatalyst under visible light (400 < λ < 800 nm) in an aqueous Fe₂(SO₄)₃ solution (2 mM as Fe³⁺, 120 mL, pH 2.4).

Conclusions

In the present work, we studied a Sillén phase SrBi₃O₄Cl₃ with single/double halogen layers to investigate the structure–property relationship of layered bismuth oxyhalides. Rietveld refinement and STEM-EDX mapping of SrBi₃O₄Cl₃ confirmed the asymmetric occupation of Bi in the fluorite [Sr_0.5Bi_1.5O₂] layer, which was caused by the coexistence of two different halogen layers. Interestingly, only SrBi₃O₄Cl₃ showed a visible light response (bandgap: 2.68 eV), whereas SrBiO₂Cl and BiOCl did not. Theoretical investigations suggest that both the Bi–Bi interaction across the single halogen layer and the electrostatic destabilization of Cl in the double halogen layer lead to the narrow bandgap of SrBi₃O₄Cl₃. TRMC measurements and DFT calculations suggested another advantage of possessing two different halogen layers in SrBi₃O₄Cl₃: the spatial separation of the CBM and VBM based on the coexistence of halogen layers probably promotes charge-carrier separation. The loading of RuO₂ cocatalysts significantly improved the O₂ evolution activity of SrBi₃O₄Cl₃, and the RuO₂-loaded SrBi₃O₄Cl₃ sample functioned as an O₂-evolving photocatalyst for Z-scheme water splitting under visible light. This study provides another option for engineering the band structures and charge-carrier dynamics of layered oxyhalides. In particular, the Bi–Bi interlayer interactions may have already played an important role in previously reported oxyhalide photocatalysts with a single halogen layer.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by JST CREST (JPMJCR1421), the JSPS Core-to-Core Program (JPJSCCA20200004), JSPS KAKENHI (JP20H00398) in Grant-in-Aid for Scientific Research (A), JSPS KAKENHI (JP16H06438) in Grant-in-Aid for Scientific Research on Innovative Area “Mixed Anion”, JSPS KAKENHI (JP17H06439 and JP17H06438) in Scientific Research on Innovative Area “Innovations for Light-Energy Conversion (I4LEC)” and KAKENHI (JP21K14719 and JP21K20556) in Grant-in-Aid for Early-Career Scientists and Research Activity Start-up. This work was also supported by the Tokuyama Science Foundation and Yashima Environment Technology Foundation. A part of this work was supported by the Advanced Characterization Platform and AIST Nanocharacterization Facility (ANCF) Platform as a program of the “Nanotechnology Platform” (JPMXP09A20KU0355). We are also grateful to Takaaki Toriyama of Kyushu University for helpful support in the STEM analysis. Synchrotron experiments were performed at SPring-8 BL02B2 of JASRI (2020A0824 and 2022A1081). The authors are also indebted to the technical division of the Institute for Catalysis, Hokkaido University, for their help in building the experimental equipment.

Notes and references

1. L. G. Sillen, Naturwissenschaften, 1938, 26, 612–613.
2. L. G. Sillen, Naturwissenschaften, 1942, 30, 318–324.
3. D. O. Charkin, O. S. Morozov, E. A. Ul'yanova, P. S. Berdonosov, V. A. Dolgikh, C. Dickinson, W. Zhou and P. Lightfoot, Solid State Sci., 2006, 8, 1029–1034.
4. A. Pfitzner and P. Pohla, Z. Anorg. Allg. Chem., 2009, 635, 1157–1159.
5. J. F. Vigier, C. Renard, N. Henry, A. Laplace and F. Abraham, Inorg. Chem., 2012, 51, 4352–4358.
6. D. O. Charkin, P. S. Berdonosov, V. A. Dolgikh and P. Lightfoot, J. Solid State Chem., 2003, 175, 316–321.
7. M. Schmidt, H. Oppermann, C. Hennig, R. W. Henn, E. Gmelin, N. Söger and M. Binnewies, Z. Anorg. Allg. Chem., 2000, 626, 125–135.
8. D. H. Nam and K. S. Choi, J. Am. Chem. Soc., 2017, 139, 11055–11063.
9. Y. Myung, J. Choi, F. Wu, S. Banerjee, E. H. Majzoub, J. Jin, S. U. Son, P. V. Braun and P. Banerjee, ACS Appl. Mater. Interfaces, 2017, 9, 14187–14196.
10. D.-H. Nam, M. A. Lumley and K.-S. Choi, Chem. Mater., 2019, 31, 1460–1468.
11. J. Sun, Q. Sun, A. Plewa, Y. Wang, L. He, F. Zheng, C. Chen, W. Zajac, J. Molenda, K. Zeng and L. Lu, Angew. Chem., Int. Ed., 2020, 59, 8991–8997.
12. J. Wang, B. Wang and B. Lu, Adv. Energy Mater., 2020, 10, 2000884.
13. V. Ahuja, S. Singh, R. Vengarathody and P. Senguttuvan, J. Phys. Chem. C, 2021, 125, 17622–17628.
14. H. Li, F. Qin, Z. Yang, X. Cui, J. Wang and L. Zhang, J. Am. Chem. Soc., 2017, 139, 3513–3521.
15. Q. Yuan, W. Gong, Y. Ye, J. Liu, Y. Lin, C. Chen, H. Zhang, P. Li, W. Cheng, X. Wei and C. Liang, ChemCatChem, 2019, 11, 2676–2682.
16. W. Ueda, S.-W. Lin and I. Tohmoto, Catal. Lett., 1997, 44, 241–245.
17. L. S. Wen, K. Y. Chul and U. Wataru, Bull. Chem. Soc. Jpn., 1998, 71, 1089–1094.
18. N. Kijima, K. Matano, M. Saito, T. Oikawa, T. Konishi, H. Yasuda, T. Sato and Y. Yoshimura, Appl. Catal., A, 2001, 206, 237–244.
19. A. Aliev, J. Olchowka, M. Colmont, E. Capoen, C. Wickleder and O. Mentré, Inorg. Chem., 2013, 52, 8427–8435.
20. J. Olchowka, H. Kabbour, M. Colmont, M. Adlung, C. Wickleder and O. Mentre, Inorg. Chem., 2016, 55, 7582–7592.
21. Y. Kitagawa, J. Ueda, K. Arai, H. Kageyama and S. Tanabe, J. Appl. Phys., 2021, 129, 183104.
22. X. Zhang, Y. Zhang, Z. Feng, J. Zhao, Z. Yang, X. Wang and W. Wang, Chem. Eng. J., 2022, 428, 131235.
23. H. Kunioku, M. Higashi and R. Abe, Sci. Rep., 2016, 6, 32664.
24. R. L. Z. Hoye, L. C. Lee, R. C. Kurchin, T. N. Huq, K. H. L. Zhang, M. Sponseller, L. Nienhaus, R. E. Brandt, J. Jean, J. A. Polizzotti, A. Kursumovic, M. G. Bawendi, V. Bulovic, V. Stevanovic, T. Buonassisi and J. L. MacManus-Driscoll, Adv. Mater., 2017, 29, 1702176.
25. T. N. Huq, L. C. Lee, L. Eyre, W. Li, R. A. Jagt, C. Kim, S. Fearn, V. Pecunia, F. Deschler, J. L. MacManus-Driscoll and R. L. Z. Hoye, Adv. Funct. Mater., 2020, 30, 1909983.
26. J. Jiang, K. Zhao, X. Xiao and L. Zhang, J. Am. Chem. Soc., 2012, 134, 4473–4476.
27. M. Guan, C. Xiao, J. Zhang, S. Fan, R. An, Q. Cheng, J. Xie, M. Zhou, B. Ye and Y. Xie, J. Am. Chem. Soc., 2013, 135, 10411–10417.
28. L. Liu, Y. Sun, X. Cui, K. Qi, X. He, Q. Bao, W. Ma, J. Lu, H. Fang, P. Zhang, L. Zheng, L. Yu, D. J. Singh, Q. Xiong, L. Zhang and W. Zheng, Nat. Commun., 2019, 10, 4472.
29. G. Qiu, T. Wang, X. Li, X. Tao and B. Li, Ind. Eng. Chem. Res., 2020, 59, 11517–11526.
30. B. Li, L. Shao, R. Wang, X. Dong, F. Zhao, P. Gao and Z. Li, J. Mater. Chem. A, 2018, 6, 6344–6355.
31. F. Xie, Y. Zhang, X. He, H. Li, X. Qiu, W. Zhou, S. Huo and Z. Tang, J. Mater. Chem. A, 2018, 6, 13236–13243.
32. Y. Wu, B. Yuan, M. Li, W. H. Zhang, Y. Liu and C. Li, Chem. Sci., 2015, 6, 1873–1878.
33. H. Li, J. Shang, H. Zhu, Z. Yang, Z. Ai and L. Zhang, ACS Catal., 2016, 6, 8276–8285.
34. M. Li, S. Yu, H. Huang, X. Li, Y. Feng, C. Wang, Y. Wang, T. Ma, L. Guo and Y. Zhang, Angew. Chem., Int. Ed., 2019, 58, 9517–9521.
35. Y. Shiraishi, M. Hashimoto, K. Chishiro, K. Moriyama, S. Tanaka and T. Hirai, J. Am. Chem. Soc., 2020, 142, 7574–7583.
36. Y. Shi, J. Li, C. Mao, S. Liu, X. Wang, X. Liu, S. Zhao, X. Liu, Y. Huang and L. Zhang, Nat. Commun., 2021, 12, 5923.
37. Y. Yu, C. Cao, H. Liu, P. Li, F. Wei, Y. Jiang and W. Song, J. Mater. Chem. A, 2014, 2, 1677–1681.
38. T. B. Li, G. Chen, C. Zhou, Z. Y. Shen, R. C. Jin and J. X. Sun, Dalton Trans., 2011, 40, 6751–6758.
39. L. Sun, L. Xiang, X. Zhao, C.-J. Jia, J. Yang, Z. Jin, X. Cheng and W. Fan, ACS Catal., 2015, 5, 3540–3551.
40. H. Kunioku, M. Higashi, O. Tomita, M. Yabuuchi, D. Kato, H. Fujito, H. Kageyama and R. Abe, J. Mater. Chem. A, 2018, 6, 3100–3107.
41. H. Suzuki, H. Kunioku, M. Higashi, O. Tomita, D. Kato, H. Kageyama and R. Abe, Chem. Mater., 2018, 30, 5862–5869.
42. C. Zhong, D. Kato, K. Ogawa, C. Tassel, F. Izumi, H. Suzuki, S. Kawaguchi, T. Saito, A. Saeki, R. Abe and H. Kageyama, Inorg. Chem., 2021, 60, 15667–15674.
43. A. Nakada, D. Kato, R. Nelson, H. Takahira, M. Yabuuchi, M. Higashi, H. Suzuki, M. Kirsanova, N. Kakudou, C. Tassel, T. Yamamoto, C. M. Brown, R. Dronskowski, A. Saeki, A. Abakumov, H. Kageyama and R. Abe, J. Am. Chem. Soc., 2021, 143, 2491–2499.
44. H. Suzuki, S. Kanno, M. Hada, R. Abe and A. Saeki, Chem. Mater., 2020, 32, 4166–4173.
45. H. Suzuki, M. Higashi, H. Kunioku, R. Abe and A. Saeki, ACS Energy Lett., 2019, 4, 1572–1578.
46. L. G. Sillén, Naturwissenschaften, 1940, 28, 396–397.
47. R. Konta, T. Ishii, H. Kato and A. Kudo, J. Phys. Chem. B, 2004, 108, 8992–8995.
48. Y. Sasaki, A. Iwase, H. Kato and A. Kudo, J. Catal., 2008, 259, 133–137.
49. A. Altomare, C. Cuocci, C. Giacovazzo, A. Moliterni, R. Rizzi, N. Corriero and A. Falcicchio, J. Appl. Crystallogr., 2013, 46, 1231–1235.
50. F. Izumi and K. Momma, Solid State Phenom., 2007, 130, 15–20.
51. K. Momma and F. Izumi, J. Appl. Crystallogr., 2011, 44, 1272–1276.
52. K. Momma, T. Ikeda, A. A. Belik and F. Izumi, Powder Diffr., 2013, 28, 184–193.
53. S. J. Clark, M. D. Segall, C. J. Pickard, P. J. Hasnip, M. I. J. Probert, K. Refson and M. C. Payne, Z. Kristallogr., 2005, 220, 567–570.
54. P. Giannozzi, S. Baroni, N. Bonini, M. Calandra, R. Car, C. Cavazzoni, D. Ceresoli, G. L. Chiarotti, M. Cococcioni, I. Dabo, A. Dal Corso, S. de Gironcoli, S. Fabris, G. Fratesi, R. Gebauer, U. Gerstmann, C. Gougoussis, A. Kokalj, M. Lazzeri, L. Martin-Samos, N. Marzari, F. Mauri, R. Mazzarello, S. Paolini, A. Pasquarello, L. Paulatto, C. Sbraccia, S. Scandolo, G. Sclauzero, A. P. Seitsonen, A. Smogunov, P. Umari and R. M. Wentzcovitch, J. Phys. Condens. Matter, 2009, 21, 395502.
55. S. Maintz, V. L. Deringer, A. L. Tchougreeff and R. Dronskowski, J. Comput. Chem., 2016, 37, 1030–1035.
56. H. Suzuki, M. Higashi, O. Tomita, Y. Ishii, T. Yamamoto, D. Kato, T. Kotani, D. Ozaki, S. Nozawa, K. Nakashima, K. Fujita, A. Saeki, H. Kageyama and R. Abe, Chem. Mater., 2021, 33, 9580–9587.
57. Y. Yang, G. Z. Wang, Q. Deng, S. H. Kang, D. H. L. Ng and H. J. Zhao, CrystEngComm, 2014, 16, 3091–3096.
58. S. M. Fray, C. J. Milne and P. Lightfoot, J. Solid State Chem., 1997, 128, 115–120.
59. D. Kato, K. Hongo, R. Maezono, M. Higashi, H. Kunioku, M. Yabuuchi, H. Suzuki, H. Okajima, C. C. Zhong, K. Nakano, R. Abe and H. Kageyama, J. Am. Chem. Soc., 2017, 139, 18725–18731.
60. Z. K. Tang, W. J. Yin, L. Zhang, B. Wen, D. Y. Zhang, L. M. Liu and W. M. Lau, Sci. Rep., 2016, 6, 32764.
61. Y. G. Zhou, L. J. Zhou, J. J. He and T. Frauenheim, J. Phys. Chem. Lett., 2020, 11, 3095–3102.
62. D. Ozaki, H. Suzuki, K. Ogawa, R. Sakamoto, Y. Inaguma, K. Nakashima, O. Tomita, H. Kageyama and R. Abe, J. Mater. Chem. A, 2021, 9, 8332–8340.
63. Y. Miseki, H. Kusama, H. Sugihara and K. Sayama, Chem. Lett., 2010, 39, 846–847.
64. H. Suzuki, O. Tomita, M. Higashi and R. Abe, J. Mater. Chem. A, 2017, 5, 10280–10288.
65. K. Ogawa, A. Nakada, H. Suzuki, O. Tomita, M. Higashi, A. Saeki, H. Kageyama and R. Abe, ACS Appl. Mater. Interfaces, 2019, 11, 5642–5650.
66. Y. O. Wang, H. Suzuki, J. J. Xie, O. Tomita, D. J. Martin, M. Higashi, D. Kong, R. Abe and J. W. Tang, Chem. Rev., 2018, 118, 5201–5241.
67. H. Kato, Y. Sasaki, N. Shirakura and A. Kudo, J. Mater. Chem. A, 2013, 1, 12327–12333.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3ta00906h

‡ H. S. and D. O. contributed equally to this work.

---