New insights into the synthesis of Sillén–Aurivillius oxyhalides: molten salts induce interlayer halogen competing reaction

Abstract

Numerous studies disclosed that molten salts provide a special liquid reaction environment, accelerate the diffusion rate of active elements, and promote the formation of high-purity materials in the molten-salt synthesis method. Herein, the interlayer halogen-competing reactions between molten salts and halogen layers during the formation of BaBi₄TiNbO₁₁X (X = Cl, Br) oxyhalides were discovered for the first time. The liquid environment provided by molten salts can facilitate the full migration of halogen atoms between molten salts and halogen layers and form an even distribution of halogen atoms, which is ascribed to the weak van der Waals forces of the halogen layers between [BaBiO₂]⁺ and [Bi₂O₂]²⁺ blocks. By employing this mechanism, BaBi₄TiNbO₁₁(Cl_1−xBr_x) with different x contents could be easily obtained. DFT calculations showed that the mixed halogen oxyhalides, such as BaBi₄TiNbO₁₁(Cl_0.5Br_0.5), introduce new chemical bonds between the O and Bi atoms, which increase their contribution to lowest occupied/unoccupied orbitals, thus regulating the energy band and accelerating the separation and transfer of photogenerated charge carriers. Consequently, the optimized BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) sample showed a superior degradation rate of RhB compared to the traditional P25 photocatalyst, and its O₂ evolution performance was also enhanced compared with other blank samples.

---

1. Introduction

With the acceleration of modern industrialization and the increase in global energy consumption, the energy crisis and environmental pollution have become increasingly prominent, seriously endangering the survival and development of human beings.^1–3 Thus, the photocatalytic technology, an emerging technology, that utilizes solar energy to drive chemical reactions, such as water splitting, hydrogen evolution, oxygen evolution, CO₂ reduction, degradation of organic pollutants, and reduction of heavy metal ions, has attracted attention from researchers around the world due to its advantages of mild reaction conditions and clean and abundant energy sources.^4–6 In the past few decades, the most widely studied catalyst is titanium dioxide (TiO₂), which has the advantages of strong oxidation capacity, low toxicity, low cost, and easy synthesis.^7,8 However, the band gap (E_g) of traditional anatase TiO₂ is wide, i.e., about 3.2 eV, and only responds to UV light, severely limiting the generation of photogenerated charge carriers. In this case, one promising method to extend its light absorption is narrowing the E_g values of semiconductors.^9,10 In general, this often causes a mismatch between the band energy position and the redox potential of the target substances, which will weaken the reaction driving force of semiconductors. Therefore, it is of great research value to explore new visible light catalysts.

Nowadays, the majority of photocatalysts that satisfy the requirements of wide spectral absorption range and matching redox potential are nitrides/nitrogen oxides and sulfides/sulfur oxides.^11,12 The non-oxide elements of these semiconductor materials will react with the holes generated under light irradiation near the valence band maximum (VBM), resulting in self-decomposition, deactivation and reduction in their photocatalytic stability. Over the past few decades, a range of semiconductor materials has been widely explored for solar energy applications with the aim of developing highly efficient and stable photocatalysts.^13–15 Among them, bismuth (Bi)-based Sillén–Aurivillius oxyhalides, such as Bi₄MO₈X (M = Nb, Ta; X = Cl, Br, and I), have been confirmed as promising photocatalysts owing to their unique electronic structure and excellent catalytic performance.^16–20 On the one hand, the VBM in Sillén–Aurivillius oxyhalides is mainly composed of O 2p orbitals, not Cl 3p, Br 4p, and I 5p orbitals, making them robust against self-decomposition during the photocatalytic process. On the other hand, the stronger hybridization between Bi 6s and O 2p orbitals can promote the VBM to shift to a more negative level, which lowers the E_g value and broadens the spectral absorption of Sillén–Aurivillius oxyhalides. Furthermore, the non-centrosymmetric (NSC) structure of Sillén–Aurivillius oxyhalides originating from the distortion of the BO₆ octahedron in the perovskite blocks can form internal electric fields (IEF), facilitate the separation of photogenerated charge carriers and lessen the recombination of electron–hole pairs.^21,22 Thus, based on the above-mentioned advantages, Sillén–Aurivillius oxyhalides have attracted extensive attention.

At present, the mainstream route for growing Sillén–Aurivillius oxyhalides is solid-state reaction (SSR), which is significantly influenced by the level of halogen precursor and calcination temperature.^16,22 When the stoichiometric precursor materials are calcined at high temperature, highly crystalline Sillén–Aurivillius oxyhalides can be formed. However, high temperature also induces the loss of halogen [X] elements, creating deep-level defects, which serve as recombination centers, hindering the transport of photogenerated charge carriers and reducing the photocatalytic performance.^23,24 Based on this, the molten-salt synthesis (MSS) method has been widely used for the fabrication of Sillén–Aurivillius oxyhalides given that it can offer a special liquid reaction environment that expedites the diffusion rate of reactive elements and motivates the formation of high-purity and uniform reaction products.^25,26 Because Sillén–Aurivillius oxyhalides contain halogen elements (F, Cl, Br, and I), most Bi-based materials are grown with the same molten salt as halogen. For example, Zhizhong Chen et al. used NaCl/CsCl as the molten salt to synthesize Bi₄NbO₈Cl.²⁷ This method can avoid the influence of the halogen elements of the molten salt based on the synthetic target substance. In addition, other researchers adopted Cl-based flux (NaCl/CsCl, NaCl/KCl) to fabricate Bi₄NbO₈Br from a mixture of Ta₂O₅, BiOBr, and Bi₂O₃ at high calcination temperature.^28,29 However, a higher temperature during the calcination process causes the full diffusion of the elements, which will inevitably affect synthetic materials.²⁵ Therefore, it is necessary to study the effect of the chosen molten salt material on the target product.

In this work, a series of novel Sillén–Aurivillius oxyhalides, BaBi₄TiNbO₁₁X (X = Cl, Br), based on different molten salt materials was investigated to confirm the competing reaction of the interlayer halogen. Initially, we determined the existence of halogen competing reactions in the synthesis of BaBi₄TiNbO₁₁X by X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and X-ray photoelectron spectroscopy (XPS). In combination with the investigation of material properties for Sillén–Aurivillius oxyhalides, a possible mechanism of interlayer halogen competing reaction between molten salts and the halogen layers was proposed. Furthermore, the experimental characterization and theoretical calculations confirmed that the hybrid halogen in Sillén–Aurivillius oxyhalides can optimize their energy band structure to facilitate the separation and transfer of the photogenerated charge carriers, promoting their photocatalytic activities in O₂ evolution and the degradation of organic pollutants. This study aimed to offer new insights into the interlayer halogen competing reaction between molten salts and the halogen layers.

2. Experimental methods

2.1. Materials

All the chemical reagents employed in this work were of analytical grade and directly use without further purification, and all purchased from Shanghai Aladdin Reagent Co., LTD.

2.2. Preparation of photocatalysts

2.2.1. Synthesis of BaBiO₂X precursors. In this experiment, BiOX (X = Cl, Br) precursors were prepared via a chemical deposition process at room temperature with the precursor materials of KX (X = Cl, Br) and Bi(NO₃)·5H₂O, and more details about this process can be found in our previous works.^9,23,30 Then, BiOX (X = Cl, Br) and BaCO₃ were mixed in the stoichiometric ratio of 1 : 1 to prepare the BaBiO₂X precursor at 700 °C for 10 h by the traditional SSR method.

2.2.2. Synthesis of Bi₃TiNbO₉ precursors. Bi₃TiNbO₉ was synthesized using a stoichiometric mixture of Nb₂O₅, Bi₂O₃, and TiO₂via the MSS method. Simultaneously, the concentration ratio between the solute and solvent was about 6.25%. Before the calcination of Bi₃TiNbO₉ at high temperature, the mixture was ground thoroughly and transferred to a common porcelain crucible. Then, the samples were calcined at 800 °C for 10 h. After the calcination process, the yellow powder was filtered with deionized water three times and dried at 60 °C for 10 h in an oven.

2.2.3. Synthesis of BaBi₄TiNbO₁₁X oxyhalides. The MSS method was used to prepare BaBi₄TiNbO₁₁X oxyhalides employing the raw materials of BaBiO₂X and Bi₃TiNbO₉. The detailed parameters of this process are similar to that for the preparation of Bi₃TiNbO₉. The BaBi₄TiNbO₁₁X samples based on molten salts of NaCl/CsCl or NaBr/CsBr were calcined at 600–700 °C for 10 h, respectively, and more details about the synthesis of BaBi₄TiNbO₁₁X are shown in Fig. 1. Samples with different molar ratios of precursor materials ([Br]/([Br] + [Cl])) (0%, 9%, 33.3%, 50%, and 100%) were prepared.

Fig. 1 Schematic illustration of the preparation process of BaBi₄TiNbO₁₁X.

2.3. Characterization

The surface morphologies of the BaBi₄TiNbO₁₁X samples were determined via scanning electron microscopy (SEM). X-ray diffraction (XRD) with a Cu K_α source was employed to characterize the crystal structure of the as-prepared samples. The microscopic characteristics of the BaBi₄TiNbO₁₁X samples were analyzed via transmission electron microscopy (TEM). The information on surface chemical elements (Ba, Bi, Ti, Nb, O, Cl, Br, and C elements) was collected via X-ray photoelectron spectroscopy (XPS) with an Al Kα source. The light absorption properties of the BaBi₄TiNbO₁₁X samples were measured by UV-vis diffuse reflectance spectroscopy (DRS). The structural composition of the BaBi₄TiNbO₁₁X oxyhalides was determined by Raman spectroscopy using a 514 nm laser. Fourier transform infrared (FTIR) spectroscopy was used to confirm the characteristic functional groups present in the BaBi₄TiNbO₁₁X oxyhalides. The separation efficiency of photogenerated electrons and holes was analyzed by photoluminescence spectroscopy (PL, Hitachi F-4600), transient photocurrent responses, and electrochemical impedance spectroscopy (EIS) (CHI-660D Electrochemical Workstation, Chenhua Instruments Co., Ltd, China). In addition, the electrochemical workstation was also employed for the electrochemical and photoelectrochemical measurements of the BaBi₄TiNbO₁₁X oxyhalides, as described in the previous studies.^9,22,31

2.4. Photocatalytic activity measurements

The photocatalytic performance of the BaBi₄TiNbO₁₁X samples was evaluated by oxygen (O₂) evolution and RhB degradation under visible light. The light source in the experiment was a 300 W Xe lamp with a 420 nm cutoff filter (Beijing China Education Au-light Co., Ltd). In the O₂ evolution experiment, about 100 mg BaBi₄TiNbO₁₁X powder was mixed with an AgNO₃ solution (250 mL, 5 mmol L⁻¹) and thoroughly stirred at room temperature. Before visible light irradiation, the reactor for O₂ evolution was vacuumized five times and purified with Ar (99.999%) to remove the residual O₂. Then, the Xe lamp was turned on. The rate of O₂ evolution was measured from the O₂ concentration by liquid chromatography once every hour.

For the process of RhB degradation under visible light, 100 mg BaBi₄TiNbO₁₁X powder was suspended in an RhB aqueous solution (100 mL, 5 mg L⁻¹). Firstly, the as-obtained mixed solution was kept in a dark room with magnetic stirring for 0.5 h. Subsequently, the reactor containing the mixed solution was moved under an Xe lamp for the photocatalytic degradation experiments. Furthermore, the reaction solution of about 4 mL was withdrawn from the reactor once every 15 min to determine the degradation efficiencies of RhB. In the trapping experiment, 0.5 mmol of EDTA-2Na, tert-butyl alcohol (t-BuOH and TBA), and benzoquinone (BQ) was employed to capture the h⁺, ˙OH, and ˙O₂⁻, respectively, and this process was similar to that of the RhB degradation process.

2.5. Computational methodology

All DFT calculations were performed using the Vienna ab initio simulation package (VASP).^32,33 The generalized gradient approximation (GGA) was employed to describe the electron exchange and correlation in the Perdew–Burke–Ernzerhof (PBE) functional.³⁴ The PBE functional was used only for structure optimization but for the band structure calculation we used the Heyd–Scuseria–Ernzerhof (HSE06) functional due to its accurate description of the electronic structure.^35,36 The plane-wave kinetic energy cutoff was set to 520 eV. The considered conversion criteria for the energies and the forces on the individual atoms were 10⁻⁵ eV and 10⁻³ eV Å⁻¹, respectively. In the partial density of states (PDOS) calculations, Gaussian smearing was used. A gamma-centered K-point mesh of 8 × 8 × 2 was considered for the optimization and PDOS calculations. For better visualization of the structure properties, we measured scanning tunneling microscopy (STM) images with a tip distance of 1.5 Å.

3. Results and discussion

3.1. Material properties of BaBi₄TiNbO₁₁X (X = Cl, Br) nanoplates

XRD was employed to characterize the phase purity and crystal structure of the as-prepared samples. The XRD patterns of BaBi₄TiNbO₁₁Cl synthesized from the raw materials of BaBiO₂Cl and Bi₃TiNbO₉ based on NaCl/CsCl molten salts are presented in Fig. 2a. Clearly, the main diffraction peaks of all the BaBi₄TiNbO₁₁Cl samples located at the 2θ values of (003), (011), (014), (210), (217), and (414) are consistent with the DFT results reported in our previous work.²² The results suggest that BaBi₄TiNbO₁₁Cl oxyhalides can be synthesized over a wide temperature range. To determine the optimal synthesis temperature, the value of peak intensity ratio (I_(014)/(210)) between the primary peak (014) and the secondary peak (210) was calculated for the different samples. The values of I_(014)/(210) for the BaBi₄TiNbO₁₁Cl samples fabricated at 625 °C, 650 °C, and 675 °C are 3.64, 3.78, and 2.44, respectively. The stronger peak intensity ratio of the samples indicates that the BaBi₄TiNbO₁₁Cl oxyhalide fabricated at 650 °C grows along the (014) crystal faces easily. Simultaneously, the full width at half maximum (FWHM) of the (014) peak for the BaBi₄TiNbO₁₁Cl samples fabricated at 625 °C, 650 °C, and 675 °C is 0.2128, 0.2021, and 0.3443, respectively, and thus the crystallinity of the sample prepared at 650 °C is the best. Based on the above-mentioned discussion, it can be concluded that the optimal synthesis temperature for the BaBi₄TiNbO₁₁Cl oxyhalides is 650 °C. To reveal the morphological characteristics of the as-prepared samples, SEM was applied to the BaBi₄TiNbO₁₁Cl oxyhalide obtained at 750 °C. It can be observed that the sample displayed irregular rectangular nanoplates, as shown in Fig. 2b. This phenomenon is ascribed to the melted NaCl/CsCl molten salts, which provide fast transfer channels for the reaction elements and motivates the formation of high-purity and uniform reaction products. Furthermore, TEM was used to analyze the microstructure characteristics of the oxyhalide material. The TEM image in Fig. 2c shows that BaBi₄TiNbO₁₁Cl possessed the typical layered stack structure. Then, the area inside the yellow box was magnified. The corresponding high-resolution TEM image of this sample in Fig. 2d shows that it possesses uniform and clear lattice fringes of 0.276 nm, consistent with the d-spacing value of the (210) crystal plane. In addition, according to the TEM-EDS elemental mappings in Fig. 2g–l, Ba, Bi, Ti, Nb, O, and Cl, respectively, are uniformly distributed throughout the nanoplate, confirming that these elements are present on the photocatalysts and the successful preparation of BaBi₄TiNbO₁₁Cl.

Fig. 2 (a) XRD patterns of BaBi₄TiNbO₁₁Cl sample calcined from MSS in the temperature range of 625–675 °C for 12 h and (b) corresponding SEM image of BaBi₄TiNbO₁₁Cl prepared at the calcination temperature of 650 °C. (c) TEM and (d) HRTEM images of BaBi₄TiNbO₁₁Cl. (e) Raman spectra of BaBi₄TiNbO₁₁Cl and BaBi₄TiNbO₁₁Br. (f) XPS peaks for Cl 2p in BaBi₄TiNbO₁₁Cl and BaBi₄TiNbO₁₁Br samples. The corresponding TEM-EDS elemental mappings of (g) Ba, (h) Bi, (i) Ti, (j) Nb, (k) O, and (l) Cl in (c).

Afterwards, the raw material BaBiO₂Cl was replaced with BaBiO₂Br to grow BaBi₄TiNbO₁₁Br oxyhalides via the same process. The crystal lattice structure and diffraction patterns are shown in Fig. S1.† The diffraction peaks of the samples are consistent with the DFT results for BaBi₄TiNbO₁₁Br, suggesting that no impurities were formed in BaBi₄TiNbO₁₁Br at 650 °C. Combined with the SEM and TEM results in Fig. S2 and S3,† respectively, it was proven that BaBi₄TiNbO₁₁Br was successfully synthesized. In addition, the vibrational signature of BaBi₄TiNbO₁₁Br and BaBi₄TiNbO₁₁Cl was described in the range of 100 to 900 cm⁻¹ by Raman spectroscopy, as shown in Fig. 2e. The Raman shifts at 160 and 455 cm⁻¹ can be related to the Bi–O bond, while that at 223, 455, and 685 cm⁻¹ are ascribed to the Bi–X (X = Cl, Br) bond, [Bi₂O₂]²⁺ layers, and Bi–Ti/Nb–O bonds, respectively.^37–39 Meanwhile, the Raman shifts at 560 and 826 cm⁻¹ are assigned to the Nb–O bonds in the NbO₆ octahedra, further confirming the successful synthesis of the BaBi₄TiNbO₁₁X (X = Cl, Br) oxyhalides.

However, the Cl element was found in the TEM-EDS elemental mappings of BaBi₄TiNbO₁₁Br, as shown in Fig. S4.† Simultaneously, the component of the Cl element, as presented in Table S1 (the components of oxyhalide materials from TEM-EDS in Fig. S4),† is about 6.87%, which is about 15.27 times higher than that of the Br element. This indicates that both Cl and Br elements are present in the BaBi₄TiNbO₁₁X oxyhalides prepared with BaBiO₂Br and Bi₃TiNbO₉ as the precursor materials and NaCl/CsCl as the molten salt. This phenomenon is consistent with the results from the XPS pattern shown in Fig. 2f. Based on previous conclusions, the molten salt only acts as a solvent to provide a liquid phase environment and facilitate the full diffusion of the elements.^10,28,29 However, this contradicts the actual results obtained in this work. The reason for this phenomenon is that a large number of Cl elements from the NaCl/CsCl molten salt substituted the halogen elements (Br) in the precursor of BaBiO₂Br to participate in the formation of the BaBi₄TiNbO₁₁X oxyhalides.

The crystal structure of BaBi₄TiNbO₁₁X is presented in Fig. 3a. It can be found that BaBi₄TiNbO₁₁X (X = Cl, Br) is a Sillén–Aurivillius perovskite oxyhalide material, which consists of a unit cell comprised of slabs of fluorite-type [Bi₂O₂]²⁺ layer inter-grown with [BiTiNbO₇]²⁻, [BaBiO₂]⁺, and [X]⁻ layers along the P2₁cn space group. Simultaneously, along the c-axis direction, the different layers between [BaBiO₂]⁺ and [Bi₂O₂]²⁺ are connected by X ions via van der Waals forces.⁴⁰ As is known, van der Waals forces, which are formed from the distribution of electric charges, are one of the weakest forces between molecules.⁴¹ Therefore, during the preparation of the Sillén–Aurivillius perovskite oxyhalide by high-temperature calcination in previous works, the halogens, such as Cl, Br, and I, are easily lost, forming defects.^16,22,42 Thus, to further confirm the influence of the molten salts on the Sillén–Aurivillius perovskite oxyhalide, different molar ratios of precursor materials were used to fabricate BaBi₄TiNbO₁₁X, and further details on the precursors and naming are shown in Table S1.† In the case of BaBi₄TiNbO₁₁Cl, the raw materials of BaBiO₂Cl and Bi₃TiNbO₉ were used to fabricate BaBi₄TiNbO₁₁X with molten salts of NaCl/CsCl. For the other samples, the raw materials were BaBiO₂Br and Bi₃TiNbO₉ and the content of molten salts of NaCl/CsCl was also decreased from 20 mmol to 0.5 mmol. Furthermore, the BaBi₄TiNbO₁₁Br sample was synthesized from the raw materials of BaBiO₂Br and Bi₃TiNbO₉ based on the molten salts of NaBr/CsBr.

Fig. 3 (a) Crystal structure of Sillén–Aurivillius oxyhalide BaBi₄TiNbO₁₁X. (b) XRD patterns and (c) FTIR spectra of BaBi₄TiNbO₁₁X samples with different molar ratios of precursor materials. (d) XPS survey spectra of pristine BaBi₄TiNbO₁₁X samples. XPS peaks of (e) Cl 2p and (f) Br 3d core levels.

Then, the crystal structures and phase purity of the as-prepared samples were characterized by XRD, as shown in Fig. 3b and S5.† The main peaks located at the 2θ values of (014), (210), and (414) for all the samples are consistent with the XRD pattern calculated using DFT, implying that the synthetic BaBi₄TiNbO₁₁X possessed high crystallinity and purity. In addition, the FTIR spectra of the as-synthesized BaBi₄TiNbO₁₁X samples showed primary absorption bands in the range of 400–4000 cm⁻¹, as depicted in Fig. 3c. The peaks observed at about 3442 cm⁻¹, 2927 cm⁻¹, and 1629 cm⁻¹ can be attributed to the O–H stretching vibration.⁴³ The Bi–O stretching vibrations were identified based on the peak at 586 cm⁻¹.⁴⁴ The other peak located at 808 cm⁻¹ can be attributed to the Nb–O–Nb stretching vibration.⁴⁵ The similar peaks in the FTIR spectra of the BaBi₄TiNbO₁₁X samples fabricated using different molar ratios of precursor materials suggest that the functional groups in the Sillén–Aurivillius perovskite oxyhalide did not change.

XPS analysis was performed to investigate the surface chemical state of the elements and confirm the effect of the molten salts on the formation of the Sillén–Aurivillius perovskite oxyhalides. The survey scan of all the samples, as shown in Fig. 3d, show that the BaBi₄TiNbO₁₁X perovskite oxyhalide is composed of Ba, Bi, Ti, Nb, O, and Cl (or Br) elements, and the high-resolution XPS spectra of the evolution behaviors of C 1s, Ba 3d, Bi 4f, Ti 2p, Nb 3d, O 1s, and Cl 2p (Br 3d) are also presented in Fig. 3e, f and S6–S11.† The binding energy (BE) of the BaBi₄TiNbO₁₁X samples was calibrated by the C 1s peaks at 284.8 eV, as shown in Fig. S6.† The peaks of Ba 3d_5/2, Ba 3d_3/2, Bi 4f_7/2, Bi 4f_5/2, Ti 2p_3/2, Nb 3d_5/2, Nb 3d_3/2, O 1s, Cl 2p_3/2, Cl 2p_1/2, Br 3d_5/2, and Br 3d_3/2 were located at approximately 779.5 eV, 794.8 eV, 159.0 eV, 164.3 eV, 458.0 eV, 206.6 eV, 209.3 eV, 529.7 eV, 198.1 eV, 199.7 eV, 68.5 eV, and 69.6 eV, respectively, which further confirm the existence of Ba²⁺, Bi³⁺, Ti⁴⁺, Nb⁵⁺, O²⁻, Cl⁻, and Br⁻ in BaBi₄TiNbO₁₁X. We also found that with an increase in the content of Br in the precursor materials, the intensity of the Cl 2p peak gradually decreased, as shown in Fig. 3e, while the Br 3d peak increased correspondingly, as shown in Fig. 3f, and these results suggest that the content of Br in the BaBi₄TiNbO₁₁X perovskite oxyhalide material halogen oxide product is related to the proportion of [Cl]/[Br] in the precursor materials. Simultaneously, the x value in the BaBi₄TiNbO₁₁(Cl_1−xBr_x) samples was calculated from the XPS peak table based on the content of elements. The x value of the BaBi₄TiNbO₁₁X samples was calculated to be 0, 9.8, 35.6, 50.8, and 100, which is close to that of the Br content in the precursor materials (0, 9, 33.3, 50, and 100), respectively. This calculated result suggests that the ratio of halogens in the precursor materials (molten salts and solutes) determined the x value of the final product. Namely, the stoichiometric ratios of the samples with different halogen molar ratios of precursor materials ([Br]/([Br] + [Cl])) (0%, 9%, 33.3%, 50%, and 100%) was BaBi₄TiNbO₁₁Cl, BaBi₄TiNbO₁₁(Cl_0.91Br_0.09), BaBi₄TiNbO₁₁(Cl_0.67Br_0.33), BaBi₄TiNbO₁₁(Cl_0.5Br_0.5), and BaBi₄TiNbO₁₁Br, respectively.

3.2. Halogen-competing reaction in BaBi₄TiNbO₁₁(Cl_1−xBr_x) Sillén–Aurivillius perovskite oxyhalides

In general, molten salts, such as NaCl/CsCl and NaBr/CsBr, can provide a special liquid reaction environment that expedites the diffusion rate of reactive elements and induces the formation of high-purity and uniform reaction products. During this process, the molten salt only plays an auxiliary role in the synthesis of the materials and does not participate in the reaction. According to our previous work,²² a suitable calcination temperature can facilitate the evolution and formation of layered BaBi₄TiNbO₁₁X perovskite oxyhalide via the MSS method in a muffle furnace. Herein, a novel method was proposed using the BaBiO₂X structure as the template to form the BaBi₄TiNbO₁₁(Cl_1−xBr_x) perovskite oxyhalide by intercalating the [BiTiNbO₇] octahedron between the [BaBiO₂]⁺ layer and the halogen [X]⁻ atomic layer through the MSS method, as depicted in Fig. 4a. Due to the proportion of [Cl]/[Br] in the precursor materials affecting the x value, the growth mechanisms for BaBi₄TiNbO₁₁(Cl_1−xBr_x) were proposed to explain the halogen-competing reactions in the perovskite halogens, as shown in Fig. 4b–g.

Fig. 4 (a) Schematic illustration of the material intercalation process of Sillén–Aurivillius BaBi₄TiNbO₁₁X oxyhalide. (b–g) Schematic illustration of halogen-competing reaction in BaBi₄TiNbO₁₁(Cl_1−xBr_x) Sillén–Aurivillius perovskite oxyhalide.

Herein, the BaBiO₂Br and Bi₃TiNbO₉ solutes and NaCl/CsCl solvents (molten salts) were used as representatives to clarify the halogen diffusion path during the formation of BaBi₄TiNbO₁₁(Cl_1−xBr_x). If the NaCl/CsCl molten salts only provide the liquid reaction environment and do not participate in the chemical reaction, the final production of perovskite oxyhalide is BaBi₄TiNbO₁₁Br, as shown in Fig. 4b and c. However, the halogen [X] layer in the crystal structure of the Sillén–Aurivillius perovskite oxyhalide connects the [BaBiO₂]⁺ and [Bi₂O₂]²⁺ layers by van der Waals force along the c-axis direction. van der Waals forces are one of the weakest forces between molecules, which induce the formation of unstable [X] atoms in the halogen layer and cause [X] vacancies (V_X), and thus some Br atoms spill out of the Sillén–Aurivillius perovskite oxyhalide into NaCl/CsCl molten salts (process I). In addition, during the high-temperature calcination process, a large number of molten salt atoms, such as Na, K, and Cl, surrounded the synthetic BaBi₄TiNbO₁₁Br products. Given that the Cl and Br atoms are located in the same main group and have similar chemical properties, the highly active Cl atoms will diffuse from the molten salt and occupy the Br atomic sites (V_X) in the halogen oxides (process II), as described in Fig. 4d and e. Simultaneously, the high calcination temperature of the samples also promotes the migration of atoms currently located in the halogen lattice site into the molten salts (process III). Furthermore, the halogen atoms that migrate to the molten salts, such as Br, may also transition to V_X again (process IV). Processes I–V occur continuously between the molten salts and perovskite oxyhalide during the high-temperature calcination process, resulting in the same proportion of [Cl]/[Br] in the molten salts as that in the halogen [X] atomic layer, as presented in Fig. 4f and g. Briefly, the weak van der Waals forces between the [BaBiO₂]⁺ and [Bi₂O₂]²⁺ layers of BaBi₄TiNbO₁₁(Cl_1−xBr_x) induce the full interaction of halogens to diffuse inside and outside the perovskite oxyhalide lattice, which form an even distribution of halogen atoms between the molten salts and the halogen layers.

3.3. Optical and electrochemical properties of Sillén–Aurivillius perovskite oxyhalide

Generally, the band gap (E_g) of photocatalysts plays a significant role in the efficiency of their catalytic performance, especially for light absorption and the recombination of photogenerated carriers.²² Thus, the UV-vis spectra of the BaBi₄TiNbO₁₁X samples were recorded to analyze their optical properties. As demonstrated in Fig. 5a, the absorption edges of all the BaBi₄TiNbO₁₁X samples are approximately 460 nm, indicating that the BaBi₄TiNbO₁₁X oxyhalide is a type of visible light photocatalyst and halogen substitution (from Cl to Br) has little effect on its light absorption. To better compare the E_g values of the samples with different halogen contents, the conventional Kubelka–Munk function was applied to the BaBi₄TiNbO₁₁X oxyhalides, as follows:^9,30,46

(αℏν)^1/m = A(ℏν − E_g) (1)

where ℏ, α, and ν are the Planck constant, absorption coefficient, and light frequency, respectively. The m value of the samples was correlated with the transition properties of the semiconductor materials. According to our previous work, the BaBi₄TiNbO₁₁X oxyhalide is an indirect bandgap semiconductor, where m is equal to 2. Accordingly, the calculated E_g of all the samples was fitted to be 2.510 eV, 2.513 eV, 2.518 eV, 2.521 eV, and 2.579 eV, as displayed in Fig. 5b, reflecting that the E_g value of BaBi₄TiNbO₁₁(Cl_1−xBr_x) increased with the x value (by substituting Br for Cl in oxyhalide material). One possible reason for this phenomenon is that the ionic radius of Br is larger than that of Cl. In addition, the enlarged E_g from BaBi₄TiNbO₁₁Cl to BaBi₄TiNbO₁₁Br is only about 0.069 eV (2.579 eV − 2.5210 eV = 0.069 eV), indicating that a variation in the halogen has little effect on the E_g of Sillén–Aurivillius oxyhalides.

Fig. 5 (a) UV-vis diffuse reflectance spectra (DRS) of different BaBi₄TiNbO₁₁X samples. (b) Band gaps of the pristine samples. (c) Mott–Schottky curves of BaBi₄TiNbO₁₁(Cl_0.5Br_0.5). (d) Corresponding band structures of the different samples. (e) Transient photocurrent densities and (f) EIS Nyquist plots of the different samples under VL irradiation.

Afterwards, the energy band structure of the as-prepared BaBi₄TiNbO₁₁(Cl_1−xBr_x) material was determined based on the Mott–Schottky (M–S) curves between the liquid and semiconductor material. As presented in Fig. 5c and S12,† all the sample curves show positive slopes of capacitance, and this the BaBi₄TiNbO₁₁X oxyhalides are n-type semiconductors.⁴⁷ According to the M–S diagram, the flat band (E_fb) potential of the BaBi₄TiNbO₁₁Cl, BaBi₄TiNbO₁₁(Cl_0.91Br_0.09), BaBi₄TiNbO₁₁(Cl_0.67Br_0.33), BaBi₄TiNbO₁₁(Cl_0.5Br_0.5), and BaBi₄TiNbO₁₁Br samples were calculated to be −0.267, −0.275, −0.281, −0.282, and −0.287 eV vs. Ag/AgCl, respectively. According to previous works,^22,48 the conduction band (CB) of a semiconductor material is about 0.1 eV more negative than that of its E_fb position. Based on this, the CBM of the BaBi₄TiNbO₁₁Cl, BaBi₄TiNbO₁₁(Cl_0.91Br_0.09), BaBi₄TiNbO₁₁(Cl_0.67Br_0.33), BaBi₄TiNbO₁₁(Cl_0.5Br_0.5), and BaBi₄TiNbO₁₁Br samples was −0.367, −0.375, −0.381, −0.382, and −0.387 eV, respectively. The valence band (VB) potential can be calculated using the following equation:

E_VB = E_CB + E_g (2)

where E_CB and E_VB represent the CB and VB potentials, respectively. Hence, the calculated E_VB of the BaBi₄TiNbO₁₁Cl, BaBi₄TiNbO₁₁(Cl_0.91Br_0.09), BaBi₄TiNbO₁₁(Cl_0.67Br_0.33), BaBi₄TiNbO₁₁(Cl_0.5Br_0.5), and BaBi₄TiNbO₁₁Br samples was 2.143, 2.138, 2.137, 2.139, and 2.192 eV, respectively. The band positions of the BaBi₄TiNbO₁₁X samples are outlined in Fig. 5d. Clearly, the E_VB of the semiconductor material is more positive than the redox potential of H₂O/O₂ at 0.82 eV, and the photo-generated holes can react with H₂O to form O₂. Therefore, it can be deduced that the BaBi₄TiNbO₁₁(Cl_1−xBr_x) oxyhalides are suitable photocatalysts for O₂ evolution.

Simultaneously, transient photocurrent and electrochemical impedance spectroscopy (EIS) measurements were performed to disclose the separation and transport performance of photo-generated charges (electrons and holes). The diffused electrons from photoexcited carrier separation can generate a photocurrent signal, and thus transient photocurrent measurement can be used to demonstrate the separation efficiency of charge carriers. The photocurrent responses of all the samples, as shown in Fig. 5e, indicate that the BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) sample has the highest photocurrent intensity and the fastest charge separation efficiency in response to visible light, which can be related to the substituting of Br for Cl in the BaBi₄TiNbO₁₁(Cl_1−xBr_x) oxyhalides, and more details about this result will be discussed through DFT calculations later. Then, the transfer behavior of electron–hole pairs was examined by EIS, as shown in Fig. 5f. Among the samples, BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) also displayed the smallest Nyquist semicircle diameter, which reveals the lowest charge transfer resistance of the BaBi₄TiNbO₁₁(Cl_1−xBr_x) oxyhalides. The photoelectrochemical results of transient photocurrent and EIS confirm that the energy band structure of BNB is conducive to facilitate the separation and transfer performance of the photogenerated charge carriers, which is expected to promote the photocatalytic reactions for BaBi₄TiNbO₁₁(Cl_0.5Br_0.5).

3.4. Photocatalytic activities of BaBi₄TiNbO₁₁(Cl_1−xBr_x) oxyhalides

The photocatalytic performance of the as-prepared BaBi₄TiNbO₁₁(Cl_1−xBr_x) photocatalysts was assessed using the time course of O₂ evolution in aqueous 5 mM AgNO₃ solution under visible light irradiation. As indicated in Fig. 6a, the BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) sample exhibited the best photocatalytic oxygen evolution efficiency among samples, indicating that halogen substitution in Sillén–Aurivillius oxyhalide enhances their photo-absorption, resulting in a more efficient photocatalytic performance. For a clearer comparison of the differences in performance, the corresponding photocatalytic rates of O₂ production were estimated, as shown in Fig. 6b. BaBi₄TiNbO₁₁Cl and BaBi₄TiNbO₁₁Br showed an O₂ production rate of 2.64 and 3.57 μmol g⁻¹ h⁻¹, respectively. Comparatively, the BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) sample displayed the highest photocatalytic activity among the oxyhalide photocatalysts, and its rate is 4.21 μmol g⁻¹ h⁻¹. This enhanced photocatalytic performance can be ascribed to the excellent separation and transfer of photogenerated charges, as exhibited in Fig. 5.

Fig. 6 (a) Oxygen evolution efficiency of different BaBi₄TiNbO₁₁X samples. (b) Corresponding calculated O₂ production rates of the pristine samples. (c) Photocatalytic degradation performance of RhB via BaBi₄TiNbO₁₁(Cl_1−xBr_x) samples under VL irradiation. (d) Deduced ln(C₀/C) plots of samples based on first-order reaction kinetics. (e) Corresponding k values of the samples. (f) Degradation rate of RhB over BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) with the addition of BQ, EDTA-2Na, and TBA.

Furthermore, the degradation of 5 mg per L rhodamine B (RhB) under visible light irradiation was also employed to evaluate the photocatalytic activities of the Sillén–Aurivillius oxyhalide. Fig. 6c shows the photocatalytic efficiencies for the degradation of RhB by the BaBi₄TiNbO₁₁(Cl_1−xBr_x) samples, where the kinetic analysis of RhB photo-degradation was followed using the equation η = (C₀ − C)/C₀ × 100% (C₀ and C are the initial and reacted concentration of RhB solution at different sampling times). It can be seen that the catalyst-free samples exhibited no obvious variation over time, which means that the RhB organic pollutant is relatively stable under solar irradiation. Although all the samples could degrade the RhB pollutant within 90 min, respectively, a significant enhancement in photocatalytic efficiency for the BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) sample was observed. Then, the reaction kinetic constants of the different samples in Fig. 6d were determined according to the following equation:

Ln(C₀/C) = kt (3)

where k is the fitted pseudo-first-order kinetic equation coefficient of the samples. The degradation efficiency for RhB solution of the BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) sample in Fig. S13† is 0.2326 min⁻¹, which is about 6.05 and 3.06 times higher than that of BaBi₄TiNbO₁₁Cl (0.03848 min⁻¹) and BaBi₄TiNbO₁₁Br (0.07611 min⁻¹) samples, respectively. Simultaneously, this result is much higher than that of commercial photocatalyst P25,²² demonstrating that the Sillén–Aurivillius oxyhalides BaBi₄TiNbO₁₁X (X = Cl, Br) mixed with halogen is a potential commercial photocatalyst material. This significant promotion of degradation efficiency for BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) is due to the enhanced separation and transfer of photogenerated charges, similar to O₂ evolution in 5 mM AgNO₃ aqueous solution under visible light irradiation, and more details about the contribution of the halogens based on DFT calculation are presented in a later section.

The degradation mechanisms of organic pollutants were investigated by radical scavenging experiments. Each scavenger was added before the photodegradation experiment. Among the many scavengers, TBA, BQ, and EDTA-2Na were incorporated as a quencher of hydroxyl radicals, singlet oxygens, and holes, respectively, and the process is similar to that for degrading RhB solution. The results of the trapping experiment are provided in Fig. S14.† After the addition of the EDTA-2Na quencher, the degradation efficiency for the organic pollutant decreased sharply, and it was only degraded by 16.9% in 120 min. This observation in the trapping experiment suggests that h⁺ as the active species plays a major role in the RhB degradation process. Thus, the degradation of organic pollutants for Sillén–Aurivillius oxyhalide materials is driven by photocatalytic oxidation rather than dye sensitization in this work. Furthermore, the degradation efficiency of the RhB solution also decreased by half after the addition of BQ, which reveals that the ˙O₂⁻ radical was generated and played a role in the process. Meanwhile, it can be found that the photocatalytic degradation efficiency of RhB has no significant change when TBA was supplied as the scavenger. Hence, the ˙OH radicals hardly affected the degradation process of RhB solution. The reason for this phenomenon is that the redox potential of H₂O/˙OH (2.73 eV) is more positive than the E_VB positive of BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) (2.139 eV) in the reaction solution, and thus the photogenerated h⁺ cannot drive the production of ˙OH radicals, as revealed in Fig. 5d. Furthermore, electron spin resonance (ESR) was further employed to evaluate the active species (˙OH and ˙O₂⁻) during the photocatalytic process. By adding the trapping agent of DMPO, the radicals of ˙OH and ˙O₂⁻ were captured. As depicted in Fig. 6c and d, the signals of the ˙OH and ˙O₂⁻ radicals were not detected in the dark (0 min). When light was applied to the surface of the photocatalyst, the ˙OH radicals had no significant change and the ˙O₂⁻ radicals were clearly observed. This phenomenon proves that the ˙O₂⁻ radicals played an important role in the photocatalytic process, which is in agreement with the results of the trapping experiment.

3.5. Photocatalytic activities in BaBi₄TiNbO₁₁(Cl_1−xBr_x) oxyhalides

Herein, we studied the electronic properties of three Sillén–Aurivillius oxyhalides, namely, BaBi₄TiNbO₁₁X (X = Br, Cl, and Cl_0.5Br_0.5). The partial density of states (PDOS), the highest occupied crystal orbital (HOCO), and the lowest unoccupied crystal orbital (LUCO) of BaBi₄TiNbO₁₁Br are shown in Fig. 7a–c, while that for BaBi₄TiNbO₁₁Cl is given in Fig. 7d–f and BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) in Fig. 7g–i, respectively. To account for both the Cl and Br in the Sillén–Aurivillius oxyhalides, we constructed a superlattice from 38 atoms with one Cl and one Br atom, as shown in Fig. 7h. The peaks of the PDOS in Fig. 7a show that the high energy-occupied bands are mainly contributed by the O atoms, with very low contribution from the Br and Ti atoms. Alternatively, the unoccupied bands (hole states) are mainly contributed by the Bi atoms, followed by a lower contribution by O and Br atoms. The same was also observed from the PDOS of BaBi₄TiNbO₁₁Cl, as shown in Fig. 7d. The distributions of HOCO on the O and Ti atoms (see HOCO in Fig. 7b and e) and the LUCO on the Bi and O atoms (see LUCO Fig. 7b and e) confirm the results obtained from PDOS.

Fig. 7 Partial density of states (a), highest occupied/lowest unoccupied crystal orbital (HOCO/LUCO) (b), and scanning tunneling microscopy (STM) image (c) of BaBi₄TiNbO₁₁Br. The same for BaBi₄TiNbO₁₁Cl are given in (d–f) and for BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) (g–i), respectively.

Regarding the band gap, it was noted that the band gap decreases from E_g = 2.56 eV in BaBi₄TiNbO₁₁Br to E_g = 2.41 eV in BaBi₄TiNbO₁₁Cl. The lower electronegativity of Br compared to that of Cl leads to a higher electronegativity difference (between O and Br atoms) in BaBi₄TiNbO₁₁Br, which increases its ionicity, decreases the orbital overlap, and eventually increases the band gap. Thus, by decreasing the electronegativity of the halogen, the band gap decreases. We tested this assumption by calculating the energy gaps of BaBi₄TiNbO₁₁F and BaBi₄TiNbO₁₁I, which were found to be 2.21 and 2.93 eV, respectively.

The presence of two halogens with equal ratios as in BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) resulted in noticeable changes in the structure and electronic properties of the samples. It was observed from the supercell, as shown in Fig. 7h, that the O atoms around the transition metals (Ti and Nb) form new chemical bonds with the Bi atoms, which did not exist in the structures with a single halogen. These new bonds enhanced the contribution of the O-atoms (see the lower part of the supercell in Fig. 7h) to the HOCO, and consequently boosted the interactivity of these atoms. Another difference observed in the LUCO is that it is distributed on the Bi atoms on both sides of the halogens (see LUCO in Fig. 7h). Therefore, incorporating two halogens increases the contribution of the different O and Bi atoms in HOCO and LUCO, respectively, and then increases the interactivity of the whole structure. According to the density of states, the band gap of BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) equals 2.54 eV, which is higher than that of BaBi₄TiNbO₁₁Cl but lower than that of BaBi₄TiNbO₁₁Br, in accordance with the experimental results.

Imaging surfaces using STM is an important characterization tool given that it provides important information about the material structure. The STM images of the three Sillén–Aurivillius compounds are shown in Fig. 7c, f and i. It can be seen that the O-atoms produce bright spots in the images, which indicate that their electronic states dominate the occupied levels. Also, around the Bi- and Nb-atoms, empty black spots inside bright ones are observed, which correspond to the Bi- and Nb-atoms between the O-atoms, respectively. The Br, Cl, and Cl_0.5Br_0.5 halogens could also be identified by STM through their less bright but recognizable spots, as seen in the plots. The brightest spots observed in all the cases (see the bright spot surrounded by lower bright four spots in the upper part of Fig. 7c) represent the electronic density from both the Ti- and O-atoms.

3.6. Proposed mechanism for the degradation of BaBi₄TiNbO₁₁X oxyhalides

According to the above-mentioned experimental and calculated results, the potential mechanism for the enhancement in the photo-catalytic performance of the BaBi₄TiNbO₁₁X oxyhalides is shown in Fig. 8. Under visible light irradiation, photons with energy larger than the E_g of the semiconductor can produce e⁻–h⁺ pairs in the semiconductor material (eqn (4)). Afterward, the e⁻ is excited to the CB, and the h⁺ is transported to the VB of the semiconductor material. The hybridization of the halogens in oxyhalides can increase the contribution to the low-energy orbitals and facilitate the separation and transfer performance of the photogenerated charge carriers, promoting the photocatalytic reaction. Given that the E_VB of the semiconductor material (about 2.14 eV) is more positive than that of the H₂O/O₂ redox potential (0.82 eV), and the photo-generated h⁺ can promote the reaction of H₂O to form O₂ (eqn (5)), the excess e⁻ is quenched by the adjuvant material. This is the mechanism for the photocatalytic process of O₂ evolution by BaBi₄TiNbO₁₁X.

Fig. 8 Possible photocatalytic mechanisms of BaBi₄TiNbO₁₁X under visible light irradiation.

Furthermore, given that the E_CB of BaBi₄TiNbO₁₁X at about −0.38 eV is more negative than the redox potential of O₂/˙O₂⁻ (−0.33 eV vs. NHE), the excited e⁻ will combine with O₂ to form ˙O₂⁻ (eqn (7)) during the degradation process of RhB pollutant. Simultaneously, the redox potential of H₂O/˙OH (2.73 eV) is more positive than the E_VB of the semiconductor material, as shown in Fig. 5d, indicating that H₂O can be converted to ˙OH, which was confirmed by the radical scavenging experiments. Finally, the e⁻ in the CB, the generated ˙O₂⁻, and the h⁺ in the VB will participate in the redox reaction to degrade the organic pollutant molecules (eqn (8)). This is the mechanism for the photocatalytic degradation of organic pollutants by the BaBi₄TiNbO₁₁X oxyhalides. Based on the above-mentioned analyses, the photocatalytic mechanism of the BaBi₄TiNbO₁₁X oxyhalides can be described as follows:

BaBi₄TiNbO₁₁X + hν → e⁻ + h⁺ (4)

2H₂O + 4h⁺ → O₂ + 4H⁺ (5)

e⁻ + adjuvant → H₂O (6)

O₂ + e⁻ → ˙O₂⁻ (7)

˙O₂⁻ + h⁺ + e⁻ + organic pollutant → intermediate product + CO₂ + H₂O (8)

4. Conclusions

In this contribution, novel Sillén–Aurivillius BaBi₄TiNbO₁₁X (X = Cl, Br) oxyhalides with a tunable halogen ratio were prepared via the molten-salt synthesis method. Both Cl and Br elements were detected in BaBi₄TiNbO₁₁Br fabricated with BaBiO₂Br and Bi₃TiNbO₉ as precursor materials and NaCl/CsCl as the molten salt demonstrated that the molten salt not only provides a liquid reaction environment but also participates in the chemical reaction. This is ascribed to the weak van der Waals forces between the [BaBiO₂]⁺ and [Bi₂O₂]²⁺ layers of BaBi₄TiNbO₁₁(Cl_1−xBr_x), which induced the full interaction of halogens to diffuse inside and outside the perovskite oxyhalide lattice, forming an even distribution of halogen atoms between the molten salts and the halogen layers in oxyhalides. By employing this mechanism, BaBi₄TiNbO₁₁(Cl_1−xBr_x) with different x contents could be easily obtained. Simultaneously, DFT calculations were performed to investigate the structural and electronic properties of the three Sillén–Aurivillius oxyhalides, namely, BaBi₄TiNbO₁₁X (X = Br, Cl, and Cl_0.5Br_0.5). The HSE calculations showed that the structures are semiconductors with band gaps equal to 5.41, 2.54, and 2.56 eV, respectively, which is agreement with the experimental E_g of the samples. The partial density of states and the visualization of the highest occupied crystal orbitals show that the low energy occupied orbitals are mainly from O, and then from the Ti atoms. Alternatively, the low energy unoccupied ones are contributed mainly by Bi followed by O atoms. Employing two chalcogens as in BaBi₄TiNbO₁₁Cl_0.5Br_0.5 led to the formation of additional Bi–O bonds and slight change in structure. These additional bonds enhanced the contributions from the different O/Bi atoms to the highest occupied/lowest unoccupied orbitals, which boosted the interactivity of the whole structure. The simulated scanning tunneling microscopy images were used for the characterization and identification of the prepared samples. Thus, the BaBi₄TiNbO₁₁(Cl_0.5Br_0.5) sample exhibited an excellent RhB degradation performance compared with the commercial photocatalyst P25. Meanwhile, it also showed a significant photocatalytic O₂ evolution performance owning to the reinforced separation and transfer performance of the photogenerated charge carriers. By the rational design of the energy band structure via the interlayer halogen-competing reaction induced by the precursor molten salts, this work developed an exceptional Sillén–Aurivillius photocatalyst family under visible light irradiation.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (No. 12274361), the Natural Science Foundation of Jiangsu Province (BK20211361), the Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX21_3145) and the school-level research projects of Yancheng Institute of Technology (xjr2019028, xjr2019059).

Notes and references

1. H. Mai, D. Chen, Y. Tachibana, H. Suzuki, R. Abe and R. A. Caruso, Chem. Soc. Rev., 2021, 50, 13692–13729.
2. Y. Zhang, L. Shi, Z. Wang, H. Dai, Z. Hu, S. Zhou, H. Chen, X. Feng, J. Zhu, Y. Sun, W. Liu and Q. Zhang, Sol. Energy, 2021, 227, 334–342.
3. Y. Zhang, Y. Zhang, X. Chen, S. Wang, Q. Gao, M. Wu, Z. Wang, J. Ao, Y. Sun, W. Liu and Q. Zhang, Mater. Sci. Semicond. Process., 2022, 140, 111569.
4. C. Sun, J. Yang, M. Xu, Y. Cui, W. Ren, J. Zhang, H. Zhao and B. Liang, Chem. Eng. J., 2022, 427, 131564.
5. C. Liu, Y. Zhang, J. Wu, H. Dai, C. Ma, Q. Zhang and Z. Zou, J. Mater. Sci. Technol., 2022, 114, 81–89.
6. X. Lu, L. Quan, H. Hou, J. Qian, Z. Liu and Q. Zhang, J. Alloys Compd., 2022, 925, 166552.
7. H. Zhang, J. Cai, Y. Wang, M. Wu, M. Meng, Y. Tian, X. Li, J. Zhang, L. Zheng, Z. Jiang and J. Gong, Appl. Catal., B, 2018, 220, 126–136.
8. S. Lee and S. Park, J. Ind. Eng. Chem., 2013, 19, 1761–1769.
9. J. Chen, Y. Gu, S. Xu, Y. Zhang, Z. Zhang, L. Shi, Z. Mu, C. Zhou, J. Zhang and Q. Zhang, Catalysts, 2023, 13, 751.
10. X. Tao, Y. Zhao, L. Mu, S. Wang, R. Li and C. Li, Adv. Energy Mater., 2018, 8, 1701392.
11. J. Akter, M. A. Hanif, M. A. Islam, K. P. Sapkota, I. Lee and J. R. Hahn, J. Environ. Chem. Eng., 2021, 9, 106831.
12. Y. Chen, W. Zhong, F. Chen, P. Wang, J. Fan and H. Yu, J. Mater. Sci. Technol., 2022, 121, 19–27.
13. X. Sun, L. Shi, H. Huang, X. Song and T. Ma, Chem. Commun., 2020, 56, 11000–11013.
14. Z. Zhu, F. Chen, N. Tian, Y. Zhang and H. Huang, Curr. Opin. Green Sustainable Chem., 2022, 37, 100669.
15. A. Kumar, A. Kumar and V. Krishnan, ACS Catal., 2020, 10, 10253–10315.
16. 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.
17. A. Sharma, S. K. Kanth, S. Xu, N. Han, L. Zhu, L. Fan, C. Liu and Q. Zhang, J. Alloys Compd., 2022, 895, 162576.
18. J. Sun, N. Han, Y. Gu, X. Lu, L. Si and Q. Zhang, Catalysts, 2020, 10, 1425.
19. C. Hu, F. Chen and H. Huang, Angew. Chem., Int. Ed., 2023, e202312895.
20. Z. Zhu, H. Huang, L. Liu, F. Chen, N. Tian, Y. Zhang and H. Yu, Angew. Chem., Int. Ed., 2022, 61, e202203519.
21. M. Banoo, R. S. Roy, M. Bhakar, J. Kaur, A. Jaiswal, G. Sheet and U. K. Gautam, Nano Lett., 2022, 22, 8867–8874.
22. S. Xu, Y. Zhang, Y. Wang, J. Chen, C. Zhou, Z. Mu, Z. Zhang, J. Zhang, J. Wang and Q. Zhang, J. Alloys Compd., 2023, 952, 169932.
23. Y. Gu, F. Yu, J. Chen and Q. Zhang, Catalysts, 2022, 12, 221.
24. L. Shi, X. Wu, C. Xu, Q. Bai, Z. Nie, C. Ji, Y. Zhang, Z. Yin, S. Zhang, X. Qu and F. Du, J. Mater. Chem. C, 2021, 9, 2784–2792.
25. A. Majumdar, U. Ghosh and A. Pal, J. Alloys Compd., 2022, 900, 163400.
26. Y. Xu, Y. You, H. Huang, Y. Guo and Y. Zhang, J. Hazard. Mater., 2020, 381, 121159.
27. Z. Chen, R. Xu, S. Ma, Y. Ma, Y. Hu, L. Zhang, Y. Guo, Z. Huang, B. Wang, Y. Y. Sun, J. Jiang, R. Hawks, R. Jia, Y. Xiang, G. C. Wang, E. A. Wertz, J. Tian, D. Gall, X. Chen, V. Wang, L. Gao, H. Zhu and J. Shi, Adv. Funct. Mater., 2022, 32, 2206343.
28. R. Xu, Y. Wang, D. Cui, K. Du, Z. Shi, H. Zhang, Z. Xu, W. Hao and Y. Du, Sol. RRL, 2022, 6, 2200365.
29. C. Hu, H. Huang, F. Chen, Y. Zhang, H. Yu and T. Ma, Adv. Funct. Mater., 2019, 30, 1908168.
30. Y. Zhang, S. Xu, Z. Mu, K. Liu, J. Chen, C. Zhou, Y. Yao, X. Chen, L. Shi, Z. Wang, Y. Sun, W. Liu and Q. Zhang, Vacuum, 2022, 206, 111569.
31. N. Han, S. Xu and Q. Zhang, J. Mater. Sci., 2022, 57, 2870–2882.
32. G. Kresse and J. Furthmüller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169–11186.
33. G. Kresse and J. Furthmüller, Comput. Mater. Sci., 1996, 6, 15–50.
34. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868.
35. J. Heyd, G. E. Scuseria and M. Ernzerhof, J. Chem. Phys., 2003, 118, 8207–8215.
36. F. Fuchs, J. Furthmüller, F. Bechstedt, M. Shishkin and G. Kresse, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 115109.
37. S. S. Kim, J. W. Kim and C. M. Raghavan, New Phys.: Sae Mulli, 2015, 65, 542–549.
38. A. Prasetyo, B. Mihailova, V. Suendo and A. A. Nugroho, J. Appl. Phys., 2015, 117, 064102.
39. Z. Wei, J. Liu, W. Fang, Z. Qin, Z. Jiang and W. Shangguan, Catal. Sci. Technol., 2018, 8, 3774–3784.
40. X. Wu, C. Zhang, Z. Nie, Y. Zhang, Q. Bai, Z. Yin, S. Zhang, X. Qu, F. Du and L. Shi, Sep. Purif. Technol., 2022, 292, 120969.
41. Q. Li, V. Rudolph, B. Weigl and A. Earl, Int. J. Pharm., 2004, 280, 77–93.
42. N. Han and Q. Zhang, Appl. Surf. Sci., 2021, 562, 150215.
43. X. Qu, M. Liu, H. Zhai, X. Zhao, L. Shi and F. Du, J. Alloys Compd., 2019, 810, 151919.
44. N. Thanabodeekij, E. Gulari and S. Wongkasemjit, Powder Technol., 2005, 160, 203–208.
45. X. Qu, M. Liu, Z. Gao, H. Zhai, W. Ren, L. Shi and F. Du, Appl. Surf. Sci., 2020, 506, 144688.
46. Y. Zhang, Z. Hu, S. Lin, S. Cheng, Z. He, C. Wang, Z. Zhou, Y. Sun and W. Liu, ACS Appl. Energy Mater., 2020, 3, 9963–9971.
47. Z. Liu, H. Wang, J. Duan and B. Hou, Ceram. Int., 2022, 48, 24777–24787.
48. S. Li, B. Xue, J. Chen, Y. Liu, J. Zhang, H. Wang and J. Liu, Sep. Purif. Technol., 2021, 254, 117579.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3ta05438a

---