Dual halogen layers diversify band engineering in Sillén oxyhalide photocatalysts: electronic structure control of SrBi₃O₄X₃ (X = Cl, Br, I) via halogen substitution

Abstract

Halogen substitution from X = Cl to Br or I in a Sillén oxyhalide SrBi₃O₄X₃ with single/double halogen layers results in both narrower bandgaps and more negative conduction band edges for various reduction reactions. This band tuning is achieved as the two halogen layers play distinct roles in shifting the band edges, offering a new strategy for band engineering in Sillén oxyhalides.

---

Sillén-type oxyhalides, consisting of alternating fluorite-type [M₂O₂] layers and halogen [X_n] (n = 1, 2, and 3) layers, possess wide structural diversity and highly controllable physicochemical properties, making them attractive materials in various fields.^1–7 Among these, BiOX (X = Cl, Br, and I) compounds comprising [Bi₂O₂] and double halogen [X₂] layers (Fig. 1a) have been widely investigated as photocatalysts.^6,8,9 The introduction of Br or I with higher p-orbital energies—derived from their lower electron affinities than Cl—leads to a substantial negative shift of the valence band maximum (VBM) (Fig. 1b), providing narrower bandgaps (BiOCl: 3.5 eV, BiOBr: 3.0 eV, BiOI: 2.0 eV) and thereby an ability to harvest a wide range of visible light. However, substituting Cl with Br or I leads to positive shift in the conduction band minimum (CBM),¹⁰ weakening the reducing ability of the BiOX system. This positive shift is presumably due to the polarization effect of adjacent halogen atoms on the Bi-6p orbitals that constitute the CBM (Fig. 1c). Consequently, the simultaneous achievement of visible-light absorption and a sufficiently negative CBM is inherently challenging in the application of BiOX systems in various target reduction reactions. This dilemma has likely limited the number of successful applications of visible-light-responsive BiOBr and BiOI as more attractive alternatives to BiOCl. In fact, owing to its substantially negative CBM, BiOCl exhibits high activities for various reactions, such as water splitting,¹¹ environmental purification,^6,12 and CO₂ reduction,^13,14 but only under UV light.

Fig. 1 Crystal structures and orbital distributions of the VBM (blue dots) and CBM (orange dots), estimated by DFT calculations (isosurface values: 0.04 e Å⁻³ for BiOCl and 0.0075 e Å⁻³ for SrBi₃O₄Cl₃), for (a)–(c) BiOX (X = Cl, Br, I) and (d)–(f) SrBi₃O₄X₃.

In stark contrast, the Sillén-type oxyhalide SrBi₃O₄Cl₃ (Fig. 1d) possesses a narrow bandgap (2.7 eV), even in its Cl form, and exhibits photocatalytic water oxidation activity under visible light.^15,16 As shown in Fig. 1a and d, SrBi₃O₄Cl₃ consists of two types of halogen layers: single [Cl] layers and double [Cl₂] layers, whereas BiOCl contains only double [Cl₂] layers. Density functional theory (DFT) calculations have revealed that the VBM of SrBi₃O₄Cl₃ is mainly composed of Cl-3p orbitals from the [Cl₂] layer, along with O-2p orbitals (Fig. 1e). Despite the similar and dominant contribution of Cl-3p in [Cl₂] layers to the VBMs of both SrBi₃O₄Cl₃ and BiOCl, the VBM of SrBi₃O₄Cl₃ is much more negative (ca. 2.4 V vs. the standard hydrogen electrode (SHE) at pH 2) than that of BiOCl (ca. 3.1 V vs. SHE at pH 2), implying that SrBi₃O₄Cl₃ has an intrinsically favorable framework for a much more negative VBM.¹⁵ The CBM of SrBi₃O₄Cl₃ mainly consists of Bi-6p orbitals adjacent to the single [Cl] layers (Fig. 1f). Recently, we also reported that substituting Cl for the larger I ion in Sillén–Aurivillius-type oxyhalides (e.g., Bi₄NbO₈X) with only a single [Cl] layer can weaken the interlayer Bi–Bi orbital interaction across the single layer, negatively shifting the CBMs.¹⁷ Such interlayer Bi–Bi interactions are also present in the single halogen layers of SrBi₃O₄Cl₃, implying the possibility of a similar negative shift in the CBM. These two findings, i.e., the significant impact of double halogen layers on VBMs and that of single layers on CBMs, motivated us to establish a new band engineering strategy by utilizing two distinct halogen layers in the SrBi₃O₄X₃ system, which cannot be realized in either BiOX with only double [X₂] or Bi₄NbO₈X with only single [X] layers.

While a commercial BiOCl reagent was used, BiOBr and BiOI particles were synthesized via a soft liquid deposition method¹⁸ using Bi(NO₃)₃·5H₂O and NaX (X = Br, I) as starting materials (Fig. S1a). SrBi₃O₄X₃ (X = Cl, Br, or I) was synthesized via a solid-state reaction based on previously reported methods^16,19 (Fig. S1b). The optimized procedures are as follows. SrBi₃O₄Cl₃ was synthesized by calcining a thoroughly mixed precursor consisting of SrCO₃ and BiOCl (1 : 3 molar ratio) at 973 K for 12 h in air, whereas SrBi₃O₄Br₃ was synthesized by calcining a thoroughly mixed precursor consisting of SrCO₃ and BiOBr (1 : 3 molar ratio) at 1073 K for 12 h under vacuum. To prepare SrBi₃O₄I₃, SrI₂, Bi₂O₃, and BiOI were mixed in a 1.1 : 1 : 1 molar ratio, followed by calcination at 1073 K for 12 h under vacuum. X-ray diffraction (XRD) analysis confirmed that all the SrBi₃O₄X₃ samples were obtained as nearly single-phase products (Fig. S1b). Energy-dispersive X-ray spectroscopy indicated that the Sr/Bi and X/Bi ratios in these samples were close to the stoichiometric values, as summarized in Table S1. Their particle morphologies were similar, independent of the halogen species, and exhibited a plate-like shape (Fig. S2).

The band levels of BiOX and SrBi₃O₄X₃ were determined by combining the bandgap estimated from the absorption edges in the UV-vis diffuse reflectance spectra (Fig. 2a and Fig. S3) and flat band potential obtained by Mott–Schottky (MS) plots (Fig. S4). In both systems, the halogen substitution from Cl to Br or I narrowed the bandgap owing to the substantial negative shift of the VBMs. Although the CBMs of BiOX shifted positively upon substitution of Cl with Br and I, as reported previously,¹⁰ those of SrBi₃O₄X₃ shifted negatively. Consequently, and importantly, the CBM level of SrBi₃O₄I₃ is obviously more negative (−0.42 V) than BiOI (−0.32 V) and nearly identical to that of BiOCl (−0.43 V), while the bandgap of SrBi₃O₄I₃ (2.10 eV) is similar to that of BiOI (2.01 eV). Additionally, the bandgap of SrBi₃O₄I₃ is substantially narrower than that of Bi₄NbO₈I (2.44 eV), which has only a single [X] layer. These findings strongly suggest that the substitution of Cl by I in the SrBi₃O₄X₃ system allows both a red-shift of the absorption edge up to around the 600 nm region and suppression of the positive CBM shift, which is difficult to achieve in conventional BiOX and Bi₄NbO₈X systems with only double or single halogen layers, respectively.

Fig. 2 (a) UV-vis diffuse reflectance spectra of SrBi₃O₄X₃ (X = Cl, Br, I). (b) Band levels of BiOX and SrBi₃O₄X₃ estimated from the diffuse reflectance spectra and MS plots.

The origin of the band position changes in SrBi₃O₄X₃ (X = Cl, Br, I) upon halogen substitution was investigated via DFT calculations. The calculated models of SrBi₃O₄X₃ assumed that Bi³⁺ and Sr²⁺ are ordered at the Bi2/Sr2 site in a supercell (Fig. 3a and Fig. S5). The calculated bandgaps using the model decrease in the order of Cl (1.419 eV) > Br (1.229 eV) > I (0.605 eV). This decreasing trend is consistent with the experimental observations (Cl: 2.68 eV > Br: 2.61 eV > I: 2.10 eV). The total and partial densities of state (DOS and PDOS, respectively) for each composition are shown in Fig. 3b, c and Fig. S6–S8. In all cases, the PDOS near the VBM indicated that the halogen atoms (X2) in the double halogen layer mainly contributed to the VBM, with additional contributions from oxygen (O1) (Fig. 3b and Fig. S7, S9). Notably, the VBMs of Br and I are almost entirely composed of X2. To clarify the origin of this prominent X2 contribution, we compared the ionic orbital energy levels of each anion,²⁰ which were estimated from the sum of the Madelung potential at each crystallographic site and the electron affinity (Fig. 4a). The X2 anions exhibited higher energy levels than X1 in SrBi₃O₄X₃, and the X2 level shifted progressively upward from Cl to Br to I. These trends indicate that the X2 anion was electrostatically destabilized relative to the X1 anion owing to the stronger repulsion between the X anions in the double layer. Additionally, the halogen p-orbital energy increases for heavier halogens owing to their lower electron affinities. Taken together, these results indicate that the upward shift of the VBM in SrBi₃O₄X₃ upon halogen substitution originates from the increasing contribution of the electrostatically destabilized X2 sites in the double halogen layer. Notably, SrBi₃O₄Cl₃ exhibits an elevated VBM due to the unique interaction between Cl2 and O1 mediated by Bi2,¹⁵ and therefore the upward shift of the VBM for SrBi₃O₄X₃ from Cl to Br is modest.

Fig. 3 (a) Structural models of SrBi₃O₄X₃ (X = Cl, Br, I) used for the DFT calculations, viewed along the [100] direction of the original structures. DOS and PDOS near the (b) VBM and (c) CBM of SrBi₃O₄X₃.

Fig. 4 (a) Ionic energy levels of each halogen site in SrBi₃O₄X₃ (X = Cl, Br, I), calculated from the sum of the Madelung potentials at the X site and electron affinity. Orbital distributions of the CBM (orange dots), estimated by DFT calculations for SrBi₃O₄X₃ (X = (b) Cl, (c) Br, (d) I) (isosurface values: 0.0075 e Å⁻³). (e) Schematic of the interlayer Bi1 6p_z–Bi1 6p_z interaction near the CBM in SrBi₃O₄X₃.

The Bi atoms adjacent to the single halogen layer (Bi1) contributed significantly to the formation of the CBM; the most dominant Bi1 contribution was observed for SrBi₃O₄Cl₃ (Fig. 3c and Fig. S8). This large Bi1 contribution is due to the interactions between the Bi-6p_z orbitals across a single halogen layer, a feature previously reported for oxyhalides with a single halogen layer.¹⁵ Substituting Cl with Br or I increases the distance between Bi1 atoms across the single halogen layer (Cl: 3.9 Å → Br: 4.1 Å → I: 4.6 Å), thereby weakening the interaction between their 6p_z orbitals. This weakening narrows the conduction bandwidth formed by the interaction, thereby shifting the CBM upward and enabling a higher reduction potential. This trend is supported by the orbital distributions near the CBM (Fig. 4b–d and Fig. S10), indicating that the overlap between the Bi-6p_z orbitals becomes less pronounced in X = Br and I than in X = Cl. Furthermore, crystal orbital Hamilton population (COHP) analysis revealed weaker interactions between the Bi1-6p_z orbitals in SrBi₃O₄Br₃ and SrBi₃O₄I₃ than in SrBi₃O₄Cl₃ (Fig. S11). These results indicate that the upward shift of the CBM in SrBi₃O₄X₃ upon halogen substitution originates from the reduced orbital interaction between Bi1 orbitals across the single halogen layer (Fig. 4e). Notably, the polarizability effect that lowers the CBM in BiOX is probably less operative in SrBi₃O₄X₃, where the longer Bi1–X1 bonds (Cl: 3.39 Å, Br: 3.49 Å, I: 3.67 Å vs. Bi–X in BiOX: 3.06/3.20/3.37 Å) reduce Bi1–X1 interactions and thereby lessen polarizability-driven stabilization.

In summary, this study demonstrated that halogen substitution in SrBi₃O₄X₃ (X = Cl, Br, I), which contains both single and double halogen layers, enables simultaneous red-shifted visible-light absorption and CBM elevation, in contrast to conventional BiOX, which features only double layers. In addition, the bandgap of SrBi₃O₄I₃ is narrower than that of the single-layered Bi₄NbO₈I system. This unique tunability arises because the CBM and VBM are primarily derived from the orbitals located near different halogen layers. Specifically, replacing Cl in SrBi₃O₄Cl₃ with Br or I increases the contribution of halogens in the double halogen layer to the VBM, thereby shifting it upward, while simultaneously weakening the interaction between the Bi-6p_z orbitals across the single halogen layer, which shifts the CBM upward. These findings establish the use of two distinct halogen layers as an effective band-engineering strategy for Sillén-type layered oxyhalides, providing new opportunities for designing photocatalysts with extended visible-light responses and high reduction potentials.

Author contributions

Y. I. designed the study with advice from H. S. and R. A. All remaining experiments were performed by Y. I. All authors contributed to the discussion of the results. Y.I. wrote the manuscript, which was edited by H. S. and O. T., with discussions primarily among Y. I., H. S., and R. A.

Conflicts of interest

There are no conflicts to declare.

Data availability

All data are available in the main manuscript and the supplementary information (SI). Supplementary information: XRD patterns, calculated models and DOS&PDOS, Mott–Schottky plots and UV-vis diffuse reflectance spectra. See DOI: https://doi.org/10.1039/d5cp04120a.

Acknowledgements

This study was supported by JSPS KAKENHI (Grants-in-Aid for Scientific Research (A) JP20H00398 and (B) JP23H02061), JST SPRING (JPMJSP2110) and JST PRESTO (JPMJPR25M8). This study was supported by Samco Foundation. The computation time was provided by the SuperComputer System at the Institute for Chemical Research, Kyoto University.

References

1. 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.
2. 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. Kursumović, M. G. Bawendi, V. Bulović, V. Stevanović, T. Buonassisi and J. L. MacManus-Driscoll, Adv. Mater., 2017, 29, 1702176.
3. D.-H. Nam, M. A. Lumley and K.-S. Choi, Chem. Mater., 2019, 31, 1460–1468.
4. D.-H. Nam and K.-S. Choi, J. Am. Chem. Soc., 2017, 139, 11055–11063.
5. 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.
6. J. Jiang, K. Zhao, X. Xiao and L. Zhang, J. Am. Chem. Soc., 2012, 134, 4473–4476.
7. K. Ogawa, Y. Wakisaka, H. Suzuki, O. Tomita and R. Abe, ACS Appl. Mater. Interfaces, 2021, 13, 5176–5183.
8. Z. Luo, X. Ye, S. Zhang, S. Xue, C. Yang, Y. Hou, W. Xing, R. Yu, J. Sun, Z. Yu and X. Wang, Nat. Commun., 2022, 13, 2230.
9. J. Xie, Y. Cao, D. Jia and Y. Li, J. Colloid Interface Sci., 2017, 503, 115–123.
10. A. M. Ganose, M. Cuff, K. T. Butler, A. Walsh and D. O. Scanlon, Chem. Mater., 2016, 28, 1980–1984.
11. L. Zhang, Z. Han, W. Wang, X. Li, Y. Su, D. Jiang, X. Lei and S. Sun, Chem. – Eur. J., 2015, 21, 18089–18094.
12. 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.
13. 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.
14. Q. Liu, C. Bai, C. Zhu, W. Guo, G. Li, S. Guo, D. Kripalani, K. Zhou and R. Chen, Adv. Sci., 2024, 11, 2400934.
15. H. Suzuki, D. Ozaki, Y. Ishii, O. Tomita, D. Kato, S. Nozawa, K. Nakashima, A. Saeki, H. Kageyama and R. Abe, J. Mater. Chem. A, 2023, 11, 15159–15167.
16. Y. Ishii, H. Suzuki, K. Ogawa, O. Tomita, A. Saeki and R. Abe, Sustainable Energy Fuels, 2022, 6, 3263–3270.
17. K. Ogawa, H. Suzuki, A. Walsh and R. Abe, Chem. Mater., 2023, 35, 5532–5540.
18. R. D. Shannon and R. K. Waring, J. Phys. Chem. Solids, 1985, 46, 325–330.
19. L. G. Sillén and E. Jörnstad, Z. Anorg. Allg. Chem., 1942, 250, 173–198.
20. D. Kato, K. Hongo, R. Maezono, M. Higashi, H. Kunioku, M. Yabuuchi, H. Suzuki, H. Okajima, C. Zhong, K. Nakano, R. Abe and H. Kageyama, J. Am. Chem. Soc., 2017, 139, 18725–18731.

---