PENG Xi, XIE Peng, SUN Jiajie, XIE Zicong, YU Menghuai, WANG Xiaotong, WEI Zhishun

(School of Materials and Chemical Engineering, Hubei University of Technology, Wuhan430068, Hubei, China)

Extended abstract:

[Background and purposes] Photocatalytic technology, which leverages solar energy for environmental remediation and clean energy production, holds great potential for addressing global energy shortages and environmental pollution issues. Among various semiconductor materials, bismuth oxyhalides (BiOX, X=Cl, Br, I) have attracted significant attention, due to their unique layered structure, which facilitates efficient separation of photogenerated charge carriers. Specifically, bismuth oxybromide (BiOBr) and bismuth oxyiodide (BiOI) are valued for their suitable band structures and excellent chemical stability. However, in traditional halogen substitution systems, enhanced light absorption is often accompanied by a reduction in the redox potential of charge carriers and there is insufficient understanding of the underlying structure-property relationships. In this study, pure-phase BiOBr and BiOI were synthesized by using a facile coprecipitation method. Their crystal structures, electronic properties and optical characteristics were systematically studied by using XRD, XPS, DRS and Mott-Schottky analysis. Furthermore, the intrinsic reasons for their distinct photocatalytic performances were elucidated by correlating halogen-induced variations in lattice parameters, local electron density and band edge positions. Finally, the differences in activity were quantitatively evaluated through photocatalytic degradation experiments under visible light irradiation.

[Methods] In this study, a simple ethanol solvent co-precipitation method has been used to synthesize bromo-bismuth oxide (BiOBr) and iodobismuth oxide (BiOI) nanomaterials, with Bi(NO₃)₃·5H₂O and NaBr/NaI as raw materials, CTAB as the surfactant, and ethylene glycol as the solvent. Crystal structure and phase composition of the materials were characterized by using X-ray diffraction (XRD, PANalytical Empyrean). Surface morphology of the materials was observed by using an ultra-high-resolution field emission scanning electron microscope (FE-SEM, Hitachi SU8010, Japan). Absorbance of the materials was measured by using a dual-beam ultraviolet-visible spectrophotometer with an integrating sphere (UV-DRS, Mecy UV-1800PC China). Valence state of elements on the material surface was examined by using an X-ray photoelectron spectrometer (XPS, ULVAC–PHI5000 VesaProbe Ⅲ, Japan).

[Results] XRD diffraction peaks of the samples are in good agreement with those of the standard data of BOB and BOI. The ionic radius of I^– (~220 pm) is significantly larger than that of Br⁻ (~196 pm). When larger I^– ions are incorporated between the layers, the adjacent [Bi₂O₂]²⁺ layers are "stretched" to accommodate their volume, which is manifested as a systematic shift of the BOI of diffraction peaks toward lower angles, corresponding to lattice expansion. In SEM images, the BOB sample appears as spheres formed by close stacking of flake-like structures in the same direction, while the BOI sample is more similar to a flower-like sphere, with its flakes overlapping more tightly in different directions. The underlying reason for this morphological difference may be the variations in radius, electronegativity and other physicochemical properties between I^– and Br^–. These differences lead to inconsistent bonding tightness between flakes during the solvothermal process, resulting in discrepancies during self-assembly and thus presenting distinct flower-like spherical structures. Furthermore, the Bi 4f binding energy of BiOBr is approximately 0.1 eV higher than that of BiOI. This phenomenon is attributed to the higher electronegativity of Br, which reduces the electron cloud density of Bi in the Bi-O-Br bond. This serves as direct experimental evidence for the differences in their electronic structures. DRS results indicate that the absorption edge of BiOI exhibits a significant red shift, giving it a broader visible light response range. The band gaps (Eg) of the two materials were approximately 2.70 eV for BiOBr and approximately 1.75 eV for BiOI, which is one of the direct reasons for their difference in performances. The energy band structure diagrams of both materials were obtained through Mott-Schottky tests. BiOBr has a more positive valence band position, resulting in stronger hole oxidation ability, while BiOI has a more negative conduction band position, leading to stronger electron reduction ability. Ultimately, these differences manifest as variations in light absorption capacity, carrier separation and migration efficiency and redox potentials, which in turn result in distinctly different photocatalytic performances. Degradation results indicate that BOI achieves a TC removal rate of up to 77.0% within 60 min, which is significantly higher than that of BOB (47.2%).

[Conclusions] In the experiment of visible light-driven photocatalytic degradation of tetracycline (TC), iodine-doped bismuth oxide (BOI) exhibited outstanding activity, with a removal rate of 77.0% within 60 min. Its pseudo-first-order reaction rate constant (k1= 0.0298 min⁻¹) was approximately 2.3 times that of bromine-doped bismuth oxide (BOB, k1= 0.0118 min⁻¹, removal rate 47.2%), fully demonstrating that BOI has significantly enhanced photocatalytic performance. This difference is mainly attributed to the fundamental differences in their optical absorption properties. The narrow band gap (1.75 eV) of BOI enables it to effectively capture a wider range of visible light, thereby generating much larger number of photogenerated electron-hole pairs than BOB (band gap 2.70 eV). Although BOB may possess stronger hole oxidation ability due to its more appropriate valence band position, in this study's visible light irradiation system, the number of photogenerated carriers is more crucial for the overall activity. The Mott-Schottky test results confirmed that both are n-type semiconductors, while the conduction band of BOI is located at −0.41 V (vs. NHE), slightly lower than that of BOB (−0.38 V), indicating that the photogenerated electrons of BOI have a slightly stronger thermodynamic reduction tendency. In the process of photocatalytic degradation of organic substances, holes and reactive oxygen species derived from electrons (such as ·O₂⁻) usually play a dominant role. The excellent activity of BOI indicates that its photogenerated carriers have more efficient separation and transmission efficiency, thereby inhibiting the recombination of electrons and holes. In conclusion, the lattice expansion and valence band top shift caused by replacing bromine with iodine significantly narrowed the material's band gap, thereby greatly enhancing the absorption ability of BOI for visible light. Although BOB may have an advantage in hole oxidation potential, in the visible light condition, the significant advantage of BOI in the number of photogenerated carriers becomes the decisive factor for its superior photocatalytic performance. The key influence of halogen regulation on the photocatalytic activity of bismuth-oxygen halides has been clarified, from the perspective of band structure and carrier behavior.

Key words: bismuth oxohalide; halogen regulation; photocatalysis; microstructure