Boosting Photocatalytic CO₂ Reduction Over In‐Doped Bi₄O₅Br₂: Synergistic Bandgap Engineering and Charge Separation

ABSTRACT

Photocatalytic CO₂ reduction represents a viable approach to achieving carbon neutrality through fuel production. Nonetheless, the practical deployment of photocatalysts is frequently hampered by the rapid recombination of photogenerated charge carriers. In this investigation, a one‐pot solvothermal method was used to synthesize In‐doped Bi₄O₅Br₂ photocatalysts. Comprehensive characterizations demonstrate that the electronic structure is effectively modulated by In³⁺ doping, resulting in a narrowed bandgap. Crucially, the recombination of photogenerated electron–hole pairs is suppressed through the introduction of dopant‐induced energy levels. The 15In‐Bi₄O₅Br₂ sample displays the optimal photocatalytic CO₂ reduction activity, producing CO at a rate of 4.67 µmol g⁻¹ h⁻¹, which is approximately 2.3 times higher than that of unmodified Bi₄O₅Br₂. This remarkable enhancement is attributed to the synergistic effects of an optimized band structure and improved charge separation efficiency induced by In doping. This work provides valuable insights into the rational design of bismuth‐based materials for sustainable CO₂ conversion.

In‐doped Bi₄O₅Br₂ unlocks a boosted efficiency in photocatalytic CO₂ reduction via dual‐function doping: In³⁺ substitution shrinks bandgaps for broader photon capture while creating electron‐transfer stepping stones that suppress recombination and extend carrier lifetimes.

1. Introduction

Photocatalytic CO₂ reduction represents a sustainable approach for converting CO₂ into value‐added fuels (CH₄ and CO) [1, 2, 3]. This process harnesses solar energy to drive redox reactions, in which photogenerated electrons reduce CO₂ while holes oxidize H₂O, simultaneously addressing energy demands and carbon neutrality goals [4, 5]. However, the practical application of this technology is hindered by limitations in light absorption and charge separation efficiency in conventional photocatalysts [6, 7, 8]. Consequently, developing efficient and stable photocatalysts is imperative for advancing the practical application of photocatalytic CO₂ reduction.

Bismuth‐based oxyhalides have emerged as promising candidates for photocatalytic CO₂ reduction due to their unique layered crystal structures, excellent visible‐light responsiveness, and abundant surface‐active sites [9]. In particular, Bi₄O₅Br₂ stands out for its distinctive crystal architecture, consisting of alternating [Bi₂O₂]²⁺ layers and double Br⁻ layers. This layered architecture facilitates the exposure of Bi−O active sites and enhances CO₂ adsorption and activation [10]. Moreover, it generates an internal electric field between the [Bi₂O₂]²⁺ and Br⁻ layers, which promotes the separation of photogenerated charge carriers. With a narrow bandgap, Bi₄O₅Br₂ also enables efficient visible‐light absorption, while its appropriately positioned conduction band (CB) meets the thermodynamic requirements for CO₂ conversion [11, 12]. Despite these advantages, the practical efficiency of Bi₄O₅Br₂ in CO₂ photoreduction is still constrained by inefficient charge separation and slow surface reaction kinetics.

To address these limitations, heteroatom doping has been identified as a viable pathway to enhance the photocatalytic performance [13, 14]. Specifically, metal ion doping is a well‐established technique for modifying the electronic and structural properties of photocatalysts. This modification is achieved through lattice distortion, the introduction of impurity energy levels, and the optimization of band alignment. All of which contribute to improved light absorption, reduced charge recombination, and enhanced CO₂ activation [15]. Indium (In³⁺) emerges as a suitable dopant for Bi₄O₅Br₂ due to its +3 valence state, which aligns with that of Bi³⁺, thereby preventing charge imbalance. Additionally, the similar ionic radius of In³⁺ facilitates its stable incorporation into the Bi−O lattice. The introduction of In³⁺ can also create new active sites on the surface of Bi₄O₅Br₂ [16]. These synergistic effects render In³⁺ doping a promising strategy to address the inherent limitations of Bi₄O₅Br₂ and enhance its photocatalytic CO₂ reduction performance.

Driven by this rationale, In³⁺ ions are implemented in Bi₄O₅Br₂ to increase its effectiveness in photocatalytic CO₂ reduction. A series of In‐doped Bi₄O₅Br₂ catalysts were synthesized via a one‐step solvothermal method. The effects of In³⁺ doping concentration on structural evolution and charge carrier dynamics are systematically investigated. Furthermore, the synergistic process that improves the photocatalytic CO₂ reduction efficiency of In‐doped Bi₄O₅Br₂ is elucidated, providing a foundation for rational design of advanced bismuth‐based photocatalysts.

2. Experimental Section

2.1. Synthesis

The pure Bi₄O₅Br₂ was synthesized via a one‐step solvothermal method. Briefly, 0.73 g of Bi(NO₃)₃·5H₂O was dissolved in 10 mL of ethylene glycol under ultrasonication for 30 min. Separately, 0.12 g KBr was dissolved in 30 mL ethanol under magnetic stirring. Bi(NO₃)₃·5H₂O solution was then dropwise added to KBr Solution under continuous stirring. To fabricate the In‐doped variants, precisely measured amounts of In(NO₃)₃·xH₂O (5, 10, 15, 20 and 25 mg) were introduced into the precursor mixture. The homogeneous solution underwent solvothermal treatment at 180 °C for 16 h. Following natural cooling, the precipitate underwent washing with deionized water and ethanol, followed by drying at 70 °C for 12 h. The resulting catalysts were labeled as 5In‐Bi₄O₅Br₂, 10In‐Bi₄O₅Br₂, 15In‐Bi₄O₅Br₂, 20In‐Bi₄O₅Br₂, and 25In‐Bi₄O₅Br₂, respectively. Further details of the materials are presented in the Supporting Information.

2.2. Characterization

Details about the characterization methods and in situ diffuse reflectance infrared Fourier transform (DRIFTS) measurement can be found in the Supporting Information.

2.3. Photoelectrochemical Measurements

A CHI660E electrochemical workstation (Chenhua Instrument Co., Shanghai) was employed to systematically characterize the photoelectrochemical properties. A standard three‐electrode configuration was utilized for the measurements, comprising a platinum wire as the counter electrode, an Ag/AgCl electrode (saturated KCl) as the reference electrode, and the synthesized catalyst/FTO electrode as the working electrode. To fabricate the working electrode, 30 mg of catalyst powder was thoroughly mixed with 3 mg of ethyl cellulose in an agate mortar. Subsequently, 5 mL of absolute ethanol was added and the mixture was ground for 30 min to achieve a homogeneous slurry. This slurry was then uniformly applied onto fluorine‐doped tin oxide (FTO) conductive glass substrates and dried at 150 °C for 3 h. In the course of photocurrent response and Mott–Schottky (MS) measurements, a 300 W xenon lamp served as the light source, while the electrolyte was a 0.5 M Na₂SO₄ solution. A 0.1 M K₃[Fe(CN)₆]/K₄[Fe(CN)₆] redox couple was used to conduct electrochemical impedance spectroscopy (EIS).

2.4. Photocatalytic CO₂ Reduction

The evaluation of photocatalytic CO₂ reduction performance utilized a closed system (Labsolar‐III AG, Beijing Perfect Light Source Technology Co., Ltd., China). A quartz plate coated with 50 mg of the photocatalyst was positioned within the reactor. After adding 10 mL of deionized water, the reactor was sealed with vacuum grease to eliminate residual air and saturate the aqueous phase. Subsequently, 1 mL of high‐purity CO₂ was injected, and a magnetic stirrer was operated to maintain gas homogeneity. The reaction was carried out for 2 h under irradiation with a 300 W xenon lamp (PLS‐SXE300, Beijing Perfect Light Source Technology Co., Ltd., China). Gas samples (1 mL) were extracted at 30‐min intervals and quantified via gas chromatography (GC, Zhejiang Fuli Analytical Instrument Co., Ltd.).

3. Results and Discussion

3.1. Structural and Morphological Characterization

X‐ray diffraction (XRD) analysis is performed on both Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂ to elucidate their crystalline phases and structural changes induced by In doping. All diffraction peaks are matched with the standard Bi₄O₅Br₂ pattern (JCPDS: 37–0699), and no impurity phases are detected (Figure 1a) [17]. These results indicate that the host crystal structure of Bi₄O₅Br₂ is preserved after In³⁺ doping. However, the peak intensities for 15In‐Bi₄O₅Br₂ are diminished relative to the pristine sample, indicating a reduction in crystallinity attributable to In³⁺ doping [18]. Moreover, a closer examination of the peak positions reveals a slight shift toward lower angles in 15In‐Bi₄O₅Br₂ compared to pristine Bi₄O₅Br₂, which can be attributed to lattice contraction caused by In³⁺ substitution for Bi³⁺.

FIGURE 1.

(a) XRD patterns of Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂, (b) TEM image, (c) HRTEM images, and (d–g) EDS mapping of 15In‐Bi₄O₅Br₂.

The microscopic morphologies of Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂ are characterized using scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As depicted in Figure S1, the pristine Bi₄O₅Br₂ manifests a uniform three‐dimensional (3D) hollow spherical structure, which is self‐assembled from nanosheets. The formation of this hollow architecture is presumably driven by the Ostwald ripening process during hydrothermal synthesis [19]. In this process, smaller nanocrystals dissolve and recrystallize on the surfaces of larger aggregates to minimize surface energy, culminating in the characteristic hollow structure. This morphological feature is maintained by the15In‐Bi₄O₅Br₂ sample, demonstrating that macroscopic morphology is not altered by In³⁺ doping (Figure S2). TEM images further elucidate the microstructure of 15In‐Bi₄O₅Br₂ (Figure 1b). A distinct lattice fringe with a spacing of 0.321 nm is resolved by high‐resolution TEM (HRTEM) image, which are assigned to the (212) plane of Bi₄O₅Br₂ (Figure 1c). Additionally, successful incorporation of In³⁺ into the Bi₄O₅Br₂ lattice is confirmed by energy‐dispersive x‐ray spectroscopy (EDS) elemental mapping, which reveals homogeneous spatial distribution of Br, O, Bi, and In throughout the 15In‐Bi₄O₅Br₂ sample (Figure 1d–g). The doping amount of indium in 15In‐Bi₄O₅Br₂ is determined to be approximately 0.484% by inductively coupled plasma optical emission spectrometry (ICP‐OES) (Table S1).

3.2. Chemical Composition Characterization

The elemental composition and chemical states of Bi₄O₅Br₂ and 15In‐samples are analyzed using x‐ray photoelectron spectroscopy (XPS). The calibration of the C 1s peak position employs the specific work function (ΦSA) of Bi₄O₅Br₂ (ΦSA = 5.14 eV) [20]. According to the binding energy (BE) calibration equation (BE = 289.58−ΦSA), the carbon peak is calibrated to 284.44 eV. All XPS spectra of Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂ are subsequently aligned to this calibrated C 1s value. The survey XPS spectrum confirms the presence of Bi 4f, O 1s, and Br 3d characteristic peaks in pristine Bi₄O₅Br₂ (Figure S3) [21]. In the 15In‐Bi₄O₅Br₂ sample, additional In 3d peaks are observed, verifying successful incorporation of In³⁺ ions [22]. The high‐resolution Bi 4f XPS spectra show that the Bi 4f_7/2 and Bi 4f_5/2 peaks for 15In‐Bi₄O₅Br₂ are located at 158.99 and 164.31 eV, respectively, which are consistent with the standard Bi³⁺ oxidation state (Figure 2a) [23]. The O 1s spectrum exhibits two distinct peaks at 531.07 and 529.67 eV, assigned to adsorbed oxygen species (O−H) and lattice oxygen (Bi−O), respectively (Figure 2b) [24]. The Br 3d spectrum features symmetric doublets at 68.29 eV (Br 3d_5/2) and 69.32 eV (Br 3d_3/2), with a stable full width at half maximum below 1.5 eV, confirming the presence of Br⁻ in the crystal lattice (Figure 2c) [25]. The In 3d spectrum (Figure 2d) exhibits characteristic spin‐orbit splitting doublets at 452.18 eV (In 3d_3/2) and 444.62 eV (In 3d_5/2) [26]. The symmetrical peak shapes and the absence of satellite peaks suggest uniform doping of In³⁺, whereas the peak observed at 442.03 eV is attributed to Bi 4d_5/2 [27]. Notably, shifts of Bi 4f, O 1s, and Br 3d peaks toward lower binding energies are detected in 15In‐Bi₄O₅Br₂. This shift is ascribed to the synergistic effect of increased electron cloud density and lattice distortion. Specifically, the lower electronegativity of In³⁺ compared to Bi³⁺ results in an increased electron density around Bi, O, and Br atoms. This enhanced electron density weakens the binding force of the nucleus on inner electrons, effectively reducing the experienced nuclear charge and thus lowering the binding energy [28, 29]. Furthermore, the smaller ionic radius of In³⁺ compared to Bi³⁺ induces lattice distortion upon the formation of new In─O bonds after doping. This subsequent electron redistribution further increases the electron cloud density, contributing to the reduction in atomic binding energy.

FIGURE 2.

(a–c) Bi 4f, O 1s and Br 3d XPS spectra of Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂. (d) In 3d XPS spectra of 15In‐Bi₄O₅Br₂.

3.3. Optical Properties and Band Structure Analysis

Ultraviolet–visible (vis) diffuse reflectance spectroscopy (UV–vis DRS) is employed to characterize the light absorption properties of Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂. As illustrated in Figure 3a, the absorption edge of pristine Bi₄O₅Br₂ is situated at 467 nm. Following In³⁺ doping, a redshift to 486 nm is observed in 15In‐Bi₄O₅Br₂, confirming an extended range of visible‐light response [30]. The band gap (E _g) of each material can be calculated using the Tauc plot of (αhν)^1/2 vs. photon energy (hν). The E _g value is obtained by extrapolating the linear region of the plot to the abscissa (Figure 3b). The E _g is determined to be 2.64 eV for Bi₄O₅Br₂ and 2.57 eV for 15In‐Bi₄O₅Br₂. These results indicate that In³⁺ doping effectively modulates band structure of Bi₄O₅Br₂, inducing band gap narrowing and a consequent improvement in light absorption performance.

FIGURE 3.

(a) The UV–vis DRS spectra, (b) Tauc plots, (c) MS plots, and (d) band structure of Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂.

MS measurement (Figure 3c) curves are then employed to determine the CB positions of Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂. The positive slopes for both samples at frequencies of 1000 and 1500 Hz confirm their n‐type semiconductor nature [31]. In comparison to pristine Bi₄O₅Br₂, the M‐S slope of 15In‐Bi₄O₅Br₂ is markedly reduced, indicating a substantial increase in carrier concentration after doping. Concurrently, a significant negative shift in flat‐band potential (E _fb, relative to the Ag/AgCl electrode) from −0.35 to −0.75 V is observed after doping, corresponding to −0.15 V and −0.55 V when compared to the normal hydrogen electrode (NHE), respectively [32]. Taking into account that n‐type semiconductors have conduction band positions (E _CB) about 0.1 V more negative than the E _fb, the E _CB values of Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂ are estimated to be −0.25 and −0.65 V, respectively. Based on the band gap formula of E _g = E _{valence band (VB)}−E _CB, the E _VB values of Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂ are calculated as 2.39 and 1.92 V, respectively [33]. Based on these derived E _VB and E _CB values, the corresponding band structures are schematically illustrated in Figure 3d. This upward shift of E _CB on 15In‐Bi₄O₅Br₂ enhances the reducing power of photogenerated electrons, providing stronger thermodynamic driving force for multi‐electron CO₂ reduction reactions [34, 35].

3.4. Charge‐Carrier Dynamics Analysis

Based on the optimized band structure, a comprehensive investigation of charge‐carrier dynamics is conducted to elucidate the separation efficiency of In‐doped Bi₄O₅Br₂. Photocurrent response measurements indicate that the photocurrent density of 15In‐Bi₄O₅Br₂ under illumination is significantly higher than that of pristine Bi₄O₅Br₂, thereby confirming superior charge separation (Figure 4a) [36]. Electrochemical impedance spectroscopy (EIS) results reveal a markedly smaller semicircular arc in the Nyquist plot for 15In‐Bi₄O₅Br₂ compared to the pristine sample (Figure 4b) [37]. A smaller arc radius is a hallmark of reduced charge transfer resistance, confirming that faster electron transport is facilitated by In³⁺ doping.

FIGURE 4.

(a) Photocurrent response curves, (b) EIS plots, (c) PL spectra, and (d) time‐resolved fluorescence decay spectra of Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂.

Photoluminescence (PL) spectroscopy is commonly employed to characterize photogenerated carrier recombination. A significantly strong PL intensity is observed for pristine Bi₄O₅Br₂ than for 15In‐Bi₄O₅Br₂. Since higher PL intensity correlates with more severe recombination, the weaker signal in 15In‐Bi₄O₅Br₂ indicates a lower electron‐hole pair recombination rate (Figure 4c). The suppression is attributed to by the introduction of the impurity energy levels within the band structure of Bi₄O₅Br₂, induced by the incorporation of In³⁺ ions [38]. These dopant‐induced energy levels act as effective electron traps. Upon photoexcitation, instead of migrating directly back to the VB to recombine with holes, the CB electrons are captured by these localized impurity states. This capture process physically sequesters the electrons, thereby preventing their immediate recombination and increasing the population of spatially separated charge carriers.

Furthermore, time‐resolved photoluminescence (TRPL) spectroscopy indicates an increase in the average fluorescence lifetime (τ _ave) of photogenerated electrons from 0.48 ns in pristine Bi₄O₅Br₂ to 0.55 ns in 15In‐Bi₄O₅Br₂ (Figure 4d). This extension signifies prolonged retention of electrons in the CB of 15In‐Bi₄O₅Br₂ [39, 40, 41]. This observation aligns with the proposed trapping model, wherein electrons occupying impurity levels exhibit a delayed recombination pathway, thereby statistically prolonging their overall lifetime. Collectively, these electrochemical results corroborate that charge separation efficiency is enhanced by In³⁺ doping.

3.5. Photocatalytic Performance of CO₂ Reduction and Mechanism Investigation

The photocatalytic activity of In‐doped Bi₄O₅Br₂ is assessed by measuring the production rates of CO and CH₄ during CO₂ reduction experiments. As shown in Figure 5a, the yields of both CO and CH₄ initially increase then decline with increasing In doping concentration (ranging from 0 to 25 mg). The CO generation rate of 15In‐Bi₄O₅Br₂ is determined to be 4.67 µmol g⁻¹ h⁻¹, representing the highest activity and being roughly 2.3 times higher than the undoped Bi₄O₅Br₂ (2.03 µmol g⁻¹ h⁻¹). The observed decline in activity at higher doping levels (>15 mg) can be attributed to severe lattice distortion and defect‐induced recombination centers that impede carrier mobility [42, 43]. Time‐dependent product profiles demonstrate linear CO accumulation but limited CH₄ formation over 15In‐Bi₄O₅Br₂ (Figure S4). This product distribution aligns with reaction kinetics, where the two‐electron pathway for CO production is kinetically more favorable than the eight‐electron pathway for CH₄ formation [44, 45]. To ascertain the origin of the products in the CO₂ reduction process, control experiments under specific atmospheric conditions were performed on Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂. It reveals that neither CO nor CH₄ is detected in the absence of CO₂ feedstock, light illumination, or photocatalyst, providing evidence that the generated CO and CH₄ originate exclusively from photocatalytic CO₂ conversion (Figures 5b and S5). The XRD patterns confirm that the crystal structure of 15In‐Bi₄O₅Br₂ before and after reaction remains unchanged, with no observable phase degradation or transformation (Figure S6). Furthermore, the SEM images of 15In‐Bi₄O₅Br₂ demonstrate that the characteristic 3D morphology is also maintained after reaction (Figure S7). These results verify the excellent structural and morphological stability of 15In‐Bi₄O₅Br₂ during the reaction.

FIGURE 5.

(a) CO and CH₄ generation rate. (b) Atmosphere control experiment on 15In‐Bi₄O₅Br₂. (c) In situ DRIFTS spectra of 15In‐Bi₄O₅Br₂.

To gain a deeper understanding of the photocatalytic CO₂ reduction process, in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) measurements are performed on Bi₄O₅Br₂ and 15In‐Bi₄O₅Br₂ samples to investigate the intermediates generated during the reaction (Figures 5c and S8). The peak of HCO³⁻ (1454 cm⁻¹) in the spectrum can be attributed to the interaction between H₂O and CO₂. The peak detected at 1616 cm⁻¹ can be assigned to the *COOH group, which is a common key intermediate for the reduction of CO₂ to CO and CH₄. Meanwhile, the peak at 1045 cm⁻¹ can be ascribed to *CHO, and the characteristic peak of *CH₃O appears at 1076 cm^{− 1} [46]. Based on these observations, the reaction pathway can be tentatively deduced as *CO₂→*COOH→*CO→*CHO→*CH₂O→*CH₃O→*CH₃OH→*CH₃→*CH₄.

More importantly, the characteristic peak intensities for these intermediates, particularly *COOH and *CHO, are significantly stronger on the 15In‐Bi₄O₅Br₂ sample compared to the pristine Bi₄O₅Br₂. This notable enhancement in signal intensity suggests that the In³⁺ doping effectively strengthens the adsorption and stabilizes the accumulation of these key intermediates on the surface. This improved surface interaction provides a direct experimental rationale for the enhanced CO₂ reduction activity observed in 15In‐Bi₄O₅Br₂, as it facilitates the subsequent multi‐electron transfer steps.

4. Conclusions

In summary, an In‐doped Bi₄O₅Br₂ photocatalyst has been successfully fabricated using a one‐step solvothermal approach, facilitating the synergistic modulation of its electronic and photocatalytic properties. The substitutional incorporation of In³⁺ ions into the Bi−O lattice induces lattice distortion, thereby optimizing the electronic configuration. This alteration enhances visible‐light absorption capacity and elevates the reduction potential of conduction band electrons. Concurrently, impurity energy levels are introduced, which promote the separation of photogenerated charge carriers. The CO production rate of the optimized 15In‐Bi₄O₅Br₂ achieves 4.67 µmol g⁻¹ h⁻¹, representing a 2.3‐fold increase over pristine Bi₄O₅Br₂. This performance enhancement is attributed to the combined effects of band structure engineering and suppressed carrier recombination. The study establishes a doping strategy for rationally designing advanced photocatalysts for converting greenhouse gases into value‐added chemicals.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting Information file 1: asia70540‐sup‐0001‐SuppMat.docx

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.