English

• Photoelectrochemical (PEC) water splitting for hydrogen production is a promising solution for future humans to cope with the rapid consumption of global energy and the transformation of energy structure [1–3]. Up to now, scientists have made great efforts in exploring high-efficiency PEC water splitting semiconductor photoanode materials, such as Fe₂O₃ [4], TiO₂ [5], WO₃ [6], BiVO₄ [7], TaON [8] and Ta₃N₅ [9,10], have been developed for PEC water splitting hydrogen production. Wherein, the band gap (2.4 eV) and band edge position of monoclinic BiVO₄ are extremely matched with the PEC water splitting potential, so its theoretical photocurrent density is as high as 7.5 mA/cm² [11,12]. However, severe electron and hole recombination and slow carrier migration have become fatal factors limiting the improvement of PEC water splitting efficiency [13–16].

  Recently, various strategies for modifying BiVO₄ have been proposed to overcome above issues and enhance its PEC water splitting efficiency. These strategies include defect engineering (such as constructing oxygen vacancies [17], plasma treatment [18], and element doping [19,20]), construction of heterojunctions [21–23], surface modification (like loading of cocatalysts [24–26]), and interface engineering (such as constructing a hole transport layer [27,28] and interfacial charge regulation [29]). Interfacial charge regulation enhances photogenerated charge separation by injecting or extracting carriers, thereby effectively mitigating interfacial electron and hole recombination. This process essentially facilitates the directional migration of electrons and holes within the bulk phase of semiconductor photoanodes [29]. On the one hand, the medium involved in regulation should possess an energy level that is compatible with the semiconductor photoanode and be capable of close contact with the semiconductor, as these factors determine the direction of electron or hole migration. On the other hand, this medium should have good electrical conductivity, which is crucial for ensuring rapid migration of electrons or holes [30]. Boron nitride quantum dots (BNQDs), serving as an interfacial charge-regulating medium, can regulate the band gap and band edge positions by their strong quantum effect. This, in turn, facilitates the formation of an efficient interface between BNQDs and semiconductors [31,32]. Moreover, the BNQDs can be used as a hole extraction agent in photocatalysis. It possesses the capability to swiftly extract and transfer photogenerated holes, thereby facilitating charge separation, and effectively mitigating the recombination of electrons and holes. For instance, the g-C₃N₄/BNQDs electrode exhibits exceptional photocatalytic activity [33]. BNQDs with negative charge functional groups can effectively attract photogenerated holes, thereby accelerating charge separation processes and facilitating the activation of molecular oxygen. Similarly, the F: α-Fe₂O₃/BNQDs photoanode demonstrates remarkable PEC properties [34]. Herein, BNQDs serve as highly efficient mediators for hole extraction, with their exceptional conductivity further enhancing the separation and migration of carriers within the α-Fe₂O₃.

  In addition, the overall water splitting efficiency is also usually hindered by the slow kinetics of the oxygen evolution reaction (OER), which is caused by the multi-step proton-coupled electron transfer inherent in the PEC water oxidation process [35]. Oxygen evolution cocatalysts (OECs) frequently collaborate synergistically with semiconductors, markedly augmenting the activity and selectivity of composite materials [36]. Moreover, OECs offer abundant active sites that facilitate surface reactions and significantly expedite the transfer of electrons and holes at the catalyst/electrolyte interface, thereby enhancing overall PEC performance [37,38].

  Cobalt borate (CoBi) ultrathin nanosheets exhibit remarkable efficacy as highly active OECs for the OER in both alkaline and neutral environments [39]. This exceptional performance is attributed to the direct coordination between the borate group and the cobalt ion, leading to enhanced electron transfer capabilities and a remarkable synergistic coupling effect [40]. The formation of higher oxidation state of Co³⁺ during photoelectric deposition procedure exhibits distinctive properties of low spin configuration, transforming into a catalytically active species [41].

  Based on the above considerations, we have rationally designed and prepared BiVO₄/BNQDs/CoBi photoanode to promote the directional migration of electrons and holes in BiVO₄, thereby inhibiting their recombination. BNQDs acted as interfacial charge regulation medium facilitates the hopping transfer of holes from BiVO₄ to BNQDs and then to CoBi. Satisfyingly, the photocurrent density of BiVO₄/BNQDs/CoBi is much higher than that of pure BiVO₄ photoanode.

  The BiVO₄/BNQDs/CoBi photoanodes were prepared by impregnation and photoelectric deposition methods. The preparation flow chart is shown in Fig. 1a. Firstly, porous BiVO₄ nano-arrays photoanodes were prepared on fluorine-doped tin oxide (FTO) conductive glass substrate by electrodeposition [42]. Secondly, BNQDs were prepared by hydrothermal method [43], and then BiVO₄ photoanodes were impregnated in the solution of BNQDs to form BiVO₄/BNQDs photoanodes. Finally, CoBi was deposited as a co-catalyst layer on the surface of the BiVO₄/BNQDs photoanodes by the photoelectric deposition method [44], and the BiVO₄/BNQDs/CoBi composite photoanodes were obtained. Detailed synthesis processes of photoanodes are provided in the Supporting information.

  Figure 1

  

  Figure 1.  (a) Schematic illustration of the preparation of BiVO₄/BNQDs/CoBi photoanode. Scanning electron microscopy (SEM) images of photoanodes: (b) BiVO₄, (c) BiVO₄/BNQDs, (d) BiVO₄/CoBi and (e) BiVO₄/BNQDs/CoBi. (f-i) HR-TEM images and (j) TEM-EDS mapping of BiVO₄/BNQDs/CoBi photoanode.

  DownLoad: Full-Size Img PowerPoint

  As shown in Fig. 1b, BiVO₄ presents a nano worm-like structure and grows uniformly on the FTO conductive glass substrate. The surface morphology of the modified photoanodes shows no significant difference compared to the unmodified ones (Figs. 1c-e). In order to further explore the microscopic morphology of the photoanodes, the BiVO₄/BNQDs/CoBi electrodes were characterized by high-resolution transmission electron microscopy (HR-TEM). As shown in Figs. 1f-i, which are the HR-TEM images of the BiVO₄/BNQDs/CoBi samples with different magnifications. In Figs. 1g and i, we can observe that the BNQDs are uniformly distributed on the surface of BiVO₄ and the TEM images revealed that the average diameter of BNQDs was around 5 nm (as marked by the red circle). The marked region in Fig. 1h is the CoBi co-catalyst layer, and it can be observed that the amorphous film with a thickness of about 5–10 nm is tightly attached to the surface of the BiVO₄/BNQDs photoanode. In addition, the lattice stripes with spacing of 0.31 and 0.21 nm correspond to the (121) and (100) crystal surfaces of BiVO₄ and BNQDs, respectively (Fig. 1i) [43,45,46]. Fig. 1j shows the energy dispersive spectra (TEM-EDS) of the BiVO₄/BNQDs/CoBi photoanode, which clearly shows that Bi, V, O, B, N, and Co are uniformly dispersed in the composite sample with no other impurities. This preliminarily indicates the successful preparation of BiVO₄/BNQDs/CoBi photoanode.

  The X-ray diffraction (XRD) patterns of photoanodes are shown in Fig. 2a. The diffraction peaks of 26.6°, 33.7°, 37.8°, and 51.8° correspond to the (110), (101), (200), and (211) crystal planes of SnO₂ (JCPDS. No. 46–1088) on the substrate, respectively. The diffraction peaks of BiVO₄ are consistent with the structure of monoclinic scheelite (JCPDS. No. 14–0688) [47–49]. No diffraction peaks of other substances were detected, which may be related to the low loading and uniform dispersion of BNQDs and CoBi. As a supplement, the Fourier transform infrared (FT-IR) spectra of BiVO₄ and BiVO₄/BNQDs photoanodes were measured to demonstrate the presence of BNQDs in BiVO₄/BNQDs composites. As shown in Fig. 2b, the peaks at 3406 and 3225 cm⁻¹ are attributed to the stretching vibrational hydroxyl groups and secondary amines at the edge of BNQDs [50]. The three characteristic vibration peaks at 1452, 1180 and 1087 cm⁻¹ correspond to C—N, N—B—O and B—C [51]. The two peaks at 1384 and 817 cm⁻¹ correspond to the stretching and bending vibrations of B-N, respectively [52], which suggests the successful loading of BNQDs on the BiVO₄ photoanode surface.

  Figure 2

  

  Figure 2.  (a) XRD patterns of BiVO₄, BiVO₄/BNQDs, BiVO₄/CoBi and BiVO₄/BNQDs/CoBi photoanodes. (b) FT-IR spectra of BiVO₄, BiVO₄/BNQDs and BiVO₄/BNQDs/CoBi photoanodes.

  DownLoad: Full-Size Img PowerPoint

  More precise information on the elemental composition as well as elemental valence states of the BiVO₄/BNQDs/CoBi photoanodes were further obtained by X-ray photoelectron spectroscopy (XPS) measurements. All elements are displayed on the total XPS survey spectrum of the BiVO₄/BNQDs/CoBi electrode (Fig. S1 in Supporting information) and the high-resolution fine spectra of each element (Figs. 3a-f). As shown in Fig. 3a, the orbital peaks at binding energies of 158.9 eV and 164.2 eV correspond to Bi 4f_7/2 and Bi 4f_5/2, respectively, which confirms the presence of Bi³⁺ [53]. The orbital peaks at 516.7 eV and 524.3 eV can confirm the presence of V⁵⁺ (Fig. 3b) [54]. As shown in Fig. 3c for the N 1s spectral region, the peaks at binding energies of 400.6 eV and 399.7 eV correspond to N—O and N—B bonds, respectively. The B 1s spectral region can be observed in Fig. 3d from BNQDs, in agreement with previous literature reports [43,50,55]. For the O 1s spectrum (Fig. 3e), the peak at a binding energy of 529.8 eV is associated with O²⁻ (O_L) in the lattice, while the signal peak at a binding energy of 531.6 eV originates from the hydroxyl group (O_v) [56]. Four peaks can be fitted in the Co 2p spectrum (Fig. 3f). Two main peaks at 780.6 eV and 796.7 eV correspond to Co 2p_3/2 and Co 2p_1/2, respectively. In addition, two satellite peaks (denoted as sat.) at 786.2 eV and 804.4 eV can be observed, which are characteristic peaks of cobalt oxides, suggesting that Co³⁺ and Co²⁺ coexist in the CoBi layer [57]. The above results indicate that BiVO₄/BNQDs/CoBi photoanodes were successfully prepared.

  Figure 3

  

  Figure 3.  High-resolution XPS spectra of BiVO₄/BNQDs/CoBi photoanode: (a) Bi 4f, (b) V 2p, (c) N 1s, (d) B 1s, (e) O 1s and (f) Co 2p.

  DownLoad: Full-Size Img PowerPoint

  Based on the results of the above investigations, the PEC water oxidation performance of the prepared photoanodes were measured in a PEC water splitting device (Fig. S2 in Supporting information). From the recorded linear scanning voltammetry (LSV) curves, the pure BiVO₄ photoanode (Fig. 4a) shows a low current density of 1.5 mA/cm² at 1.23 V vs. RHE and that of BiVO₄/BNQDs photoanode can reach 4.0 mA/cm² (Fig. 4a and Fig. S3a in Supporting information, the maximum current density was reached at 2 h of impregnation). While the photocurrent density of BiVO₄/BNQDs/CoBi prepared at optimal conditions (Fig. S3b in Supporting information) reaches the highest photocurrent density value of 5.1 mA/cm², which is 3.4 times that of the pure BiVO₄ photoanode. BiVO₄/CoBi photoanode have a photocurrent density of 3.1 mA/cm², indicating CoBi has excellent catalytic performance. It is noted that the photocurrent density of the BiVO₄/BNQDs/CoBi photoanode is higher than that of the BiVO₄/CoBi electrode, suggesting that the introduction of the BNQDs layer between the BiVO₄ and CoBi can promote the transfer of photogenerated holes and thus improve the PEC performance. As can be seen from the LSV curves under dark conditions in Fig. S4 (Supporting information), the BiVO₄/BNQDs/CoBi exhibits a lower onset potential compared to the pure BiVO₄ photoanode, indicating that the BiVO₄/BNQDs/CoBi has excellent oxygen evolution reaction activity. As shown in Fig. S5 (Supporting information), all the photoanodes exhibit fast photocurrent responses at the instant of turning on and off the light. The steady-state current of the BiVO₄/BNQDs/CoBi photoanode is significantly higher than that of the others, which is consistent with the LSV results.

  Figure 4

  

  Figure 4.  (a) LSV curves, (b) ABPE spectra, (c) IPCE spectra, (d) APCE spectra of BiVO₄, BiVO₄/BNQDs, BiVO₄/CoBi and BiVO₄/BNQDs/CoBi photoanodes.

  DownLoad: Full-Size Img PowerPoint

  To further investigate the underlying reasons for the aforementioned observations, we evaluated the photoelectric conversion efficiency of the photoanodes during water oxidation. Firstly, the applied bias photon-to-current efficiency (ABPE) for different electrodes was calculated by LSV curves. As shown in Fig. 4b, it is clearly observed that the ABPE of the BiVO₄/BNQDs/CoBi photoanodes reach 2.1% at 0.60 V vs. RHE. The ABPE values for the BiVO₄/BNQDs (1.3% at 0.7 V vs. RHE) and BiVO₄/CoBi (1.0% at 0.67 V vs. RHE) photoanodes are both higher than that of pure BiVO₄ (0.31% at 0.84 V vs. RHE). This suggests that the incorporation of BNQDs and the cocatalyst CoBi contributes significantly to enhancing the photon-to-current conversion efficiency of BiVO₄. Secondly, the incident photon-to-current conversion efficiency (IPCE) of the electrodes was measured as shown in Fig. 4c. All photoanodes exhibit similar profiles to pure BiVO₄ in the range of 350–600 nm. It is noteworthy that the BiVO₄/BNQDs/CoBi photoanode exhibits a largest IPCE value of 64%, while the BiVO₄/BNQDs, BiVO₄/CoBi and BiVO₄ photoanodes have maximum IPCE values of 55%, 46% and 30%, respectively. The prepared composite photoanodes are all higher than that of pure BiVO₄ photoanode, which is partly attributed to the enhancement of the light harvesting efficiency (LHE, Fig. S6 in Supporting information). In addition, the absorbed photon-to-current conversion efficiency (APCE) was obtained from IPCE and LHE calculations. As shown in Fig. 4d, the APCE of the BiVO₄/BNQDs/CoBi photoanode is 2.5 times higher than that of BiVO₄. The above results indicate that the BiVO₄/BNQDs/CoBi photoanode can absorb light more efficiently and significantly enhance the photoelectric conversion efficiency.

  In addition to the measurement of photoelectric conversion efficiency, the transport behavior of carriers of different photoanodes was also studied. Firstly, electrochemical impedance spectra (EIS) of the photoanodes were measured. As shown in Fig. 5a, compared to the BiVO₄, BiVO₄/CoBi and BiVO₄/BNQDs photoanodes, the Nyquist curves of the BiVO₄/BNQDs/CoBi photoanode exhibits the smallest arc radius, which suggests that it has smaller charge transfer resistance. The charge transfer resistance (R_ct) values of different photoanodes fitted according to the equivalent circuit diagram are consistent with the above results (Table S1 in Supporting information). We also performed a Mott-Schottky (M-S) analysis at 1 kHz under dark conditions. The flat band potential (E_FB) is the intercept value estimated by extrapolating the M-S plot to the x-axis. As shown in Fig. 5b, a negative shift of the E_FB indicates an increase in the band bending of the integrated photoanode, which leads to a stronger interfacial charge transfer drive [58]. The addition of BNQDs and CoBi can significantly increase the carrier density (N_d) of BiVO₄, which implies the enhancement of electrical conductivity as well as the improvement of charge separation performance (Table S2 in Supporting information) [59]. The above results indicate BNQDs can improve charge transfer.

  Figure 5

  

  Figure 5.  (a) EIS curves, (b) Mott-Schottky plots, (c) charge injection efficiency curves and (d) charge separation efficiency curves of BiVO₄, BiVO₄/BNQDs, BiVO₄/CoBi and BiVO₄/BNQDs/CoBi photoanodes.

  DownLoad: Full-Size Img PowerPoint

  The charge injection efficiency (η_inj) and charge separation efficiency (η_sep) values were calculated from the LSV curves (Fig. S7 in Supporting information) with sodium sulfite (Na₂SO₃) as the hole scavenger and the maximum theoretical photocurrent density J_abs (Fig. S8 in Supporting information) under AM 1.5 G illumination (100 mW/cm²). As shown in Fig. 5c, the η_inj value of the BiVO₄/BNQDs photoanode is 67% at 1.23 V vs. RHE, which is about 2.3 times higher than that of the pure BiVO₄ (29%). In addition, the η_sep value of the BiVO₄/BNQDs photoanode reached 89% at 1.23 V vs. RHE (Fig. 5d). The η_inj and η_sep values are further improved to 82% and 93% by coating CoBi cocatalysts on the surface of the BiVO₄/BNQDs photoanode. The above results likewise indicate that loading BNQDs and CoBi can simultaneously promote charge separation and migration. At the same time, the BiVO₄/BNQDs/CoBi exhibits more hydrophilic behavior (Fig. S9 in Supporting information) and active surface area (Fig. S10 in Supporting information), which is beneficial for holes to participate in the water oxidation reaction, thereby improving the injection efficiency.

  To gain deeper insights into the role of BNQDs as a hole extractor in enhancing carrier separation and facilitating rapid charge transport during the PEC water oxidation process of a BiVO₄/BNQDs/CoBi photoanode, we conducted measurements of the energy band structures of BiVO₄, BNQDs, and CoBi. Their optical properties were evaluated using UV–vis diffuse reflectance spectroscopy (UV–vis DRS), and the optical band gaps were calculated based on the Tauc plot, yielding values of 2.51 eV for BiVO₄, 2.75 eV for BNQDs, and 4.25 eV for CoBi, respectively (Fig. 6a). The energy level structures of BiVO₄, BNQDs and CoBi were characterized using ultraviolet photoelectron spectroscopy (UPS) (Figs. 6b and c), which will determine the charge separation and transfer efficiency [60]. The work functions (Φ) were determined by subtracting the secondary electron cutoff energies E_cutoff from the He I excitation energy (21.22 eV), and then the Fermi level (E_f), valence band and conduction band positions of BiVO₄, BNQDs and CoBi were obtained. The calculation results (Tables S3 and S4 in Supporting information) were plotted as a schematic diagram of the charge flow path of BiVO₄/BNQDs/CoBi photoanode (Fig. 6d). Under the irradiation of AM 1.5 G simulated sunlight, electron-hole pairs were generated in the BiVO₄/BNQDs/CoBi photoanode. The photogenerated electrons were excited from VB to CB, while the holes remained in the VB of BiVO₄ and CoBi. Since the Fermi level of BNQDs is much lower than the CB of CoBi, electrons are captured from CB of CoBi to flow through BNQDs to BiVO₄, and finally reach the Pt electrode through an external circuit for hydrogen evolution reaction. At the same time, holes transfer from the VB of BiVO₄ to the VB of BNQDs, and finally transfers to the VB of CoBi and participates in the water oxidation reaction. As a hole extractor, BNQDs effectively promote the bulk charge separation and the OER water oxidation kinetics by the transfer driving force from BiVO₄ to CoBi.

  Figure 6

  

  Figure 6.  (a) UV–vis DRS and Tauc plots of BiVO₄, BNQDs and CoBi powders. (b, c) UPS plots of BiVO₄, BNQDs and CoBi powders. (d) Charge flow paths for BiVO₄/BNQDs/CoBi photoanode.

  DownLoad: Full-Size Img PowerPoint

  Simultaneously, the average carrier lifetime (τ_A) of the photoanodes were characterized by time-resolved photoluminescence (TRPL). BiVO₄/CoBi/BNQDs photoanode has longest average carrier lifetime of 17.86 ns compared to that of the BiVO₄ (6.54 ns), BiVO₄/CoBi (11.61 ns) and BiVO₄/BNQDs (12.91 ns) photoanodes (Fig. 7a and Table S5 in Supporting information). Similarly, BiVO₄/CoBi/BNQDs photoanode has the lowest fluorescence intensity in the range of 470–600 nm, indicating that there is less carrier recombination within the electrode (Fig. S11 in Supporting information). The transient photovoltage (TPV) spectra (Fig. 7b) show that all photoanodes exhibit positive photovoltage signals. Notably, the photovoltage values of the BiVO₄/CoBi/BNQDs and BiVO₄/BNQDs photoanodes are significantly higher than those of BiVO₄ and BiVO₄/CoBi. This confirms that the photogenerated holes were effectively transferred to the surface of the photoanodes, indicating that the BNQDs possess hole-acceptor characteristics, thereby enhancing the electron-hole separation efficiency in BiVO₄ [61,62]. Additionally, it is worth noting that the times (t_max) for the TPV signals of the BiVO₄/CoBi/BNQDs and BiVO₄/BNQDs photoanodes to reach their maximum values and then decay back to the baseline are significantly delayed compared to those of the BiVO₄ and BiVO₄/CoBi photoanodes. This is attributed the inhibition of photogenerated electrons and holes recombination in composite photoanode, especially for BiVO₄/CoBi/BNQDs and BiVO₄/BNQDs [63].

  Figure 7

  

  Figure 7.  (a) TRPL spectroscopy, (b) TPV spectroscopy, (c) OCP-derived carrier transfer lifetimes, (d) IMPS spectra, (e) the average charge transport time τ_t and (f) charge transfer and recombination rate constant of BiVO₄, BiVO₄/BNQDs, BiVO₄/CoBi and BiVO₄/BNQDs/CoBi photoanodes.

  DownLoad: Full-Size Img PowerPoint

  The open circuit potential (OCP) transient decay profile (Fig. S12 in Supporting information) can provide information of photogenerated charges [64]. The carrier lifetime (τ_n) derived from OCP can be used to quantitatively investigate the charge separation behavior. As presented in Fig. 7c, the trend of carrier lifetime values of different photoanodes when illumination was removed is as follows: BiVO₄ (0.755 s) > BiVO₄/CoBi (0.639 s) > BiVO₄/BNQDs (0.524 s) > BiVO₄/BNQDs/CoBi (0.249 s). The carrier lifetime of BiVO₄/CoBi when loaded with BNQDs is shortened to a minimum, which suggests that the charge separation efficiency and transfer dynamics of BiVO₄ are significantly enhanced due to the rapid extraction of holes by the BNQDs. More importantly, intensity modulated photocurrent spectroscopy (IMPS) was used to study the relative rate constants of charge transfer (k_ct) and charge recombination (k_re). Fig. 7d shows the IMPS Nyquist plots of BiVO₄, BiVO₄/CoBi, BiVO₄/BNQDs and BiVO₄/BNQDs/CoBi photoanodes at 1.23 V vs. RHE, from which the average charge transport time (τ_t) of the photoanodes were obtained (Fig. 7e, and Table S6 in Supporting information). As shown in Fig. 7e, the τ_t of BiVO₄/BNQDs/CoBi photoanode (0.175 ms) is much lower than that of BiVO₄ (0.303 ms), BiVO₄/CoBi (0.252 ms), BiVO₄/BNQDs (0.210 ms) photoanodes. It is worth noting that the τ_t of BiVO₄/BNQDs (0.210 ms) is smaller than that of BiVO₄/CoBi (0.252 ms), indicating that BNQDs contribute more significantly to enhancing the charge transport dynamics of the BiVO₄ photoanode. Specifically, the results of photoanode charge transport and recombination rate constants (Fig. 7f and Fig. S13 in Supporting information) show that the k_re of BiVO₄/BNQDs is only slightly smaller than that of BiVO₄/BNQDs, while the k_ct of BiVO₄/BNQDs is significantly higher than that of BiVO₄/CoBi. This means BNQDs and CoBi play a similar role for inhibiting charge recombination, while BNQDs play significant role for improving the charge transfer rate than CoBi.

  Finally, the generated H₂ and O₂ gas by BiVO₄/BNQDs/CoBi photoanode were detected. As shown in Fig. S14 (Supporting information), the molar ratio of H₂ and O₂ gas is about 2:1, with calculated Faraday efficiency of around 96% for both gases, indicating that the majority of the photogenerated charge is used for water splitting to produce O₂ and H₂.

  In summary, we have constructed efficient BiVO₄/BNQDs/CoBi photoanode for PEC water splitting, in which BNQDs as hole extractors and CoBi as a cocatalyst. The photocurrent densities of BiVO₄/BNQDs/CoBi reach 5.1 mA/cm² at 1.23 V vs. RHE, which is 3.4 times that of the pure BiVO₄. BiVO₄/BNQDs/CoBi exhibits a largest IPCE value of 64%, which is much greater than BiVO₄/BNQDs, BiVO₄/CoBi and BiVO₄ photoanodes. The significant enhancement in PEC performance is attributed to the simultaneous promotion of charge separation and migration, facilitated by the presence of BNQDs and CoBi. IMPS and TPV demonstrated that BNQDs and CoBi play a similar role for inhibiting charge recombination while BNQDs play significant role for improving the charge transfer rate than CoBi. This work provides insights into the efficient separation and directional transfer of carriers, which is crucial for the development of high-performance PEC water splitting systems.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  CRediT authorship contribution statement

  Jingwei Huang: Writing – review & editing, Supervision, Resources. Zhanghao Yang: Writing – original draft, Methodology, Investigation. Xiaoli Yuan: Investigation, Data curation. Lei Wang: Software. Houde She: Methodology. Qizhao Wang: Conceptualization.

  Acknowledgments

  This work was financially supported by the National Natural Science Foundation of China (No. 52173277), and the Young Teachers' Research Ability Improvement Project of Northwest Normal University (No. NWNU-LKQN2020–01).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2025.111062.

  
  
• 1. [1]

     G. Zeng, T.A. Pham, S. Vanka, et al., Nat. Mater. 20 (2021) 1130–1135. doi: 10.1038/s41563-021-00965-w

  2. [2]

     L. Gan, G. He, Y. Liu, et al., Appl. Catal. B: Environ. 349 (2024) 123895. doi: 10.1016/j.apcatb.2024.123895

  3. [3]

     L. Wang, Z. Liu, J. Zhang, et al., Chin. Chem. Lett. 34 (2023) 108007. doi: 10.1016/j.cclet.2022.108007

  4. [4]

     Y. Su, W. Yu, L. Liao, et al., Adv. Funct. Mater. 33 (2023) 2303976. doi: 10.1002/adfm.202303976

  5. [5]

     H. Eidsvåg, S. Bentouba, P. Vajeeston, et al., Molecules 26 (2021) 1687. doi: 10.3390/molecules26061687

  6. [6]

     G. Zheng, J. Wang, G. Zu, et al., J. Mater. Chem. A 7 (2019) 26077–26088. doi: 10.1039/c9ta09188b

  7. [7]

     D. Xiao, G. Cheng, H. She, et al., J. Phys. Chem. C 128 (2024) 15266–15276. doi: 10.1021/acs.jpcc.4c03542

  8. [8]

     M.H. Ryu Abe, Kazunari Domen, J. Am. Chem. Soc. 132 (2010) 11828–11829. doi: 10.1021/ja1016552

  9. [9]

     L. Wang, X. Zhou, N.T. Nguyen, et al., Adv. Mater. 28 (2016) 2432–2438. doi: 10.1002/adma.201505312

  10. [10]

      Y. Li, L. Zhang, A. Torres-Pardo, et al., Nat. Commun. 4 (2013) 2566. doi: 10.1038/ncomms3566

  11. [11]

      M. Barawi, M. Gomez-Mendoza, F.E. Oropeza, et al., ACS Appl. Mater. Interfaces 14 (2022) 33200–33210. doi: 10.1021/acsami.2c07451

  12. [12]

      S. Wang, X. Wang, B. Liu, et al., Nanoscale 13 (2021) 17989–18009. doi: 10.1039/d1nr05691c

  13. [13]

      Y. Miao, J. Liu, L. Chen, et al., Chem. Eng. J. 427 (2022) 131011. doi: 10.1016/j.cej.2021.131011

  14. [14]

      S.A. Shabbir, M. Haris, R. Ajmal, et al., Mater. Lett. 360 (2024) 135932. doi: 10.1016/j.matlet.2024.135932

  15. [15]

      J. Huang, T. Liu, R. Wang, et al., J. Colloid Interface Sci. 570 (2020) 89–98. doi: 10.3390/ijerph18010089

  16. [16]

      J. Huang, W. Luo, X. Yuan, et al., J. Alloys Compd. 930 (2023) 167397. doi: 10.1016/j.jallcom.2022.167397

  17. [17]

      J.B. Pan, B.H. Wang, J.B. Wang, et al., Angew. Chem. Int. Ed. 60 (2020) 1433–1440.

  18. [18]

      P. Yue, H. She, L. Zhang, et al., Appl. Catal. B: Environ. 286 (2021) 119875. doi: 10.1016/j.apcatb.2021.119875

  19. [19]

      M. Chen, X. Chang, C. Li, et al., J. Colloid Interface Sci. 640 (2023) 162–169. doi: 10.1016/j.jcis.2023.02.096

  20. [20]

      S.G. Shim, J. Tan, H. Lee, et al., Chem. Eng. J. 430 (2022) 133061. doi: 10.1016/j.cej.2021.133061

  21. [21]

      H. Khan, M.J. Kim, J.H. Baek, et al., ACS Appl. Energy Mater. 5 (2022) 15788–15798. doi: 10.1021/acsaem.2c03345

  22. [22]

      X. Chen, C. Zhen, N. Li, et al., Small Methods 7 (2023) 2201611. doi: 10.1002/smtd.202201611

  23. [23]

      K. Ren, J. Zhou, Z. Wu, et al., Small 20 (2023) 2304835.

  24. [24]

      X. Hu, Q. Wang, Y. Li, et al., J. Colloid Interface Sci. 607 (2022) 219–228. doi: 10.3390/cancers15010219

  25. [25]

      Y. Qi, J. Zhang, Y. Kong, et al., Nat. Commun. 13 (2022) 484. doi: 10.1038/s41467-022-28146-6

  26. [26]

      Q. Sun, T. Cheng, Z. Liu, et al., Appl. Catal. B: Environ. 277 (2020) 119189. doi: 10.1016/j.apcatb.2020.119189

  27. [27]

      M.A. Gaikwad, U.P. Suryawanshi, U.V. Ghorpade, et al., Small 18 (2021) 2105084.

  28. [28]

      X. Wei, J. Zhang, L. Wang, et al., Chem. Eng. J. 482 (2024) 149114. doi: 10.1016/j.cej.2024.149114

  29. [29]

      X. Ning, B. Lu, Z. Zhang, et al., Angew. Chem. Int. Ed. 58 (2019) 16800–16805. doi: 10.1002/anie.201908833

  30. [30]

      S. Zhang, P. Du, H. Xiao, et al., Angew. Chem. Int. Ed. 63 (2023) e202315763. doi: 10.1002/anie.202315763

  31. [31]

      H. Zhang, L.V. Besteiro, J. Liu, et al., Nano Energy 79 (2021) 105416. doi: 10.1016/j.nanoen.2020.105416

  32. [32]

      L. Li, H. Feng, X. Wei, et al., Nanomaterials 10 (2020) 253. doi: 10.3390/nano10020253

  33. [33]

      Y. Yang, C. Zhang, D. Huang, et al., Appl. Catal. B: Environ. 245 (2019) 87–99. doi: 10.1016/j.apcatb.2018.12.049

  34. [34]

      T.K. Sahu, M.K. Mohanta, M. Qureshi, J. Power Sources 445 (2020) 227341. doi: 10.1016/j.jpowsour.2019.227341

  35. [35]

      D.K. Bediako, Y. Surendranath, D.G. Nocera, J. Am. Chem. Soc. 135 (2013) 3662–3674. doi: 10.1021/ja3126432

  36. [36]

      S. Zhong, Y. Xi, S. Wu, et al., J. Mater. Chem. A 8 (2020) 14863–14894. doi: 10.1039/d0ta04977h

  37. [37]

      B. An, W. Liu, J. Dong, et al., Chin. J. Catal. 65 (2024) 113–125. doi: 10.1016/S1872-2067(24)60124-5

  38. [38]

      J. Huang, K. Li, L. Wang, et al., Chin. Chem. Lett. 33 (2022) 3787–3791. doi: 10.1016/j.cclet.2021.11.028

  39. [39]

      E.A. Turhan, S.V.K. Nune, E. Ülker, et al., Chem. : Eur. J. 24 (2018) 10372–10382. doi: 10.1002/chem.201801412

  40. [40]

      P. Chen, K. Xu, T. Zhou, et al., Angew. Chem. Int. Ed. 55 (2016) 2488–2492. doi: 10.1002/anie.201511032

  41. [41]

      M.D. Yogesh Surendranath, D.G. Nocera, J. Am. Chem. Soc. 131 (2009) 2615–2620. doi: 10.1021/ja807769r

  42. [42]

      T.W. Kim, K.S. Choi, Science 343 (2014) 990–994. doi: 10.1126/science.1246913

  43. [43]

      Z. Feng, S. Wang, G. Yin, et al., Nanotechnology 33 (2022) 255602. doi: 10.1088/1361-6528/ac5a85

  44. [44]

      S. Wang, P. Chen, J.H. Yun, et al., Angew. Chem. Int. Ed. 56 (2017) 8500–8504. doi: 10.1002/anie.201703491

  45. [45]

      X. Chang, T. Wang, P. Zhang, et al., J. Am. Chem. Soc. 137 (2015) 8356–8359. doi: 10.1021/jacs.5b04186

  46. [46]

      K. Chu, X. Li, Q. Li, et al., Small 17 (2021) e2102363. doi: 10.1002/smll.202102363

  47. [47]

      B. Zhang, S. Yu, Y. Dai, et al., Nat. Commun. 12 (2021) 6969. doi: 10.1038/s41467-021-27299-0

  48. [48]

      J.B. Pan, B.H. Wang, S. Shen, et al., Angew. Chem. Int. Ed. 62 (2023) e202307246. doi: 10.1002/anie.202307246

  49. [49]

      M. Wang, Z. Wang, B. Zhang, et al., ACS Catal. 10 (2020) 13031–13039. doi: 10.1021/acscatal.0c03671

  50. [50]

      Y. Liu, M. Wang, Y. Nie, et al., Anal. Chem. 91 (2019) 6250–6258. doi: 10.1021/acs.analchem.9b00965

  51. [51]

      M. Sheng, L. Yu, Y. Peng, et al., Anal. Chem. 96 (2024) 41–48. doi: 10.1021/acs.analchem.3c02545

  52. [52]

      Y. Liu, X. Chen, M. Wang, et al., Green Chem. 20 (2018) 5520–5527. doi: 10.1039/c8gc03010c

  53. [53]

      X. Li, J. Wan, Y. Ma, et al., Chem. Eng. J. 404 (2021) 127054. doi: 10.1016/j.cej.2020.127054

  54. [54]

      H. Bai, X. Li, Y. Zhao, et al., Appl. Surf. Sci. 538 (2021) 148150. doi: 10.1016/j.apsusc.2020.148150

  55. [55]

      R. Kumar, R.K. Singh, S.K. Yadav, et al., J. Alloys Compd. 683 (2016) 38–45. doi: 10.5958/2249-7307.2016.00032.3

  56. [56]

      S. Wang, T. He, J.H. Yun, et al., Adv. Funct. Mater. 28 (2018) 1802685. doi: 10.1002/adfm.201802685

  57. [57]

      L. Deng, S.F. Hung, S. Zhao, et al., Energy Environ. Sci. 16 (2023) 5220–5230. doi: 10.1039/d3ee01856c

  58. [58]

      X. An, C. Hu, H. Liu, et al., J. Mater. Chem. A 5 (2017) 24989–24994. doi: 10.1039/C7TA08809D

  59. [59]

      S. Wang, T. He, P. Chen, et al., Adv. Mater. 32 (2020) 2001385. doi: 10.1002/adma.202001385

  60. [60]

      L. Li, R. Zhang, Y. Lin, et al., Chem. Eng. J. 453 (2023) 139970. doi: 10.1016/j.cej.2022.139970

  61. [61]

      H. Yuan, W. Shi, J. Lu, et al., Chem. Eng. J. 454 (2023) 140442. doi: 10.1016/j.cej.2022.140442

  62. [62]

      Y. Zhao, L. Xu, X. Wang, et al., Nano Today 43 (2022) 101428. doi: 10.1016/j.nantod.2022.101428

  63. [63]

      Q. Wu, K. Zhang, D. Wang, et al., Electrochim. Acta 449 (2023) 142241. doi: 10.1016/j.electacta.2023.142241

  64. [64]

      H. Zhang, D. Li, W.J. Byun, et al., Nat. Commun. 11 (2020) 4622. doi: 10.1038/s41467-020-18484-8

• 

  Figure 1  (a) Schematic illustration of the preparation of BiVO₄/BNQDs/CoBi photoanode. Scanning electron microscopy (SEM) images of photoanodes: (b) BiVO₄, (c) BiVO₄/BNQDs, (d) BiVO₄/CoBi and (e) BiVO₄/BNQDs/CoBi. (f-i) HR-TEM images and (j) TEM-EDS mapping of BiVO₄/BNQDs/CoBi photoanode.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (a) XRD patterns of BiVO₄, BiVO₄/BNQDs, BiVO₄/CoBi and BiVO₄/BNQDs/CoBi photoanodes. (b) FT-IR spectra of BiVO₄, BiVO₄/BNQDs and BiVO₄/BNQDs/CoBi photoanodes.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  High-resolution XPS spectra of BiVO₄/BNQDs/CoBi photoanode: (a) Bi 4f, (b) V 2p, (c) N 1s, (d) B 1s, (e) O 1s and (f) Co 2p.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  (a) LSV curves, (b) ABPE spectra, (c) IPCE spectra, (d) APCE spectra of BiVO₄, BiVO₄/BNQDs, BiVO₄/CoBi and BiVO₄/BNQDs/CoBi photoanodes.

  下载: 全尺寸图片 幻灯片
  

  Figure 5  (a) EIS curves, (b) Mott-Schottky plots, (c) charge injection efficiency curves and (d) charge separation efficiency curves of BiVO₄, BiVO₄/BNQDs, BiVO₄/CoBi and BiVO₄/BNQDs/CoBi photoanodes.

  下载: 全尺寸图片 幻灯片
  

  Figure 6  (a) UV–vis DRS and Tauc plots of BiVO₄, BNQDs and CoBi powders. (b, c) UPS plots of BiVO₄, BNQDs and CoBi powders. (d) Charge flow paths for BiVO₄/BNQDs/CoBi photoanode.

  下载: 全尺寸图片 幻灯片
  

  Figure 7  (a) TRPL spectroscopy, (b) TPV spectroscopy, (c) OCP-derived carrier transfer lifetimes, (d) IMPS spectra, (e) the average charge transport time τ_t and (f) charge transfer and recombination rate constant of BiVO₄, BiVO₄/BNQDs, BiVO₄/CoBi and BiVO₄/BNQDs/CoBi photoanodes.

  下载: 全尺寸图片 幻灯片
•