An Inorganic–Organic Hybrid Polymer Cocatalyst for Photoelectrochemical Water Oxidation with Dual Functions of Accelerating Kinetics and Improving Charge Transfer

Open AccessCCS ChemistryRESEARCH ARTICLE13 Apr 2021An Inorganic–Organic Hybrid Polymer Cocatalyst for Photoelectrochemical Water Oxidation with Dual Functions of Accelerating Kinetics and Improving Charge Transfer Wenlong Guo, Si Shu, Tong Zhang, Yanlin Tao, Yinqiong Xie and Xi Liu Wenlong Guo Chongqing Key Laboratory of Green Synthesis and Applications, College of Chemistry, Chongqing Normal University, Chongqing 401331 , Si Shu Chongqing Key Laboratory of Green Synthesis and Applications, College of Chemistry, Chongqing Normal University, Chongqing 401331 , Tong Zhang Chongqing Key Laboratory of Green Synthesis and Applications, College of Chemistry, Chongqing Normal University, Chongqing 401331 , Yanlin Tao Chongqing Key Laboratory of Green Synthesis and Applications, College of Chemistry, Chongqing Normal University, Chongqing 401331 , Yinqiong Xie Chongqing Key Laboratory of Green Synthesis and Applications, College of Chemistry, Chongqing Normal University, Chongqing 401331 and Xi Liu *Corresponding author: E-mail Address: [email protected] Chongqing Key Laboratory of Green Synthesis and Applications, College of Chemistry, Chongqing Normal University, Chongqing 401331 https://doi.org/10.31635/ccschem.021.202100785 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesTrack Citations ShareFacebookTwitterLinked InEmail Oxygen evolution cocatalysts (OECs) play important roles in improving the efficiency of photocatalysts in solar water splitting. Inorganic–organic hybrid polymers (IOHPs), which have good electrolyte accessibility and evenly distributed active sites, are expected to be promising OECs. Here, a novel IOHP [Co(Bpn)2(SCN)2]n ( 1, Bpn = 2,6-bis(4-pyridyl)-naphthalene, SCN = thiocyanate ion) exhibited a two-dimensional (2D) layer structure with (4, 4) topology, was constructed by Bpn ligands connecting Co(II) ions, and was decorated on BiVO4 photoanodes for photoelectrochemical (PEC) water oxidation. The 1/BiVO4 hybrid electrode showed significantly negative onset potential and approximately 3.7 times higher photocurrent density at 1.23 V versus reversible hydrogen electrode (RHE) compared with the bare BiVO4. The mechanisms for the improved PEC efficiency were investigated and mainly ascribed to enhanced water oxidation kinetics and increased charge separation and transfer properties. This work provides a promising OEC candidate for PEC water oxidation and sheds light on the attractive application prospect of IOHPs for solar water splitting. Download figure Download PowerPoint Introduction Due to a four-electron four-proton-coupled reaction, sluggish kinetics for the oxygen evolution reaction mainly limits the photoelectrochemical (PEC) water oxidation performance of photoanodes.1,2 Development and improvement of oxygen evolution cocatalysts (OECs) for PEC water splitting have driven much research efforts in recent years.3–5 Some OECs based on Fe, Co, and Ni (FeOOH,5 NiOOH,1 CoPi,4 Co3O4,6 etc.) have been synthesized and deposited on several well-studied metal oxide photoanodes (α-Fe2O3,7 WO3,8 BiVO4,5 etc.). With these OECs, more photogenerated holes are extracted from the bulk material and effectively injected into the electrolyte to participate in the water oxidation reactions. Nevertheless, there is still a large proportion of carriers that recombine in the interface between the light absorber and OECs and on the surface of OECs. Therefore, it is quite necessary to develop new types of OECs with high efficiency to further improve the PEC performance of photoanodes. Inorganic–organic hybrid polymers (IOHPs), which are constructed by inorganic solids and functional organic ligands, have received widespread attention.9–16 These materials have the combined advantages of inorganic, organic, and polymeric materials. The concomitant but separate molecular designs for inorganic parts with small dimensions are allowed. Meanwhile, the organic parts can serve as spacers, controllers, and templates at the molecular-engineering level.17–20 These features give IOHPs the advantages of high surface area, high designability, porosity and tunable properties, good electrolyte accessibility, and evenly distributed active sites.10,11,21,22 IOHPs, therefore, have great potential to be a new family of OECs with high efficiencies. To date, Fe-based [MIL-100(Fe),23 MIL-53(Fe),24 MIL-101(Fe),25,26 and NH2-MIL-101(Fe)25], Co-based [ZIF-8,27 ZIF-67,28 ZIF-8/67,29 and Co2(bim)430], and Ti-based [NH2-MIL-125(Ti)31,32] IOHPs have been applied to photoanodes for PEC water oxidation. These hybrid electrodes achieve PEC performances superior to the pristine electrodes, demonstrating the positive effect for PEC water oxidation of these IOHPs and the fascinating potential of IOHPs as OECs. However, according to our knowledge, not many IOHPs are used for PEC water oxidation to date and further improvement of efficiencies remains for those studied IOHPs. Therefore, to develop new IOHPs to serve as OECs is of great significance for broadening the application of IOHPs and further improving the PEC efficiencies of solar water splitting. In this work, a novel IOHP [Co(Bpn)2(SCN)2]n ( 1, Bpn = 2,6-bis(4-pyridyl)-naphthalene, SCN = thiocyanate ion) is synthesized using the hydrothermal method. IOHP 1 exhibits a two-dimensional (2D) layer structure with (4, 4) topology and is constructed by Bpn ligands connecting Co(II) ions. Then, 1 is deposited on BiVO4, a promising metal oxide photoanode that suffers from sluggish water oxidation kinetics,1,5 to evaluate its effects as an OEC. The 1/BiVO4 photoanode shows a remarkable negative onset potential and approximately 3.7 times higher photocurrent density compared with the pristine BiVO4 at 1.23 V versus reversible hydrogen electrode (RHE). By exploring the mechanisms for the increased efficiency, it is found that the existence of 1 not only improves the kinetics for water oxidation but also enhances the charge separation and transfer properties. These results reveal that 1 is a promising OEC candidate for PEC water oxidation and the application prospect of IOHPs for solar water splitting is highly attractive. Experimental Section Materials and instrumentation All chemicals were obtained from commercial sources and were used without further purification. The infrared (IR) vibration spectra were collected with a KBr pellet using an FT-IR 8400S spectrometer (Shimadzu, Kyoto, Japan) in the wavelength range of 4000–400 cm−1. A Vario EL III elemental analyzer (Elementar, Langenselbold, Germany) was used to analyze the C, H, and N elements. Powder X-ray diffraction (PXRD) measurements were conducted using an X-ray diffractometer (XRD-6100; Shimadzu, Japan) with a scan speed of 2 °/min. A HENVEN-HJ HCT-3 thermoanalyzer (Hengjiu, China) was used to conduct the thermogravimetric analyses (TGA) with a heating rate of 10 °C/min under an air atmosphere. A UV–vis spectrophotometer (PE Lambda 35; PerkinElmer, United States) was used to record the diffuse reflectance spectra at room temperature. The samples were ground into fine powder and pressed onto the holder of a thin glass slide. The standard reference was the BaSO4 plate with a 100% reflectance. ,The absorbance was derived from the reflectance spectra based on the Kubelka–Munk function,33 α/S = (1 − R)2/2R. For this function, α is the absorption coefficient, R is the reflectance, and S is the scattering coefficient. When the particle size is larger than 5 μm, S is practically wavelength independent. The morphologies and elements distribution of the films were obtained using scanning electron microscopy [SEM; FEI Inspect F50 (FSEM); Thermofisher, United States] and transmission electron microscopy (TEM; JEM-2100F, JEOL, Japan). Synthesis of 1 Method A: A mixture of Co(NO3)2·6H2O (5.8 mg, 0.02 mmol), Bpn (5.6 mg, 0.02 mmol), potassium thiocyanate (KSCN) (11.7 mg, 0.12 mmol), and CH3CN/H2O (12 mL, v/v = 3∶1) was sealed in a 25 mL Teflon-lined stainless-steel reactor. The reactor was held at 120 °C for 3 days, and then was cooled to room temperature at a rate of 10 °C/h. Red block crystals were obtained in 53% yield (based on Bpn). Anal. Calcd for C42H28CoN6S2 (739.75, %): C, 68.19; H, 3.81; N, 11.36. Found: C, 67.92; H, 3.69; N, 11.12. FT-IR (KBr pellet, ν/cm−1): 3053 (w), 2099 (s), 1609 (m), 1547 (w), 1488 (w), 1420 (m), 1319 (w), 1222 (m), 1066 (m), 1035 (w), 1012 (w), 911 (m), 815 (s), 806 (s), 695 (s), 628 (w), 594 (w), 534 (s), 468 (s). ν(C≡N) vibration of SCN− anions: 2099 (s). Method B: A mixture of Co(NO3)2·6H2O (5.8 mg, 0.02 mmol), Bpn (5.6 mg, 0.02 mmol), KSCN (5.6 mg, 0.06 mmol), KBr (7.1 mg, 0.06 mmol), and CH3CH2OH/H2O (6 mL, v/v = 3∶1) was sealed in a 25 mL Teflon-lined stainless-steel reactor. The reactor was held at 120 °C for 4 days, and then was cooled to room temperature at a rate of 10 °C/h. Red block crystals were obtained in 61% yield (based on Bpn). X-ray single-crystal structure determinations First, we selected measurable single crystals of 1 using an optical microscope and glued them to thin glass fibers. A diffractometer (Bruker APEX3; Bruker, Germany) equipped with a graphite-monochromated Mo Kα radiation (λ = 0.71073 Å) at 293 K was used to collect the data. A ω-scan technique was performed to record the intensity data sets which were then reduced using Bruker APEX3 software.34 Direct methods were used to solve the structures, and the full-matrix least-squares techniques were performed to refine them. The difference Fourier maps were used to locate the nonhydrogen atoms which were subjected to the anisotropic refinement. Per the theoretical models, hydrogen atoms were added. The crystallographic software (Siemens SHELXTL version 5 package; Siemens(Bruker), Germany) was used to perform all the calculations.35 Crystallographic data and structural refinements for 1 are summarized in Supporting Information Table S3. Selected bond lengths and angles are summarized in Supporting Information Table S4. More details of the crystallographic studies and the atom displacement parameters are provided in the Supporting Information. Crystallographic data for 1 C42H28CoN6S2, Mr = 739.75, crystal size: 0.16 × 0.09 × 0.03 mm3, orthorhombic, space group, P1, a = 7.656(6), b = 9.229(3), c = 13.485(5) Å, V = 851.5(8) Å3, T = 293(2) K, Z = 1, Dcalcd = 1.443 g cm−3, μ = 0.668 mm−1, F(000) = 381 e. A total of 26,918 reflections, with 3174 unique, were measured in the range 2.860 < θ < 25.499°(Rint = 0.0687). Structure solution and refinement was based on 3174 observed reflections with I < 2 σ(I) and 232 refined parameters, 0 restraints gave R1 = 0.0648, wR2 = 0.1662, and S = 1.000 (R1 = 0.0752, wR2 = 0.1757, and S = 1.000 for all reflections), Δρ(max/min) = 1.486/−0.965 e Å−3. The supplementary crystallographic data for this paper are contained in The Cambridge Crystallographic Data Centre (CCDC). These data can be obtained for free from CCDC via www.ccdc.cam.ac.uk/data_request/cif. Synthesis of BiVO4 and 1/BiVO4 photoanodes BiVO4 films were synthesized per the literature procedure.36 Briefly, 15 mM Bi(NO3)3, 400 mM KI, and 30 mM lactic acid were dissolved in 50 mL deionized water; 46 mM p-benzoquinone was dissolved in 20 mL ethanol. The latter solution was slowly added into the former one. The pH of the mixed solution was adjusted to 3.4 ± 0.05 by adding HNO3. BiOI films were deposited on the fluorine-doped tin dioxide (FTO) glass substrate by potentiostatic electrodeposition in this mixed electrolyte (−0.35 V vs Ag/AgCl for 20 s and −0.1 V vs Ag/AgCl for 1020 s). Then, the BiOI electrodes were converted to BiVO4 with a thermal treatment in air at 450 °C for 2 h by dropping 50 μL cm−2 of a 200 mM VO(acac)2 dimethyl sulfoxide solution on the film surface. After the thermal treatment, excess V2O5 was removed by soaking the film in 1 M NaOH solution for 30 min with stirring. The 1/BiVO4 photoanode was prepared by soaking a BiVO4 film in the Teflon-lined stainless-steel reactor containing solution used to fabricate 1 (Method B). The reactor was held at 120 °C for 1, 3, 5, and 7 h, respectively. After the reaction, the film was taken out and rinsed with ethanol and water. Results and Discussion Characterizations and structure analysis of complex 1 As described in the Experimental Section, complex 1 can be obtained in a mixed solvent of CH3CN and H2O or a mixed solvent of CH3CH2OH and H2O. However, the red crystals obtained in a mixed solvent of CH3CH2OH and H2O have better crystallinity than those in a mixed solvent of CH3CN and H2O. The crystals obtained have the same phase under both conditions, which is confirmed by the elemental analyses, IR spectroscopy, and PXRD. As shown in Supporting Information Figure S3, the characteristic vibration bands of the aromatic rings are located at 1609–1420 cm−1, near 1222 cm−1, and between 1066–911 cm−1, and strong absorption bands of the SCN− ligands appear at ca. 2099 cm−1. Furthermore, the phase purity of the bulk samples of 1 is confirmed through comparing their PXRD data with the simulated data from the CIF file ( Supporting Information Figure S4). In addition, complex 1 is stable in air and can maintain its crystallinity at room temperature for several days. The TGA curve shows that 1 decomposes at the decomposition point Tonset of approximately 286 °C ( Supporting Information Figure S5), indicating that it is a stable material and suitable for various practical applications. The optical property of 1 is investigated by the relatively diffuse reflection and a normalized absorbance is derived from the spectrum based on the Kubelka–Munk function ( Supporting Information Figure S6). An onset wavelength of ∼700 nm is observed from the absorption spectrum, suggesting an optical band gap of ∼1.77 eV. The band structure calculated by density functional theory (DFT) also reveals the semiconductor character of 1 (for more details, see Supporting Information Figure S2). Complex 1 exhibits a 2D layer structure with (4, 4) topology, which is constructed by Bpn ligands connecting Co(II) ions. As shown in Figure 1, the asymmetric unit consists of one half Co(II) cation, two halves of Bpn ligands, and one thiocyanate group. The Co1 cation locates in a compressed centrosymmetric octahedral coordination geometry, which consists of two Nthiocyanate atoms from two distinct SCN− groups, and four Npyridyl atoms from four distinct Bpn ligands. The bond length of Co–Nthiocyanate is 2.068(4) Å, while the average bond length of Co–Npyridyl is 2.212(4) Å (Figure 1a). The SCN− groups coordinate to Co(II) cations as terminal ligands via N atoms. The Bpn ligands bridge Co(II) ions to form a 2D planar layer approximately paralleling crystal plane [1 1 1] with (4, 4) topological nets when the Co(II) cations are regarded as four-connected nodes and the Bpn ligands are considered bridging linkers (Figure 1b). These planar layers are further stacked together in a complementary way to form a three-dimensional (3D) structure via abundant weak π–π supramolecular interactions (Figures 1c and 1d and Supporting Information Table S5). It should be noted that no solvent molecules reside in these layers although a ∼4.5% void space of the crystal is calculated using the PLATON program. A 3D network of 1 along the [1 1 1] direction is provided in Supporting Information Figure S1. Figure 1 | (a) View of the coordination environment of Co(II) cation in 1. Symmetry codes: A: −x + 1, −y + 1, −z; B: −x + 2, −y + 1, −z − 1; C: −x + 1, −y + 2, −z − 1. (b) 2D (4, 4) topological layer constructed by Bpn ligands bridging Co(II) cations. (c) Packing view of 2D layers in 1. (d) View of the 3D network of 1 along the [1 1 1] direction. All H atoms are omitted for clarity. Download figure Download PowerPoint Characterizations of the hybrid 1/BiVO4 photoanode To evaluate the performance of this IOHP as an OEC, 1 is deposited on BiVO4 because BiVO4 is a promising photoanode that suffers sluggish water oxidation kinetics, which makes it an excellent test platform.1,5 According to the XRD patterns (Figure 2a), the pure phase of the BiVO4 film is obtained. For the 1/BiVO4 electrode, the positions and intensities of the characteristic diffraction peaks corresponding to BiVO4 are almost unchanged. This implies that the hydrothermal process has no effect on the phase and crystallinity of BiVO4. Obviously, several characteristic peaks of 1 are observed, demonstrating the growth of the crystalline 1 on BiVO4. IR spectra also confirm the formation of 1 on the surface of BiVO4. As shown in Figure 2b, the vibration absorption peaks of aromatic rings and the SCN− ligands belonging to 1 are collected from 1/BiVO4. SEM measurements show that the morphology of BiVO4 is porous, which is consistent with the previous work (Figures 3a and 3b).36 After the deposition of 1 by the hydrothermal method, the morphology of the hybrid electrode is similar to the pristine BiVO4 (Figures 3c and 3d). However, some of the particles on the surface become smaller and agglomerate. For comparison, BiVO4 films are put into the reactor with pure solvent, and after the hydrothermal process, a similar change of the morphology is observed for the BiVO4 film ( Supporting Information Figure S7). This indicates that the solvent has a slight etching effect on the material under the hydrothermal environment, and the resulting small particles are agglomerated, probably due to the surface effect. TEM and high-resolution TEM (HRTEM) were conducted to further demonstrate the formation of 1 on BiVO4. A lattice fringe spacing of ∼3.09 nm from the HRTEM image originates from the (112) plane of BiVO4, and an approximately 10 nm thin layer of complex 1 wraps on the surface of BiVO4 (Figures 3e and 3f). Furthermore, the distributions of C, N, S, and Co elements on 1/BiVO4 are tested using the HRTEM elemental mapping analysis, and the results confirm the existence of 1 on the surface of BiVO4 (Figures 3g–3k). Figure 2 | (a) XRD patterns of BiVO4 and 1/BiVO4 films. The red square indicates characteristic diffraction peaks of 1. The clover symbol indicates the diffraction peaks of the tin oxide. (b) The FT-IR spectra of BiVO4 and 1/BiVO4 films. Download figure Download PowerPoint Figure 3 | SEM morphologies of BiVO4 (a and b) and 1/BiVO4 (c and d). HRTEM of 1/BiVO4 at two different test areas (e and f). High-angle annular dark-field scanning TEM (HAADF-STEM) (g) and the corresponding HRTEM elemental mappings of 1/BiVO4 (h−k). Download figure Download PowerPoint PEC water oxidation performance The linear sweep voltammetry (LSV) is used to test the PEC efficiency of BiVO4 and 1/BiVO4 in 0.2 M Na2SO4 aqueous solution. The comparisons of the bare BiVO4 and the 1/BiVO4 photoanodes deposited for various times suggest that 3 h is the optimal hydrothermal time ( Supporting Information Figure S8). Hereafter, 1/BiVO4 represents the BiVO4 film coated with 1 for a 3 h reaction time. As shown in Figure 4a, 1/BiVO4 shows much higher photocurrent density and dramatically negative shift onset potential compared with BiVO4. Particularly, the photocurrent density of 1/BiVO4 (∼2.60 mA/cm2) is 3.7 times higher than that of BiVO4 (∼0.70 mA/cm2) at 1.23 V versus RHE. Comparisons of the PEC performances between 1 and other IOHPs as OECs reported in previous work are summarized in Supporting Information Table S2. By adding 1 M sulfite into the electrolyte as a hole scavenger, 1/BiVO4 also exhibits higher current density than BiVO4, suggesting that the coating of 1 makes more holes reach the surface (Figure 4b). By calculations, the hole-injection efficiencies for water oxidation reveal that the existence of 1 significantly boosts the efficiency for PEC water oxidation of BiVO4 (Figure 4c). Applied bias photon-to-current efficiency (ABPE) of 1/BiVO4 is dramatically enhanced with a value calculated to be 0.62% at 0.8 V versus RHE, which is approximately 10 times higher than that of BiVO4 (0.06%, Figure 4d). In addition, the incident photon to current efficiency (IPCE) of BiVO4 significantly increases after the coating with complex 1 (Figure 4e). Specifically, the efficiency measured in the sulfate solution at 1.23 V versus RHE is enhanced from 6.65% to 40.12% at the wavelength of 425 nm. The power densities of the incident light at various wavelengths for the IPCE measurements are provided in Supporting Information Figure S15. In addition, the amount of O2 generated during the PEC process is measured, and the calculated faradaic efficiencies for 1/BiVO4 are shown in Figure 4f. These results suggest that 1 is highly selective for PEC water oxidation. Additionally, 1/BiVO4 shows good photostability in 0.2 M Na2SO4 electrolyte ( Supporting Information Figure S9). Figure 4 | LSV curves of BiVO4 and 1/BiVO4 measured in 0.2 M Na2SO4 solution (a) without and (b) with 1 M Na2SO3 as a hole scavenger (pH 7). Illumination source: air mass (AM) 1.5G (100 mW/cm2). Scan rate: 25 mV/s. (c) Calculated charge injection efficiency of BiVO4 and 1/BiVO4. (d) ABPE curves of BiVO4 and 1/BiVO4 as a function of the applied potentials. (e) IPCE plots of BiVO4 and 1/BiVO4. (f) Time course of generated O2 amount and the faradaic efficiency of the 1/BiVO4 photoanode tested in 0.2 M Na2SO4 at 1.23 V vs RHE under an AM 1.5G illumination (100 mW/cm2). Download figure Download PowerPoint Mechanisms investigations To investigate the mechanisms of the improved efficiency of the 1/BiVO4 photoanode, electrochemical impedance spectroscopy (EIS) is applied to examine the charge-transfer properties across the electrode/electrolyte interface (Figure 5a).37,38 The arc in the Nyquist diagram for 1/BiVO4 is much smaller than that for BiVO4, suggesting that 1/BiVO4 has a smaller charge-transfer resistance. Through the analysis of the equivalent circuit, the value of Rct, which is the resistance of charge transfer between the surface and the electrolyte, is fitted to be 363 and 154 Ω for BiVO4 and 1/BiVO4, respectively ( Supporting Information Table S1). This shows that 1 can accelerate the kinetics for water oxidation of BiVO4. This conclusion is further confirmed by the LSV measurements performed in dark condition in which 1/BiVO4 exhibits an obvious cathodic shift relative to BiVO4 for the electrocatalytic water oxidation ( Supporting Information Figure S16). To exclude the influence of the films' morphologies, the electrochemical capacitance of BiVO4 and 1/BiVO4 are measured using cyclic voltammetry in a non-faradaic potential range, and the results show that these two electrodes have similar electrochemically-active surface areas ( Supporting Information Figure S10). Figure 5 | (a) EIS plots of BiVO4 and 1/BiVO4 tested in 0.2 M Na2SO4 electrolyte at 1.23 V vs RHE under AM 1.5 G illumination (100 mW/cm2). (b) Calculated charge separation efficiencies of BiVO4 and 1/BiVO4. (c) Mott–Schottky plots of the 1 film measured in 0.2 M Na2SO4 at various frequencies. The inset is the calculated band positions of 1 and BiVO4 with respect to the RHE level. Plots of the CIMPS complex plane of (d) BiVO4 and (e) 1/BiVO4 measured in a 0.2 M Na2SO4 solution with different light intensity at 1.23 V versus RHE. (f) Calculated electron transport time of BiVO4 and 1/BiVO4 as a function of the light intensity. Download figure Download PowerPoint In addition to the boosted kinetics, the charge separation and transfer properties, which also have crucial effect on the PEC performance of photoanodes, are explored. As discussed above (Figure 4b), 1/BiVO4 shows higher current density than BiVO4 for sulfite oxidation, which indicates that more holes reach the electrode's surface for 1/BiVO4. This can be ascribed to the photoexcitation of 1 as a semiconductor with a band gap of ∼1.77 eV ( Supporting Information Figure S6) or the enhanced charge separation and transfer for the hybrid 1/BiVO4 electrode. To verify the first conjecture, 1 is coated on the FTO glass substrate to test the PEC performance. In fact, the 1/FTO electrode shows negligible current densities measured in both sulfate and sulfite electrolytes ( Supporting Information Figure S12). Additionally, IPCE values measured in Na2SO4 solution with and without Na2SO3 (Figure 4e and Supporting Information Figure S13) reveal that the absorbed photons for 1/BiVO4 beyond the wavelength of 520 nm caused by 1 are not effectively utilized for the water or sulfite oxidation reactions. In contrast, 1/BiVO4 shows a higher separation efficiency than BiVO4 by calculations (Figure 5b). The absorbed photon flux for BiVO4 and 1/BiVO4 films is integrated to be 4.37 and 4.68 mA, respectively ( Supporting Information Figure S11). This demonstrates the second conjecture is one of the main reasons for the improved PEC performance of 1/BiVO4. To further explain this property, the flat-band potential of 1 is evaluated using the Mott–Schottky measurements at various frequencies (Figure 5c). The plots indicate that the flat-band potential of 1 is approximately −0.07 V versus RHE, and the positive slopes of the Mott–Schottky curves suggest that 1 is an n-type semiconductor. The conduction band energy (ECB) of an n-type semiconductor is generally 0.1–0.2 eV more negative than its Fermi level (the flat band potential).39 Therefore, the ECB of 1 is calculated to be located between −0.17 and −0.27 eV, and its valence band energy (EVB) should be within the range of 1.6 to 1.5 eV considering the band gap of 1 (∼1.77 eV). The flat band potential of BiVO4 is estimated to be about 0.07 V versus RHE ( Supporting Information Figure S14). Therefore, given a ∼2.41 band gap (the inset of Supporting Information Figure S11a), the ECB and EVB of BiVO4 is estimated to be within the range of −0.03 to −0.13 eV and 2.38 to 2.28 eV, respectively, consistent with the previous work.40,41 Apparently, the ECB and EVB of 1 are, respectively, higher than the corresponding values of BiVO4 (the inset of Figure 5c). Thermodynamically, this matching type of the band positions is beneficial for the separation and transport of photogenerated carriers.42–44 Additionally, controlled intensity-modulated photocurrent spectroscopy (CIMPS) is conducted to further investigate the charge separation and transfer properties of BiVO4 and 1/BiVO4.45–47 The CIMPS Nyquist plots measured at various light intensities using a 365 nm light-emitting diode (LED) are depicted in Figures 5d and 5e. By comparison, the radius of the semicircular shape at certain light intensity for 1/BiVO4 is larger than the corresponding one for BiVO4. This suggests that more holes are collected for 1/BiVO4 during the PEC water oxidation, which accords with the results of the PEC efficiencies. Furthermore, the electron transfer times (τ) of BiVO4 and 1/BiVO4 at various light intensities are calculated based on the frequency of the lowest point in the CIMPS complex plane, as shown in Figure 5f. 1/BiVO4 has smaller τ values than BiVO4, demonstrating that 1/BiVO4 processes a better carrier mobility. Conclusion In this work, a novel IOHP [Co(Bpn)2(SCN)2]n ( 1, Bpn = 2,6-bis(4-pyridyl)-naphthalene) is synthesized using a hydrothermal method and applied on the BiVO4 photoanode as an OEC. The hybrid electrode 1/BiVO4 shows remarkable negative onset potential and significant higher photocurrent density compared with the pristine BiVO4. Particularly, the photocurrent density of 1/BiVO4 is 3.7 times higher than the bare BiVO4 at 1.23 V versus RHE. By investigating the mechanisms, it is found that the existence of 1 highly boosts the hole-injection efficiency into the electrolyte by accelerating the water oxidation kinetics. In addition, 1 can improve the charge separation and transfer properties due to the suitable match of the band positions with BiVO4, which is further confirmed using the CIMPS measurements. Our work provides a promising OEC and investigates the mechanisms of the improved efficiency, which highlights the advantages of IOHPs as OECs and their potential applications for solar water splitting. Supporting Information Supporting Information is available and includes electrochemical and PEC measurements, DFT calculation approach methodology, view of the 3D network of 1, electronic band structure of 1, FT-IR spectrum, experimental and simulated PXRD patterns, TGA curve, relative diffuse reflectance and absorption spectrum, SEM morphologies, PEC performances, stability test, cyclic voltammograms, absorbance spectra and absorbed photon flux, current-time curves, IPCE plots, Mott–Schottky plots, light power density spectrum, LSV, values of the equivalent circuit elements, and crystal and structure refinement data. Declaration of Competing Interest The authors declare that they have no known competing financial interest