Photoelectrochemical H₂ evolution on WO₃/BiVO₄ enabled by single-crystalline TiO₂ overlayer modulations

Abstract

Tungsten oxide/bismuth vanadate (WO₃/BiVO₄) has emerged as a promising photoanode material for photoelectrochemical (PEC) water splitting owing to its facilitated charge separation state differing significantly from single phase materials. Practical implementation of WO₃/BiVO₄ is often limited by poor stability arising from the leaching of V⁵⁺ from BiVO₄ during PEC operations. Herein, we demonstrate that the synthesis of a tungsten oxide/bismuth vanadate/titanium oxide (WO₃/BiVO₄/TiO₂) heterostructure onto a fluorine-doped tin oxide-coated glass substrate through a combined simple hydrothermal-spin coating strategy will advance PEC performance while slowing water oxidation kinetics and improving photostability. We show that surface postmodification with a nanometer-thick layer of (1 0 1) monofacet-selective single-crystalline TiO₂ provides stable photocurrent density, up to 1.04 mA cm⁻² at 1.23 V (compared to a reversible hydrogen electrode in 0.5 M Na₂SO₄), with excellent quantum efficiency (45% at 460 nm) and long-term photostability (24 h). Interestingly, crystalline TiO₂ activation layers behave differently from previous TiO₂ amorphous layers, blocking surface defects while improving corrosion resistance, photostability, and the electron transfer process. These results indicate a ≈2.5 times enhancement in photoelectrocatalytic activity related to referenced WO₃/BiVO₄ photoanodes, encouraging the use of single-crystalline TiO₂ modulations to develop a range of materials for PEC/photocatalytic applications.

---

Eunoak Park

Eunoak Park received her B.S. degree from the Kyungpook National University, Republic of Korea in 2019. After that, she started her master's degree in Kyungpook National University and moved to Inha University, Republic of Korea in 2021. Her research interests are focused on anodization, synthesis of nanostructure materials with heterojunction, photoelectrochemistry, photocatalyst, water electrolysis (hydrogen evolution), and various other applications.

Santosh S. Patil

Dr Santosh S. Patil received his Ph.D. in Chemistry from Shivaji University, Kolhapur (India) in 2016. After that, he conducted his post-doctoral studies at Chonnam National University, POSTECH, Kyungpook National University and Inha University (S. Korea). His research interests are focused on the synthesis of nanostructured materials, thin films, and photocatalysts, structure–composition–property control and investigating the fundamental process of charge-carrier dynamics at the interface of these materials to increase the overall efficiency of operations particularly for catalysis, energy conversion (hydrogen generation), waste water treatment and various other applications.

Hyeonkwon Lee

Hyeonkwon Lee fulfilled a bachelor's and master's degree at Kyungpook National University in South Korea where he studied electrocatalytic and capacitive properties of metal oxides. Subsequently, seeking to meet his scientific interests, he joined the University of Cologne in Germany. His current research topic is to find magnetic fields effect on the growth of metal oxides and their electrochemical properties.

Vijay S. Kumbhar

Dr Vijay S. Kumbhar obtained his PhD investigating the supercapacitive properties of samarium based rare earth chalcogenides from Shivaji University, Kolhapur, India, in 2015. Then, he has a wide range of postdoctoral experience in energy storage at several materials science laboratories in South Korea. He has been associated with Professor Kiyoung Lee since 2018. He is a recipient of the National Postdoctoral Fellowship from the Science and Engineering Research Board, India as well as the Japan Society for the Promotion of Science (JSPS), Japan. Dr Vijay has published about 40 research and review articles. His current research area mainly focusses on the formation of core–shell based materials for energy generation and storage such as electrocatalysis, photoelectrocatalysis, oxygen reduction reaction, and supercapacitors.

Kiyoung Lee

Kiyoung Lee studied Chemical Engineering at Inha University, Korea and obtained his Ph.D. from the University of Erlangen-Nuremberg, Germany in 2013. From 2013 to 2015, he worked at the University of Erlangen-Nuremberg, Germany and the University of California, Riverside, US as a postdoctoral researcher. From 2016 to 2021, he was an assistant and associate professor at the School of Nano & Materials Science and Engineering of Kyungpook National University, Korea. Since 2021, he has been an associate professor at the Department of Chemical Engineering, Inha University, Korea. His current research interests include functional nanomaterial synthesis by electrochemical methods and their use in energy conversion and storage devices.

---

Introduction

Converting solar energy into hydrogen (H₂) energy via photoelectrochemical water splitting is a cost effective and sustainable approach for mitigating overreliance on fossil fuels and global warming issues.^1,2 As H₂ energy is renewable, it can be stored, transported, and even used in fuel cells to generate power and/or is an important feedstock in the chemical industry (e.g., in the catalytic conversion of CO₂ into hydrocarbons, such as methanol, formic acid, methane, or ammonia synthesis).^3–5 Photoelectrochemical (PEC) water splitting (H₂O → H₂ + 1/2 O₂) is based on photogenerated electron–hole pairs in a semiconductor that subsequently catalyze the oxidation and reduction of a H₂O molecule into H₂ and O₂ fuels.⁶ However, the practical implementation of PEC cells remains hindered by stringent material challenges that simultaneously require (i) excellent light harvesting capability, (ii) long-term stability, and (iii) charge-carrier separation and transfer. Since Fujishima Honda's⁷ pioneering work on titanium dioxide (TiO₂) photocatalysts, various visible-light responsive metal oxides (such as tungsten oxide (WO₃), bismuth vanadate (BiVO₄), and iron(III) oxide (Fe₂O₃)) have been examined to absorb a major fraction of the solar spectrum and produce green H₂. However, those single materials are insufficient to deliver high solar-hydrogen-conversion efficiency for practical application.

A number of strategies have been developed by combining two or more semiconductors, co-catalysts for constructing heterostructures, integrated material designs, or composites that provide an opportunity to achieve multiple catalytic functionalities from different counterparts; these have enhanced PEC/photocatalytic performance and stability regarding water splitting and environmental remediation applications. For instance, a combination of WO₃ and BiVO₄ creates a type II heterojunction and enhances PEC performance,^8,9 which mainly stems from the excellent transference of photogenerated electrons to produce a charge-separated state.^10–12 Nonetheless, while WO₃/BiVO₄ is a relatively stable material, it often suffers from severe V⁵⁺ leaching out of the BiVO₄ core during photoelectrocatalytic operations.¹³

Postmodification of a WO₃/BiVO₄ heterojunction with a surface protecting promoter is indispensable to realize potentially high PEC performance with long-term durability. Previous efforts have mainly relied on building an additional passivation layer onto a WO₃/BiVO₄ semiconductor pair.^14–17 For example, Khoomortezaei et al.¹⁴ reported tungsten oxide/bismuth vanadate/bismuth ferrite (WO₃/BiVO₄/BiFeO₃) heterojunctions with a layered coating of BiFeO₃ that lowered the overpotential and charge transfer resistance of the photoelectrode. Baek et al.¹¹ made a tin oxide/tungsten oxide/bismuth vanadate (SnO₂/WO₃/BiVO₄) triple-layer planar photoelectrode that exhibited superior photoconversion efficiency during water decomposition. Recently, Seo et al.¹⁸ added a photocatalytically inactive amorphous TiO₂ layer onto a WO₃/BiVO₄ electrode that reduced charge-carrier recombination by passivating the surfaces. Although previous studies have made remarkable developments, they used amorphous TiO₂ overlayers and met with little success in attaining sufficient stability. Simple solution-processed synthesis of crystalline TiO₂ can offer a means of creating intricately designed WO₃/BiVO₄ heterostructures that otherwise defy creation. We anticipate that depositing a nanometer-thick layer of single-crystalline TiO₂ onto WO₃/BiVO₄ may grant proficient catalytic operations with long-term durability.

Herein, we present the synthesis of a WO₃/BiVO₄/TiO₂ heterojunction using a simple hydrothermal-spin coating method. Postmodification with crystalline TiO₂ was accomplished and played a key role in blocking surface defects in BiVO₄ nanoparticles,¹⁹ leading to substantial improvements in electronic properties and photostability. The WO₃/BiVO₄/TiO₂ heterojunction showed the best photocurrent density of 1.04 mA cm⁻² at 1.23 V when compared to a reversible hydrogen electrode (RHE), an incident photon-to-current conversion efficiency (IPCE) of 25% in response to visible light at 460 nm, and exceptionally high durability (24 h) in 0.5 M Na₂SO₄ at 1.23 V_RHE under standard 1 sun illumination (100 mW cm⁻²). According to gas chromatography analysis, the utmost H₂ evolution rate of 78.8 μL cm⁻² was achieved during PEC water splitting.

Results and discussion

The surface morphologies of WO₃, WO₃/BiVO₄, and WO₃/BiVO₄/TiO₂ electrodes synthesized through a hydrothermal-spin coating method were evaluated with scanning electron microscopy (SEM) images (Fig. 1). The prepared WO₃ showed rough nanoplate shapes with relatively dense distributions ranging from nanometer to submicrometer scales (Fig. 1a). Interconnected nanoplates were grown nearly vertical to the fluorine-doped tin oxide (FTO) surface, creating several void spaces between them. The average length and thickness of the individual nanoplates were ∼400–600 nm and 100–160 nm, respectively, as confirmed by the cross-section and top view SEM images (Fig. 1a and d). Fig. 1c and f depict the top and cross-section views of SEM images after the deposition of BiVO₄ nanoparticles (20 cycles) onto WO₃via a spin coating method and subsequent air annealing at 500 °C. While BiVO₄ nanoparticles with diameters ∼50–80 nm were uniformly coated onto WO₃ nanoplate surfaces, overlayer thicknesses seemed irregular, indicating a possible overgrowth of BiVO₄ nanoparticles (Fig. S1b and c in the ESI†). A thin top layer of TiO₂ was formed onto the WO₃/BiVO₄ by spin coating a titanium butoxide (Ti(OCH₂CH₂CH₂CH₃)₄) solution and subsequent annealing at 500 °C for 1 h (Fig. 1c and f).

Fig. 1 (a–c) Top view and (d–f) cross-sectional FE-SEM images of WO₃, WO₃/BiVO₄, and WO₃/BiVO₄/TiO₂, respectively.

Crystal structures were analyzed further by acquiring X-ray diffraction (XRD) patterns. Fig. 2a shows the prominent diffraction peaks at 23.0°, 23.5°, 24.3°, 28.7°, 34.0°, 41.6°, 49.9°, 55.7°, and 76.4°; these can be assigned to the (0 0 2), (0 2 0), (2 0 0), (1 1 2), (2 2 0), (2 3 2), (−2 4 1), (4 0 2), and (0 6 1) planes, respectively, representing monoclinic WO₃ (JCPDS #89-4476).^20,21 The WO₃/BiVO₄ heterojunction showed preservation of the WO₃ phase and Bragg diffractions at 18.8°, 28.9°, 30.4°, and 39.8°; these correspond to the (1 0 1), (1 1 2), (0 0 4), and (2 1 1) planes, respectively, suggesting the generation of a monoclinic BiVO₄ form (JCPDS #75-1866).^22,23 As illustrated in Fig. 2b (a magnified view of an XRD pattern), the one-cycle coated TiO₂ shows a Bragg peak at 25.3°, confirming the generation of (1 0 1) selective single-crystalline anatase TiO₂ (JCPDS #89-4921) nanoparticles.^24,25 The XRD pattern of a five-cycle spin coated TiO₂ (Fig. S3†) revealed a diffraction peak at 48.1°, implying the (2 0 0) plane of anatase TiO₂ together with the heightened peak intensity of the (1 0 1) crystal plane.

Fig. 2 XRD patterns of (a) FTO, WO₃, WO₃/BiVO₄, and WO₃/BiVO₄/TiO₂ samples and (b) the corresponding magnified view.

High-angle annular dark-field STEM (HAADF-STEM) was employed to investigate crystalline contacts between WO₃/BiVO₄/TiO₂ components. The area marked A in Fig. 3a illustrates a high-resolution transmission electron microscopy (TEM) image with measured lattice spacings of 0.263 nm and 0.164 nm; this agrees well with the (2 2 0) and (4 0 2) planes of WO₃. The selected area electron diffraction (SAED) pattern (Fig. 3d) confirms a polycrystalline nature, and the (2 2 0) and (4 0 2) planes correspond to monoclinic WO₃. Fig. 3e (the area marked B in Fig. 3a) shows clear lattice fringes with lattice spacings of 0.308 nm and 0.349 nm that correspond with the (1 1 2) and (1 0 1) planes of monoclinic BiVO₄ and anatase TiO₂, respectively. The corresponding SAED pattern (Fig. 3f) reveals that single-crystalline TiO₂ is oriented along the (1 0 1) facet. Energy dispersive X-ray spectroscopy (EDS) elemental maps were acquired (Fig. 3b1–b5) to investigate the composition of WO₃/BiVO₄/TiO₂, demonstrating that five elements (W, O, Bi, V, and Ti) were uniformly distributed throughout the entire structure.

Fig. 3 (a) TEM image recorded in the HADDF mode of WO₃/BiVO₄/TiO₂; (b1–b5) corresponding elemental analysis by EDS mapping of W, O, Bi, V, and Ti levels; (c) HR-TEM images of WO₃/BiVO₄/TiO₂ for areas marked A in 3 (a); (d) corresponding SAED pattern; (e) HR-TEM images of WO₃/BiVO₄/TiO₂ for areas marked B in 3(a); and (f) corresponding SAED pattern.

X-ray photoelectron spectroscopy (XPS) was performed to verify chemical states and compositions. Fig. 4a depicts the survey spectra, confirming the presence of Ti, W, Bi, V, and O elements in the WO₃/BiVO₄/TiO₂ heterojunction. High-resolution XPS spectra of W 4f (Fig. 4b) show main peaks at binding energies of 35.9 and 38 eV, which can be attributed to the respective W 4f_7/2 and W 4f_5/2 states of the W⁶⁺ valence state.²⁶Fig. 4c shows two peaks at binding energies of 159.6 and 164.8 eV, demonstrating that the respective Bi 4f_7/2 and Bi 4f_5/2 orbits exhibit a 3+ valence state in BiVO₄.²⁷ The convoluted signals of V 2p_3/2 and V 2p_1/2 peaks (shown in Fig. 4d) at 517.1 and 524.6 eV, respectively, clearly reveal an oxidation state of V⁵⁺ in BiVO₄.²⁸ The spin orbit coupling of Ti 2p_3/2 and Ti 2p_1/2 peaks (see Fig. 4e) showed binding energies of 459.4 and 465 eV, respectively, demonstrating the presence of TiO₂ with a valence state of Ti⁴⁺.²⁹ The asymmetric peak of O 1s at ∼530.7 eV, shown in Fig. 4f, can be assigned to lattice oxygen (O²⁻) and surface hydroxyl species.³⁰

Fig. 4 XPS survey of (a) full spectra, (b) W 4f, (c) Bi 4f, (d) V 2p, (e) Ti 2p, and (f) O 1s region of WO₃/BiVO₄/TiO₂.

We employed ultraviolet–visible (UV-vis) absorption spectroscopy to understand the light absorption characteristics. As seen from Fig. 5a, the strong light absorption within a range of 325–460 nm was revealed for pristine WO₃ electrodes, whereas WO₃/BiVO₄ and WO₃/BiVO₄/TiO₂ exhibited longer wavelength light absorption in the range of 360–500 nm, demonstrating enhanced light absorption ability in the latter. IPCE spectra vs. wavelength was measured with various band-pass filters in a xenon (Xe) lamp, showing a red shift in the light absorption (Fig. 5b). The band gap values were ∼2.60 eV, 2.46 eV, and 2.44 eV, for WO₃, WO₃/BiVO₄ and WO₃/BiVO₄/TiO₂, respectively, as shown by the plot in Fig. 5c of (IPCE·hν)^1/2vs. hν; this is in close agreement with band gap values estimated by the Tauc plot (ESI Fig. S4†) acquired from previous optical absorption spectroscopy (Fig. 5a).

Fig. 5 Optoelectronic characterization of (a) UV-vis absorption spectra, (b) IPCE spectra, (c) (IPCE·hν)^1/2vs. hν plots, and (d) Mott–Schottky plots of WO₃, WO₃/BiVO₄, and WO₃/BiVO₄/TiO₂ photoelectrodes.

Considering the absorption in the visible region, we measured photoelectrochemical performances of WO₃/BiVO₄/TiO₂ electrodes for water splitting and compared them with those of referenced WO₃/BiVO₄ and WO₃ electrodes in 0.5 M Na₂SO₄ under standard simulated sunlight illumination (1 sun) at 1.23 V_RHE. Fig. 6a shows the linear sweep voltammetry (LSV) curves of the prepared photoanodes, which were recorded at room temperature under chopped light steps (10 s). The photocurrent density at 1.23 V_RHE followed the order WO₃/BiVO₄/TiO₂ (1.04 mA cm⁻²) > WO₃/BiVO₄ (0.43 mA cm⁻²) > WO₃ (0.23 mA cm⁻²), indicating a substantial improvement in the photoelectrocatalytic performance with addition of BiVO₄ and TiO₂ layers. This ∼2.5 times enhancement can be associated with interparticle electron transfer between integrated electrode materials and suppressed charge recombination due to the heterojunction effect.^31,32 In addition, WO₃/BiVO₄/TiO₂ showed an onset potential of 500 mV vs. RHE, which was lower than that of WO₃ (800 mV vs. RHE) and WO₃/BiVO₄ (550 mV vs. RHE). The shift of the onset potential toward the negative further facilitates the electron drive from the conduction band (CB) of TiO₂ to BiVO₄ and then BiVO₄ to WO₃.³³ This enhancement in anodic photocurrents was consistent with IPCE results (see Fig. 5b) of considerably higher values (45%) for the WO₃/BiVO₄/TiO₂ electrode, indicating that the photocurrent is dependent on light absorption and consistent with improved performance. To determine the flat band potentials and donor density, a Mott–Schottky analysis was performed in 0.5 M Na₂SO₄. As seen in Fig. 5d, the electrode has a positive slope, suggesting n-type semiconductor behavior.³⁴ The flat band potential values can be estimated using eqn (4); they were 0.35 V, 0.39 V, and 0.46 V vs. RHE for the WO₃/BiVO₄/TiO₂, WO₃/BiVO₄, and WO₃ electrodes, respectively.

Fig. 6 Photoelectrochemical performance of WO₃, WO₃/BiVO₄, and WO₃/BiVO₄/TiO₂ photoelectrodes: (a) Linear sweep voltammetry curves, (b) Nyquist plots and equivalent circuit, (c) plot of H₂ evolution as a function of time, and (d) chronoamperometric current density–voltage (J–T) curves.

As a control experiment, an amorphous TiO₂ layer was coated onto our WO₃/BiVO₄ electrode by adopting a previously reported procedure.^18,35 Results showed reduced photocurrent density, representing weaker interaction between host (WO₃/BiVO₄) and guest (amorphous-TiO₂) counterparts (Fig. S5†). This reduction in the photocurrent density of WO₃/BiVO₄/TiO₂ (amorphous) was presumably due to the low electric conductivity and defective amorphous TiO₂ reported previously.^36,37 In addition, a 5-cycle coated TiO₂ electrode exhibited low photocurrent density (Fig. S6†), presumably due to the presence of high thickness and defective polycrystalline TiO₂, as evident from previous XRD results.³⁷

To gain more insights into charge transfer kinetics at the interface of the photoanode/electrolyte, electrochemical impedance spectroscopy (EIS) was performed. Fig. 6b shows the EIS Nyquist plots of WO₃, WO₃/BiVO₄, and WO₃/BiVO₄/TiO₂, which were fitted to the equivalent circuit (Fig. 6b) and consisted of solution resistance (R_s), charge transfer resistance (R_ct), and constant phase element (Q_CPE).³⁸ The equivalent circuit-fitted values are summarized in Table 1. It is apparent that WO₃/BiVO₄/TiO₂ displays a smaller R_ct value (228.3 Ω), when compared to WO₃ (2063 Ω) and WO₃/BiVO₄ (244.1 Ω). This indicates that charge transfer resistance was reduced after a coating of TiO₂ nanoparticles was applied.

Table 1 Fitted values of each parameter in EIS results

Parameter	WO3	WB	WBT
R s (Ω)	42.47	29.05	14.96
R ct (Ω)	2063	244.1	228.3
Q CPE (F)	42.32 × 10^−6	0.18623 × 10^−3	69.16 × 10^−6

---

These features suggest that creating a heterojunction with BiVO₄ and crystalline TiO₂ is an effective way to optimize the separation efficiency of electron–hole pairs, which would promote overall photoelectrocatalytic performance.³⁵ Photoelectrochemical H₂ gas evolution was performed under standard sunlight irradiation (1 Sun) for a 3 h duration. The H₂ evolution rates were 78.8 μL cm⁻² and 52.9 μL cm⁻² for WO₃/BiVO₄/TiO₂ and WO₃/BiVO₄ photoelectrodes, respectively, at 1.23 V vs. RHE under simulated solar light (AM 1.5, 100 mW cm⁻²); the larger HER rate represents the positive effect of crystalline TiO₂ deposition (Fig. 6c). The durability test (J–T curve) of WO₃/BiVO₄/TiO₂ displayed almost unchanged initial photocurrent density after 24 h of operation (Fig. 6d), whereas unmodified WO₃/BiVO₄ exhibited an initial drastic loss in photocurrent that the latter managed to retain after 0.5 h of operation, demonstrating robustness of ternary WO₃/BiVO₄/TiO₂ as a photoelectrocatalyst for water splitting. The stable photocurrent density value of ∼0.76 mA cm⁻² (Fig. 6d) obtained in this study is comparable to that of state-of-the-art WO₃/BiVO₄ photoanodes and other BiVO₄-based photoanodes, such as WO₃/BiVO₄ (0.6 mA cm⁻² at 1.23 V_RHE, 0.1 M Kpi),³⁹ SnO₂/WO₃/BiVO₄ (1.5 mA cm⁻² at 1.23 V_RHE, Kpi with 0.1 M Na₂SO₃),⁴⁰ WO₃/BiVO₄/BiFeO₃ (1.1 mA cm⁻² at 1.5 V_RHE, 0.5 M Na₂SO₄),¹⁴ and SnO₂/BiVO₄/Co-Pi (0.8 mA cm⁻² at 1.23 V_RHE, 0.1 M Kpi).⁴¹ Please see Table S1† for more details.

Guided by observations and experimental results, a plausible electron transfer mechanism of a WO₃/BiVO₄/TiO₂ photoelectrode is illustrated in Fig. 7. The band gap alignment was assigned on the basis of estimated optical band gap values and flat band potentials. Pristine WO₃ is a relatively weak n-type material that can possess an optical absorption onset ∼2.7 eV, while WO₃/BiVO₄ has strong n-type behavior due to the heterojunction effect, producing large built-in potential at the interfaces and extended light absorption over wide band gap ranges (2.4–2.7 eV). When BiVO₄ is in strong contact with WO₃, a type II band offset can be produced that transfers electrons from the BiVO₄ CB (+0.02 V vs. RHE) to the WO₃ CB (+0.41 V vs. RHE) and facilitates electron–hole separation.⁴² Interestingly, since the CB edge potential is more negative in a WO₃/BiVO₄/TiO₂ heterostructure, TiO₂ would begin by inducing an electron transfer from the TiO₂ CB to the BiVO₄ CB and then inject into WO₃ electrode's CB. Thus, accumulated electrons would be injected into a current collector and migrate to the metallic cathode (Pt) electrode to participate in the hydrogen evolution reaction. We assume that introducing a thin crystalline TiO₂ overlayer would not allow hole accumulation driving forces (higher valence band potential). However, it could block surface defects that are centric to electron–hole recombination and lower the overpotential of photoelectrochemical reactions, leading to improved reaction kinetics. As a result, hole transfer going from the BiVO₄ valence band (VB) to the WO₃ VB would occur at the photoelectrode/electrolyte interface, leading to an oxygen evolution reaction with water molecules. In an account, the as-formed WO₃/BiVO₄/TiO₂ photoelectrode with porous surface morphology produced from nearly vertical nanoplate-like structures provides a more accessible surface area, a type II heterojunction, and a thin crystalline TiO₂ layer, which should synergistically boost the charge-carrier separation efficiency, photostability, and overall PEC performance without the aid of any co-catalyst or sacrificial agents.

Fig. 7 Electron transfer mechanism of a WO₃/BiVO₄/TiO₂ photoelectrode.

Conclusions

We demonstrated an integrated design of WO₃/BiVO₄/TiO₂ photoanode material by using a simple hydrothermal-spin coating strategy, which exhibited enhanced photoelectrocatalytic activity and photostability under simulated sunlight illumination. The WO₃/BiVO₄/TiO₂ heterostructure displayed an outstanding photocurrent density of 1.04 mA cm⁻² at 1.23 V vs. RHE and photostability over 24 h of operation. Postmodification with crystalline (1 0 1) TiO₂ resulted in a ≈2.5 times improvement in PEC performance, substantially better than referenced WO₃/BiVO₄. This improvement in PEC performance can be attributed to fast electron transfer at the interface due to a type II band alignment, a high light harvesting ability, and excellent corrosion robustness (characterized through EIS, HR-TEM, IPCE, and chronoamperometry analysis). A thin crystalline TiO₂ overlayer was beneficial to the blocking of surface defects, increasing the surface charge-carrier separation efficacy to hypothetically behave differently from previous amorphous-TiO₂ passivation overlayers during PEC water splitting operations. This work offers new avenues for employing crystalline TiO₂ overlayer modulations when developing commercially plausible visible-light driven photocatalyst materials.

Experimental

Methods and materials

All chemicals were of reagent grade and used without further purification. Sodium tungsten dihydrate (Na₂WO₄·2H₂O, ≥99%), oxalic acid (C₂H₂O₄, 98%), sodium sulfate (Na₂SO₄, ≥99%), ammonium metavanadate (NH₄VO₃, ≥99%), bismuth(III) nitrate pentahydrate (Bi(NO₃)₃·5H₂O, 98%), citric acid (C₆H₈O₇, ≥99.5%), polyvinyl alcohol ((C₂H₄O)_x), and titanium(IV) butoxide (Ti(OCH₂CH₂CH₂CH₃)₄, 97%) were bought from Sigma Aldrich. Hydrochloric acid (HCl, 35–37%), nitric acid (HNO₃, 60%), and isopropyl alcohol (C₃H₈O. 99.5%) were purchased from Samchun Chemicals.

Synthesis of nanoplate-like WO₃

WO₃ nanostructures were fabricated on FTO coated glass substrates using a hydrothermal method. Prior to deposition, the substrates (2 × 4 cm²) were cleaned with acetone, ethanol, and deionized (DI) water using an ultrasonicator before being dried with N₂ gas. They were then placed in a polyparaphenol (PPL)-lined reactor with their conducting sides facing downward. In typical synthesis, 0.16 g of Na₂WO₄·2H₂O was dissolved in 20 mL DI water before adding 4 mL of 3 M HCl. Next, 20 mL of 35 mM C₂H₂O₄ and 0.05 g of Na₂SO₄ (morphological agent) was added. The entire solution (30 mL) was then transferred into a PPL-lined autoclave and maintained at 140 °C for 2 h. After the reaction, the reactor was allowed to cool naturally to room temperature. The WO₃-coated FTO electrode was cleaned using DI water and air annealed at 500 °C for 2 h.

Synthesis of WO₃/BiVO₄ heterojunction photoelectrodes

A BiVO₄ layer was deposited onto premade WO₃/FTO (2 × 2 cm²) by spin coating the precursor solution at 3000 rpm for 30 s (20 cycles). In a typical synthesis, 1.17 g of NH₄VO₃, 10 mL of HNO₃, and 20 mL of DI water were mixed together and stirred for 30 min to dissolve the precursors. Then, 4.85 g of Bi(NO₃)₃·5H₂O and 1.92 g of C₆H₈O₇ were added to the mixture before stirring to allow dissolution. When the solution color turned blue, 6 mL of the solution was mixed with 2 mL of (C₂H₄O)_x solution (0.4 g in 40 mL DI water) and kept under sonication for 5 min. The as-obtained films were dried at 100 °C for 5 min and air annealed at 500 °C for 30 min.

Synthesis of TiO₂ thin layers onto the WO₃/BiVO₄ heterojunction

TiO₂ precursor solution was prepared by dissolving 0.1 M Ti(OCH₂CH₂CH₂CH₃)₄ in a mixed solvent of 20 mL C₃H₈O, 0.18 g DI water, and 0.331 g HNO₃. The resultant solution was ultrasonicated for 1 h to allow complete dissolution. TiO₂ was then deposited onto the as-obtained WO₃/BiVO₄ by spin coating the above solution mixture at 3000 rpm for 40 s and annealing it at 500 °C for 1 h.

Characterization

The surface morphologies were examined by FE-SEM (JSM-6701 F, JEOL). Crystalline structures were characterized using an X-ray diffractometer (X'Pert Pro MRD, PANalytical) with a Cu-Kα radiation source and a high-resolution transmission electron microscope (HR-TEM; Titan G2 ChemiSTEM, FEI Co.). XPS (NEXSA, ThermoFisher) was used to analyze the elemental composition and oxidation states of the elements present within a heterojunction. Optical absorption spectra were recorded with a UV-vis spectrophotometer (UV-1900i, Shimadzu).

PEC measurements

All PEC measurements were performed with a potentiostat (SP150, Bio-logic) under a solar simulator (LCS-100, Oriel Instruments) with a 100 W xenon arc lamp (100 mW cm⁻²). The electrolyte used was a 0.5 M Na₂SO₄ solution and the light exposed area of the photoanode was 1.5 cm². A standard three-electrode cell, consisting of silver/silver chloride (Ag/AgCl) and Pt mesh, was used as reference and counter electrodes. The as-prepared photoanodes were used as working electrodes. LSV was performed at a scan rate of 5 mV s⁻¹. IPCE (%) was measured at 1.23 V vs. RHE, using a light source and monochromator (MonoRa200, DONGWOO OPTRON) and the following equation:where h is the Planck constant, c is the velocity of light, J_ph is the photocurrent density, λ is the incident wavelength, and P is the power density of light which is measured at each wavelength.

The measured potentials were converted into the RHE using the following equations:

and

E_Ag/AgCl (3 M KCl) = 0.1976 V at 25 °C. (3)

The Mott–Schottky plots were obtained using

where C is the space charge layer capacitance, e is electron charge, ε is the dielectric constant, ε₀ is the vacuum permittivity, N_d is the electron donor density, V is the applied potential, V_fb is the flat band potential, k is the Boltzmann constant, C⁻² indicates applied potential, and T is the temperature. The x intercept and the linear fit determine the flat band potential.

EIS measurements were performed within the frequency range of 50 kHz–100 MHz at 1.21 V vs. RHE. To examine evolved H₂ gas during the PEC process, 200 μL of gas was taken from the sealed quartz tube system every half hour and injected into a gas chromatograph (Agilent 6850, Agilent Technologies).

Author contributions

E. Park: conceptualization, data curation, formal analysis, methodology, visualization, writing – original draft, and writing – review and editing; S. S. Patil: formal analysis, methodology, and writing – review and editing; H. Lee: formal analysis, methodology, and writing – review and editing; V. S. Kumbhar: conceptualization, supervision, writing – original draft, and writing – review and editing; K. Lee: conceptualization, supervision, writing – original draft, writing – review and editing, and funding acquisition.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was supported by the National Research Foundation of Korea [grant numbers NRF-2019R1l1A3A01041454], funded by the South Korean Government particularly the Ministry of Education.

References

1. J. H. Kim and J. S. Lee, BiVO₄-Based heterostructured photocatalysts for solar water splitting: A review, Energy Environ. Focus, 2014, 3(4), 339–353.
2. M. G. Lee, J. S. Park and H. W. Jang, Solution-processed metal oxide thin film nanostructures for water splitting photoelectrodes: A review, J. Korean Ceram. Soc., 2018, 55(3), 185–202.
3. B. S. Kalanoor, H. Seo and S. S. Kalanur, Recent developments in photoelectrochemical water-splitting using WO₃/BiVO₄ heterojunction photoanode: A review, Mater. Sci. Energy Technol., 2018, 1(1), 49–62.
4. T. Hisatomi and K. Domen, Introductory lecture: Sunlight-driven water splitting and carbon dioxide reduction by heterogeneous semiconductor systems as key processes in artificial photosynthesis, Faraday Discuss., 2017, 198(0), 11–35.
5. F. E. Osterloh, Inorganic nanostructures for photoelectrochemical and photocatalytic water splitting, Chem. Soc. Rev., 2013, 42(6), 2294–2320.
6. V. S. Kumbhar, H. Lee, J. Lee and K. Lee, Interfacial growth of the optimal BiVO₄ nanoparticles onto self-assembled WO₃ nanoplates for efficient photoelectrochemical water splitting, J. Colloid Interface Sci., 2019, 557, 478–487.
7. F. A. Honda, Electrochemical Photolysis of Water at a Semiconductor Electrode, Nature, 1972, 238, 37–38.
8. V. O. Smilyk, S. S. Fomanyuk, G. Y. Kolbasov, I. A. Rusetskyi and V. S. Vorobets, Electrodeposition, optical and photoelectrochemical properties of BiVO₄ and BiVO₄/WO₃ films, Res. Chem. Intermed., 2019, 45(8), 4149–4161.
9. C. Liu, Y. Yang, J. Li, S. Chen, W. Li and X. Tang, An in situ transformation approach for fabrication of BiVO₄/WO₃ heterojunction photoanode with high photoelectrochemical activity, Chem. Eng. J., 2017, 326, 603–611.
10. L. Xia, J. Bai, J. Li, Q. Zeng, X. Li and B. Zhou, A highly efficient BiVO₄/WO₃/W heterojunction photoanode for visible-light responsive dual photoelectrode photocatalytic fuel cell, Appl. Catal., B, 2016, 183, 224–230.
11. J. H. Baek, B. J. Kim, G. S. Han, S. W. Hwang, D. R. Kim, I. S. Cho and H. S. Jung, BiVO₄/WO₃/SnO₂ Double-heterojunction photoanode with enhanced charge separation and visible-transparency for bias-free solar water-splitting with a perovskite solar cell, ACS Appl. Mater. Interfaces, 2017, 9(2), 1479–1487.
12. P. Chatchai, Y. Murakami, S.-y. Kishioka, A. Y. Nosaka and Y. Nosaka, Efficient photocatalytic activity of water oxidation over WO₃/BiVO₄ composite under visible light irradiation, Electrochim. Acta, 2009, 54(3), 1147–1152.
13. Y. Park, K. J. McDonald and K.-S. Choi, Progress in bismuth vanadate photoanodes for use in solar water oxidation, Chem. Soc. Rev., 2013, 42(6), 2321–2337.
14. S. Khoomortezaei, H. Abdizadeh and M. R. Golobostanfard, Triple layer heterojunction WO₃/BiVO₄/BiFeO₃ porous photoanode for efficient photoelectrochemical water splitting, ACS Appl. Energy Mater., 2019, 2(9), 6428–6439.
15. Z. Zhang, B. Chen, M. Baek and K. Yong, Multichannel charge transport of a BiVO₄/(RGO/WO₃)/W18O49 three-storey anode for greatly enhanced photoelectrochemical efficiency, ACS Appl. Mater. Interfaces, 2018, 10(7), 6218–6227.
16. I. Grigioni, A. Corti, M. V. Dozzi and E. Selli, Photoactivity and Stability of WO₃/BiVO₄ Photoanodes: Effects of the contact electrolyte and of Ni/Fe oxyhydroxide protection, J. Phys. Chem. C, 2018, 122(25), 13969–13978.
17. X. Zhang, X. Wang, D. Wang and J. Ye, Conformal BiVO₄-Layer/WO₃-Nanoplate-array heterojunction photoanode modified with cobalt phosphate cocatalyst for significantly enhanced photoelectrochemical performances, ACS Appl. Mater. Interfaces, 2019, 11(6), 5623–5631.
18. S. S. Kalanur, I.-H. Yoo, J. Park and H. Seo, Insights into the electronic bands of WO₃/BiVO₄/TiO₂, revealing high solar water splitting efficiency, J. Mater. Chem. A, 2017, 5(4), 1455–1461.
19. T. Zhang, J. Su and L. Guo, Morphology engineering of WO₃/BiVO₄ heterojunctions for efficient photocatalytic water oxidation, CrystEngComm, 2016, 18(46), 8961–8970.
20. Y. Zhang, D. Zhang, X. Xu and B. Zhang, Morphology control and photocatalytic characterization of WO₃ nanofiber bundles, Chin. Chem. Lett., 2018, 29(9), 1350–1354.
21. C. Wang, X. Li, C. Feng, Y. Sun and G. Lu, Nanosheets assembled hierarchical flower-like WO₃ nanostructures: Synthesis, characterization, and their gas sensing properties, Sens. Actuators, B, 2015, 210, 75–81.
22. J. Yin, S. Huang, Z. Jian, Z. Wang and Y. Zhang, Fabrication of heterojunction SnO₂/BiVO₄ composites having enhanced visible light photocatalytic activity, Mater. Sci. Semicond. Process., 2015, 34, 198–204.
23. Z. Jiao, H. Yu, X. Wang and Y. Bi, Ultrathin BiVO₄ nanobelts: controllable synthesis and improved photocatalytic oxidation of water, RSC Adv., 2016, 6(77), 73136–73139.
24. F. Bella, A. B. Muñoz-García, G. Meligrana, A. Lamberti, M. Destro, M. Pavone and C. Gerbaldi, Unveiling the controversial mechanism of reversible Na storage in TiO₂ nanotube arrays: Amorphous versus anatase TiO₂, Nano Res., 2017, 10(8), 2891–2903.
25. M. Tahir, B. Tahir and N. A. S. Amin, Gold-nanoparticle-modified TiO₂ nanowires for plasmon-enhanced photocatalytic CO₂ reduction with H₂ under visible light irradiation, Appl. Surf. Sci., 2015, 356, 1289–1299.
26. D. a. e. Nagy, Effect of the morphology and phases of WO₃ nanocrystals on their photocatalytic efficiency, R. Soc. Chem., 2016, 6, 33743–33754.
27. Z. Ma, K. Song, L. Wang, F. Gao, B. Tang, H. Hou and W. Yang, WO₃/BiVO₄ Type-II Heterojunction arrays decorated with oxygen-deficient ZnO passivation layer: A highly efficient and stable photoanode, ACS Appl. Mater. Interfaces, 2019, 11(1), 889–897.
28. P. Guan, In situ anchoring Ag through organic polymer for configuring efficient plasmonic BiVO₄ photoanode, Chem. Eng. J., 2019, 358, 658–665.
29. L. Shi, C. Xu, X. Sun, H. Zhang, Z. Liu, X. Qu and F. Du, Facile fabrication of hierarchical BiVO₄/TiO₂ heterostructures for enhanced photocatalytic activities under visible-light irradiation, J. Mater. Sci., 2018, 53(16), 11329–11342.
30. J. Z. Yin, S. B. Huang, Z. C. Jian, M. L. Pan, Y. Q. Zhang, Z. B. Fei and X. R. Xu, Enhancement of the visible light photocatalytic activity of heterojunction In₂O₃/BiVO₄ composites, Appl. Phys. A, 2015, 120(4), 1529–1535.
31. S. S. Gujral, A. N. Simonov, M. Higashi, R. Abe and L. Spiccia, Optimization of titania post-necking treatment of TaON photoanodes to enhance water-oxidation activity under visible-light irradiation, ChemElectroChem, 2015, 2(9), 1270–1278.
32. N. Hirayama, H. Nakata, H. Wakayama, S. Nishioka, T. Kanazawa, R. Kamata, Y. Ebato, K. Kato, H. Kumagai, A. Yamakata, K. Oka and K. Maeda, Solar-driven photoelectrochemical water oxidation over an n-type lead–titanium oxyfluoride anode, J. Am. Chem. Soc., 2019, 141(43), 17158–17165.
33. A. P. Singh, N. Kodan, B. R. Mehta, A. Held, L. Mayrhofer and M. Moseler, Band edge engineering in BiVO₄/TiO₂ heterostructure: Enhanced photoelectrochemical performance through improved charge transfer, ACS Catal., 2016, 6(8), 5311–5318.
34. C. Liu, J. Su and L. Guo, Comparison of sandwich and fingers-crossing type WO₃/BiVO₄ multilayer heterojunctions for photoelectrochemical water oxidation, RSC Adv., 2016, 6(33), 27557–27565.
35. D. Eisenberg, H. S. Ahn and A. J. Bard, Enhanced photoelectrochemical water oxidation on bismuth vanadate by electrodeposition of amorphous titanium dioxide, J. Am. Chem. Soc., 2014, 136(40), 14011–14014.
36. N. Wu, J. Wang, D. N. Tafen, H. Wang, J.-G. Zheng, J. P. Lewis, X. Liu, S. S. Leonard and A. Manivannan, Shape-enhanced photocatalytic activity of single-crystalline anatase tio₂ (101) nanobelts, J. Am. Chem. Soc., 2010, 132(19), 6679–6685.
37. U. Shaislamov and B. L. Yang, CdS-sensitized single-crystalline TiO₂ nanorods and polycrystalline nanotubes for solar hydrogen generation, J. Mater. Res., 2013, 28(3), 418–423.
38. D. Hongxing, L. Qiuping and H. Yuehui, Preparation of nanoporous BiVO₄/TiO₂/Ti film through electrodeposition for photoelectrochemical water splitting, R. Soc. Open Sci., 2018, 5(9), 180728.
39. D. Coelho, J. Gaudêncio, S. A. Carminati, F. W. P. Ribeiro, A. F. Nogueira and L. H. Mascaro, Bi electrodeposition on WO₃ photoanode to improve the photoactivity of the WO₃/BiVO₄ heterostructure to water splitting, Chem. Eng. J., 2020, 399, 125836.
40. S. S. M. Bhat, S. A. Lee, J. M. Suh, S.-P. Hong and H. W. Jang, Triple planar heterojunction of SnO₂/WO₃/BiVO₄ with enhanced photoelectrochemical performance under front illumination, Appl. Sci., 2018, 8(10), 1765.
41. J. Liu, J. Li, M. Shao and M. Wei, Directed synthesis of SnO₂@BiVO₄/Co-Pi photoanode for highly efficient photoelectrochemical water splitting and urea oxidation, J. Mater. Chem. A, 2019, 7(11), 6327–6336.
42. I. Grigioni, K. G. Stamplecoskie, E. Selli and P. V. Kamat, Dynamics of photogenerated charge carriers in WO₃/BiVO₄ heterojunction photoanodes, J. Phys. Chem. C, 2015, 119(36), 20792–20800.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1nr04763a

---