Photoelectrochemical H₂ evolution using TiO₂-coated CaFe₂O₄ without an external applied bias under visible light irradiation at 470 nm based on device modeling

Abstract

CaFe₂O₄ (CFO) can be used as a photocathode to evolve H₂ from water in a photoelectrochemical cell. However, CFO degrades during operation and an external voltage is necessary for PEC H₂ evolution because the onset potential is less than the potential required for water oxidation considering the overpotential at the counter electrode. In order to develop a reliable CFO electrode with a greater onset potential, improvement of chemical stability and suppression of surface recombination is necessary. In this study, a chemically stable electrode structure with a greater onset potential was achieved by coating the [00l]-oriented CFO with a thin layer of titanium dioxide (TiO₂). A CFO|TiO₂ electrode was designed using a device simulator. The simulation results predict that coating CFO with TiO₂ produces a positive or negative shift in the onset potential under visible and ultraviolet light irradiation, respectively. The experimental onset potentials matched the simulation prediction. The observed onset potential for TiO₂-coated CFO was around (1.6 V vs. RHE) under visible light (470 nm) and 0.9 V under ultraviolet light (300 nm), compared to 1.2–1.3 V vs. RHE for a bare CFO electrode. The onset potential (1.6 V) under visible light irradiation is the most positive onset potential among the oxide photocathodes ever reported for PEC water splitting. Using the TiO₂-coated CFO as the photocathode and RuO₂-loaded Pt as the anode, stable photocurrent was observed under 470 nm excitation without an external voltage and evolution of H₂ from the system was confirmed.

---

Introduction

Hydrogen (H₂) is an ideal energy carrier for CO₂-free combustion. However, most H₂ is produced from fossil fuels by steam reforming, which results in substantial CO₂ emission. Photoelectrochemical (PEC) water-splitting is a promising solar-driven, carbon-free H₂ production method.^1–3 The first report of PEC H₂ production utilized a wide gap n-type TiO₂–Pt system limited to UV absorption.⁴ Subsequent studies have replaced TiO₂ with materials that can absorb longer wavelengths of light such as TaON,^5,6 BiVO₄,^7–10 Ta₃N₅,^6,11,12 SiC,^13,14 CuO_x,¹⁵ Rh-doped SrTiO₃,¹⁶ Fe₂O₃,^17–21 CaFe₂O₄,^22–30 LaFeO₃³¹ and ZnFe₂O₄.³² However, PEC systems have more practical problems to be addressed such as efficiency, stability, and cost.

CaFe₂O₄ (CFO) is an attractive material composed of Earth-abundant elements with low handling and purification costs.³³ CFO is a p-type semiconductor (bandgap: 1.9 eV) that can function as a photocathode in a PEC cell.³⁴ The onset potential for CFO photocathodic current is reported to be ∼1.2 V vs. a reference hydrogen electrode (RHE), which is higher than other p-type semiconductors such as CuO_x, SiC, and chalcogenide-based materials.^35–38 However, an external applied voltage is necessary to decompose water using a two-electrode system such as CFO and Pt. To produce H₂ and O₂ without an external applied bias, the onset potential for the CFO electrode must be increased to ∼1.5–1.6 V vs. RHE considering the overpotential required for water oxidation at the counter electrode.

Coating a photoelectrode with a metal oxide such as TiO₂^39,40 or ZnO¹⁵ is often used to shift the onset potential for p-type semiconducting materials in addition to effectively protecting the surface. However, depending on the deposition conditions of the metal oxide,⁴¹ the onset potential may be reduced rather than increased. The magnitude and direction of the induced shift depends on the generation, recombination, and charge transport across the metal oxide, which are sensitive to electrical properties such as mobility and charge carrier density.

In this study, a PEC system consisting of a TiO₂-coated CFO photocathode (CFO|TiO₂) and Pt counter electrode was designed and demonstrated to evolve H₂ without an external applied bias. The design was developed using wxAMPS – a free device physics simulation package previously used to model the performance of dye-sensitized solar cells and functionalized Si(111) photoelectrodes, among other applications.^42–47 The simulation results predicted that excitation with wavelengths exclusively >420 nm (i.e. avoiding excitation of the TiO₂) would increase the onset potential of the CFO|TiO₂ photocathode past ∼1.5–1.6 V vs. RHE and produce H₂ without an external applied bias. The prediction was validated experimentally for a CFO|TiO₂–Pt system under 470 nm excitation.

Experimental

Materials synthesis

Materials. Ca(CH₃COO)₂·H₂O (99.0% Wako Pure Chemical Industries, Ltd.), Fe(NO₃)₃·9H₂O (99.0% Wako Pure Chemical Industries Ltd.), polyethylene glycol (99.5%, Wako Pure Chemical Industries Ltd.) and TiO₂ (99.0% Wako Pure Chemical Industries Ltd.) were used to prepare the samples.

Preparation of CaFe₂O₄ and TiO₂-coated CaFe₂O₄. CaFe₂O₄ (CFO) and TiO₂-coated CFO electrodes (CFO|TiO₂) were prepared as follows. CFO powder was prepared by mixing Ca(CH₃COO)₂·H₂O and Fe(NO₃)₃·9H₂O in deionized water, followed by the addition of an aqueous solution of 5 wt% polyethylene glycol. The solution was heated and held at 150 °C while stirring until the powder was fully dried. This was then calcined in a mantle heater at 1100 °C for 3 h. CFO electrodes were prepared using a suspension of 100 mg of the CFO powder in ethanol, which was then applied to a Pt substrate (1 × 2 cm²) and dried at 100 °C. The CFO-coated substrates were then calcined at 1230 °C for 3 h. For CFO|TiO₂ electrodes, a thin layer of TiO₂ (20 nm) was deposited on the surface by pulsed laser deposition (PLD) from a dense TiO₂ pellet using a PASCAL PLD-7 under 0.67 Pa O₂. The excimer laser power and frequency were 180 mJ per pulse at 10 Hz, respectively. An insulated wire was soldered to the back of the CFO electrodes and the entire back side and edges were covered with epoxy resin for PEC measurements.

Electrochemical measurement. PEC and Mott–Schottky experiments were conducted in a conventional three-electrode quartz electrochemical cell with a Pt plate counter electrode and a reference electrode. PEC experiments were performed in a 0.1 M NaOH aqueous solution under irradiation from a solar simulator (AM1.5), a 300 nm (15 mW cm⁻²), a 470 nm (LED lamp, 90 mW cm⁻²), or a 532 nm (green laser, 1 W cm⁻²). The PEC H₂ evolution experiments were conducted in a 0.1 M NaOH aqueous solution (Ar gas atmosphere) under 90 mW cm⁻² of 470 nm illumination without an external applied voltage. The amount of H₂ evolved was measured by an online gas chromatograph system (Shimadzu, GC-8A).

Characterization and equipment. The crystal structure of the TiO₂ films was analyzed using in-plane and/or out-of-plane X-ray diffraction (XRD; Rigaku, SmartLab) with Cu Kα radiation. The surface morphology of the electrode was analyzed using scanning electron microscopy (SEM; JEOL, SU8000). X-ray photoelectron spectroscopy (XPS) spectra were obtained using a Shimadzu AXIS 165.

Steady-state simulation model. A drift-diffusion solid state device simulator, wxAMPS, was utilized to model CFO and CFO|TiO₂ photoelectrodes.^42–47 wxAMPS is a variant of AMPS, a solar cell simulation program developed by Fonash et al. at the Pennsylvania State University.⁴⁵ wxAMPS uses the method of finite differences to calculate the concentration of charge carriers throughout the device as a function of position and applied potential by solving Poisson's equation, the continuity equation for free electrons, and the continuity equation for free holes simultaneously in 1-dimension. A detailed description of the model and its application to PEC cells is included in previous work.⁴⁶ The onset potential was determined from wxAMPS by simulating the photocurrent versus voltage curve for a specified potential range and illumination conditions and determining the zero current condition.

Results and discussion

Fig. 1a shows the current–potential curves of a bare CFO electrode in 0.1 M NaOH solution under chopped AM1.5 solar illumination. The onset potential for photocathodic current was around 1.30 V vs. RHE. Although the onset potential is slightly larger than the water oxidation potential (1.23 V) it is still not large enough to realize PEC H₂ evolution without an external applied voltage in a two-electrode system such as CFO–Pt. The necessary onset potential would be ∼1.5–1.6 V vs. RHE due to the overpotential of the water oxidation at the Pt electrode. However, the valence band position determined from the photoelectron spectroscopy in air spectrum was 1.58 V vs. NHE as shown in Fig. 1b. This indicates that increasing the onset potential to >1.5 V is indeed possible in theory, implying that PEC H₂ evolution without an external applied voltage can be realized.

Fig. 1 (a) Current–potential curve of a bare CFO electrode in 0.1 M NaOH solution under chopped illumination (solar simulator, AM1.5) and (b) photoelectron spectroscopy in air spectrum and band structure model of CaFe₂O₄.

To develop a method of achieving a larger (more positive) onset potential, a drift-diffusion solid state device simulator, wx-AMPS was used to predict the performance. In this simulation, information of band structure (bandgap: 1.9 eV, CBM: −0.32 V vs. NHE, VBM: 1.58 V vs. NHE), carrier concentration, mobility, defects, and surface recombination rate is necessary. The input values for the carrier concentration (N_A = 10¹⁶ cm³) and hole mobility (μ_p = 0.05 cm² V⁻¹ s⁻¹) were estimated from the Mott–Schottky plot as shown in Fig. S1.† To model trap states in the TiO₂, donor defects were added at midgap with a concentration of 10¹⁶ cm⁻³ and capture cross sections of 5 × 10¹² cm² for both electrons and holes. In general, it was difficult to measure the actual surface recombination rate. Therefore, in this study, the surface recombination rate of CFO was estimated by comparing the actual onset potential for photocathodic current of CFO with the simulated open circuit voltage as a function of surface recombination rate. Fig. 2a shows the simulated relationship between open circuit voltage and surface recombination rate obtained using wxAMPS. The open circuit voltage decreased with increasing surface recombination rate, which indicates that less surface recombination will improve the open circuit voltage. For a surface recombination rate between 10⁴ and 10⁸ cm², the open circuit voltage remained constant at about 1.3 V vs. NHE, which corresponded to the actual onset potential (1.30 V vs. NHE) for photocathodic current of CFO in Fig. 1a. In the following simulations, the surface recombination rate was set as 10⁷ cm s⁻¹, which is approximately equal to the thermal velocity of an electron. Fig. 2b shows current–potential curves with or without light illumination of CaFe₂O₄ simulated using the above-mentioned input parameters with the photocurrent normalized at 0.0 V as a reference. Additional input parameters are summarized in Table S1.†

Fig. 2 (a) Simulated open circuit voltage (onset potential) as a function of surface recombination rate of CaFe₂O₄ and (b) simulated current–potential curve of CaFe₂O₄ (surface recombination rate: 10⁷ cm s⁻¹).

In this study, a TiO₂ passivation layer was used to improve the onset potential for photocathodic current. First, current–potential curves of TiO₂-coated CFO at different wavelengths of excitation were simulated using wxAMPS. The simulated photocurrent versus applied potential under 300, 470, and 532 nm excitations are shown in Fig. 3. The calculated band diagrams of CFO and TiO₂-coated CFO photoelectrodes are shown in ESI Fig. S2 and S3.† The photocurrents were normalized using the photocurrent at 0.0 V as a reference. The open circuit voltage (V_oc) for bare CFO was ∼1.35 V vs. NHE for all three wavelengths of excitation, which agrees well with the experimentally observed onset potential of 1.30 V. For the TiO₂-coated CFO photoelectrode, the V_oc was strongly influenced by the wavelength of excitation. The V_oc for 300 nm, 470 nm, and 532 nm excitation was 1.06 V, 1.92, and 1.87 V vs. NHE, respectively. Therefore, the TiO₂ induces a decrease in V_oc for 300 nm excitation (UV light) and an increase in the V_oc for 470/532 nm excitation (visible light). The simulation results indicate that the V_oc of TiO₂-coated CFO under longer wavelength irradiation is sufficient to realize PEC H₂ evolution without an external applied voltage.

Fig. 3 Simulated photocurrent versus applied voltage for CFO and TiO₂-coated CFO electrodes under (a) 300, (b) 470, and (c) 500 nm excitation.

We propose the following explanation for the change in sign of the photo-induced shift in V_oc by the TiO₂ coating under UV and visible light irradiation. Under UV illumination, the photon energy is sufficient to generate carriers in the TiO₂. These accumulate in surface traps resulting in an electric field in the TiO₂ that reduces the V_oc. Under visible light illumination, the photon energy is insufficient to excite carriers in the TiO₂ and all the carrier generation occurs in the CFO. The field in the CFO efficiently separates the electron–hole pairs, and the conduction and valence band offsets between the CFO and TiO₂ create an electron-selective contact, resulting in a large V_oc. The proposed mechanism for the TiO₂-coated CFO photoelectrode under visible light excitation is shown in Fig. 4. Additionally, the simulations shown in Fig. 2a reveal that a lower V_oc may be caused by a high surface recombination rate. In solar cell technology, it is well known that introducing a passivation layer is effective at reducing the surface recombination rate. Therefore, the presence of TiO₂ may contribute to a higher V_oc by lowering the surface recombination rate of the electrode.

Fig. 4 Proposed charge transfer mechanism for TiO₂-coated CaFe₂O₄ photoelectrode under visible light illumination.

To validate the simulation prediction, the PEC performance of a TiO₂ (20 nm)-coated CFO electrode was experimentally observed. Fig. 5 shows a structural model and typical SEM images of the CFO|TiO₂ electrode. A dense 50 μm film of CFO was prepared on a Pt substrate and 20 nm of TiO₂ was deposited on top. The resistivity of TiO₂ films deposited by PLD at room temperature was 0.13 Ωm, which is much lower than that of bulk TiO₂ (>10⁹ Ωm). These results indicate that a TiO₂ layer with a high concentration of donor-like defects was prepared. Fig. 6a shows out-of-plane XRD pattern of TiO₂ (20 nm)-coated CFO electrode. One strong diffraction peak assigned to (004) and/or (302) faces of CFO was observed (ICSD-16695, CaFe₂O₄). Fig. 6b shows pole figure for diffraction peak at 33.4°. Many diffraction spots were observed at around 60 degrees. Since the (004) face is inclined at 60° with respect to the (302) face, the diffraction spots at around 60° were assigned to (004) and/or (302) faces. A spot-like pattern in the pole figure means that the CFO film is composed of crystal grains with a relatively large lateral size. To confirm whether the CFO film is oriented to the [004] or [302] direction, wide-angle reciprocal space mapping of the CFO film was measured. Fig. 6c shows the reciprocal space mapping of the CFO film. Characteristic diffraction spots assigned to (112), (113), (114), (115) and (116) faces were observed in the mapping, which is a characteristic reciprocal space mapping assigned to (004)-oriented CFO. This indicates that the diffraction pattern at around 33.4° in the XRD pattern (Fig. 6a) is assigned to the (004) face. Thus, it was found that the CFO film is mainly oriented to [001] direction as shown in the structural image of the CFO film in the inset of Fig. 6a.

Fig. 5 Structural model and SEM image of TiO₂-coated CFO electrode.

Fig. 6 (a) Out-of-plane XRD pattern with an inset of the structural image of CFO, (b) pole figure for 33.4°, (c) wide-angle reciprocal mapping.

With regard to the crystal structure of the TiO₂ film prepared by PLD, the diffraction pattern assigned to TiO₂ was not observed in the out-of-plane XRD. Therefore, the crystal structure of the prepared TiO₂ was analyzed by in-plane XRD. Fig. 7 shows the in-plane XRD pattern of the TiO₂-coated CFO electrode. Although the diffraction patterns assigned to (210), (410) and (610) faces for CFO were observed, there is no clear XRD pattern for the TiO₂, indicating that it has an amorphous structure.

Fig. 7 In-plane XRD pattern of TiO₂ (20 nm)-coated CaFe₂O₄.

The surface coverage of the TiO₂ film was analyzed using XPS. Fig. 8 shows survey and narrow Ti 2p XPS spectra of CFO and TiO₂-coated CFO. The bare CFO electrode showed signals assigned to Fe, Ca, and O (Fig. 8a). The TiO₂-coated CFO electrode showed signals assigned only to Ti and O, a very weak Ca signal, and no signal for Fe (Fig. 8b). This indicates that most of the surface of the CFO was coated with TiO₂. The binding energy of the Ti 2p_3/2 peak was 458.4 eV, which is attributed to Ti⁴⁺ species in the TiO₂ matrix.

Fig. 8 Survey and narrow Ti 2p XPS spectra of (a) CFO and (b) TiO₂-coated CFO.

Fig. 9 shows experimental current–potential curves of CFO and TiO₂-coated CFO electrodes in 0.1 M NaOH solution under chopped illumination (300, 470, 532 nm). Under 300 nm light, the onset potentials of bare CFO and TiO₂-coated CFO were 1.27 and 0.92 V vs. RHE, respectively, as shown in Fig. 9a and b. Thus, under 300 nm excitation, TiO₂ induces a negative shift in the onset potential which agrees very well with the simulated current–potential curve as shown in Fig. 3a. When the electrode is illuminated with 300 nm light, which corresponds to the bandgap excitation of TiO₂, photo-excited carriers are generated in the TiO₂ layer. Thus, an electric field is formed in the TiO₂ by the photo-excited carriers that accumulate in trap states, which decreases the onset potential.

Fig. 9 Current–potential curves of CFO and TiO₂-coated CFO electrodes in 0.1 M NaOH solution under chopped light illumination (300, 470, 532 nm); black line: CFO, red line: TiO₂-coated CFO.

With 470 nm excitation the onset potentials of bare CFO and CFO|TiO₂ were 1.21 and 1.60 V vs. RHE, respectively, as shown in Fig. 9c and d. Although the incident photon to current conversion efficiency (IPCE) at 1.23 V for TiO₂-coated CFO was 0.002%, the IPCEs from 1.2 to 1.6 V was apparently improved compared with bare CFO (Fig. S4†). The onset potentials (vs. RHE) in 0.1 M NaOH was the almost same as those in 0.1 M Na₂SO₄ or 0.1 H₂SO₄ aqueous solution as shown in Fig. S5.† For 532 nm excitation the onset potentials of bare CFO and CFO|TiO₂ were 1.18 V and 1.57 V vs. RHE (Fig. 9e and f). However, the actual onset potentials are at minimum 1.60 and 1.57 V vs. RHE for 470 and 532 nm, respectively. For example, in the case of 470 nm excitation, photocathodic current is observed at 1.70 V vs. RHE as shown in Fig. S6,† but background current due to water electrolysis interferes with the exact determination of the onset potential. Thus, under visible irradiation, TiO₂ induces an increase in the onset potential, as predicted by the simulation results as shown in Fig. 3b and c. As mentioned above, in this system, the TiO₂ layer acts as a passivation layer for the CFO surface and as an electron-selective contact between CFO and the electrolyte, which results in a decrease in the surface recombination rate of CFO and an increase in the onset potential.

In the case of powdered photocatalysis, TiO₂ usually requires a co-catalyst for photocatalytic H₂ evolution. However, in the present case, adding a co-catalyst is not necessary to produce H₂ on TiO₂. When TiO₂-coated CFO is illuminated by visible light, carriers are only generated in the CFO and not in the TiO₂. The degree of band bending in the CaFe₂O₄ combined with proper band-edge alignment between TiO₂ and CaFe₂O₄ produces a driving force sufficient enough to evolve H₂ without requiring a catalyst.

To the best of our knowledge, the TiO₂-coated CFO has the most positive onset potential among the oxide photocathodes ever reported for PEC water splitting. Another important result is that the onset potential is strongly affected on the excitation wavelength. In the case of TiO₂-coated CFO, the experimentally observed potential difference between 300 nm illumination and 470 nm illumination was 680 mV. This result indicates that the excitation wavelength for PEC H₂ evolution should be considered according to the optical property of the passivation layer.

Next, PEC H₂ evolution without an external applied voltage was attempted for a TiO₂-coated CFO and RuO₂–Pt two-electrode system. RuO₂-loaded Pt electrode was used as the anode electrode for O₂ evolution because the overpotential on RuO₂-loaded Pt was slightly lower than that of pure Pt. Fig. 10a shows the I–V curve of a photocell with TiO₂-coated CaFe₂O₄ and RuO₂–Pt electrodes in 0.1 M NaOH aqueous solution under 470 nm light illumination. The short-circuit current (V = 0) was 18 μA cm⁻².

Fig. 10 (a) Current–potential curve of a photocell with TiO₂-coated CaFe₂O₄ and RuO₂–Pt electrode electrodes in 0.1 M NaOH aqueous solution under 470 nm-light illumination and (b) CFO–Pt electrodes system under visible light irradiation (470 nm) in 0.1 M NaOH aqueous solution.

The photocurrent was still observed at 1.6 V, although it is difficult to see due to large background currents for water decomposition. The onset potential of TiO₂-coated CaFe₂O₄ under 470 nm light irradiation in Fig. 9 was around 1.6 V, which matches the voltage observed photocurrent in the TiO₂-coated CFO and RuO₂–Pt electrode system shown in Fig. 10a. Fig. 10b shows the current–time curve under 470 nm light irradiation without an external applied bias. Although a rapid decrease in photocurrent was observed initially, the system showed a stable photocurrent (around 500 nA cm⁻²) for over 15 h. On the other hand, in the case of bare CFO, the photocurrent decreased immediately and became negligible (Fig. 10b).

Fig. 11 shows the time evolution of H₂ production from a TiO₂-coated CFO and RuO₂–Pt electrode system without an external applied bias under 470 nm irradiation. H₂ gas evolution was confirmed by gas chromatography, indicating that PEC H₂ evolution reaction occurred at the TiO₂-coated CFO electrode. The amount of evolved H₂ corresponds to the theoretical amount estimated from Faraday's law. Evolution of H₂ was observed for over 70 h indicating that the TiO₂-coated CFO electrode is stable during the PEC reaction. With regard to oxygen, reliable results for oxygen evolution (nanomole amount of oxygen) were not obtained due to contamination of very small amount of air into the reaction chamber. In the case of bare CFO electrode, H₂ evolution was not observed without an external bias. Thus, a positive shift in the onset potential of CFO from 1.3 V to 1.6 V by coating the electrode with TiO₂ results in H₂ evolution without an external applied voltage under visible light illumination.

Fig. 11 Time evolution of H₂ production from photoelectrochemical cell using TiO₂ coated CFO electrode without an external applied bias in 0.1 M NaOH aqueous solution under visible light irradiation (470 nm); anode: RuO₂-loaded Pt electrode.

To analyze the chemical stability of the TiO₂-coated CFO electrode, PEC H₂ production was conducted under irradiation from a solar simulator (400 mW cm⁻²) applying an external bias (1.2 V) to accelerate the PEC reaction on the electrode surface. For the bare CFO (cathode)–Pt (anode) system, the photocurrent decreased with increasing reaction time as shown in Fig. 12a due to the decomposition of the CFO surface. Before the reaction, CFO has a flat surface; yet after the PEC reaction, many cracks were observed on the surface of the CFO as shown in Fig. 13a. Also, the color of the electrode surface changed from black to brown due to the loss of Ca²⁺ ions into the aqueous solution. For the TiO₂ coated-CFO (cathode)–Pt (anode) system, although the initial photocurrents were smaller than that for the bare CFO, the decrease in the photocurrent was suppressed (Fig. 12a). The photocurrent increased during the first 4 hours of PEC H₂ production but then remained constant throughout the reaction. The initial increase in photocurrent may be due to a change in the surface condition of the TiO₂ or the interfacial structure between the amorphous-like TiO₂ layer and CFO during the PEC reaction. With regard to chemical stability, no cracks were evident on the electrode surface of TiO₂-coated CFO after 14 h of H₂ production as shown in Fig. 13b.

Fig. 12 (a) Current–time evolution for the CaFe₂O₄ (cathode)–Pt (anode) electrodes in a 0.1 M NaOH aqueous solution under irradiation from a solar simulator (400 mW cm⁻²) applying an external bias (1.2 V) and (b) time evolution of H₂ and O₂ production from TiO₂-coated CFO and Pt electrode system.

Fig. 13 FE-SEM images electrode surface after PEC reaction for 14 h of (a) CaFe₂O₄ and (b) TiO₂-coated CaFe₂O₄ electrodes.

The time evolution of H₂ and O₂ production for the TiO₂-coated-CFO (cathode)–Pt (anode) system is shown in Fig. 12b. Faraday efficiencies for the evolved H₂ and O₂ were 92% and 70%, respectively. Backward reaction may be a main reason why the amounts of H₂ and O₂ evolved did not correspond to the quantity of electricity that flowed. The total amount of evolved O₂ was not sufficiently much and then the solubility in water of oxygen is larger than that of hydrogen, which might be related to the low faradaic efficiency of oxygen compared with hydrogen. Thus, it was confirmed that coating with TiO₂ is effective to improve the chemical stability of CFO and onset potential under visible light irradiation.

Conclusions

A TiO₂-coated CFO photocathode was designed to achieve H₂ evolution without an external voltage using a device simulator. The simulation results predicted that coating CFO with TiO₂ gives an increase or decrease in the onset potential under visible and ultraviolet irradiation, respectively. Under 300 nm light the observed onset potential shifted from 1.27 to 0.92 V vs. RHE while under visible irradiation (470 & 532 nm) the onset potential shifted from 1.2 to 1.6 V vs. RHE. Thus, the simulation results predicted the actual performance of the TiO₂-coated CFO electrode. Using TiO₂-coated CFO as the photocathode and RuO₂-loaded Pt as the anode, stable photocurrent was observed under 470 nm excitation without an external voltage. Evolution of H₂ from the system was confirmed. Chemical stability of CFO was also improved by coating the surface with TiO₂. To our knowledge, TiO₂-coated CFO is the first example of a photocathode that exhibits a greater onset potential than that required for photocathodic water reduction.

Acknowledgements

This work was supported by the JST PRESTO program, JSPS KAKENHI Grant Numbers 15H00879. We thank Prof. T. Nakanotani for discussion about STEM data.

Notes and references

1. A. J. Bard and M. A. Fox, Acc. Chem. Res., 1995, 28, 141–145.
2. R. Abe, J. Photochem. Photobiol., C, 2010, 11, 179–209.
3. F. E. Osterloh, Chem. Soc. Rev., 2013, 42, 2294–2320.
4. A. Fujishima and K. Honda, Nature, 1972, 238, 37–38.
5. R. Abe, M. Higashi and K. Domen, J. Am. Chem. Soc., 2010, 132, 11828–11829.
6. M. Higashi, K. Domen and R. Abe, Energy Environ. Sci., 2011, 4, 4138–4147.
7. A. Kudo, K. Omori and H. Kato, J. Am. Chem. Soc., 1999, 121, 11459–11467.
8. K. Sayama, A. Nomura, T. Arai, T. Sugita, R. Abe, M. Yanagida, T. Oi, Y. Iwasaki, Y. Abe and H. Sugihara, J. Phys. Chem. B, 2006, 110, 11352–11360.
9. Y. H. Ng, A. Iwase, A. Kudo and R. Amal, J. Phys. Chem. Lett., 2010, 1, 2607–2612.
10. W. Luo, Z. Yang, Z. Li, J. Zhang, J. Liu, Z. Zhao, Z. Wang, S. Yan, T. Yu and Z. Zou, Energy Environ. Sci., 2011, 4, 4046–4051.
11. J. Hou, Z. Wang, C. Yang, H. Cheng, S. Jiao and H. Zhu, Energy Environ. Sci., 2013, 6, 3322–3330.
12. Y. Li, T. Takata, D. Cha, K. Takanabe, T. Minegishi, J. Kubota and K. Domen, Adv. Mater., 2013, 25, 125–131.
13. T. Inoue and T. Yamase, Chem. Lett., 1985, 7, 869.
14. M. Kato, T. Yasuda, K. Miyake, M. Ichimura and T. Hatayama, Int. J. Hydrogen Energy, 2014, 39, 4845–4849.
15. A. Paracchino, V. Laporte, K. Sivula, M. Grätzel and E. Thimsen, Nat. Mater., 2011, 10, 456–461.
16. K. Iwashina and A. Kudo, J. Am. Chem. Soc., 2011, 133, 13272–13275.
17. C. Leygraf, M. Hendewerk and G. A. Somorjai, J. Phys. Chem., 1982, 86, 4484–4485.
18. J. E. Turner, M. Hendewerk and G. A. Somorjai, Chem. Phys. Lett., 1984, 105, 581–585.
19. A. Duret and M. Grätzel, J. Phys. Chem. B, 2005, 109, 17184–17191.
20. D. K. Zhong, J. Sun, H. Inumaru and D. R. Gamelin, J. Am. Chem. Soc., 2009, 131, 6086–6087.
21. W. B. Ingler Jr, J. P. Baltrus and S. U. M. Khan, J. Am. Chem. Soc., 2004, 126, 10238–10239.
22. S. Ida, K. Yamada, T. Matsunaga, H. Hagiwara, Y. Matsumoto and T. Ishihara, J. Am. Chem. Soc., 2010, 132, 17343–17345.
23. Y. Matsumoto, M. Omae, K. Sugiyama and E. I. Sato, J. Phys. Chem., 1987, 91, 577–581.
24. Z. Liu, Z. G. Zhao and M. Miyauchi, J. Phys. Chem. C, 2009, 113, 17132–17137.
25. J. Cao, T. Kako, P. Li, S. Ouyang and J. Ye, Electrochem. Commun., 2011, 13, 275–278.
26. S. Ida, K. Yamada, T. Matsunaga, H. Hagiwara, T. Ishihara, T. Taniguchi, M. Koinuma and Y. Matsumoto, Electrochemistry, 2011, 79, 797–800.
27. S. Ida, K. Yamada, M. Matsuka, H. Hagiwara and T. Ishihara, Electrochim. Acta, 2012, 82, 397–401.
28. J. Cao, J. Xing, Y. Zhang, H. Tong, Y. Bi, T. Kako, M. Takeguchi and J. Ye, Langmuir, 2013, 29, 3116–3124.
29. E. S. Kim, N. Nishimura, G. Magesh, J. Y. Kim, J. W. Jang, H. Jun, J. Kubota, K. Domen and J. S. Lee, J. Am. Chem. Soc., 2013, 135, 5375–5383.
30. K. Sekizawa, T. Nonaka, T. Arai and T. Morikawa, ACS Appl. Mater. Interfaces, 2014, 6, 10969–10973.
31. Q. Yu, X. Meng, T. Wang, P. Li, L. Liu, K. Chang, G. Liu and J. Ye, Chem. Commun., 2015, 51, 3630–3633.
32. K. J. McDonald and K. S. Choi, Chem. Mater., 2011, 23, 4863–4869.
33. M. G. Ahmed, T. A. Kandiel, A. Y. Ahmed, I. Kretschmer, F. Rashwan and D. Bahnemann, J. Phys. Chem. C, 2015, 119, 5864–5871.
34. Y. Matsumoto, J. Solid State Chem., 1996, 126, 227–234.
35. K. Ramanathan, M. A. Contreras, C. L. Perkins, S. Asher, F. S. Hasoon, J. Keane, D. Young, M. Romero, W. Metzger, R. Noufi, J. Ward and A. Duda, Prog. Photovoltaics Res. Appl., 2003, 11, 225–230.
36. I. Repins, M. A. Contreras, B. Egaas, C. DeHart, J. Scharf, C. L. Perkins, B. To and R. Noufi, Prog. Photovoltaics Res. Appl., 2008, 16, 235–239.
37. P. Jackson, D. Hariskos, E. Lotter, S. Paetel, R. Wuerz, R. Menner, W. Wischmann and M. Powalla, Prog. Photovoltaics Res. Appl., 2011, 19, 894–897.
38. L. Zhang, T. Minegishi, J. Kubota and K. Domen, Phys. Chem. Chem. Phys., 2014, 16, 6167–6174.
39. B. Seger, T. Pedersen, A. B. Laursen, P. C. K. Vesborg, O. Hansen and I. Chorkendorff, J. Am. Chem. Soc., 2013, 135, 1057–1064.
40. B. Seger, S. D. Tilley, T. Pedersen, P. C. K. Vesborg, O. Hansen, M. Grätzel and I. Chorkendorff, J. Mater. Chem. A, 2013, 1, 15089–15094.
41. Y. Lin, R. Kapadia, J. Yang, M. Zheng, K. Chen, M. Hettick, X. Yin, C. Battaglia, I. D. Sharp, J. W. Ager and A. Javey, J. Phys. Chem. C, 2015, 119, 2308.
42. Y. Liu, Y. Sun and A. Rockett, Sol. Energy Mater. Sol. Cells, 2012, 98, 124.
43. Solar Cells Simulation, https://wiki.cites.illinois.edu/wiki/display/solarcellsim, accessed March 2, 2016.
44. M. Chitambar, Z. Wang, Y. Liu, A. Rockett and A. Maldonado, J. Am. Chem. Soc., 2012, 134(25), 10670.
45. S. J. Fonash, J. Arch, J. Cuiffi, J. Hou, W. Howland, P. J. McElheny, A. Moquin, M. Rogosky, T. Tran, H. Zhu and F. Rubinelli, A Manual for AMPS-1D for Windows 95/NT, The Pennsylvania State University, Happy Valley, PA, 1997.
46. K. L. Kearney and A. Rockett, J. Electrochem. Soc., 2016, 163(7), H598.
47. H. J. Kim, K. L. Kearney, L. H. Le, L. Z. J. Haber, A. A. Rockett and M. J. Rose, J. Phys. Chem. C, 2016, 120(45), 25697–25708.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7se00084g

---