Revealing long-lived electron–hole migration in core–shell α/γ-Fe₂O₃/FCP for efficient photoelectrochemical water oxidation

Abstract

The oxygen evolution reaction (OER) of Fe₂O₃ is limited by its low photocarrier separation efficiency in photoelectrochemical (PEC) water splitting. How to construct an effective photocarrier transmission route in Fe₂O₃ has become an important bottleneck for enhancing OER performance. Herein, we exploit a core–shell nanorod structure loaded with FeCo Prussian blue (FCP) to boost the water oxidation kinetics and charge transfer efficiency for the first time. As expected, the optimal γ-Fe₂O₃/α-Fe₂O₃ photoanode exhibits a remarkable photocurrent density of 2.4 mA cm⁻² at 1.23 V vs. RHE; when a cocatalyst FCP is introduced as a hole-transport layer, it shows a photocurrent density of 3.5 mA cm⁻² at 1.23 V vs. RHE, which is 8.7 times higher than that of the pure α-Fe₂O₃. The outstanding photochemical performance could be attributed to the highest separation efficiency. A further study on the carrier lifetime was performed and clarified that the photocarrier lifetime of the γ-Fe₂O₃/α-Fe₂O₃/FCP photoanode is prolonged (∼50.64 ps) as compared to that of the pure α-Fe₂O₃ photoanode (∼21.00 ps) using femtosecond time-resolved absorption spectroscopy (fs-TAS). This work successfully explains the photocatalytic water oxidation mechanism in the γ-Fe₂O₃/α-Fe₂O₃/FCP photoanode and provides an effective insight into designing a photocarrier delivery channel for the outstanding water oxidation.

---

1. Introduction

Photoelectrochemical (PEC) water splitting is regarded as the promising strategy to solve the energy crisis.^1–3 However, the rate of the photoanode reaction is lower than that of the water reduction reaction.^4,5 Therefore, improving the efficiency of the water oxidation reaction is the key to optimize the process of water splitting. In particular, α-Fe₂O₃ is an excellent candidate for sunlight-driven PEC water oxidation with good sunlight harvesting capability, outstanding stability and nontoxic nature.^6–8 Theoretically, hematite possesses a high solar-to-hydrogen efficiency (16.8%) and a maximum photocurrent of 12.6 mA cm⁻².^9–11 However, the strong recombination of photo-generated carriers and the short hole-diffusion length of pure α-Fe₂O₃ make it have poor water oxidation ability to meet practical applications. Therefore, various remarkable strategies have been progressed to achieve effective water oxidation kinetics and charge separation and these involve ion doping (Ti⁴⁺, Ta⁴⁺, Sn⁴⁺, etc.),^12–15 heterojunctions,^16–18 morphology control, oxygen vacancies,¹⁹ and oxygen-evolution cocatalysts (InOOH, Fh, and CoPi).^20–23

Among these strategies, constructing heterojunctions is able to make use of the interfacial electric field for further enhancing spatial charge separation, such as Co₃O₄/Ti–Fe₂O₃,²⁴ Fe₂O₃/CdS,^25,26 Fe₂O₃/g-C₃N₄,²⁷ and V₂O₅/ZnO.²⁸ However, the interfacial charge transfer in heterojunctions is hindered due to the presence of lattice matching with different degrees between different semiconductors. As an alternative, formation of phase junctions with the same composition was found to be an effective strategy to improve efficient charge separation. Accordingly, like WO₃,²⁹ TiO₂³⁰ and poly(1,4-diethynylbenzene)³¹ have been successfully fabricated and showed excellent charge separation efficiency. Moreover, few research studies provided an in-depth explanation on the dynamic electron and hole transfer processes using fs-TAS. Therefore, fs-TAS is necessary to study the role of phase junctions and cocatalysts in the photocatalytic OER process.

In this work, a Ti⁴⁺-doped phase junction with a hydrophilic ultrathin FeCo Prussian blue (FCP) photoanode was synthesized via a simple hydrothermal method.^10,32 As expected, the optimal γ-Fe₂O₃/α-Fe₂O₃/FCP photoanode exhibits a photocurrent density of 3.5 mA cm⁻² at 1.23 V vs. RHE, which is 8.7 times higher than that of pure α-Fe₂O₃. More importantly, the fs-TAS results demonstrate that the excellent synergies between the phase junction and FCP could not only facilitate the photoexcited hole transfer, but also maintain long-lived charge separation states (∼50.64 ps). This work provides distinct insights into designing excellent catalysts in water oxidation.

2. Experimental section

2.1 Materials

Ferric chloride hexahydrate (FeCl₃·6H₂O; 97.0–102%) was bought from Alfa Aesar. Potassium ferricyanide (K₃[Fe(CN)₆]), titanium tetrachloride (TiCl₄) and sodium nitrate (NaNO₃) were purchased from Sinopharm Reagent. Cobalt chloride (CoCl₂) and cobalt nitrate hexahydrate (Co(NO₃)₂·6H₂O) were bought from Shanghai Reagent. All materials were used without further purification.

2.2 Synthesis of the Ti–Fe₂O₃ bottom layer

According to the previous report, Ti–Fe₂O₃ nanorod arrays could be successfully hydrothermally prepared on FTO substrates.³³ The FTO glass was ultrasonically cleaned with acetone and ethanol to make its surface clean. To obtain the hydrothermal precursor solution, 0.5063 g FeCl₃·6H₂O and 1.0625 g NaNO₃ were mixed in deionized water (100 mL) under stirring for 20 min. Subsequently, different volume amounts of TiCl₄ solution (75 μL, 175 μL, and 275 μL) were added to the precursor solution. Then, the mixed aqueous solutions with different doping volumes of TiCl₄ solution were transferred into a 50 mL Teflon-lined autoclave inserted with FTO, and then heated to 100 °C for 12 h in an oven. After the oven was cooled down to room temperature, the obtained Ti–FeOOH was taken out and annealed at 550 °C for two hours under an air atmosphere. On the basis of different Ti doping volumes, the samples were denoted as 75Fe₂O₃ (75 μL), 175Fe₂O₃ (175 μL) and 275Fe₂O₃ (275 μL). The main reactions are listed as follows:³⁴

Fe³⁺ + 2H₂O = FeOOH + 3H⁺

2FeOOH = α-Fe₂O₃ + H₂O

2.3 Synthesis of the γ-Fe₂O₃/α-Fe₂O₃ phase junction

2.025 g FeCl₃·6H₂O and 4.25 g NaNO₃ were dissolved in deionized water, and the TiCl₄ ethanol solution (75 μL) was mixed with the above solution (100 mL) to form a hydrothermal precursor solution. Then, a piece of the Ti–Fe₂O₃ bottom layer was placed into the autoclave with its conducting side facing down. The autoclave was maintained at 100 °C for 12 h and allowed to cool naturally. Subsequently, the obtained samples were washed by water several times and annealed at 550 °C for 2 h. Interestingly, several nano-sized γ-Fe₂O₃ formed during the second calcination process, and the conversion mechanism was evaluated using the equations below:³⁵

α-Fe₂O₃ + 6H⁺ → 2Fe³⁺ + 3H₂O

2Fe³⁺ + 6NO₃⁻ → 2γ-Fe₂O₃ + 6NO_x

Fe³⁺ + 2H₂O → FeOOH + 3H⁺

γ-Fe₂O₃/FeOOH → γ-Fe₂O₃/α-Fe₂O₃ + H₂O

Therefore, after the second hydrothermal treatment and calcination, the core–shell γ-Fe₂O₃/α-Fe₂O₃ was successfully formed. In addition, according to the different Ti doping volumes, the composite samples were denoted as 75Ti–Fe₂O₃/75Ti–Fe₂O₃, 175Ti–Fe₂O₃/75Ti–Fe₂O₃ and 275Ti–Fe₂O₃/75Ti–Fe₂O₃. It was worth noting that in the following description, we denoted 175Ti–Fe₂O₃/75Ti–Fe₂O₃ as γ-Fe₂O₃/α-Fe₂O₃.

2.4 Synthesis of the γ-Fe₂O₃/α-Fe₂O₃/FCP photoanode

The amorphous FeCo Prussian blue (FCP) cocatalyst was loaded by sequentially dipping the γ-Fe₂O₃/α-Fe₂O₃ phase junction in 0.02 M K₃[Fe(CN)₆] aqueous solution and 0.04 M CoCl₂ aqueous solution through a method described previously.³⁶ First, the phase junction was immersed in K₃[Fe(CN)₆]³⁻ solution for 15 min and CoCl₂ solution for another 15 min, alternately, followed by rinsing with deionized water and drying in air. It is necessary to repeat this step several times (two to five) to obtain the optimized photoanode. Furthermore, the obtained samples are heated at 25, 75 and 300 °C to generate the best performance.

2.5 Synthesis of the γ-Fe₂O₃/α-Fe₂O₃/Co-Pi photoanode

A representative three-electrode system was used for the electrodeposition of cobalt phosphate (CoPi) on γ-Fe₂O₃/α-Fe₂O₃.²² Herein, this electrodeposition process was carried out at a constant potential of 1 V (vs. Ag/AgCl), in which the obtained γ-Fe₂O₃/α-Fe₂O₃ substrate served as the working electrode, a Pt sheet as the counter electrode, and Ag/AgCl as the reference electrode. In addition, the electrolyte was prepared by dissolving 0.1 M potassium phosphate and 0.5 mM Co(NO₃)₂·6H₂O in water (pH = 7). In this work, the electrodeposition time was 200 s to obtain the photoanode.

2.6 Structural characterization

X-ray diffraction with Cu Kα radiation (XRD, D/MAX2250 diffractometer) was applied to analyse the crystal structure from 10° to 80° at a scan rate of 10° min⁻¹. Field-emission scanning electron microscopy (SEM; Hitachi Company) was conducted to examine the surface morphology and film thickness. High-resolution transmission electron microscopy (HR-TEM; FEI Tecnai G2 F20) and the related energy-dispersive X-ray spectrometry (EDX) were used to further analyse the elemental composition. X-ray photoelectron spectroscopy (XPS) was performed to explore the surface chemical state of elements by using the C 1s peak as a correction. A UV-vis-NIR spectrophotometer (Shimadzu UV-3600) recorded the UV-vis absorption spectrum of the samples with BaSO₄ as the reference. The lock-in-based SPV measurements and TPV measurements on home-made equipment were applied to measure the separation and kinetic behaviours of photogenerated charge carriers.

A Helios pump/probe setup (Ultrafast Systems LLC) could be used to perform transient absorption (TA) spectroscopy. 800 nm fs laser pulses were generated from the amplifier (Coherent Libra, 50 fs, 1 kHz). Besides, we use a 3 mm sapphire crystal to product the white light continuum probe, and use a BBO crystal with frequency doubling the 800 nm to produce 400 nm pump pulses. In addition, the diameter of the pump beam spot is 2 mm, the pump power is 30 μW, and probe beam is 50 μm. The probe beam passing through the sample was collected using a UV-vis detector (CCD sensor).

2.7 PEC measurements

The PEC measurements of the composite phase junction were evaluated in a standard three-electrode system. The working electrode, counter electrode and reference electrode are the as-synthesized phase junction, the Pt sheet, and Ag/AgCl, respectively, which make up the CHI 660E electrochemical workstation. The illuminated area was controlled to 0.283 cm², and 1 M KOH aqueous solution (pH = 13.6) was applied as the electrolyte. A 300 W xenon lamp (Microsolar 300, Beijing Perfect light) was used to simulate AM 1.5 G illumination (100 mW cm⁻²).

The applied potential versus Ag/AgCl was translated into potential versus RHE based on the Nernst equation:

E_RHE = E_Ag/AgCl + 0.059 pH + E^θ_Ag/AgCl

The electrolyte in our PEC measurements was 1 M KOH (pH = 13.6) unless otherwise specified, and the active area was 0.283 cm². The current vs. potential curve was obtained by linear sweep voltammetry (LSV) at a scan rate of 10 mV s⁻¹.

Applied bias photon to current efficiency (ABPE) was calculated via the following equation:

where J_light and J_dark referred to the current densities of the synthesized photoanodes under illumination and in the dark, respectively. V_RHE referred to the applied potential versus RHE, and P_light referred to the incident light intensity (100 mW cm⁻²).

The injection efficiency of the photoanodes was obtained through employing 0.5 M H₂O₂ as a hole scavenger in the 1 M KOH electrolyte, and the oxidation of H₂O₂ is thermodynamically and kinetically more favorable than that of water. The injection efficiency (η_inj) was determined by using the following equation:

η_inj = J_H2O/J_H2O2

where J_H2O and J_H2O2 referred to the photocurrent density for PEC H₂O oxidation and PEC H₂O₂ oxidation, respectively.

The separation efficiency of photogenerated charge carriers during the reaction was characterized by the difference values of open circuit potential between the dark and illumination conditions.

Electrochemical impedance spectroscopy (EIS) was conducted with a frequency range of 10⁵ Hz–0.05 Hz at 1.0 V vs. RHE under illumination (100 mW cm⁻²) to ascertain the charge transport of the photoanode. The EIS spectrum was fitted using the ZView software according to the following fitted model:

The incident photon to current conversion efficiency (IPCE) was measured at 1.23 V vs. RHE with the help of a monochromator, and the intensity of monochromatic light was measured using a calibrated Si photodiode. The IPCE was obtained according to the equation:
Mott–Schottky analysis: in the M–S plot, the flat band potential of the photoelectrode is measured according to following equation:
in which C is the space charge capacitance, e is the electron charge, ε is the vacuum permittivity (8.85 × 10⁻¹² F m⁻¹), ε₀ is the relative dielectric constant of hematite (ε₀ = 80), N_D is the charge donor density (cm⁻³), E is the electrode applied potential, E_FB is the flat band potential, κ is the Boltzmann constant (1.38 × 10⁻²³ J K⁻¹) and T is the absolute temperature (in K).

3. Results and discussion

3.1 Composition and chemical state

X-ray diffraction (XRD) patterns could be measured to examine the crystalline structure of the samples (Fig. 1a). After the first hydrothermal treatment, the typical diffraction peaks located at 35.6° and 64° are indexed to the pure α-Fe₂O₃ (JCPDS No. 33-0664).³⁷ After the second hydrothermal reaction, besides the characteristic peaks of α-Fe₂O₃, a new characteristic peak of 31.2° could be found from the enlarged XRD pattern as shown in Fig. 1b, which could be attributed to the characteristic peak of γ-Fe₂O₃ (JCPDS No. 39-1346).³⁸ To further confirm the crystalline structure of γ-Fe₂O₃, the XRD pattern of the sample powder fabricated by the same method was obtained. The powder XRD patterns of γ-Fe₂O₃/α-Fe₂O₃ were obtained as shown in Fig. S1,† and the major diffraction peaks at 35.5° (311), 43.5° (400), 57.5° (511), and 63° (440) are in good agreement with γ-Fe₂O₃ (JCPDS No. 39-1346). Nevertheless, there is no evident diffraction peaks corresponding to the FeCo Prussian blue (FCP), which may result from its amorphous structure and ultrathin thickness. In order to further study the chemical states, X-ray photoelectron spectroscopy (XPS) is carried out. Fig. 1c reveals that there are two characteristic peaks at 710.5 eV and 724.1 eV, which correspond to Fe 2p_3/2 and Fe 2p_1/2. For pure α-Fe₂O₃, four characteristic peaks (709.9, 710.7, 711.5 and 713.4 eV) of Fe 2p_3/2 could be fitted, while two new characteristic peaks (713.1 and 711.4 eV) belonging to γ-Fe₂O₃ are present. After loading FCP on the γ-Fe₂O₃/α-Fe₂O₃ sample, another peak (714.2 eV) could be assigned to an Fe–Co bond that emerged.^22,39–41 In Fig. 1d, the spectrum of the O 1s region exhibits two peaks (531.6 and 529.5 eV) corresponding to the lattice oxygen (OH⁻) and hydroxyl oxygen (O²⁻).⁴² Moreover, the peaks located at 796.7 and the six characteristic peaks (779.9, 782.0, 783.8, 786.2, 788.4 and 790.7 eV) of γ-Fe₂O₃/α-Fe₂O₃/FCP and γ-Fe₂O₃/α-Fe₂O₃ are found to be the splitting signals of Co 2p_1/2 and Co 2p_3/2 (Fig. 1e), respectively.⁴³ As illustrated in Fig. 1f, two peaks located at 464.3 eV (Ti 2p_1/2) and 458.3 eV (Ti 2p_3/2) for Ti 2p are observed, which can correspond to Ti⁴⁺.⁴⁴ In addition, the survey scan spectrum of the γ-Fe₂O₃/α-Fe₂O₃/FCP and γ-Fe₂O₃/α-Fe₂O₃ photoanodes is shown in Fig. S2,† revealing that Fe, O, Co, and Ti are all detected. Moreover, the Raman spectra of γ-Fe₂O₃/α-Fe₂O₃ and α-Fe₂O₃ were measured as shown in Fig. S3,† and this also confirms the existence of γ-Fe₂O₃.

Fig. 1 (a) XRD patterns of the α-Fe₂O₃-based photoanodes; (b) enlarged XRD patterns in the range of 20–40°. XPS spectra of the α-Fe₂O₃-based photoanodes: Fe 2p (c); O 1s (d); Co 2p (e); Ti 2p (f).

Fig. 2a shows the schematic of the overall hydrothermal procedure of the core–shell structure. To study the morphologies of the as-prepared samples, the top-view and cross-sectional images from scanning electron microscopy (SEM) are shown in Fig. 2b and c; herein, the γ-Fe₂O₃/α-Fe₂O₃/FCP sample is made up of uniform and homogeneous nanorod arrays. There is no change in the surface morphology between γ-Fe₂O₃/α-Fe₂O₃ and the pristine α-Fe₂O₃ (Fig. S4).† Moreover, the high-resolution transmission electron microscopy (HR-TEM) image clearly reveals that the obtained photoanodes possess a core–shell structure, and the amorphous FCP was simultaneously formed on their surfaces (Fig. 2d). The typical lattice fringes of the phase junction (0.24 nm) belong to the (110) plane of γ-Fe₂O₃, while another lattice fringe (0.22 nm) could correspond to the (113) plane of α-Fe₂O₃ (Fig. 2e).⁴⁵ In addition, the Fe, O, Co and Ti elements in γ-Fe₂O₃/α-Fe₂O₃/FCP demonstrate that the FCP nanoparticles were uniformly dispersed in the γ-Fe₂O₃/α-Fe₂O₃ sample (Fig. 2f).

Fig. 2 (a) Fabrication diagram of the nanoarrays. (b) Superficial SEM image of γ-Fe₂O₃/α-Fe₂O₃/FCP. (c) Cross-sectional image of γ-Fe₂O₃/α-Fe₂O₃/FCP. (d) Magnified image of the core–shell structure of γ-Fe₂O₃/α-Fe₂O₃/FCP. (e) HR-TEM image of γ-Fe₂O₃/α-Fe₂O₃/FCP. (f) EDX elemental mapping of γ-Fe₂O₃/α-Fe₂O₃/FCP.

3.2 The synergistic effect of the phase junction and FCP in PEC water-oxidation performance

The PEC measurement is a basic strategy to systematically investigate the performance of the photoanode with the solar simulated light illumination (AM 1.5 G, 100 mW cm⁻²). Firstly, to verify the photocatalysis potential, the photoanodes are optimized by adjusting different Ti-doping volumes in the Fe₂O₃ bottom layer. As shown in Fig. S5a,† the core–shell phase junction with the 175 μL doping volume of Ti⁴⁺ in the Fe₂O₃ bottom layer possesses the best photocurrent response (denoted as γ-Fe₂O₃/α-Fe₂O₃). In addition, the ABPE and IPCE measurements (Fig. S5b and c†) also indicate that this phase junction exhibits the highest IPCE value (58% at 390 nm) and ABPE value (0.169%). Particularly, when modifying with the ultrathin FCP layer (Fig. 3a), the γ-Fe₂O₃/α-Fe₂O₃/FCP sample achieves excellent PEC performance; it is worth noting that the reaction time and temperature of FCP were optimized to obtain the optimal sample (Fig. S6);† the optimized current could reach 3.5 mA cm⁻² at 1.23 V vs. RHE, which is 8.5 times greater than that of pure α-Fe₂O₃. And the sample exhibits a ∼90 mV cathodic shift of the onset potential (Fig. S7†). In addition, FCP has a better photocatalytic performance than cobalt phosphate (CoPi) (Fig. S8†). The above analysis implies that the synergistic effect of the phase junction and FCP could optimize the performance in water oxidation. In order to comprehensively understand the role of the phase junction and FCP in improving the PEC performance, transient photocurrent response measurements under chopped light with a frequency of 0.25 Hz are employed (Fig. 3b). It shows that the photocurrent densities follow the trend γ-Fe₂O₃/α-Fe₂O₃/FCP > γ-Fe₂O₃/α-Fe₂O₃ > α-Fe₂O₃/FCP > α-Fe₂O₃, suggesting that the surface decoration of FCP and γ-Fe₂O₃ is unable to improve the water oxidation capacity. Further, chronoamperometric measurements based on i–t curves at 1.23 V vs. RHE are also performed (Fig. 3c), which is also consistent with the LSV results. Meanwhile, γ-Fe₂O₃/α-Fe₂O₃/FCP shows the highest ABPE as compared to others, and its value could reach up to 0.38% at 1.02 V_RHE (Fig. 3d). And the γ-Fe₂O₃/α-Fe₂O₃/FCP sample exhibits the highest IPCE value of 87% at 390 nm (Fig. 3e). Moreover, after 2 h photostability measurements, the γ-Fe₂O₃/α-Fe₂O₃/FCP and γ-Fe₂O₃/α-Fe₂O₃ samples could maintain about 94% and 97% of the initial photocurrent density (Fig. 3f), respectively. Interestingly, the contact angle of γ-Fe₂O₃/α-Fe₂O₃/FCP and α-Fe₂O₃ (Fig. 3f) was also measured to elucidate the liquid surface adhesion. Herein, a relatively smaller contact angle is obtained for γ-Fe₂O₃/α-Fe₂O₃/FCP, indicating that γ-Fe₂O₃/α-Fe₂O₃/FCP possesses the outstanding hydrophilic properties in the observed excellent PEC performance.

Fig. 3 (a) Current density–potential curves. (b) LSV curves under chopped light. (c) i–t curves at 1.23 V_RHE. (d) ABPE spectra. (e) IPCE plots at 1.23 V vs. RHE. (f) Results of the PEC OER stability test at 1.23 V vs. RHE in KOH electrolyte. The inset shows the water contact angle under ambient conditions.

Typically, three key factors in developing suitable photocatalysts could be attributed to solar light harvesting, photocarrier separation and injection efficiency.^46,47 First, the UV-vis diffuse reflectance spectra of the photoanode are measured (Fig. S9a),† and the light absorption wavelength edge of all the samples is around 600 nm. The bandgaps (E_g) of α-Fe₂O₃ could be estimated to be 2.11 eV based on the corresponding Tauc analysis (Fig. S9b).† These results indicate that both phase junction and FCP have no significant effect on the absorption properties. Consequently, it is reasonable to deduce that the enhanced charge separation and injection efficiency are the main reasons for the enhanced PEC performance.

Based on the above analysis and assumptions, open-circuit photovoltage (OCP) measurements are carried out to verify the effects on the behaviour of generated charges (Fig. 4a). In this work, the open circuit voltage (OCP) is the difference value between the open circuit voltage in the dark state (V_dark) and the open circuit voltage under the light state (V_light). The open circuit voltages in the dark state and the light state used here are both stable voltage values. The γ-Fe₂O₃/α-Fe₂O₃/FCP sample exhibits the highest OCP values, suggesting that the composite photoanode could offer an additional intrinsic built-in electric field to boost the driving force of charge separation. Besides, the injection efficiency of the photoanodes in 1 M KOH with or without 0.5 M H₂O₂ could be used to estimate the charge transfer efficiency at the interface between the phase junction and FCP (Fig. 4b). The hole injection efficiency is significantly improved when the phase junction and FCP are introduced. Meanwhile, M–S plot analysis is conducted to obtain underlying information about the charge transfer process, in which a much flatter slope for γ-Fe₂O₃/α-Fe₂O₃/FCP could be achieved as shown in Fig. 4c. Particularly, the N_D values of α-Fe₂O₃ and γ-Fe₂O₃/α-Fe₂O₃/FCP are 4 × 10¹⁹ cm⁻³ and 1.15 × 10²⁰ cm⁻³, respectively (see the inset in Fig. 4c). This result implies a substantially higher charge carrier density, which leads to the effective charge separation and transfer. Moreover, we investigate the driving force at the interfacial electric field between α-Fe₂O₃ and the FCP through the WF measurements (Fig. S10),† which directly reveals the contact potential difference (CPD) between the sample and the Au probe. The WF values are calculated based on the following equation:

Φ_sample = Φ_Au + eCPD (Φ_Au = 4.8 eV)

Fig. 4 (a) Open circuit potentials (OCPs) in 1 M KOH electrolyte. (b) Injection efficiencies. (c) M–S plots. The inset shows the enlarged Mott–Schottky curves. (d) Nyquist plots at 1.0 V vs. RHE.

Herein, the WF values of α-Fe₂O₃ and the FCP were determined to be 4.71 and 5.07 eV, respectively. The results indicated that the Fermi level of α-Fe₂O₃ is higher than that of FCP, and the free electrons would transfer from α-Fe₂O₃ to FCP until the Fermi level is aligned when FCP is deposited on α-Fe₂O₃. It has been clearly demonstrated earlier that the interfacial electric field is oriented from α-Fe₂O₃ to FCP, which could facilitate the hole transfer from α-Fe₂O₃ to FCP.

To further illustrate the impressive role of the phase junction and FCP in the charge transfer process, electrochemical impedance spectroscopy (EIS) analysis was conducted. It should be noted that the small and the large semicircle are relative to the charge transfer resistance (R₂) at the solid/solid interface and the charge injection resistance (R₃) at the photoanode/electrolyte interface, respectively.^48,49 As shown in Fig. 4d, the typical EIS Nyquist plots of the obtained four photoanodes are fitted using the equivalent circuit model. Their results are listed in Table S1† and the equivalent circuit is shown in Fig. S11.† It is worth mentioning that the charge transfer resistance (R₂) at the interface of γ-Fe₂O₃ and α-Fe₂O₃ dramatically lowered, which suggests that the phase junction is expected to accelerate the charge separation. It is apparent that a drastically decreased radius of γ-Fe₂O₃/α-Fe₂O₃/FCP is found relative to γ-Fe₂O₃/α-Fe₂O₃, which means that the R₃ of photogenerated holes is dramatically reduced due to the superior catalytic properties of FCP.

3.3 The charge carrier dynamics of the α-Fe₂O₃-based photoanodes

To deeply understand the transfer and separation of photogenerated charges, a surface/transient photovoltage (SPV/TPV) test is conducted (introduction of instruments in the ESI†). Theoretically, the SPV signal could be used to deduce the separation efficiency of the photocatalyst. And the stronger signal indicates better separation efficiency.^50–53 As shown in Fig. 5a, pure α-Fe₂O₃ exhibits a relatively weak SPV signal while γ-Fe₂O₃/α-Fe₂O₃/FCP shows the strongest SPV values. Besides, the conventional TPV signal of the phase junctions under 532 nm laser pulse is implemented to confirm the transfer processes of photogenerated charge carriers. The increasing signal suggests the outstanding separation efficiency. As shown in Fig. 5b, the γ-Fe₂O₃/α-Fe₂O₃/FCP photoanode prolongs the lifetime of photogenerated charges and effectively suppresses charge recombination. Moreover, we employed photoluminescence (PL) spectroscopy to further explore the charge recombination behaviors. The peaks at 633 nm (Fig. 5c) indicated that the synergistic effect of FCP and the core–shell phase junction was effective in suppressing the charge recombination. In addition, time-resolved photoluminescence (TRPL) was conducted to further understand the role of FCP and the phase junction in the dynamics of charge carriers. And the PL decay spectrum (Fig. 5d) was fitted using the biexponential function for obtaining the first fast decay and the later slow decay (inset in Fig. 5d). Herein, the fast lifetime (τ₁) of the samples decreased from 0.42 to 0.32 ns when γ-Fe₂O₃/α-Fe₂O₃/FCP was formed, and the slow lifetime (τ₂) of the γ-Fe₂O₃/α-Fe₂O₃/FCP photoanode drastically reduced from 4.96 to 4.52 ns; double-exponential fitting is suitable for the PL decay curve probed at 485 nm (Fig. S12).† These results fully explained that the introduction of FCP and the phase junction could effectively improve the charge separation efficiency.

Fig. 5 (a) SPV signal. The inset shows the schematic of the SPV measurements. (b) TPV signal from front side illumination. The inset shows the schematic of the TPV measurements. (c) Steady-state photoluminescence (PL) spectra at the excitation wavelength of 485 nm produced by a xenon lamp. (d) Results of time-resolved photoluminescence (TRPL) excited by a 485 nm pumped pulsed laser.

Femtosecond time-resolved transient absorption spectroscopy (fs-TAS) is applied to investigate the electron–hole separation efficiency and understand the kinetics process of oxygen production on the PEC performance. Herein, fs-TAS is used to track the real-time photogenerated hole dynamics under 400 nm excitation.^13,54–57 All samples obtained a broad negative absorption band as shown in Fig. 6a–d, due to the presence of the state filling of photobleaching and the stimulated emission. Compared with pure α-Fe₂O₃, the more intensive negative signal of γ-Fe₂O₃/α-Fe₂O₃ suggests that more photogenerated holes in the phase junction could be extracted from bulk. Moreover, the further improvement of the signal is mainly due to the ultrathin catalyst FCP; it is obvious that the FCP has a positive effect on the PEC performance. In addition, as shown in Fig. 6e–i, double-exponential fitting is suitable for the TAS decay curve probed at 585 nm in order to study the decay kinetics of photo-generated holes, and the related parameters are shown in Table S2.† Theoretically, the short lifetime (τ₁) and long lifetime (τ₂) components could be used to estimate the recombination degree. The prolonged lifetimes of the γ-Fe₂O₃/α-Fe₂O₃ and γ-Fe₂O₃/α-Fe₂O₃/FCP photoanodes are largely increased (τ₁: 0.94 ± 0.014 ps; τ₂: 33.45 ± 0.50 ps and τ₁: 1.31 ± 0.029 ps; τ₂: 50.64 ± 0.78 ps, respectively), as compared to that of pure α-Fe₂O₃ (τ₁: 0.81 ± 0.019 ps; τ₂: 21.00 ± 0.58 ps). As we know, during the PEC water splitting reaction, more longer-lived photogenerated holes on the surface are more beneficial for the PEC water splitting reaction, and the lifetimes of the different photoanodes follow the order γ-Fe₂O₃/α-Fe₂O₃/FCP > γ-Fe₂O₃/α-Fe₂O₃ > α-Fe₂O₃/FCP > α-Fe₂O₃, which is consistent with the results of the PEC performance. In addition, the pseudocolor TA plots (Fig. S13†) of the α-Fe₂O₃-based photoanodes indicate that the active ultralong lived charge separation state facilitates the water oxidation.

Fig. 6 TAS spectra of a) γ-Fe₂O₃/α-Fe₂O₃/FCP, b) γ-Fe₂O₃/α-Fe₂O₃, c) α-Fe₂O₃/FCP, and d) α-Fe₂O₃ pumped by a 400 nm femtosecond laser. e) Time profiles of the normalized TAS at 585 nm. Experimental decay kinetics fitted at 585 nm: f) γ-Fe₂O₃/α-Fe₂O₃/FCP, g) γ-Fe₂O₃/α-Fe₂O₃, h) α-Fe₂O₃/FCP and i) α-Fe₂O₃.

According to the above conclusion, Scheme 1 shows the schematic band diagram of the γ-Fe₂O₃/α-Fe₂O₃/FCP composite photoanode for water oxidation. In this system, benefitting from the enhanced charge separation, the holes of γ-Fe₂O₃ could timely transfer to α-Fe₂O₃ at the γ-Fe₂O₃/α-Fe₂O₃ interface. Meanwhile, the holes on α-Fe₂O₃ quickly captured by FCP could further participate in the water splitting reaction.

Scheme 1 The schematic band diagram illustrating the γ-Fe₂O₃/α-Fe₂O₃/FCP photoanode architecture with directions of charge separation and transportation.

4. Conclusions

In this study, it is demonstrated that both phase junction and FCP in γ-Fe₂O₃/α-Fe₂O₃/FCP could optimize the kinetics in water oxidation and improve the interfacial charge transfer for PEC water splitting. As a result, the optimal γ/α-Fe₂O₃/FCP composite photoanode exhibits the fast surface kinetics of α-Fe₂O₃ and a considerably negative onset potential, achieving a remarkable photocurrent density of up to 3.5 mA cm⁻² at 1.23 V vs. RHE, which is 8.7 times higher than that of pure α-Fe₂O₃. The excellent bi-functional effects originate from the rapid interfacial hole injection and long-lived charge separation states (∼50.64 ps) as investigated by fs-TAS, indicating that the core–shell nanorod structure loaded with FeCo Prussian blue (FCP) could boost the water oxidation kinetics and charge transfer efficiency. This work provides an effective insight into designing an outstanding water splitting photoanode.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We are grateful to the National Natural Science Foundation of China (No. 21773086, 21872063).

Notes and references

1. D. Zhang, W. He, J. Ye, X. Gao, D. Wang and J. Song, Small, 2021, 2005149.
2. Y. Xu, Z. Zhang, C. Qiu, S. Chen, X. Ling and C. Su, ChemSusChem, 2021, 14, 582–589.
3. X. Ren, Y. Ji, Y. Zhai, N. Yuan, J. Ding, Y. Li, J. Yan and S. F. Liu, J. Energy Chem., 2021, 60, 512–521.
4. W. Fu, X. Guan, Y. Si and M. Liu, Chem. Eng. J., 2021, 410, 128391.
5. S. Lin, H. Ren, Z. Wu, L. Sun, X.-G. Zhang, Y.-M. Lin, K. H. L. Zhang, C.-J. Lin, Z.-Q. Tian and J.-F. Li, J. Energy Chem., 2021, 59, 721–729.
6. H. Li, M. Yin, X. Li and R. Mo, ChemSusChem, 2021, 1–11.
7. Y. Fan, X. Ning, Q. Zhang, H. Zhao, J. Liu, P. Du and X. Lu, ChemSusChem, 2021, 14, 1414–1422.
8. C. Li, T. Wang, Z. Luo, S. Liu and J. Gong, Small, 2016, 12, 3415–3422.
9. T. H. Jeon, G.-h. Moon, H. Park and W. Choi, Nano Energy, 2017, 39, 211–218.
10. F. Feng, C. Li, J. Jian, F. Li, Y. Xu, H. Wang and L. Jia, J. Power Sources, 2020, 449, 227473.
11. T. Jiao, C. Lu, D. Zhang, K. Feng, S. Wang, Z. Kang and J. Zhong, Appl. Catal., B, 2020, 269, 118768.
12. C. Li, Z. Luo, T. Wang and J. Gong, Adv. Mater., 2018, 30, e1707502.
13. H. Zhang, D. Li, W. J. Byun, X. Wang, T. J. Shin, H. Y. Jeong, H. Han, C. Li and J. S. Lee, Nat. Commun., 2020, 11, 4622.
14. M. Li, Y. Yang, Y. Ling, W. Qiu, F. Wang, T. Liu, Y. Song, X. Liu, P. Fang, Y. Tong and Y. Li, Nano Lett., 2017, 17, 2490–2495.
15. Y. L. Tong, B. Q. Chi, D. L. Qi and W. Zhang, RSC Adv., 2021, 11, 1233–1240.
16. R. Kant, S. Pathak and V. Dutta, Sol. Energy Mater. Sol. Cells, 2018, 178, 38–45.
17. A. Helal, F. A. Harraz, A. A. Ismail, T. M. Sami and I. A. Ibrahim, Appl. Catal., B, 2017, 213, 18–27.
18. P. I. Kyesmen, N. Nombona and M. Diale, J. Alloys Compd., 2021, 863, 158724.
19. Y. Makimizu, N. T. Nguyen, J. Tucek, H. J. Ahn, J. Yoo, M. Poornajar, I. Hwang, S. Kment and P. Schmuki, Chemistry, 2020, 26, 2685–2692.
20. S. Zhang, P. Shangguan, S. Tong, Z. Zhang and W. Leng, J. Phys. Chem. C, 2019, 123, 24352–24361.
21. Q. Bu, S. Li, Q. Wu, L. Bi, Y. Lin, D. Wang, X. Zou and T. Xie, ChemSusChem, 2018, 11, 3486–3494.
22. Q. Bu, S. Li, K. Zhang, Y. Lin, D. Wang, X. Zou and T. Xie, ACS Sustainable Chem. Eng., 2019, 7, 10971–10978.
23. Z. Wang, J. Rong, J. Lv, R. Chong, L. Zhang, L. Wang, Z. Chang and X. Wang, J. Energy Chem., 2021, 56, 152–161.
24. S.-S. Yi, B.-R. Wulan, J.-M. Yan and Q. Jiang, Adv. Funct. Mater., 2019, 29, 1801902.
25. R. Shen, L. Zhang, X. Chen, M. Jaroniec, N. Li and X. Li, Appl. Catal., B, 2020, 266, 118619.
26. Y. Li, Q. Wu, Q. Bu, K. Zhang, Y. Lin, D. Wang, X. Zou and T. Xie, Chin. J. Catal., 2021, 42, 762–771.
27. R. Zhang, L. Bi, D. Wang, Y. Lin, X. Zou and T. Xie, Catal. Commun., 2020, 142, 106028.
28. T.-F. Hou, M. A. Johar, R. Boppella, M. A. Hassan, S. J. Patil, S.-W. Ryu and D.-W. Lee, J. Energy Chem., 2020, 49, 262–274.
29. Y. Li, Z. Tang, J. Zhang and Z. Zhang, Appl. Catal., B, 2017, 207, 207–217.
30. H. Zhang, L. Ma, J. Ming, B. Liu, Y. Zhao, Y. Hou, Z. Ding, C. Xu, Z. Zhang and J. Long, Appl. Catal., B, 2019, 243, 481–489.
31. H. Sun, C. Neumann, T. Zhang, M. Loffler, A. Wolf, Y. Hou, A. Turchanin, J. Zhang and X. Feng, Adv. Mater., 2019, 31, e1900961.
32. J. H. Kim, J. H. Kim, J. H. Kim, Y. K. Kim and J. S. Lee, Sol. RRL, 2019, 4, 1900328.
33. Q. Wu, Q. Bu, S. Li, Y. Lin, X. Zou, D. Wang and T. Xie, J. Alloys Compd., 2019, 803, 1105–1111.
34. D. Chen, Z. Liu and S. Zhang, Appl. Catal., B, 2020, 265, 118580.
35. D. Cao, H. Li, L. Pan, J. Li, X. Wang, P. Jing, X. Cheng, W. Wang, J. Wang and Q. Liu, Sci. Rep., 2016, 6, 32360.
36. B. Moss, F. S. Hegner, S. Corby, S. Selim, L. Francàs, N. López, S. Giménez, J.-R. Galán-Mascarós and J. R. Durrant, ACS Energy Lett., 2018, 4, 337–342.
37. Y. Zhang, J. He, Q. Yang, H. Zhu, Q. Wang, Q. Xue and L. Yu, J. Power Sources, 2019, 440, 227120.
38. C. Yang, J. Zhang, X. Zhu, Y. Liu, Y. Chen and C. Wang, Appl. Surf. Sci., 2020, 529, 147060.
39. H. Lan, A. Wei, H. Zheng, X. Sun and J. Zhong, Nanoscale, 2018, 10, 7033–7039.
40. M. C. Biesinger, B. P. Payne, A. P. Grosvenor, L. W. M. Lau, A. R. Gerson and R. S. C. Smart, Appl. Surf. Sci., 2011, 257, 2717–2730.
41. J. L. Ortiz-Quinonez, U. Pal and M. S. Villanueva, ACS Omega, 2018, 3, 14986–15001.
42. B. Sun, W. Zhou, H. Li, L. Ren, P. Qiao, F. Xiao, L. Wang, B. Jiang and H. Fu, Appl. Catal., B, 2018, 221, 235–242.
43. K. Fan, H. Chen, B. He and J. Yu, Chem. Eng. J., 2020, 392, 123744.
44. S. Li, Q. Zhao, D. Meng, D. Wang and T. Xie, J. Mater. Chem. A, 2016, 4, 16661–16669.
45. Y. Jin, L. Dang, H. Zhang, C. Song, Q. Lu and F. Gao, Chem. Eng. J., 2017, 326, 292–297.
46. A. Govind Rajan, J. M. P. Martirez and E. A. Carter, ACS Catal., 2020, 10, 11177–11234.
47. S. Cao, J. Low, J. Yu and M. Jaroniec, Adv. Mater., 2015, 27, 2150–2176.
48. B. Zhang, L. Chou and Y. Bi, Appl. Catal., B, 2020, 262, 258–265.
49. C. Feng, Q. Zhou, B. Zheng, X. Cheng, Y. Zhang and Y. Bi, J. Mater. Chem. A, 2019, 7, 22274–22278.
50. T. Jiang, T. Xie, W. Yang, L. Chen, H. Fan and D. Wang, J. Phys. Chem. C, 2013, 117, 4619–4624.
51. T. Jiang, T. Xie, L. Chen, Z. Fu and D. Wang, Nanoscale, 2013, 5, 2938–2944.
52. L. Bi, X. Gao, Z. Ma, L. Zhang, D. Wang and T. Xie, ChemCatChem, 2017, 9, 3779–3785.
53. Y. Li, Q. Wu, K. Zhang, B. Hu, Y. Lin, D. Wang and T. Xie, Chem. Res. Chin. Univ., 2020, 36, 1059–1067.
54. W. Wang, Y. Tao, L. Du, Z. Wei, Z. Yan, W. K. Chan, Z. Lian, R. Zhu, D. L. Phillips and G. Li, Appl. Catal., B, 2021, 282, 119568.
55. F. Liu, R. Shi, Z. Wang, Y. Weng, C. M. Che and Y. Chen, Angew. Chem., Int. Ed., 2019, 58, 11791–11795.
56. M. Sachs, J. S. Park, E. Pastor, A. Kafizas, A. A. Wilson, L. Francas, S. Gul, M. Ling, C. Blackman, J. Yano, A. Walsh and J. R. Durrant, Chem. Sci., 2019, 10, 5667–5677.
57. Y. Li, Q. Wu, Y. Chen, R. Zhang, C. Li, K. Zhang, M. Li, Y. Lin, D. Wang, X. Zou and T. Xie, Appl. Catal., B, 2021, 290, 120058.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1cy01628h

---