Constructing oxygen vacancy-enriched Fe₂O₃@NiO heterojunctions for highly efficient electrocatalytic alkaline water splitting

Abstract

Electrolysis of water to produce high-purity hydrogen is a very promising method. The development of green, high-efficiency, long-lasting and low-cost dual function electrocatalysts for oxygen evolution reaction (OER) and hydrogen evolution reaction (HER) is essential for electrocatalytic total water splitting. In this work, oxygen vacancy-enriched Fe₂O₃@NiO heterojunctions as bifunctional electrocatalysts are prepared through a facile one-step hydrothermal method followed by a calcination process. The synergistic effect of Fe₂O₃ and NiO, as well as the rich oxygen vacancies in Fe₂O₃, optimize their electronic structures, leading to an enhanced charge transfer rate and improved catalytic ability. Therefore, in both OER and HER processes, overpotentials needed for the Fe₂O₃@NiO catalyst to achieve the current density of 10 mA cm⁻² under alkaline conditions are 224 mV and 187 mV, respectively. Furthermore, the catalyst showed excellent dynamic characteristics and durability. This research provides a new strategy for regulating the electronic structure of bifunctional catalysts by heterostructures and oxygen vacancies, thereby promoting the performance of total water splitting.

---

1. Introduction

The discovery of sustainable, green and cheap new energy to replace traditional fuels has attracted extensive attention.^1–3 Hydrogen (H₂) has become a promising alternative energy source due to its high specific energy density and environmentally friendly features in many energy sources. Water splitting is a very effective strategy for producing high-purity hydrogen. The water-splitting process is divided into oxygen evolution reaction (OER) on the anode and hydrogen evolution reaction (HER) on the cathode.^4–6 Among them, OER has been regarded as an important process of water decomposition.⁷ However, OER is hampered by ultra-high potential because it is a coupled four-electron proton process with slow reaction dynamics.^8,9 The precious-metal-based catalysts such as RuO₂, Pt/C and IrO₂ are currently the most effective catalysts for OER and HER, however, their high price and low abundance greatly restrict their extensive use.^10,11 Therefore, the development of cheap and environmentally friendly non-noble metal-based catalysts has become the focus of attention. In particular, it has great potential to optimize the performance of catalysts by controlling the structure and composition using earth-abundant transition metals.^12–14

In recent years, transition metal-based catalysts (e.g. nitrides,^15,16 phosphides,^17,18 oxides,^19,20 sulfides,²¹ layered double hydroxide²²) have been widely used in research on HER and OER electrocatalysts. Among them, transition metal oxides have been widely investigated because of their simple method, environmental friendliness and stable electrocatalytic performance.^23,24 However, transition metal oxides are generally considered inefficient catalysts toward HER.²⁵ Therefore, for the sake of improving the activity of transition metal oxide bifunctional catalyst, it is an effective strategy to introduce or form a large number of defects into the catalyst.^26,27 Numerous studies have shown that structural defects can adjust the electronic structure and surface properties, and optimize the H₂O adsorption energy on the active site, thereby enhancing the catalytic activity.^28,29 In addition, the generation of oxygen vacancies as structural defects can improve the strong adsorption behavior of H₂O, which leads to the excellent electrocatalytic performance of OER and HER and realizes the overall water splitting.³⁰ According to the study by Lee et al., a large number of oxygen vacancies caused by chloride ions increased the number of active sites of (nickel foam) NF, thereby significantly improving the overall catalytic performance of water decomposition.³¹ Similarly, Liang and coworkers reported that the electronic structure of g-C₃N₄@CeO₂ photocatalyst was optimized by oxygen vacancies, which greatly improved the efficiency of charge separation and transfer and promoted the photocatalytic reduction of CO₂.³²

On the other hand, previous reports show that heterojunction electrode materials greatly improve electrocatalytic performance. Two compounds in a heterojunction are connected with each other, which produces a synergistic effect on the interface of epitaxial growth and inward growth.^33,34 Furthermore, the heterojunction can reduce the energy barrier of the water decomposition process and optimize the adsorption/desorption of intermediates.³⁵ Fe₂O₃@NiO heterojunctions have generated extensive attention in energy conversion fields because they optimize the adsorption/desorption of intermediates, enhancing active surface and efficient mass transport.³⁶ For instance, the individual p-NiO/n-Fe₂O₃ heterojunction nanowires were constructed by Singh et al. through electrodeposition of Fe and Ni NWs inside the pores of the anodic alumina template followed by controlled oxidation.³⁷ The p–n junction of NiO/P-α-Fe₂O₃ was prepared by Li et al. for photoelectrochemical water splitting.³⁸ Nevertheless, the synthesis of these Fe₂O₃@NiO heterojunctions usually requires more than two steps or templates. There is no report on the preparation of oxygen vacancy-enriched Fe₂O₃@NiO heterojunctions by a simple hydrothermal method followed by a calcination process without any template and surfactant. According to the above discussion, constructing a heterogeneous nanostructure with rich oxygen vacancies is an effective method to further increase the dual-function catalytic activities of electrocatalysts.

Herein, we report that oxygen vacancy-enriched Fe₂O₃@NiO heterojunction can be formed by a simple hydrothermal method followed by a calcination process without any template and surfactant. Moreover, the Fe₂O₃@NiO heterojunction acts as a highly active electrocatalyst for OER, HER and overall water splitting. Surprisingly, compared with the pure phase Fe₂O₃ and NiO, the oxygen vacancy-enriched Fe₂O₃@NiO catalyst exhibits prominent electrocatalytic activity and durability in 1.0 M KOH. The overpotential values at 10 and 50 mA cm⁻² for the OER process are 224, 264 mV, respectively, and the Tafel slope is only 20.0 mV dec⁻¹. Meanwhile, during the HER process, the corresponding overpotentials are 187 and 280 mV, respectively, and the Tafel slope is 53.8 mV dec⁻¹. Moreover, in the alkaline double electrode system, the total water-splitting reaction of the Fe₂O₃@NiO heterojunction requires only 1.63 V cell voltage at 10 mA cm⁻².

2. Experimental section

Materials

Polyvinylidene fluoride (PVDF) powder and acetylene black were purchased from Zhenkai Lai Zhuote Technology Co., Ltd. HCl (∼37%) and ethanol were purchased from the Sinopharm Group. Fe(NO₃)₃·9H₂O and (NO₃)₂·6H₂O were obtained from Xilong Science Co., Ltd. N,N-dimethylformamide (NMP), urea and KOH pellets were obtained from Maclean Ltd.

Synthesis of Fe₂O₃@NiO

Fe₂O₃@NiO was synthesized by dissolving 1 mmol Fe(NO₃)₃·9H₂O, 1 mmol Ni(NO₃)₂·6H₂O and 10 mmol urea in 20 ml deionized water under vigorous stirring. After fully dissolving the solids under ultrasonication, the homogeneous mixture was completely poured into an autoclave with a PTFE liner (40 ml) and maintained at 120 °C for 6 h. After the reaction was completed, the obtained products were cleaned with pure water and absolute ethanol and centrifuged 3 times. Then, the product was dried at 55 °C for 10 h. Finally, the Fe₂O₃@NiO catalyst was obtained by calcining the above product at 400 °C for 2 h, and the heating rate was 2 °C min⁻¹.

Synthesis of NiO and Fe₂O₃

The preparation method of pure NiO and Fe₂O₃ is similar to that of Fe₂O₃@NiO. In a typical experiment, 1 mmol Ni(NO₃)₂·6H₂O or Fe(NO₃)₃·9H₂O was added to the mixed solution containing 20 ml deionized water and 10 mmol urea. The homogeneous solution after the ultrasonic dissolution treatment was transferred in an autoclave to react at 120 °C for 6 h. After the reaction, the resulting solid product was washed and centrifuged with pure water and absolute ethanol several times, afterwards dried at 55 °C. Ultimately, the dried powder was annealed at 400 °C for 2 h to gain NiO and Fe₂O₃ respectively.

Fabrication of the working electrode

Firstly, 14 mg Fe₂O₃@NiO catalyst, 4 mg acetylene black (as conductive agent) and 2 mg PVDF powder were ground into a fine powder in an agate mortar. Then, 120 μL of NMP was appended to the powder to transform a uniform ink. The ink was coated on the NF with a size of 1 cm² to form a homogeneous electrode layer (mass loading: 2.8 mg cm⁻²). After natural drying, the working electrode was prepared and its electrochemical performance was tested.

Characterizations

The samples were characterized using X-ray powder diffraction (XRD, D8 advanced diffraction system) in the 2θ angle range of 10° to 80°. A field emission scanning electron microscope (Hitachi 8100) was used to obtain SEM images. High-resolution transmission electron microscopy (HRTEM, Fei Tecnai G2 F30) and transmission electron microscopy (TEM, Hitachi HT-7700) were used to characterize the lattice fringes and morphology of the samples. The specific surface area of Brunauer Emmett Teller (BET) and the pore size of BJH were measured using micromeritics ASAP 2460. EDX (Hitachi 8100) and XPS (XPS, ESCALAB 250Xi) spectra were used to determine the state and distribution of elements, respectively.

Electrochemical measurements

The electrocatalytic tests were carried out in 1.0 M KOH electrolyte at room temperature using the CHI660e electrochemical working instrument of Shanghai Chenhua Instrument Company. The Pt sheet (or graphite) and the Ag/AgCl (saturated KCl) electrodes were used as the counter and reference electrodes, respectively. The catalyst-supported NF served as the working electrode, and a three-electrode system was used for testing. Before the test, the electrolytic cell was purged with high-purity argon or oxygen for half an hour. During the tests, the scan rate of cyclic voltammetry (CV) was 50 mV s⁻¹. The polarization curve after iR compensation was obtained using linear sweep voltammetry (LSV), and the scan rate was 5 mV s⁻¹. During the electrochemical tests, the obtained potential based on the formula (E_{vs RHE} = E_{vs Ag/AgCl} + 0.059 pH + 0.197) was calibrated by a reversible hydrogen electrode (RHE). The Tafel slope of the Fe₂O₃@NiO catalyst was obtained from the compensated LSV polarization curve. The relationship between the current density and time was measured under the same conditions. Electrochemical impedance spectroscopy (EIS) was obtained by applying open-circuit voltage under the frequency range of 10⁻¹ Hz to 10⁵ Hz. The electrochemical surface area (ECSA) was assessed from the CV plots of the non-Faraday interval at different scanning rates.

Electrochemical H₂ and O₂ evolutions were measured in a homemade electrochemical device using a CHI660E electrochemical workstation. A two-electrode device was assembled using Fe₂O₃@NiO as the anode and cathode. The actual volumes of O₂ and H₂ produced in the whole water splitting process were compared with the theoretically calculated gas volume, and the Faraday efficiency was calculated.

3. Results and discussion

The Fe₂O₃@NiO heterojunction materials composed of Fe₂O₃ nanoparticles and NiO nanosheets were obtained by hydrothermal synthesis and annealing process. As illustrated in Fig. 1, Fe₂O₃@Ni(OH)₂ electrocatalysts were firstly synthesised by a hydrothermal method. In this process, under the action of urea, Ni(OH)₂ was generated through the reaction as shown in the equation at the bottom of Fig. 1.³⁹ Subsequently, Fe₂O₃@Ni(OH)₂ precursor was transformed into the Fe₂O₃@NiO heterojunction by the annealing process. During the heating process, the oxygen vacancy-enriched Fe₂O₃ was formed due to ion intermixing. The morphology of the Fe₂O₃@NiO heterojunction at different magnifications is shown in Fig. 2a. As depicted in this SEM micrograph, high-density Fe₂O₃ nanoparticles were modified on the surface of NiO nanoflakes after calcination. From the TEM image of Fe₂O₃@NiO heterojunction composites (Fig. 2b), we can also see that many Fe₂O₃ nanoparticles with 50–80 nm are anchored on the NiO nanosheets.

Fig. 1 The synthesis scheme of Fe₂O₃@NiO heterojunction and the formation reaction of Ni(OH)₂.

Fig. 2 (a) SEM images of the Fe₂O₃@NiO heterojunction. (b) TEM image. (c) HRTEM image. (d) STEM and corresponding elemental mapping of Fe, Ni and O for the Fe₂O₃@NiO heterojunction.

In addition, pure Fe₂O₃ and NiO were also synthesized, respectively. Fig. S1 and S2† display the TEM and SEM charts of single-phase Fe₂O₃ and NiO, respectively. In order to further study the formation of Fe₂O₃@NiO heterojunction, HRTEM was used to characterize the lattice fringes of the heterojunction. As seen from Fig. 2c, the two sets of lattice fringes with the spacing of 0.24 and 0.37 nm can be indexed to the (021) and (012) lattice planes of Fe₂O₃@NiO, respectively. Moreover, it can be clearly seen from the EDX elemental mapping (Fig. 2d), the as-prepared heterojunction is made up of O, Ni and Fe elements, without other impurities, and the atomic ratio of Fe/Ni is about 2 : 1 (see the ESI,† Fig. S4 and Table S1). The formation of Fe₂O₃@NiO heterogeneous composites was further confirmed by the XRD and XPS results. Furthermore, the heterogeneous interface between Fe₂O₃ and NiO is beneficial to improve the charge/electron transfer rate and expose more active sites, thus significantly improving the electrocatalytic activity of the materials.

The XRD diffraction pattern of the Fe₂O₃@NiO heterojunction is depicted in Fig. 3a. The characteristic peaks (012), (104), (110), (113), (024), (116), (214) and (300) of Fe₂O₃ (JCPDF 79-1741) can be observed at about 2θ of 24.15°, 33.17°, 35.64°, 40.87°, 49.47°, 54.08°, 62.45° and 64.02°, respectively. Meanwhile, the XRD peaks at 37.25°, 43.30°, 62.85°, 75.42° and 79.39° are assigned to (021), (202), (220), (223) and (042), respectively, of NiO (JCPDF 89-7101). It is obvious that the XRD characteristic peaks of Fe₂O₃@NiO entirely correspond to the pure phase Fe₂O₃ and NiO. Moreover, the pore size and specific surface area of the Fe₂O₃@NiO heterojunction were obtained from N₂ adsorption/desorption data. It can be seen from Fig. 3b that the prepared Fe₂O₃@NiO sample has obvious hysteresis isothermal loop. The results show that the Fe₂O₃@NiO heterojunction has a macropore volume of 0.30 cm³ g⁻¹. In addition, the Fe₂O₃@NiO heterojunction has a larger specific surface area than those of pure Fe₂O₃ and NiO, and about 72.29 m² g⁻¹. The high specific surface area can significantly expose more edge defects, promote the formation of active sites and enhance the electrical conductivity, which leads to a remarkable electrocatalytic performance.

Fig. 3 (a) XRD patterns of Fe₂O₃, NiO and Fe₂O₃@NiO. (b) Adsorption–desorption isotherm and the corresponding pore size distribution curves for Fe₂O₃@NiO, Fe₂O₃ and NiO, illustration showing the pore size distribution curve.

XPS was used for characterizing the surface elemental composition and valence of the material. The whole XPS spectra of Fe₂O₃@NiO before and after the test showed that O, Fe and Ni elements were present and the element states did not change significantly (Fig. S5†). In Fig. 4a, the two convolution peaks at 872.5 and 854.9 eV match with Ni 2p_1/2 and Ni 2p_3/2, respectively. Therefore, the spin–orbit splitting energy between these two peaks was calculated to be 17.6 eV, indicating the existence of Ni²⁺. The two peaks centered at 873.5 and 855.9 eV are ascribed to Ni³⁺, while the centers of the fitting peaks at 872.0 and 854.4 eV are attributed to the binding energy of Ni²⁺.^40,41 Meanwhile, two satellite vibration peaks (marked as “Sat.”) with smaller intensity can be observed at 879.6 and 861.3 eV.⁴² From the XPS diagram of Fe 2p for Fe₂O₃@NiO in Fig. 4b, it can be seen that there are two different peaks at 724.0 and 710.5 eV, which are typical characteristic peaks of Fe³⁺ in 2p_1/2 and 2p_3/2 orbitals, respectively. In addition, the two deconvoluted peaks at 726.9 and 710.4 eV correspond to Fe²⁺, which is due to the presence of oxygen vacancies caused by the reduction of a part of Fe³⁺ to Fe²⁺. The existence of oxygen vacancies gives the system a certain reducibility. Meanwhile, the two deconvoluted peaks at 732.8 and 717.7 eV are ascribed to the presence of their satellite vibration peaks (marked as “Sat.”).^43–45 Comparing XPS spectra of samples before and after the test, the surface elemental composition and element valence of Fe and Ni metals did not change, which further proves the chemical stability of the materials during the catalytic process. As depicted in Fig. 4c, the XPS spectra of O1s can be well deconvolved into two oxygen peaks (labeled O1 and O2), which are related to the metal–oxygen bond and oxygen vacancy, respectively. The typical metal–oxygen bond corresponds to the O1 peak at 529.6 eV, while the oxygen vacancy peak corresponding to O2 is 531.2 eV.^27,46 However, as shown in Fig. 4c, the binding energy peaks corresponding to O 1 s orbital move towards the lower binding energy regions after the test, which may be due to the slight change of Fe²⁺/Fe³⁺ ratio. The XPS results indicate the co-presence of Fe²⁺ and Fe³⁺. It is important that the formed redox couple of Fe²⁺/Fe³⁺ can accelerate the charge transfer in Fe₂O₃ with rich oxygen vacancies because Fe³⁺ is reduced to Fe²⁺. The coupling between Fe²⁺ and Fe³⁺ ions of different valences has a great influence on oxygen vacancies. Furthermore, molar ratios of Fe²⁺/Fe³⁺ before and after the test (Table S3†) were 59.3% and 44.5%, respectively; oxygen vacancies would be changed to maintain the electrostatic balance, and the density of the oxygen vacancy would decrease after the test. Hence, according to the XPS spectra, the slight change of Fe²⁺/Fe³⁺ ratio indicates rearrangement in the valence states, which could also cause a shift in the binding energy of M–O. Moreover, part of the electrons of iron will be transferred from the symmetric electronic structure to the asymmetric three-dimensional electronic structure containing the oxygen vacancies.^9,47,48 Therefore, oxygen vacancies will optimize the partial density of states of three-dimensional electrons in Fe₂O₃@NiO, and promote the reaction sites to improve the electrocatalytic activity. To further probe the possible existence of oxygen vacancies in the material, we recorded the electron paramagnetic resonance (EPR) spectrum (see Fig. S6†). According to the study of Jitianu et al., two main EPR signals (g = 2.0 and g = 4.3) can be attributed to two different Fe³⁺ sites.⁴⁹ The EPR signal (g = 2.006) can be ascribed to oxygen vacancies.⁵⁰ However, as shown in Fig. S6,† the as-prepared Fe₂O₃@NiO nanostructures show a broad and symmetric signal at g ∼ 2. This is due to Fe(III) ions in the spin pair of Fe³⁺–O²⁻–Fe³⁺ can give a broad and symmetric signal at g ∼ 2, and oxygen vacancies also appear a signal at g ~ 2.^51,52 The coexistence of Fe³⁺ and oxygen vacancies results in a broad and symmetric EPR signal at g ∼ 2. In addition, the generation of oxygen vacancies can also improve the strong adsorption behavior of H₂O. Based on the above discussion and analysis, as-prepared Fe₂O₃@NiO possesses abundant heterojunction interfaces, a large proportion of Fe²⁺/Fe³⁺ and rich O vacancies, which accelerate the charge/electron transfer rate and bring about a better electrical conductivity, thus improving the electrocatalytic performance of OER and HER and realizing the overall water splitting.

Fig. 4 XPS spectra of Fe₂O₃@NiO before and after testing: (a) Ni 2p, (b) Fe 2p and (c) O 1s.

The electrocatalytic activity of the Fe₂O₃@NiO heterojunction in 1.0 M KOH was measured by LSV. In contrast, the electrocatalytic activities of NiO, Fe₂O₃ and commercial RuO₂ under the same environment were also discussed. Before the electrochemical tests, the working electrode was scanned using cyclic voltammetry to obtain a relatively stable state. The compensated LSV curves of Fe₂O₃@NiO heterojunction and other electrodes are shown in Fig. 5a. When the current density is 10 mA cm⁻², the overpotential of the Fe₂O₃@NiO heterojunction was 224 mV, which is much smaller than the overpotential of NiO (315 mV) and Fe₂O₃ (319 mV). In addition, the catalytic activity of Fe₂O₃@NiO is better than that of commercial RuO₂, NF and published non-noble metal OER catalysts (Table S2†). Fig. 5c shows the overpotential comparison between this catalyst and other controls at different current densities. The results indicate that the Fe₂O₃@NiO heterojunction shows the lowest overpotential at the above current density. Furthermore, the Tafel slope is an important kinetic parameter for further study on electrochemical kinetic processes, and the lower Tafel slope indicates faster electrocatalytic kinetics. In Fig. 6a, the Tafel value was calculated from the measured polarization curve. We can see that the Tafel slope of Fe₂O₃@NiO heterojunction was obviously less than that of the other control groups, which indicates that the Fe₂O₃@NiO electrode is more conducive to the OER kinetic process. Besides, the charge transport characteristics of the Fe₂O₃@NiO heterojunction was researched using EIS. As shown in Fig. 6b, the EIS half-circle arc radius of Fe₂O₃@NiO is the shortest, indicating that the Fe₂O₃@NiO heterojunction has the fastest charge transfer dynamics during the electrochemical process. Moreover, the stability of the catalyst was further explored. The performance of the Fe₂O₃@NiO catalyst before and after the 4000 CV cycle is compared in Fig. 6c. It can be seen that the OER polarization curve of Fe₂O₃@NiO has almost no reduction after 4000 cycles in alkaline electrolytes. Simultaneously, the long-term OER persistence of Fe₂O₃@NiO heterojunction at 50 and 10 mA cm⁻² was studied by potentiostatic chronoamperometry. As shown in Fig. 6d, the prepared catalyst still has good OER activity after 20 hours of continuous testing and exhibits excellent stability. Moreover, the multi-step chronocurrent curves of Fe₂O₃@NiO were recorded with the current density successively increasing from 10 to 100 mA cm⁻² at regular intervals of 1200 s (Fig. 6e). At the beginning of 1.455 V, it stabilizes at 10 mA cm⁻² and stays unchanged for 1200 s. The current density gradually increases and similar results are also obtained. Fig. 6e shows that the obtained Fe₂O₃@NiO exhibits fast and stable responses at different current densities. In summary, these results indicate that the Fe₂O₃@NiO catalyst displays good OER stability and activity in the continuous electrochemical process.

Fig. 5 (a and c) LSV polarization curves and overpotentials of different catalysts after compensation in OER. (b and d) LSV polarization curves and overpotentials of different catalysts after compensation in HER.

Fig. 6 (a) Corresponding Tafel plots. (b) EIS spectra of different catalysts. (c) Cycle curves of Fe₂O₃@NiO before and after 4000 cycles at a scanning speed of 50 mV s⁻¹. (d) The relationship curves between current density and time (i–t) after 20 h of OER electrocatalysis at j = 10 and 50 mA cm⁻². (e) Multi-step chronoamperometric curves of OER over Fe₂O₃@NiO. (f) The plots of current density to scan rate (10–100 mV s⁻¹).

In order to deeply explore the mechanism and kinetics of the electrocatalytic reaction, ESCA of the catalysts was also discussed. Generally, the increase of ECSA provides more active sites and promotes the improvement of electrocatalytic activity. The ECSA value can be estimated using the electrochemical double-layer capacitance (C_dl) obtained by cyclic voltammetry (CV). The C_dl value can be calculated using the formula: J_c = vC_dl.^53,54J_c and v are the current density and scanning rate, respectively. In addition, the C_dl values can be calculated by the slope of the plot Δj vs. the scan rate and dividing by two. The ECSA value can be obtained from the C_dl value according to the following formula: ECSA = C_dl/C_s, where C_s represents the specific capacitance, and the C_s value is 0.04 mF cm⁻².⁵⁵ As shown in Fig. S7,† the cyclic voltammetry was measured at different scanning rates in the input potential range of 0–0.1 V. It can be discovered from Fig. 6f that the Fe₂O₃@NiO heterojunction shows the highest C_dl value (40.7 mF cm⁻²) compared with NF (0.6 mF cm⁻²), Fe₂O₃ (2.5 mF cm⁻²) and NiO (19.0 mF cm⁻²). This indicates that the Fe₂O₃@NiO heterojunction with a high electrochemical surface area can supply more active centers for OER. In addition, it can be seen in Fig. S3† that the morphology before and after the performance test was not changed significantly.

These results indicate that the Fe₂O₃@NiO catalyst exhibits a faster charge transfer rate and a faster kinetic process in the OER process, compared with pure Fe₂O₃ and NiO catalysts. Besides, the Fe₂O₃@NiO catalyst also has good mechanical stability and corrosion resistance.

While studying the OER activity, we also studied the HER activity of Fe₂O₃@NiO in 1.0 M KOH. Similarly, the properties of Fe₂O₃, NiO and NF were compared under the same conditions. Fig. 5b displays the HER polarization curve of various samples. It can be seen that when the current density is −10 mA cm⁻², the corresponding overpotential of the Fe₂O₃@NiO catalyst reaches 187 mV, which is evidently smaller than the value of NiO (247 mV), Fe₂O₃ (277 mV) and NF (278 mV) catalysts. Fig. 5d compares the overpotential of Fe₂O₃@NiO with other control groups at different current densities. Moreover, Table S2† depicts the comparison of the overpotential of the prepared Fe₂O₃@NiO with the most reported non-noble metal electrocatalysts in the alkaline electrolyte at −10 mA cm⁻². Remarkably, the HER catalytic activity of Fe₂O₃@NiO is superior to that of most reported non-precious metal-based electrocatalysts. Similarly, the Tafel value is calculated from the LSV curve for HER (Fig. 7a). The Tafel value of the Fe₂O₃@NiO catalyst is obviously less than 68.5 mV dec⁻¹ of Fe₂O₃, 99.6 mV dec⁻¹ of NF and 70.3 mV dec⁻¹ of NiO. Compared with other controls, the overpotential (187 mV) and Tafel value (53.8 mV dec⁻¹) of Fe₂O₃@NiO at −10 mA cm ⁻² both show the smallest values. This result shows that the material still has good electrocatalytic performance in the HER process.

Fig. 7 (a) Tafel slope. (b) EIS diagram. (c) Cycle curves before and after 4000 CV cycles at a sweep rate of 50 mV s⁻¹ for HER. (d) Multi-step chronoamperometric curve. (e) Current density vs. time (i–t) curves at 10 and 50 mA cm⁻² over 20 h-long electrocatalytic HER.

In addition, to study the charge transferability of the materials in the electrocatalytic process, the EIS test was also performed. In Fig. 7b, the diameter of the fitted impedance semicircle for Fe₂O₃@NiO is markedly smaller than those of Fe₂O₃, NiO and NF, which indicate that the Fe₂O₃@NiO exhibits the smallest charge transfer resistance (R_ct) and excellent conductivity for HER. Furthermore, we also investigated the stability of the catalyst for HER. Fig. 7c shows the comparison of CV curves of the Fe₂O₃@NiO electrode at the beginning and after 4000 cycles, which expresses that the deviation of the CV curve is small. Meanwhile, the multi-step chronopotentiometry curves for Fe₂O₃@NiO material were also obtained (Fig. 7d). The steady voltage–current response of Fe₂O₃@NiO reveals that it also possesses excellent mass transport properties for HER. Moreover, the chronoamperometric curves of the continuous test for 20 h were recorded under −10 and −50 mA cm ⁻², respectively. Surprisingly, the current density still remains above 97% after 20 h of continuous testing (Fig. 7e). As shown in Fig. S3,† there is no obvious change in the form and structure of the material before and after the performance test. In conclusion, Fe₂O₃@NiO showed excellent catalytic performance and persistence in the HER process.

The excellent HER and OER activities of the Fe₂O₃@NiO catalyst promote the study of the overall water splitting performance as an excellent electrocatalyst. Therefore, by using the Fe₂O₃@NiO catalyst as both the anode and cathode, the double electrode device of Fe₂O₃@NiO||Fe₂O₃@NiO was constructed, by which the electrocatalytic water decomposition process was studied in 1.0 M KOH solution. Fig. 8a displays the LSV curves measured with a scanning speed of 5 mV s⁻¹. The constructed Fe₂O₃@NiO||Fe₂O₃@NiO device requires 1.63 V at 10 mA cm⁻². Undoubtedly, Fe₂O₃@NiO shows better water splitting performance than that Fe₂O₃, NiO and NF. In addition, the device was subjected to a long-term continuous electrocatalysis experiment at 10 mA cm⁻². In contrast, the prepared Fe₂O₃@NiO||Fe₂O₃@NiO device had undergone a 20 h long-term measurement, and the density of the current remains unchanged, which reflects the excellent stability of the material in the overall water splitting (Fig. 8b). Additionally, the Faraday efficiency of the hydrolysis process was calculated by measuring the actual volume of O₂ and H₂ produced in the anode and cathode catalysts in an H-type electrolyzer. Fig. 8c shows the amount of gas produced during electrolysis at high current density. It can be seen that the ratio of O₂ to H₂ is almost 1 : 2. This is basically consistent with the theoretical calculation results, indicating that the Faraday efficiency of the whole water splitting process is close to 100%.

Fig. 8 (a) LSV polarization curves of the catalysts in the two-electrode system for total water splitting. (b) Fe₂O₃@NiO long-term chronoamperometric stability test in 1.0 M KOH electrolyte. (c) Comparison of the amount of H₂ and O₂ released over time and theoretical values during the water splitting of Fe₂O₃@NiO.

Based on the above description and discussion, the Fe₂O₃@NiO heterojunction catalyst with oxygen vacancies exhibits excellent OER, HER and overall water decomposition performance, which may be due to these factors: (i) the synergism of Fe₂O₃@NiO heterostructure can effectively improve the electrochemical performance by reducing the energy barrier of water decomposition process; (ii) the existence of oxygen vacancies optimizes the surface properties and electronic structure and the formed Fe²⁺/Fe³⁺ redox pair can promote charge transferability, thereby enhancing the catalytic performance; (iii) nanostructures constructed by Fe₂O₃ nanoparticles attached to NiO nanosheets form a great number of active centers and modulate electronic interaction with a faster speed, which can greatly enhance the electrochemical performance.

4. Conclusions

In short, the efficient and stable Fe₂O₃@NiO heterojunction electrocatalyst was prepared by a simple hydrothermal synthesis and calcination method. The heterojunction formed by direct integration of Fe₂O₃ on NiO nanosheets significantly reduces the water splitting barrier, optimizes the adsorption/desorption of intermediates, enhances active surface and efficient mass transport. In addition, the rich oxygen vacancies optimize the electronic structure of the composite material and improve the efficiency of charge separation and transfer, thereby promoting the catalytic efficiency of HER and OER. The overpotential values corresponding to OER and HER are 224 and 187 mV at 10 mA cm⁻², respectively. The water electrolysis cell of the two-electrode system assembled with Fe₂O₃@NiO only needs 1.63 V cell voltage to reach 10 mA cm⁻² for overall water decomposition and has excellent stability. Consequently, this work may provide a convenient and useful means for the development of non-noble metal water splitting electrocatalysts with high activity, durability, and low cost.

Author contributions

Yan Sang: supervision. Xi Cao: conceptualization, methodology, writing – original draft. Gaofei Ding: visualization, investigation. Zixuan Guo: software, validation, data curation. Yingying Xue: formal analysis, investigation. Guohong Li: methodology, investigation, validation, writing – review & editing. Runhan Yu: project administration, funding acquisition.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Natural Science Foundation of Anhui Provincial Education Department (KJ2020ZD05) and National College Students Innovation and Entrepreneurship Training Program (202010370160).

References

1. W. Ahn, M. G. Park, D. U. Lee, M. H. Seo, G. Jiang, Z. P. Cano, F. M. Hassan and Z. Chen, Adv. Funct. Mater., 2018, 28, 1802129.
2. S. Anantharaj and S. Noda, Small, 2020, 16, 1905779.
3. Y. Chen, C. Dong, J. Zhang, C. Zhang and Z. Zhang, J. Mater. Chem. A, 2018, 6, 8430–8440.
4. K. Kannimuthu, K. Sangeetha, S. S. Sankar, A. Karmakar, R. Madhu and S. Kundu, Inorg. Chem. Front., 2021, 8, 234–272.
5. K. N. Dinh, P. Zheng, Z. Dai, Y. Zhang, R. Dangol, Y. Zheng, B. Li, Y. Zong and Q. Yan, Small, 2018, 14, 1703257.
6. L. L. Feng, G. Yu, Y. Wu, G. D. Li, H. Li, Y. Sun, T. Asefa, W. Chen and X. Zou, J. Am. Chem. Soc., 2015, 137, 14023–14026.
7. J. Staszak-Jirkovsky, C. D. Malliakas, P. P. Lopes, N. Danilovic, S. S. Kota, K. C. Chang, B. Genorio, D. Strmcnik, V. R. Stamenkovic, M. G. M. G. Kanatzidis and N. M. Markovic, Nat. Mater., 2016, 15, 197–203.
8. J. Ping, Y. Wang, Q. Lu, B. Chen, J. Chen, Y. Huang, Q. Ma, C. Tan, J. Yang, X. Cao, Z. Wang, J. Wu, Y. Ying and H. Zhang, Adv. Mater., 2016, 28, 7640–7645.
9. R. Liu, Y. Wang, D. Liu, Y. Zou and S. Wang, Adv. Mater., 2017, 29, 1701546.
10. K. Hu, M. Wu, S. Hinokuma, T. Ohto, M. Wakisaka, J. Fujita and Y. Ito, J. Mater. Chem. A, 2019, 7, 2156–2164.
11. Y. Hu, H. Yang, J. Chen, T. Xiong, M. S. Balogun and Y. Tong, ACS Appl. Mater. Interfaces, 2019, 11, 5152–5158.
12. Y. Pei, Y. Ge, H. Chu, W. Smith, P. Dong, P. M. Ajayanc, M. Ye and J. Shen, Appl. Catal., B, 2019, 244, 583–593.
13. Z. Qiu, Y. Ma and T. Edvinsson, Nano Energy, 2019, 66, 104118.
14. Y. Wu, Y. Liu, G. D. Li, X. Zou, X. Lian, D. Wang, L. Sun, T. Asefa and X. Zou, Nano Energy, 2017, 35, 161–170.
15. Z. Yang, Y. Lin, F. Jiao, J. Li, W. Wang, Y. Gong and X. Jing, Appl. Surf. Sci., 2020, 502, 144147.
16. H. Liu, X.-H. Zhang, Y.-X. Li, X. Li, C.-K. Dong, D.-Y. Wu, C.-C. Tang, S.-L. Chou, F. Fang and X.-W. Du, Adv. Energy Mater., 2020, 10, 1902521.
17. Y. Ding, B.-Q. Miao, S.-N. Li, Y. C. Jiang, Y.-Y. Liu, H.-C. Yao and Y. Chen, Appl. Catal., B, 2020, 268, 118393.
18. L. Cao, Y. Hu, S. Tang, A. Iljin, J. Wang, Z. Zhang and T. Lu, Adv. Sci., 2018, 5, 1800949.
19. J. Yang, J. Zheng, M. Xu, Z. Zhuo, W. Yang, L.-W. Wang, L. Dai, J. Lu, K. Amine and F. Pan, ACS Catal., 2018, 8, 466–473.
20. X. Gao, H. Zhang, Q. Li, X. Yu, Z. Hong, X. Zhang, C. Liang and Z. Lin, Angew. Chem., Int. Ed., 2016, 55, 6290–6294.
21. Y. Gong, Y. Lin, Z. Yang, F. Jiao, J. Li and W. Wang, Appl. Surf. Sci., 2019, 476, 840–849.
22. A. Karmakar, K. Karthick, S. S. Sankar, S. Kumaravel, R. Madhu and S. Kundu, J. Mater. Chem. A, 2021, 9, 1314–1352.
23. H. Xu, J. Wei, K. Zhang, M. Zhang, C. Liu, J. Guo and Y. Du, J. Mater. Chem. A, 2018, 6, 22697–22704.
24. S. Jin, ACS Energy Lett., 2017, 2, 1937–1938.
25. J. Chen, F. Wang, X. Qi, H. Yang, B. Peng, L. Xu, Z. Xiao, X. Hou and T. Liang, Electrochim. Acta, 2019, 326, 134979.
26. H. Xu, J. Wei, M. Zhang, C. Liu, Y. Shiraishi, C. Wang and Y. Du, Nanoscale, 2018, 10, 18468–18472.
27. S. Sultana, S. Mansingh and K. M. Parida, J. Mater. Chem. A, 2018, 6, 11377–11389.
28. L. Zhuang, Y. Jia, H. Liu, Z. Li, M. Li, L. Zhang, X. Wang, D. Yang, Z. Zhu and X. Yao, Angew. Chem., Int. Ed., 2020, 59, 14664–14670.
29. Y. Zhang, J. Fu, H. Zhao, R. Jiang, F. Tian and R. Zhang, Appl. Catal., B, 2019, 257, 117899.
30. A. Karmakar, K. Karthick, S. S. Sankar, S. Kumaravel, M. Ragunath and S. Kundu, J. Mater. Chem. A, 2021, 9, 11691–11704.
31. W. H. Lee, M. H. Han, U. Lee, K. H. Chae, H. Kim, Y. J. Hwang, B. K. Min, C. H. Choi and H.-S. Oh, ACS Sustainable Chem. Eng., 2020, 8, 14071–14081.
32. M. Liang, T. Borjigin, Y. Zhang, B. Liu, H. Liu and H. Guo, Appl. Catal., B, 2019, 243, 566–575.
33. J. Su, L. Guo, N. Bao and C. A. Grimes, Nano Lett., 2011, 11, 1928–1933.
34. J. Su, G.-D. Li, X.-H. Li and J.-S. Chen, Adv. Sci., 2019, 6, 1801702.
35. L. Tao, M. Li, S. Wu, Q. Wang, X. Xiao, Q. Li, M. Wang, Y. Fu and Y. Shen, Catal. Sci. Technol., 2018, 8, 4151–4158.
36. B. H. R. Suryanto, Y. Wang, R. K. Hocking, W. Adamson and C. Zhao, Nat. Commun., 2019, 10, 5599.
37. A. K. Singh and D. Sarkar, Nanoscale, 2018, 10, 13130–13139.
38. F. Li, J. Li, J. Zhang, L. Gao, X. Long, Y. Hu, S. Li, J. Jin and J. Ma, ChemSusChem, 2018, 11, 2156–2164.
39. E. Hosono, S. Fujihara, I. Honma and H. Zhou, Adv. Mater., 2005, 17, 2091–2094.
40. S. Hao, L. Chen, C. Yu, B. Yang, Z. Li, Y. Hou, L. Lei and X. Zhang, ACS Energy Lett., 2019, 4, 952–959.
41. F. Wu, Q. Liao, F. Cao, L. Li and Y. Zhang, Nano Energy, 2017, 34, 8–14.
42. X. Cao, Y. Sang, L. Wang, G. Ding, R. Yu and B. Geng, Nanoscale, 2020, 12, 19404–19412.
43. A. K. Singh and D. Sarkar, Nanoscale, 2018, 10, 13130–13139.
44. L. Zhang, Q. Fang, Y. Huang, K. Xu, P. K. Chu and F. Ma, Anal. Chem., 2018, 90, 9821–9829.
45. G. Shen, R. Zhang, L. Pan, F. Hou, Y. Zhao, Z. Shen, W. Mi, C. Shi, Q. Wang, X. Zhang and J.-J. Zou, Angew. Chem., Int. Ed., 2020, 59, 2313–2317.
46. Y. Ma, Z. Lu, S. Li, J. Wu, J. Wang, Y. Du, J. Sun and P. Xu, ACS Appl. Mater. Interfaces, 2020, 12, 12668–12676.
47. Y. Ding, J. Zhao, W. Zhang, J. Zhang, X. Chen, F. Yang and X. Zhang, ACS Appl. Energy Mater., 2019, 2, 1026–1032.
48. T. Odedairo, X. Yan, X. Yao, K. K. Ostrikov and Z. Zhu, Adv. Mater., 2017, 29, 1703792.
49. P. F. Liu, S. Yang, B. Zhang and H. G. Yang, ACS Appl. Mater. Interfaces, 2016, 8, 34474–34481.
50. J. Dong, Y. Wang, Q. Jiang, Z. Nan, F. R. Fan and Z. Tian, J. Mater. Chem. A, 2021, 9, 20058–20067.
51. A. Jitianu, M. Crisan, A. Meghea, I. Rau and M. Zaharescu, J. Mater. Chem., 2002, 12, 1401–1407.
52. G. Alvarez, R. Font, J. Portelles, O. Raymond and R. Zamorano, Solid State Sci., 2009, 11, 881–884.
53. D. Friebel, M. W. Louie, M. Bajdich, K. E. Sanwald, Y. Cai, A. M. Wise, M.-J. Cheng, D. Sokaras and T.-C. Weng, J. Am. Chem. Soc., 2015, 137, 1305–1313.
54. Y. Gong, Z. Yang, Y. Lin, J. Wang, H. Pan and Z. Xu, J. Mater. Chem. A, 2018, 6, 16950–16958.
55. L. An, J. Feng, Y. Zhang, R. Wang, H. Liu, G.-C. Wang, F. Cheng and P. Xi, Adv. Funct. Mater., 2019, 29, 1805298.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1ce01309b

---