Prussian Blue Analogues Derived Hollow FeCoP Nanocubes for Electrocatalytic Overall Water Splitting

Abstract

Cleverly designing and synthesizing bifunctional electrocatalysts with high activity and exceptional stability for hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) hold profound importance in the realm of renewable energy technologies. Here we demonstrated the fabrication of FeCoP hollow nanocubes by precisely controlling the oxidation-phosphorization processing on Prussian blue analogues. Owing to the robust electronic interaction and hollow structure, the obtained FeCoP material, when utilized for catalyzing HER and OER in a 1 M KOH solution, requires overpotentials of merely 74 mV and 219 mV to achieve 10 mA cm⁻², respectively. Importantly, when used as a bifunctional catalyst for overall water splitting, FeCoP can achieve a current density of 10 mA cm⁻² at a low cell voltage of only 1.58 V and exhibits impressive durability. After a prolonged test of 52 hours under a constant current density, there is no significant degradation performance decay. The current method offers a broader avenue for the controllable synthesis of phosphide electrocatalysts in practical applications.

1. Introduction

As a clean and efficient energy source, hydrogen has been widely recognized as an ideal candidate to replace fossil fuels and alleviate the increasingly serious energy crisis and environmental pollution.^1,2 Electrolysis of water is generally regarded as one of the most promising methods for hydrogen production, thanks to its low energy consumption, high stability, and environmental friendliness.^3,4 However, the kinetic processes of hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) are relatively slow, reducing the efficiency of water decomposition, highlighting the urgent need to develop efficient electrocatalysts to minimize the cost of hydrogen energy. Noble metal-based materials, such as Pt, IrO_x, and RuO_x, are typically excellent catalysts for HER or OER.^5,6 Nevertheless, the limited availability and high cost of those materials always stand in the way of their industrialization. Therefore, it is urgent to develop efficient electrocatalysts that are abundant on Earth to achieve low overpotential and long-term durability in overall water splitting technology.

Until now, significant progress have been acquired in developing HER and OER electrocatalysts based earth-abundant elements.^7,8 Among them, transition metal phosphides (TMPs) have been intensively researched due to their exceptional electrochemical stability, intermediate adsorption energy that is favourable for catalysis, and minimal production expenses, regarded as viable substitutes for catalysts based on noble metals for HER or OER.^9,10 To optimize the electrochemical performance of TMPs to further adopt to the practical requirements, the transformation of monometallic into heterometallic phosphides, along with the accomplishment of secondary metal doping, has been theoretically and experimentally proven to be an effective strategy.¹¹ Doped metal atoms can induces electron rearrangement which generally leads to moderate changes in the adsorption energy of reaction intermediates leading to higher catalytic activity of bimetallic phosphides. For instance, Zhang et al. reported that Fe-doped CoP nanoframes have batter performance toward water splitting than pristine CoP.¹² Wang et al. reported that mixtures of Ni and Fe phosphide alloys have batter performance toward water splitting than pristine FeP.¹³

The microstructure and surface morphology attributes are also play a pivotal role in defining the intrinsic properties of the catalytic materials.¹⁴ Various TMPs designed with one-dimensional (1D) particles,¹⁰ 2D nanosheets¹¹ and 3D nanocubes structure¹² have been reported for enhanced catalytic applications, particularly through the development of three-dimensional porous and hollow nanocubes structures. These materials boast a substantial surface area and abundant reaction sites. Additionally, their frameworks are open, which is conducive to improved reaction kinetics and overall stability.^15,16 However, the synthesis of metal phosphides with three-dimensional porous morphology using traditional chemical methods often requires complex and tedious procedures. Prussian blue analogues (PBAs, denoted as M^3-II[M^III(CN)₆]₂·nH₂O) include transition metals (i.e., Co, Ni, Fe) as metallic nodes and CN assumed the role of linkers,¹⁷ are family of well-organized metal-organic frameworks (MOFs). Due to its diverse chemical composition and microstructure, as well as the simplicity of the preparation process, PBAs have emerged as an ideal precursor for constructing non-noble metal-based electrocatalysts with exquisite structures.¹⁸ Unfortunately, most of the active materials derived from MOFs are in the shape of powders, and the usage of polymer binders will increase the “dead mass” for the electrodes and block the active sites.^19,20 Thus, developing MOF-derived functional nanostructures on flexible current collectors, granting them a vast surface area and highly accessible active sites for promoting charge transfer, would be an attractive research direction.

Inspired by the aforementioned considerations, herein, we designed a self-supported FeCoP hollow nanostructure on carbon cloth (CC) through a controlled oxidation-phosphorization processing derived from Fe-Co PBAs precursor. The as-prepared FeCoP materials retains their distinct structure with a hollow nanocubic. Additionally, the incorporation of cobalt enhanced the electron transport and subsequently reduced the charge transfer resistance. Furthermore, the flexible and conductive carbon cloth, which is fabric tightly connected with the FeCoP nanocubes can be directly utilized as an electrode without the need for any binder additives, thereby guaranteeing high mechanical stability and conductivity. Benefiting from its distinctive structure and chemical composition, the resulting FeCoP nanocubes exhibits exceptional activity for the OER and HER in 1 M KOH, and possesses excellent long-term stability. Notably, when used in a dual-electrode alkaline electrolyzer, FeCoP nanocubes require only 1.58 V to achieve the desired current density of 10 mA cm⁻² and exhibit remarkable durability for at least 50 h. This work underscores the significant advancements in synthesizing highly active catalysts for achieving remarkable overall water decomposition performance.

2. Experimental

2.1 Materials and chemicals

All the reagents and solvents were commercially available and used without further purification. Sodium citrate (Na₃C₆H₅O₇·2H₂O) and cobalt nitrate hexahydrate (Co(NO₃)₂·6H₂O) were provided by Tianjin Fuchen Chemical Reagent Technologies Co. Ltd. Potassium hexacyanoferrate (III) (K₃[Fe(CN)₆]) were purchased from Sinopharm Chemical Reagent Co. Ltd. Sodium hypophosphite (NaH₂PO₂·H₂O) was purchased from Aladdin Reagent. Carbon cloth (CC) was provided by Shanghai Hesen Corp. Commercial IrO₂ and Commercial Pt/C (20 % Pt on Vulcan XC-72) were purchased from Alfa Aesar. Deionized water (from MilliQ system) with a resistivity of 18 MΩ cm was used in all experiments.

2.2 Synthesis of Fe-Co PBAs precursors on carbon cloth

According to the literature, Co-Fe PBAs precursors on carbon cloth was synthesized via a co-precipitation method with some modification. Specifically, 1.32 g (4.0 mmol) of K₃Fe(CN)₆ was dissolved in 200 mL of deionized water to obtain yellow solution A. Meanwhile, 1.75 g (6.0 mmol) of Co(NO₃)₂·6H₂O and 2.65 g (9.0 mmol) of sodium citrate were completely dissolved in another 200 mL of deionized water to obtain solution B. Add solution A to solution B at a steady flow rate under continuous magnetic stirring. After that, a pre-treated carbon cloth (2 × 1 cm²) was immersed into the mixture solution. After stirring magnetically for 5 min, let it age at room temperature for 24 hours. Remove the carbon cloth, rinsed it sequentially with deionized water and absolute ethanol several times, and dry at 60 °C for 12 hours under vacuum.

2.3 Synthesis of FeCoO_x composites on carbon cloth

The carbon cloth loaded with Fe-Co PBAs nanocubes were directly loaded into a muffle furnace. Firstly, heat the muffle furnace at a rate of 0.5 °C min⁻¹ to 270 °C and maintain this temperature for 1 hour. Subsequently, increase the temperature to 350 °C at a rate of 20 °C min⁻¹ and hold it there for 0.5 hours. After the muffle furnace naturally cools down to room temperature, the intermediate FeCoO_x composites on carbon cloth was obtained.

2.4 Synthesis of FeCoP composites on carbon cloth

Place the as-prepared carbon cloth loaded with FeCoO_x composites and 1.5 g of NaH₂PO₂·H₂O separately into two individual alumina crucibles. Place these crucibles into a tube furnace, with the crucible containing NaH₂PO₂·H₂O positioned upstream of the gas flow. Subsequently, heat the tube furnace at a rate of 2 °C min⁻¹ to 300 °C and maintain this temperature under an Ar atmosphere for 2 hours. After it naturally cools down to room temperature, remove the carbon cloth carrying the sample, rinse it thoroughly with ethanol and deionized water several times, and finally dry it under vacuum at 60 °C for 12 hours. The loading amount of the catalyst was determined by weighing the carbon cloth before and after catalyst deposition using a high-precision analytical balance. In this experiment, the loading amount of FeCoP nanocube composite material on the carbon cloth was approximately 0.16 mg cm⁻². For comparison, we synthesized FeP and CoP samples by directly phosphorizing PB and Co-Co PBAs nanocubes that were loaded onto the carbon cloth.

2.5 Characterizations

Powder X-ray diffraction (PXRD) data were recorded on a Rigaku D/MAXRC X-ray diffractometer (45.0 kV, 50.0 mA). The target material employed is Cu (Kα), with a voltage of 40 kV and a current of 30 mA. Scanning electron microscopic (SEM) and transmission electron microscopic (TEM) images were obtained on a HELIOS NanoLab 600i and a Tecnai F20, both manufactured by FEI Company, USA. The Raman spectroscopy measurements were conducted using a laser-based confocal micro-Raman spectrometer, model InVia, manufactured by Renishaw (UK). During the measurements, a 532 nm laser source was selected, with a laser power set at 0.1 mW and an exposure time of 10 seconds. X-ray photoelectron spectra (XPS) were obtained utilized a PHI-5400 ESCA model instrument manufactured by PerkinElmer (USA). The instrument employs an AlKα radiation source with an energy of 1486.6 eV.

2.6 Electrochemical measurements

All the electrochemical performance tests were evaluated using a standard three-electrode system. The electrochemical workstation used was manufactured by Shanghai Chenhua Instruments Co., Ltd., with a model number of CHI-660E. Where the desired samples on carbon cloth served as the working electrode (WE), a graphite rod as the counter electrode and Hg/HgO electrode as the reference electrode. Commercial Pt/C and IrO₂ ink was prepared by dispersing 4 mg IrO₂ powder into a water-ethanol solution (v/v, 3 : 1) containing 5 wt% Nafion for at least 1 h to form a homogeneous catalyst ink. Subsequently, the catalyst ink was coated onto carbon cloth with a Pt/C or IrO₂ loading mass density of approximately 0.16 mg cm⁻² and dried in air at room temperature.

The linear sweep voltammetry (LSV) was conducted at room temperature with a scan rate of 5 mV s⁻¹. Cyclic voltammetry (CV) diagrams at different scan rates (20, 40, 60, 80, 100, 120, 140, 160, 180, and 200 mV s⁻¹) within the potential range of 0.083–0.183 V were collected and used to estimate the double-layer capacitance (C_dl). Electrochemical impedance spectroscopy (EIS) measurements were performed within a frequency range of 10⁶ to 0.1 Hz. A multi-current step procedure in 1 M KOH was employed to test the mass transfer performance. The stability of the catalysts was measured using cyclic voltammetry (CV) and chronoamperometry.

3. Results and Discussion

3.1 Physical characterizations

The FeCoP hollow nanocubes supported on conductive carbon cloth (CC) were synthesized by oxidation and subsequent phosphorization process of the Fe-Co PBAs precursor, as depicted in Scheme 1.

Scheme 1.

Schematic representation of the synthesis of FeCoP hollow nanocubes.

Initiating with a straightforward chemical precipitation method, Fe-Co PBAs nanocubes were uniformly deposited onto the CC surface. The morphological and crystallographic characteristics of the precursor were meticulously analyzed using scanning electron microscopy (SEM), transmission electron microscopy (TEM), and X-ray diffraction (XRD) patterns. In contrast to the smooth carbon fibers of the pristine CC (Fig. 1a), Figs. 1b and 1d clearly demonstrate that the Fe-Co PBAs nanocubes with smooth surfaces, uniform dimensions, and particle size of approximately 200 nm were grown evenly on the surface of carbon cloth.

Figure 1.

(a) SEM images of blank carbon cloth. (b, c) SEM and (d) TEM image of Fe-Co PBAs nanocubes on carbon cloth.

The XRD pattern (Figs. 2a and 2c) reveals that excluding the peak from the characteristic peaks originating from the CC substrate, all the diffraction signals of the as-prepared Fe-Co PBAs were in excellent agreement with the standard pattern of Co₃[Fe(CN)₆]₂·10H₂O (JCPDS No. 46-0907)²¹ with face-center-cubic structure, and devoid of any impurities (Fig. 2a). The obtained Fe-Co PBAs nanocubes, serving as precursors, were transformed into hollow FeCoP nanocubes through oxidation and subsequent phosphorization processes. Thermal annealing of these Fe-Co PBAs precursor in air can result in the formation of FeCoO_x. The corresponding XRD pattern (Fig. 2b) of as-prepared nanocomposites show that a well-defined peaks ascribed to Co₃O₄²² (JCPDS No. 43-1003) and Fe₂O₃²³ (JCPDS No. 25-1402). After further low temperature (300 °C) phosphorization, the corresponding metal phosphides were successfully synthesized. The XRD pattern (Fig. 2c) confirms the presence of Co₂P²⁴ (JCPDS No. 32-0306), CoP (JCPDS No. 29-0497) and FeP₄²⁵ (JCPDS No. 34-0995). This indicates that after continuous oxidation and low-temperature phosphating, FeCoO_x nanocomposites were phosphatized to FeCoP. Moreover, Energy Dispersive Spectrometer (EDX) techniques were utilized to determine the chemical compositions of various composites. As shown in Figs. 2d and 2f, EDS spectra indeed confirm that the Fe-Co PBAs, FeCoO_x and FeCoP on carbon cloth have been successfully prepared. For comparison, pure FeP and CoP nanocubes also were prepared by the similar oxidation-phosphorization processing of a Prussian blue (PB) and Co-Co PBA precursors (Fig. 3).

Figure 2.

XRD patterns and EDS spectra of (a, d) Fe-Co PBAs, (b, e) FeCoO_x and (c, f) FeCoP on carbon cloth.

Figure 3.

XRD patterns of FeP and CoP on carbon cloth.

X-ray Photoelectron Spectroscopy (XPS) tests were further carried out to investigate the chemical valence states and surface composition of FeCoP. The high-resolution Co 2p spectrum can be fitted with two spin-orbit doublets at 778.4 eV and 793.8 eV for the Co-P bond, and then 783.8 and 799.5 eV were attributed to the oxidized Co species (Co²⁺) resulting from the surface oxidation (Fig. 4a).^26,27 Fe 2p XPS spectra in Fig. 4b showed two peaks at 707.2 eV and 720.4 eV, which can be assigned to the typical 2p_3/2 and 2p_1/2 peaks in Fe-P, respectively, and another peak at 710.7 eV and 724.6 eV were ascribed to the oxidized Fe on the surface.^28,29 The P 2p spectra can be separated into three peaks at 130.4, 133.8 and 134.5 eV, attributed to P 2p_1/2 of the metal phosphides and oxidized P, respectively (Fig. 4c).³⁰ All of these results indicate that FeCoP was successfully synthesized. Meanwhile, both Co and Fe possess relatively small positive charges, while P has a relatively small negative charge, indicating that electrons transfer from Co and Fe to P.³¹ Consequently, the FeCoP electrocatalyst exhibits excellent electron transfer properties.

Figure 4.

High resolution XPS spectra of (a) Co 2p, (b) Fe 2p and (c) P 2p levels of FeCoP.

To investigate the evolution of morphology and structure during the synthesis process of these samples, scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were employed. After oxidation-phosphorization processing, the FeCoO_x, FeCoP, FeP and CoP nanocomposites retained their robust growth on CC and maintained the overall nanocubes structure of the precursor with a slightly roughened surface (Fig. 5). This geometric confinement of nanoparticles within CC was anticipated to enhance interfacial interaction, mitigate structural disintegration and clustering, which could enhance catalytic performance and robustness.^32,33 Of note was that the particle size of FeP and CoP nanocomposites was much larger than that of FeCoP nanocomposites. TEM observation reveals that FeCoO_x sample was consist of numerous Co₃O₄ and Fe₂O₃ nanoparticles ranging from 5–20 nm with abundant pores (Figs. 6a and 6b). The apparent particle size decreased due to the reconstruction of the framework following the oxidation process. The lattice fringes of 2.51 Å and 2.46 Å observed in the high-resolution transmission electron microscopy (HRTEM) images correspond to the (110) and (311) facet of Fe₂O₃ and Co₃O₄, respectively (Fig. 6c).^34,35 All the results mentioned above indicate that FeCoO_x nanocubes have been successfully prepared.

Figure 5.

SEM images of FeCoO_x (a, b), FeCoP (c, d), FeP (e, f) and CoP (g, h) nanocubes on carbon cloth.

Figure 6.

TEM (a, b) and HRTEM images (c) of FeCoO_x nanocubes.

TEM images (Figs. 7a and 7b) of FeCoP indicate that FeCoP nanocomposites maintains a nanocube structure which were similar to FeCoO_x, but the entire nanocube becomes a hollow structure. The clear lattice stripes in HRTEM images (Fig. 7c) correspond to the Co₂P (211) plane, CoP (211) plane, FeP₄ (-132) and (-112) plane, with lattice fringe spacings of 0.208, 0.180, 0.253, and 0.297 nm, respectively.³⁶ The selected area electron diffraction (SAED) pattern reveals diffraction concentric rings that correspond to the (011) planes of CoP, (130) planes of Co₂P and the (-114) planes of FeP₄ (inset of Fig. 7c). Additionally, the elemental mapping results (Figs. 7d–7g) confirm the homogeneous distribution of Co, Fe and P within the FeCoP. Notably, the presence of P elements outside the cube validated the hollow structure of FeCoP nanocomposites.

Figure 7.

(a, b) TEM images, (c) HRTEM images (inset: SAED pattern), (d) STEM image and (e–g) elemental distribution mapping of FeCoP nanocubes.

Based on the analysis above, the formation mechanism of FeCoP hollow nanocubes was proposed. First, the FeCoO_x nanoparticles was formed by continuous oxidation FeCo-PBA nanocubes in air. Next, NaH₂PO₂·H₂O was used as the phosphorous sources to modify the FeCoO_x intermediate. PH₃ was generated in-situ upon the decomposition of hypophosphite and then adsorbed on the surface of FeCoO_x. The FeCoO_x nanoparticles can react with PH₃ to form FeCo phosphide during the thermal treatment process.³⁷ The hollow structure can be counted for a “reaction and diffusion” controlled process based on the Kirkendall effect.^38,39 The Fe, Co and O has a tendency to diffuse outward to react with PH₃ to form the FeCo phosphide. However, due to an imbalance of ion exchange, the inward diffusion rate of the Fe, Co and O was faster than the inward transport rate of PH₃, the FeCoP shell will therefore increase, and the FeCoO_x intermediate core will decrease gradually.^40,41 With increasing reaction time, thus resulting in the hollow structure at the centre.

3.2 Electrochemical performance

The HER activity of as-synthesized catalysts was initially evaluated in 1 M KOH solution. A commercial available Pt/C electrode with a mass loading of approximately ∼0.16 mg cm⁻² on CC was used as a control. Figure 8a presents the polarization curve after iR correction. It was evident that the Fe-Co PBAs precursor exhibits virtually no catalytic activity towards HER. As expected, compared with FeCoO_x, FeP and CoP, the FeCoP catalyst displays the most superior HER catalytic activity, equiring an overpotential of only 74 mV vs. RHE at a current density of 10 mA cm⁻² (η₁₀), indicating its efficiency as a cathode for hydrogen production from water. These results underscore the synergistic interactions among the different components of CoP, Co₂P and FeP₄.

Figure 8.

Electrochemical properties of the Fe-Co PBAs, FeCoO_x, FeP, CoP and FeCoP for HER in 1 M KOH. (a) iR-corrected LSV curves. (b) Tafel plots. (c) The current density vs. the scan rates of the corresponding samples. (d) Nyquist plots (inset is the equivalent circuit).

For HER in alkaline solution, water reduction was described by the following three steps:⁴²

  

\begin{equation} \text{(Volmer step)}\qquad \text{H$_{2}$O} + \text{e$^{-}$} + \text{M} \to \text{M} - \text{H} + \text{OH$^{-}$} \end{equation} (1)

  

\begin{equation} \text{(Heyrovsky step)}\quad \text{H$_{2}$O} + \text{e$^{-}$} + \text{M} - \text{H} \to \text{H$_{2}$} + \text{OH$^{-}$} + \text{M} \end{equation} (2)

  

\begin{equation} \text{(Tafel step)}\quad\qquad \text{2M} - \text{H} \to \text{H$_{2}$} + \text{2M} \end{equation} (3)

where M denotes the surface empty site.

As previously reported, Tafel slopes of 120, 40, and 30 mV dec⁻¹ were observed for Volmer, Heyrovsky, and Tafel determining steps, respectively.⁴³

The Tafel slopes of the as-prepared samples were studied to elucidate the reaction kinetics (Fig. 8b). The Tafel slope value of FeCoP (74 mV dec⁻¹) was smaller than those of FeCoO_x (85 mV dec⁻¹), CoP (102 mV dec⁻¹), and FeP (121 mV dec⁻¹) electrodes, indicating that the FeCoP electrocatalysts exhibit higher intrinsic activity compared to other samples. Additionally, the Tafel slope value falling within the range of 40–120 mV dec⁻¹, suggests that the hydrogen evolution reaction occurring on the FeCoP surface follows the Volmer-Heyrovsky mechanism, with the electrochemical desorption process being the rate-limiting step of the reaction. This appreciable HER catalytic activity can be attributed to the structural features of FeCoP, where the FeCoP nanocubes was loaded on a conductive carbon cloth enable the high electronegative P atom can attract electron from metal atom and become a negatively charged center (σ⁻), which can act as absorption sites for positively charged H protons while metal centers (σ⁺) act as a hydride acceptor.³¹ Additionally, the unique hollow nanocubes structure has a higher wettability, which could enhance the electrolyte penetration and augment the contact degree between reactants and active sites, and thus facilitate the HER kinetics.

The electrochemical double-layer capacitance (C_dl) of each catalyst was primarily utilized to estimate its electrochemically active surface area (ECSA). Cyclic voltammetry (CV) tests were performed at varying scan rates within the non-Faradaic potential window of 0.083 to 0.183 V (vs. RHE), and the C_dl values of the catalysts were then calculated from the CV curves at a common potential (0.133 V) (Fig. 9). As depicted in Fig. 8c, the C_dl values for FeCoO_x, FeP, CoP and FeCoP were determined to be 2.7 mF cm⁻², 1.6 mF cm⁻², 5.2 mF cm⁻² and 12.9 mF cm⁻² respectively. This indicates that the prepared FeCoP possesses a large electrochemically active surface area, attributed to its unique hollow nanocube structure and the interfaces formed between its various components, which collectively contribute to an abundance of active sites. This was one of the reasons for its superior HER activity.⁴⁴ Building upon this foundation, to further investigate the electrode kinetics during HER catalysis for each catalyst, EIS (Electrochemical Impedance Spectroscopy) measurements were conducted. The Nyquist plots (Fig. 8d) were fitted using an equivalent circuit (inset in Fig. 8d). The results reveal that the charge transfer resistances (R_ct) of FeCoO_x, FeP, CoP and FeCoP hollow nanocubes decrease sequentially (37.3 > 13.7 > 4.2 > 2.5 Ω). This result reinforces the following observation that FeCoP exhibits a favorable electron transfer rate at the catalyst-electrolyte interface. This was attributed to the vectorial electron transport in FeCoP hollow nanocubes and the excellent electrical connection between the various components (Co₂P, CoP and FeP₄) of FeCoP as well as conductive carbon cloth substrates.

Figure 9.

CV curves of the Fe-Co PBAs, FeCoO_x, FeP, CoP and FeCoP samples under different scan rates in the region of 0.083–0.183 V. These data were used to present the plots showing the extraction of the C_dl for different samples shown in Fig. 8c in the main text.

The good stability and durability of the electrocatalyst were also of great significance for practical application. First, the stability of FeCoP catalyst was evaluated by multi-step chronopotentiometric measurements. As shown in Fig. 10a, the current density of the sample was changed from 25 to 175 mA cm⁻² with an increase of one step 25 mA cm⁻² per 500 s⁻¹ and the change of potential during this process was recorded (without iR correction). The initial potential was approximately −0.13 V, and the voltage remained basically constant after 500 s. Similar phenomena were also observed for the subsequent steps. This indicates that FeCoP catalyst has excellent mass transfer capabilities and stability. To further explore the durability of FeCoP, a chronoamperometry test was carried out on the sample at a constant potential of 76 mV in 1 mol/L KOH (Fig. 10b), and the current density remained stable at approximately 10 mA cm⁻². This excellent stability may be attributed to the strong adhesion between FeCoP and the conductive substrate. In addition, no noticeable change in the size and morphology of FeCoP could be seen in SEM images after the HER test, confirming its robustness as a highly active HER electrocatalyst (Fig. 11).

Figure 10.

(a) Multi-step chronopotentiometric curve obtained with FeCoP electrode (without iR-correction). (b) The chronoamperometric responses of FeCoP at overpotential of 76 mV vs. RHE.

Figure 11.

SEM images of FeCoP nanocomposites after i-t test.

The OER performance of each catalyst was further tested in 1 M KOH. Similarly, an ink prepared from commercial IrO₂ powder was drop-casted onto CC (with a similar mass loading of IrO₂ at approximately 0.16 mg cm⁻²) for comparative testing. The OER polarization curves (Fig. 12a) reveal that compared to FeCoO_x (310 mV vs. RHE), FeP (390 mV vs. RHE) and CoP (410 mV vs. RHE), only 219 mV vs. RHE overpotential of FeCoP was required to drive a current density of 10 mA cm⁻². This indicates that FeCoP also exhibits the best performance for oxygen evolution. In addition, the Tafel slopes of each catalyst calculated from the LSV curve also confirm this. The OER was generally associated with O-H bond breaking and attendant O-O bond formation, which undergoes a proton-coupled electron transfer at the high equivalency of 4⁴⁵

  

\begin{equation} \text{M} + \text{OH$^{-}$} \to \text{MOH} + \text{e$^{-}$} \end{equation} (4)

  

\begin{equation} \text{MOH} + \text{OH$^{-}$} \to \text{MO} + \text{H$_{2}$O} + \text{e$^{-}$} \end{equation} (5)

  

\begin{equation} \text{MO} + \text{OH$^{-}$} \to \text{MOOH} + \text{e$^{-}$} \end{equation} (6)

  

\begin{equation} \text{MOOH} + \text{OH$^{-}$} \to \text{M} + \text{O$_{2}$} + \text{H$_{2}$O} + \text{e$^{-}$} \end{equation} (7)

where M denotes the surface empty site.

Figure 12.

Electrochemical catalytic performance (a) LSV curves after iR correction of the Fe-Co PBAs, FeCoO_x, FeP, CoP, FeCoP and IrO₂@CC catalysts for OER in 1 M KOH. (b) Tafel plots. (c) Electrochemical impedance spectroscopy (the equivalent circuit used to simulate the Nyquist plots was shown as inset). (d) Comparison of LSV curves of FeCoP at the 1st and 1000th cycle. Inset was stability of FeCoP for OER under a static overpotential of 224 mV vs. RHE over 12 h.

In our case, as shown in Fig. 12b, compared to FeCoO_x (42 mV dec⁻¹), FeP (80 mV dec⁻¹), CoP (47 mV dec⁻¹) and RuO₂ (217 mV dec⁻¹), FeCoP exhibited the smallest Tafel slope, only 31.2 mV dec⁻¹, suggesting that Reaction (2) was the rate-determining step.⁴² The phosphorus possesses lone-pair electrons in 3p orbitals and vacant 3d orbitals, which were utilized to accommodate electrons on the surface and increase local charge density.⁴⁶ As previously reported, during the oxygen evolution reaction (OER) testing, the surfaces of this metal phosphides undergoes self-oxidation at positive potentials, and the formation of PO₄³⁻ may result in a more active state to accelerate the kinetics of oxygen evolution.⁴⁷ Simultaneously, coupled Fe and Co generate hydroxyl ions. MOOH was an effective species for reaction (3),⁴⁸ which can further react with hydroxyl groups to produce oxygen (reaction (4)). For this reason, the FeCoP exhibits the fastest reaction kinetics during the OER process. Additionally, EIS testing (Fig. 12c) was used to further reveal the intrinsic activity of the catalyst. FeCoP has the smallest charge transfer resistance, further illustrates that FeCoP has the most effective charge transfer pathway and faster charge transfer rate in the catalytic OER process. This could be attributed to the directly synthesized self-supported electrode, which provides a rapid electron transport and gas diffusion channel, while the intimate contact between the support and catalyst accelerates the charge flow from the substrate to the metal components. Similarly, the FeCoP catalyst also demonstrated exceptional stability for OER. After 1000 consecutive cyclic voltammetry (CV) cycles, the polarization curve of FeCoP exhibited only minor fluctuations (Fig. 12d). Furthermore, the chronoamperometric studies revealed negligible decay in the initial current density (10 mA cm⁻²) after 12 h of testing (inset of Fig. 12d). These results collectively demonstrate the outstanding OER stability of FeCoP.

As is well known, during the OER testing process, the surfaces of most metal phosphides undergo self-oxidation and transform into oxidation intermediates (MPOx and MOx/M(OOH)x) at positive potentials, serving as actual surface active sites.^49,50 Therefore, to investigate the intrinsic active species of FeCoP, the composition and structure of the FeCoP catalyst were characterized following OER testing. Initially, we compared the Raman spectra of FeCoP before and after OER testing. As shown in Fig. 13, newly diffraction peaks at 477, 552, and 683 cm⁻¹ after OER testing can be attributed to Fe(Co)OOH,^51,52 indicating that the composition of the catalyst has indeed changed. The corresponding SEM images reveal that the nanocube morphology of FeCoP did not undergo significant changes and still grew well on the CC surface, but some minor cracks were caused by the continuous release of oxygen bubbles during the OER process (Figs. 14a and 14b). The TEM images in Fig. 14c also show well maintained nanocubes structure of the catalyst. An HRTEM study shows that the lattice fringes of FeCoP after the OER process can be assigned to the (101) and (110) crystal planes of Fe(Co)OOH, respectively. In addition, the selected-area electron diffraction (SAED) pattern of FeCoP gives clear diffraction spots of (110) and (100) planes of Fe(Co)OOH (Fig. 14d),⁵³ which demonstrates that the surface of the catalyst was oxidized after OER. Moreover, XPS was performed to explore the composition of FeCoP after OER. As shown in Fig. 15a, in the high resolution Co 2p XPS spectrum, peaks at 780.3 and 795.6 eV were related to Co³⁺, indicating the formation of cobalt oxides/oxyhydroxides that were electrochemical oxidation-formed at the surface.⁵⁴ With respect to Fe 2p (Fig. 15b), the signal from Fe-P almost disappeared, indicating the presence of Fe³⁺.⁵⁵ The peaks of P 2p also get disappeared, and the peaks at 531.91 and 531.12 eV in the O 1s spectrum were attributed to the M-O and M-OH, respectively (Fig. 15d), implying the formation of the Fe-Co (oxy)hydroxide species on the surface of the catalyst under the OER conditions.⁵⁶ Therefore, these results confirms that the FeCoP catalyst was oxidized to the Co(Fe) oxyhydroxide species, which may be the active species to boost the OER.⁵⁷

Figure 13.

Raman patterns of as-prepared FeCoP after OER test.

Figure 14.

SEM images (a, b), TEM images (c) and HRTEM images of FeCoP after OER test. The inset in (d) is the SAED pattern.

Figure 15.

High resolution XPS spectra of (a) Co 2p, (b) Fe 2p, (c) P 2p and (d) O 1s levels of FeCoP nanocomposites after OER measurement.

Based on the above results, it can be concluded that the FeCoP catalyst exhibits excellent catalytic activity and durability for both HER and OER. Therefore, the synthesized FeCoP catalyst was utilized as both the cathode and anode to assemble a two-electrode electrolytic cell to evaluate its feasibility as a bifunctional catalyst for overall water splitting. As shown in Fig. 16a, compared to the commercial IrO₂/CC // Pt/C/CC couple, the FeCoP catalyst required only 1.58 V to achieve a current density of 10 mA cm⁻². Meanwhile, during overall water splitting, continuous and stable H₂ and O₂ evolution could be clearly observed at the cathode and anode, respectively (inset in Fig. 16a). For comparison, CoP and FeP bifunctional catalysts were also studied. As expected, FeCoP still exhibited the most superior catalytic activity. Furthermore, FeCoP was also better than most reported non-noble-metal bifunctional electrocatalysts for alkaline water splitting as shown in Table 1. A chronopotentiometric test was employed to evaluate the long-term stability of FeCoP for overall water splitting. After continuous operation for 52 h, the current density decline of the FeCoP sample was negligible (Fig. 16b). The outstanding catalytic activity and durability of this catalyst make it a potential electrode material for practical applications in electrocatalytic water splitting.

Figure 16.

(a) LSV curves of FeCoP, CoP, FeP and IrO₂/CC//Pt/C/CC in two-electrode. (b) Long-time stability test of the FeCoP electrode at a constant overpotential for 52 h.

Table 1. Comparison of the electrochemical performances of FeCoP nanocubes for overall water splitting in 1.0 M KOH with recently reported bifunctional electrocatalysts.

Catalyst	Current Density (mA cm−2)	η (V vs. RHE)	Ref.
FeCoP nanocubes	10	1.58	In this work
(Co0.52Fe0.48)2P	10	1.53	58
NiCoP/Ni foam	10	1.58	59
NiCoP/rGO	10	1.59	60
Co-Ni-P nanowires/Ni foam	10	1.57	61
NiFePX@NiCo2PX	10	1.56	62
Fe-CoP	10	1.6	63
NiFe/NiCo2O4/NF	10	1.67	64
Co4Ni1P NTs	10	1.59	65
Ni-Fe-P nanocubes	10	1.68	66
FeMnP films	10	1.55	67
NiFeOx/NiFe-P	30	1.73	68
Ni1Mo1P NSs@MCNTs	10	1.601	69
CoP/NCNHP	10	1.64	70
Cu0.08Co0.92P NAs/CP	10	1.72	71
(NixFey)2P	10	1.61	72
FeCoP alloy	10	1.68	73

4. Conclusions

In summary, we successfully synthesized FeCoP hollow nanocubes on carbon cloth substrate through a controlled oxidation-phosphorization processing derived from Fe-Co PBAs precursor. These FeCoP electrode demonstrate remarkable electrocatalytic performance and stability towards both HER and OER, achieving a current density of 10 mA cm⁻² in alkaline electrolytes at mere overpotentials of 74 mV and 219 mV, respectively. The efficient catalytic activity of FeCoP for HER and OER can be attributed to the in-situ grown 3D hollow nanocube structure on CC, which could afford abundant active sites, facilitates efficient charge transfer, and ensures good mass transport. Additionally, the synergistic effect between the different components within the phosphide, coupled with the high conductivity of the CC substrate, contributes significantly to the outstanding HER and OER performance of FeCoP. Furthermore, when FeCoP electrode were assembled into a two-electrode electrolytic cell for overall water splitting, it can achieve a current density of 10 mA cm⁻² at a low battery voltage of 1.58 V. Given its excellent catalytic activity, stability, and ease of preparation, FeCoP holds great potential as a noble-metal-free electrode material for industrial-scale water electrolysis.

Acknowledgments

The authors thank the Dr. Liu Hanghang, College of Biological and Pharmaceutical Science, China Three Gorges University, for supporting this work.

CRediT Authorship Contribution Statement

Yuzhi Li: Writing – original draft (Lead)

Jinyu Wang: Investigation (Equal), Methodology (Equal)

Shuai Zou: Investigation (Equal), Methodology (Equal)

Aihong Wang: Investigation (Equal), Methodology (Equal)

Huize Jiang: Data curation (Equal), Validation (Equal)

Kaibo Wang: Conceptualization (Lead), Writing – original draft (Equal)

Conflicts of Interest

The authors declare no conflicts of interest.

Footnotes

K. Wang: Present address: Chemical General Factory Nanxing Hubei, No. 64 Jianghan Avenue, ZhiJiang City, Hubei Province, China

References

• 1)  C. W. Wang, Y. Dai, Z. Zhao, A. Wang, T. Zhang, and Z. L. Wang, Nat. Commun., 8, 88 (2017).
• 2)  N. S. Lewis, Proc. Natl. Acad. Sci. U.S.A., 103, 15729 (2006).
• 3)  S. E. Hosseini and M. A. Wahid, Renewable Sustainable Energy Rev., 57, 850 (2016).
• 4)  L. Barreto and K. Riahi, Int. J. Hydrogen Energy, 28, 267 (2003).
• 5)  N. Cheng, S. Stambula, D. Wang, M. N. Banis, J. Liu, A. Riese, B. Xiao, R. Li, T. K. Sham, L. M. Liu, G. A. Botton, and X. Sun, Nat. Commun., 7, 13638 (2016).
• 6)  L.-W. Chen and H.-W. Liang, Catal. Sci. Technol., 11, 4673 (2021).
• 7)  L. Sun, Q. Luo, Z. Dai, and F. Ma, Coord. Chem. Rev., 444, 214049 (2021).
• 8)  S. Y. Tee, K. Y. Win, W. S. Teo, L. D. Koh, S. Liu, C. P. Teng, and M. Y. Han, Adv. Sci., 4, 1600337 (2017).
• 9)  L. Yan, J. Zhu, Y. Li, P. Tsiakaras, and S. P. Kang, Appl. Catal. B, 265, 118555 (2020).
• 10)  J. Wang, W. Li, X. Chen, and A. Huang, J. Alloys Compd., 934, 167828 (2023).
• 11)  C.-J. Huang, H.-M. Xu, T.-Y. Shuai, Q.-N. Zhan, Z.-J. Zhang, and G.-R. Li, Appl. Catal. B, 325, 122313 (2023).
• 12)  Y. Yuan, K. Wang, B. Zhong, D. Yu, F. Ye, J. Liu, J. Dutta, and P. Zhang, Energy Environ. Mater., e12747 (2024).
• 13)  H. W. Huang, C. T. Zhao, X. T. Han, J. Yang, Z. B. Liu, S. F. Li, M. D. Zhang, and J. S. Qiu, Nano Energy, 34, 472 (2017).
• 14)  S. Yang, X. N. Chen, M. J. Huang, S. H. Cai, J. Y. Han, and J. S. Li, Appl. Catal. B, 304, 120914 (2022).
• 15)  Y. T. Li, J. Y. Xie, Y. Y. Mao, and W. B. Hu, J. Colloid Interface Sci., 650, 1113 (2023).
• 16)  G. Ye, Y. Sun, X. Zhou, K. Zhu, J. Zhou, and M.-O. Coppens, Chem. Eng. J., 329, 56 (2017).
• 17)  M. B. Zakaria and T. Chikyow, Coord. Chem. Rev., 352, 328 (2017).
• 18)  L.-M. Cao, D. Lu, D.-C. Zhong, and T.-B. Lu, Coord. Chem. Rev., 407, 213156 (2020).
• 19)  C. Guan, X. Liu, W. Ren, X. Li, C. Cheng, and J. Wang, Adv. Energy Mater., 7, 1602391 (2017).
• 20)  J. X. Feng, S. H. Ye, H. Xu, Y. X. Tong, and G. R. Li, Adv. Mater., 28, 4698 (2016).
• 21)  W.-J. Li, C. Han, G. Cheng, S.-L. Chou, H.-K. Liu, and S.-X. Dou, Small, 15, 1900470 (2019).
• 22)  J. C. Yu, J. H. Su, L. Li, P. Jing, X. Xu, B. C. Liu, and J. Zhang, ACS Mater. Lett., 6, 2591 (2024).
• 23)  K. B. Tan, Y. Liu, Y. Wang, S. Ali, W. Wang, J. Li, L. Shang, X. Yan, X. Zhang, and G. Zhan, Catal. Sci. Technol., 14, 3436 (2024).
• 24)  X. Chen, M. Cheng, D. Chen, and R. Wang, ACS Appl. Mater. Interfaces, 8, 3892 (2016).
• 25)  F. Meng, J. Zhu, K. Shi, X. Wang, J. Zhou, S. Jia, and Y. Yu, ACS App. Nano Mater., 7, 8756 (2024).
• 26)  A. Li, L. Zhang, F. Wang, L. Zhang, L. Li, H. Chen, and Z. Wei, Appl. Catal. B, 310, 121353 (2022).
• 27)  K. Liu, C. Zhang, Y. Sun, G. Zhang, X. Shen, F. Zou, H. Zhang, Z. Wu, E. C. Wegener, C. J. Taubert, J. T. Miller, Z. Peng, and Y. Zhu, ACS Nano, 12, 158 (2018).
• 28)  Y. Lian, H. Sun, X. Wang, P. Qi, Q. Mu, Y. Chen, J. Ye, X. Zhao, Z. Deng, and Y. Peng, Chem. Sci., 10, 464 (2019).
• 29)  X. Ding, W. Uddin, H. Sheng, P. Li, Y. Du, and M. Zhu, J. Alloys Compd., 814, 152332 (2020).
• 30)  H. Xu, L. Jin, K. Wang, L. Yang, G. He, and H. Chen, Int. J. Hydrogen Energy, 48, 38324 (2023).
• 31)  B. You, Y. Zhang, Y. Jiao, K. Davey, and S. Z. Qiao, Angew. Chem., Int. Ed., 58, 11796 (2019).
• 32)  S. X. Li, L. Wu, H. W. Zhao, L. X. Li, C. G. Sun, and B. G. An, J. Mater. Chem. A, 10, 9858 (2022).
• 33)  X. Zheng, Y. Chen, X. Bao, S. Mao, R. Fan, and Y. Wang, ACS Catal., 10, 11634 (2020).
• 34)  T. Tang, Z. Jiang, J. Deng, S. Niu, Z.-C. Yao, W.-J. Jiang, L.-J. Zhang, and J.-S. Hu, Nano Res., 15, 10021 (2022).
• 35)  P. Zheng, Y. Zhang, Z. Dai, Y. Zheng, K. N. Dinh, J. Yang, R. Dangol, X. Liu, and Q. Yan, Small, 14, 1704065 (2018).
• 36)  W. Li, Y. Chen, B. Yu, Y. Hu, X. Wang, and D. Yang, Nanoscale, 11, 17031 (2019).
• 37)  F. Chen, B. Zhao, M. Sun, C. Liu, Y. Shi, Y. Yu, and B. Zhang, Nano Res., 15, 10134 (2022).
• 38)  H. J. Fan, U. Gösele, and M. Zacharias, Small, 3, 1660 (2007).
• 39)  J. Zhang, M. Li, X. Liang, and Z. Zhuang, ACS Appl. Mater. Interfaces, 11, 1267 (2019).
• 40)  H. Zhang, L. V. Solomon, D.-H. Ha, S. Honrao, R. G. Hennig, and R. D. Robinson, Dalton Trans., 42, 12596 (2013).
• 41)  J. Joo, T. Kim, J. Lee, S. I. Choi, and K. Lee, Adv. Mater., 31, 1806682 (2019).
• 42)  K. Xu, H. Ding, M. Zhang, M. Chen, Z. Hao, L. Zhang, C. Wu, and Y. Xie, Adv. Mater., 29, 1606980 (2017).
• 43)  T. Shinagawa, A. T. Garcia-Esparza, and K. Takanabe, Sci. Rep., 5, 13801 (2015).
• 44)  Y. F. Yang, H. H. Bi, G. F. Huang, D. L. Yao, and W. Q. Huang, J. Phys. D: Appl. Phys., 55, 055502 (2022).
• 45)  J. Rossmeisl, Z. W. Qu, H. Zhu, G. J. Kroes, and J. K. Nørskov, J. Electroanal. Chem., 607, 83 (2007).
• 46)  L. A. Stern, L. Feng, F. Song, and X. Hu, Energy Environ. Sci., 8, 2347 (2015).
• 47)  R. Subbaraman, D. Tripkovic, K.-C. Chang, D. Strmcnik, A. P. Paulikas, P. Hirunsit, M. Chan, J. Greeley, V. Stamenkovic, and N. M. Markovic, Nat. Mater., 11, 550 (2012).
• 48)  H. Dau, C. Limberg, T. Reier, M. Risch, S. Roggan, and P. Strasser, ChemCatChem, 2, 724 (2010).
• 49)  M. Yang, J. Y. Xie, W. L. Yu, Y. N. Cao, B. Dong, Y. N. Zhou, F. L. Wang, Q. Z. Li, Y. L. Zhou, and Y. M. Chai, ACS Appl. Mater. Interfaces, 13, 17450 (2021).
• 50)  R.-Y. Fan, Y.-N. Zhou, M.-X. Li, J.-Y. Xie, W.-L. Yu, J.-Q. Chi, L. Wang, J.-F. Yu, Y.-M. Chai, and B. Dong, Chem. Eng. J., 426, 131943 (2021).
• 51)  X. Feng, W. Xing, L. Song, and Y. Hu, J. Mater. Chem. A, 2, 13299 (2014).
• 52)  Y. J. Ai, C. Zhang, L. Qi, M. Q. He, Z. Liang, H. B. Sun, G. Luo, and Q. L. Liang, ACS Appl. Mater. Interfaces, 10, 32180 (2018).
• 53)  B. Shang, P. Ma, J. Fan, L. Jiao, Z. Liu, Z. Zhang, N. Chen, Z. Cheng, X. Cui, and W. Zheng, Nanoscale, 10, 12330 (2018).
• 54)  N. Yu, R. N. Luan, C. Y. Gu, Y. L. Dong, G. Han, H. Hu, B. Liu, Y. M. Chai, and B. Dong, Fuel, 376, 132757 (2024).
• 55)  X. D. Wang, Y. Cao, Y. Teng, H. Y. Chen, Y. F. Xu, and D. B. Kuang, ACS Appl. Mater. Interfaces, 9, 32812 (2017).
• 56)  C. Du, L. Yang, F. Yang, G. Cheng, and W. Luo, ACS Catal., 7, 4131 (2017).
• 57)  P. F. Liu, X. Li, S. Yang, M. Y. Zu, P. Liu, B. Zhang, L. R. Zheng, H. Zhao, and H. G. Yang, ACS Energy Lett., 2, 2257 (2017).
• 58)  Y. Tan, H. Wang, P. Liu, Y. Shen, C. Cheng, A. Hirata, T. Fujita, Z. Tang, and M. Chen, Energy Environ. Sci., 9, 2257 (2016).
• 59)  A. Han, H. Chen, H. Zhang, Z. Sun, and P. Du, J. Mater. Chem. A, 4, 10195 (2016).
• 60)  J. Li, M. Yan, X. Zhou, Z.-Q. Huang, Z. Xia, C.-R. Chang, Y. Ma, and Y. Qu, Adv. Funct. Mater., 26, 6785 (2016).
• 61)  W. Li, X. Gao, X. Wang, D. Xiong, P.-P. Huang, W.-G. Song, X. Bao, and L. Liu, J. Power Sources, 330, 156 (2016).
• 62)  Z. Wang, N. Heng, X. Wang, J. He, and Y. Zhao, J. Catal., 374, 51 (2019).
• 63)  L. M. Cao, Y. W. Hu, S. F. Tang, A. Iljin, J. W. Wang, Z. M. Zhang, and T. B. Lu, Adv. Sci., 5, 1800949 (2018).
• 64)  C. Xiao, Y. Li, X. Lu, and C. Zhao, Adv. Funct. Mater., 26, 3515 (2016).
• 65)  Y. Yang, Z. Lin, S. Gao, J. Su, Z. Lun, G. Xia, J. Chen, R. Zhang, and Q. Chen, ACS Catal., 7, 469 (2017).
• 66)  C. Xuan, J. Wang, W. Xia, Z. Peng, Z. Wu, W. Lei, K. Xia, H. L. Xin, and D. Wang, ACS Appl. Mater. Interfaces, 9, 26134 (2017).
• 67)  Z. Zhao, D. E. Schipper, A. P. Leitner, H. Thirumalai, J.-H. Chen, L. Xie, F. Qin, M. K. Alam, L. C. Grabow, S. Chen, D. Wang, Z. Ren, Z. Wang, K. H. Whitmire, and J. Bao, Nano Energy, 39, 444 (2017).
• 68)  Q. Hu, X. Liu, C. Tang, L. Fan, X. Chai, Q. Zhang, J. Liu, and C. He, Sustainable Energy Fuels, 2, 1085 (2018).
• 69)  H. Xu, J. Wei, K. Zhang, Y. Shiraishi, and Y. Du, ACS Appl. Mater. Interfaces, 10, 29647 (2018).
• 70)  Y. Pan, K. Sun, S. Liu, X. Cao, K. Wu, W. C. Cheong, Z. Chen, Y. Wang, Y. Li, Y. Liu, D. Wang, Q. Peng, C. Chen, and Y. Li, J. Am. Chem. Soc., 140, 2610 (2018).
• 71)  L. Yan, B. Zhang, J. Zhu, Y. Li, P. Tsiakaras, and P. K. Shen, Appl. Catal. B, 265, 118555 (2020).
• 72)  S. Li, X. Wang, M. Li, J. Liu, C. Li, H. Wu, D. Guo, F. Ye, S. Wang, L. Cheng, and A. Liu, Electrochim. Acta, 319, 561 (2019).
• 73)  K. Sun, K. Wang, T. Yu, X. Liu, G. Wang, L. Jiang, Y. Bu, and G. Xie, Int. J. Hydrogen Energy, 44, 1328 (2019).