One-dimensional nickel–cobalt bimetallic phosphide nanostructures for the oxygen evolution reaction

Abstract

Nickel–cobalt bimetallic phosphides have widely been used in the oxygen evolution reaction (OER). However, finding effective strategies for fabricating one-dimensional nanostructures of bimetallic phosphides with enhanced electrocatalytic OER properties is still challenging. Herein, novel catalysts based on NiCoP nanofibers with exceptional electrocatalytic performances were synthesized for efficient OER. The resulting NiCoP nanofibers required only a minimum overpotential of 325 mV in 1.0 M KOH solution to drive a current density of 50 mA⁻² for oxygen production. After 5000 OER cycles, the activity of NiCoP nanofibers remained practically unaltered, showcasing exceptional stability. The proposed strategy for designing and implementing nanostructured electrocatalysts looks promising for enhancing the performance of catalysts toward the OER for better energy production devices.

---

1. Introduction

The excessive exploitation of conventional energy sources has resulted in environmental pollution, energy resource depletion, and other significant negative impacts.¹ As a result, developing efficient and environmentally friendly alternative energy sources has become an urgent matter. Electrocatalytic hydrolysis combines the hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) as promising energy conversion pathways. More importantly, electrocatalytic hydrolysis is also promising for large-scale hydrogen production, thereby being useful for overcoming current energy problems. However, the anodic oxygen evolutionary reaction (OER) is characterized by slow kinetics and a high thermodynamic energy barrier, seriously hindering the efficiency of electrochemical hydrolysis.² Currently used electrocatalysts with better activity for electrocatalytic oxygen evolution reaction (OER) consist of precious metals, such as IrO₂ and RuO₂. Nevertheless, the exorbitant cost and limited availability of these precious metal-based catalysts severely restrict their applications in water electrolysis. Therefore, exploring efficient and stable non-precious metal electrocatalysts is highly desirable for better energy outcomes.^3,4

Materials, such as transition metal-based oxides, bimetallic alloys, selenides, sulfides, and phosphates have attracted significant attention in electrochemical hydrolysis due to their diverse range of candidate material components with different catalytic efficiencies.^5,6 Among these, transition metal phosphides (TMPs) are abundant on Earth and possess exceptional catalytic activity attributed to their favorable electrical conductivity,^7,8 electron orbital properties of phosphorus atoms,^9,10 and moderate hydrogen (H) binding energy.¹¹ As a result, the catalytic properties of various forms of TMPs toward the HER, OER, and total hydrolysis were thoroughly investigated. These encompass Ni/Ni₂ hollow nanoparticles,¹² Ni₅P₄ discoides,¹³ WP nanowire arrays,¹⁴ CoP@CoMoO₄ nanotubes,¹⁵ Ce–Co nanosheets,¹⁶ hollow urchin-like FeP,¹⁷ and MoP₂ hollow nanospheres.¹⁸ However, monometallic phosphides require enhancement in terms of electrocatalytic activity and stability. By comparison, bimetallic phosphides exhibit better electrocatalytic activity than monometallic phosphides. However, finding effective strategies for fabricating one-dimensional nanostructures and bimetallic phosphides for enhanced electrocatalytic properties of transition metal phosphides is still challenging.

Herein, a novel approach synergistically integrating one-dimensional nanostructures with bimetallic phosphides was developed to significantly enhance the electrocatalytic performance of transition metal phosphides. The precursor nanofibers were obtained through an electrospinning process. Subsequently, NiCo₂O₄ nanofibers were obtained by thermal annealing and finally converted into bimetallic phosphides (NiCoP nanofibers) through a phosphorylation process. The introduction of bimetal enhanced the electrocatalytic activity of transition metal phosphides toward the OER. The obtained NiCoP nanofibers outperformed monometallic phosphides in all aspects. Additionally, the synergistic effect between phosphorus (P) and isocyanate enhanced the OER properties of NiCoP nanofibers, while the formation of P–M bonds effectively inhibited P degradation, greatly improving the catalyst stability toward the OER.^19–22

The proposed one-dimensional structure also facilitated the adsorption of numerous compounds.²² The unique fiber network structure allowed electrospinning to significantly enhance stabilization and increase the interfacial area between the catalytic electrode and the electrolyte, making it an ideal technique for catalysis. The ultra-long one-dimensional nanofibers were then treated as self-contained binderless electrodes, and the resulting NiCoP nanofibers showed ample electrocatalytically active sites and enhanced charge conduction capacity owing to their unique one-dimensional structure. Consequently, the as-obtained NiCoP nanofibers required only a minimum overpotential of 325 mV to drive a current density of 50 mA⁻² for oxygen production. Furthermore, NiCoP nanofibers exhibited outstanding durability after undergoing 5000 OER cycles, showing outstanding catalyst performances.

2. Experimental

2.1 Chemical reagents

Polyacrylonitrile (PAN) was purchased from Sigma-Aldrich. N,N-Dimethylformamide (DMF), KOH, Ni(CH₃COO)₂·4H₂O, and Co(CH₃COO)₂·6H₂O were all obtained from Zhiyuan Chemical Reagents, Tianjin. Sodium hypophosphite (NaH₂PO₂·4H₂O) was provided by Shanghai Aladdin Biochemical Technology Corporation. All chemicals were of analytical grade and used as received without additional purification.

2.2 Synthesis of NiCo₂O₄ nanofibers

The preparation process of NiCo₂O₄ nanofibers is shown in Scheme 1. First, 0.5 g of polyacrylonitrile (PAN) powder was added to a 5 mL solution of N,N-dimethylformamide (DMF). The mixture was covered with a plastic wrap to prevent evaporation and stirred to form a uniform solution with a concentration of 10 wt%. Stoichiometric amounts of 0.4 mmol Ni (CH₃COO)₂·4H₂O and 0.4 mmol Co(CH₃COO)₂·6H₂O were then added to the solution by maintaining a molar ratio of Ni/Co at 1 : 1. The mixture was magnetically stirred at room temperature for 10 h to guarantee thorough mixing and obtain a uniform clear viscous precursor solution for electrospinning. The PAN fibers were produced by pouring the mixed precursor solution into a syringe fitted with a stainless-steel needle of 40 mm in length. Afterward, high voltage was applied between the aluminum collector and the tip of the needle at an operating voltage of approximately 7 kV while keeping the collector distance at 15 cm under humidity conditions from 30% to 40%. Electrospinning begins when there is weakened balance between pressure at the needle tip and applied voltage. The droplets of the solution gradually decreased in size, and approximately 10 μL of the solution was evenly dispersed onto the collector. Upon approaching the target electrode, electrostatic repulsion acted on the fiber while facilitating solvent evaporation. After 12 h of electrostatic spinning, the resulting fiber membrane was collected and used as original nanofibers, which then were subjected to heating at a rate of 2 °C min⁻¹ until reaching 600 °C followed by calcination in air for 2 h to produce NiCo₂O₄ nanofibers. The synthesis processes of NiO and Co₃O₄ followed a similar procedure to that of NiCo₂O₄.

Scheme 1 Schematic diagram of the fabrication process of NiCo₂O₄ and NiCoP-600 nanofibers.

2.3 Synthesis of NiCoP nanofibers

The process consisted of phosphating NiCo₂O₄ nanofibers with NaH₂PO₂ as the phosphorus source to synthesize NiCoP nanofibers. Briefly, 30 mg of NiCo₂O₄ nanofibers and 600 mg of NaH₂PO₂·4H₂O were loaded into separate ceramic vessels and placed in a tube furnace. The NaH₂PO₂·4H₂O was positioned 10 cm upstream from the NiCo₂O₄ nanofibers inside the quartz tube. The furnace was then purged with nitrogen (N₂, 99.999%) for 1 h to remove air and then kept under a constant flow rate of nitrogen gas at 320 mL min⁻¹ throughout the process. Subsequently, the temperature was increased to 400 °C for 2 h at a heating rate of 2 °C min⁻¹ before allowing natural cooling to room temperature to obtain NiCoP nanofibers.

For comparison, NiCo₂O₄ nanofiber catalysts with different Ni : Co molar ratios were prepared by the same process. The obtained NiCo₂O₄ NFs and Ni : Co at ratios of 1 : 1, 1 : 2, and 2 : 1 were denoted as NCO-1 : 1, NCO-1 : 2, and NCO-2 : 1, respectively. Next, the same phosphorylation process was used to convert the precursor NiCo₂O₄ into NCO-1 : 1P, NCO-1 : 2P, and NCO-2 : 1P at phosphating temperatures of 400 °C, 500 °C, 600 °C, 700 °C, and 800 °C. The resulting samples were named NCO-1 : 1P400(NiCoP-400), NCO-1 : 1P500(NiCoP-500), NCO-1 : 1P600(NiCoP-600), NCO-1 : 1P700(NiCoP-700), and NCO-1 : 1P800(NiCoP-800), respectively. Similarly, NiCoP-1 : 2 and NiCoP-2 : 1 obtained at phosphating temperatures of 400 °C, 500 °C, 600 °C, 700 °C, and 800 °C were named NCO-1 : 2P400, NCO-1 : 2P500, NCO-1 : 2P600, NCO-1 : 2P700, NCO-1 : 2P800, NCO-2 : 1P400, NCO-2 : 1P500, NCO-2 : 1P600, NCO-2 : 1P700, and NCO-2 : 1P800, respectively. For comparison, synthetic Ni₂P/Ni₅P₄ and CoP₂ nanofibers were prepared and used as control samples.

2.4 Materials characterization

The crystal structures of the as-prepared compounds were analyzed by X-ray diffraction (XRD, D/max 2600, Rigaku, Japan). The morphologies were viewed by scanning electron microscopy (SEM, SU70, Hitachi, Japan). The atomic structures were observed by transmission electron microscopy (TEM, FEI, Tecnai TF20) and high resolution TEM (HRTEM). The surface chemical components of the composites were identified by X-ray photoelectron spectroscopy (XPS, Thermofisher Scientific).

2.5 Electrochemical measurements

The working electrodes were prepared by first mixing 15 mg of NiCoP-600 nanofibers, polyvinylidene fluoride (PVDF), and carbon nanotubes at the mass ratio of 8 : 1 : 1 with PVDF as the binder and carbon nanotubes as the conductivity agent. Next, a solution of 50 μL N-methylpyrrolidone (NMP) was added and the mixture was ultrasonically treated for 1 h to obtain a homogeneous suspension. The slurry was vigorously mixed and sonicated for an additional 1 h before being uniformly coated onto a 1 cm² carbon paper substrate. After drying in a vacuum oven at 60 °C for 10 h, the modified substrates with a loading mass of the active substance of approximately 2.5 mg cm⁻² were tested as working electrodes. The electrochemical measurements were conducted on a workstation (VMP-300, France) connected to a three-electrode cell. All electrochemical measurements were performed in a solution containing 1.0 M KOH, with the counter electrode made of a platinum sheet, reference electrode consisting of Ag/AgCl, and working electrode comprising the catalyst sample.

The reversible hydrogen electrode (RHE) was used as the reference for conversion of all measured potentials according to the formula: E_RHE = E_SCE + 0.059 pH + 0.196 V (pH = 14), with overpotential calculated by η = E_RHE − 1.23 V. To ensure reliable cyclic voltammetry curves, the electrocatalysts were activated for a total of 70 cycles at a scan rate of 20 mV s⁻¹. Linear sweep voltammetry (LSV) was performed at the scan rate of 5 mV⁻¹. The Tafel slopes were obtained using the formula: η = b log(j) + a, where η represents the overpotential, b is the Tafel slope, and j denotes the current density. Electrochemical impedance spectroscopy (EIS) measurements were performed at a potential of 500 mV and frequencies ranging from 0.01 Hz to 100 kHz. Cyclic voltammetry (CV) measurements were carried out in a non-Faraday current regime for evaluating the electrochemical surface area (ECSA). The capacitance of the electric double layer (C_dl) was studied at various CV scan rates ranging from 2 to 12 mV s⁻¹. Cycle tests consisting of 5000 cycles at a scan rate of 100 mV s⁻¹ were carried out to test the long-term stability of the electrocatalysts using the time potential method (J = 10 mA cm⁻²) with stability recorded after 15 h. All potentials were corrected by the 95%-IR compensation rule.

3. Results and discussion

The synthesis route of NiCo₂O₄ nanofibers and NiCoP-600 nanofibers is illustrated in Scheme 1. The NiCo₂O₄ nanofibers were first fabricated by electrospinning, followed by the synthesis of NiCoP-600 nanofibers by a low-temperature phosphating process.

The SEM images of NiO, Co₃O₄, NiCo₂O₄, and their phosphating products Ni₂P/Ni₅P₄, CoP₂, and NiCoP-600 nanofibers are provided in Fig. 1. For further clarification, the HRTEM images of NiCo₂O₄ and NiCoP-600 nanofibers were also included. As shown in Fig. 1a and b, fibrous structures of NiO nanofibers were formed at different magnifications, with uniform diameters of approximately 50–100 nm and slight fractures observed in the fiber structure (Fig. 1b). The SEM images in Fig. 1c and d display Co₃O₄ nanofibers with adhered fibrous structures and consistent diameters of around 200–250 nm (Fig. 1d). The SEM images in Fig. 2e and f show distinctly rough surface fibrous structures of NiCo₂O₄ nanofibers with a uniform diameter ranging from approximately 150 to 200 nm (Fig. 1f). Ni₂P/Ni₅P₄ nanofibers in Fig. 2g and h illustrate smoother surfaces than NiO nanofibers (Fig. 2h). The SEM images in Fig. 2i and j display CoP₂ nanofibers with a uniform diameter of about 400–500 nm, smooth surfaces, and good adhesion between the fibers. NiCoP-600 nanofibers in Fig. 2k and l exhibit rough surfaces and uniform diameters of about 200–250 nm (Fig. 2l). From the above SEM images, it can be seen that through the electrospinning method, we can prepare a relatively ideal one-dimensional oxide nanofiber, and the phosphide prepared on this basis can better retain the one-dimensional nanostructure. The complete one-dimensional structure of phosphide is conducive to improving its performance in the field of electrochemistry. The HRTEM images of NiCo₂O₄ nanofibers in Fig. 1m reveal a distinct lattice spacing of 0.47 nm corresponding to the (111) plane of the cubic structure, as indicated by XRD. The HRTEM images of NiCoP-600 nanofibers in Fig. 1n demonstrate a distinct lattice spacing of 0.22 nm, consistent with the XRD pattern and indicating the presence of the (111) plane within the cubic structure. The high chemical reactivity interaction between isocyanate and P formed NiCoP-600 nanofibers with densely packed structures (Fig. 1l). The elemental mappings of P, Ni, and Co in NiCoP-600 nanofibers can be found in Fig. S1 of the ESI.†

Fig. 1 SEM images of NiO nanofibers (a) and (b), Co₃O₄ nanofibers (c) and (d), NiCo₂O₄ nanofibers (e) and (f), Ni₂P/Ni₅P₄ nanofibers (g) and (h), CoP₂ nanofibers (i) and (j), and NiCoP-600 nanofibers (k) and (l). TEM and HRTEM images of NiCo₂O₄ nanofibers (m) and NiCoP-600 nanofibers (n).

Fig. 2 XRD patterns of NiO nanofibers, Co₃O₄ nanofibers, NiCo₂O₄ nanofibers, Ni₂P/Ni₅P₄ nanofibers, CoP₂ nanofibers, and NiCoP-600 nanofibers.

The crystal structures of NiO, Co₃O₄, NiCo₂O₄, Ni₂P/Ni₅P₄, CoP₂, and NiCoP-600 nanofibers obtained by XRD are depicted in Fig. 2. The diffraction peaks of NiO nanofibers and Co₃O₄ nanofibers were consistent with those of NiO (PDF#4-835) and Co₃O₄ (PDF#73-1701). The diffraction peaks of NiCo₂O₄ nanofibers can be attributed to the spinel structure of NiCo₂O₄ (PDF#20-0781). XRD analysis according to the PDF#20-0781 database indicated the presence of peaks at 31.1°, 36.6°, 44.6°, 59.0°, and 64.5°, which can be assigned to the crystal planes (220), (311), (400), (511), and (440) of typical NiCo₂O₄, respectively. Additionally, NiO was present in the XRD detection profiles. After phosphating treatment, two phosphates namely Ni₂P and Ni₅P₄ were obtained with characteristic diffraction peaks matching those reported for Ni₂P(PDF#3-953) and Ni₅P₄(PDF#18-883), respectively. Phosphating treatment of Co₃O₄ nanofibers resulted in the formation of CoP₂ nanofibers with characteristic diffraction peaks corresponding to those of CoP₂(PDF#77-263). NiCoP-600 nanofibers obtained by phosphating NiCo₂O₄ nanofibers showed minor amounts of NiP and CoP₃. The XRD pattern revealed distinct peaks at 41.02°, 44.92°, and 47.6°, attributed to the crystal planes (111), (201), and (210) of NiCoP-600 nanofibers successfully synthesized in this study (PDF#71-2366), respectively. Thus, P and isocyanate were covalently linked in the NiCoP-600 structure through Ni–P and Co–P bonds.

The chemical makeup of NiCoP-600 was analyzed by XPS on both nanofibers of NiCo₂O₄ and NiCoP-600. The presence of Ni 2p, Co p, and P 2p characteristic peaks in the XPS spectrum of NiCoP-600 nanofibers confirmed the existence of these elements, aligning well with EDS mapping results (Fig. S2a†). The complete XPS spectrum of NiCo₂O₄ is presented in Fig. S3,† while that of NiCoP-600 is given in Fig. 3a. For comparison, the Ni 2p XPS spectra of NiCo₂O₄ and NiCoP-600 are presented in Fig. 3b. The spectrum of NCO exhibited six distinct peaks, with two peaks at 853.2 eV and 854.9 eV related to Ni 2p_3/2 orbitals. Additionally, twin peaks were observed at 871.8 eV and 875.9 eV, associated with the Ni 2p_1/2 orbital, while satellite peaks can be found at approximately 860.2 eV and 878.9 eV. Compared to NiCo₂O₄, the binding energy of Ni 2p peaks in NiCoP-600 shifted by approximately 2.6 eV toward higher values. Additionally, the Ni 2p spectrum in Fig. 3b depicts a peak of P–Ni originating from NiCoP due to the formation of a P–Ni bond between isocyanate and P under an argon atmosphere.^22–24,36 The high-resolution Co 2p XPS spectra of both NiCo₂O₄ and NiCoP-600 samples in Fig. 3c exhibit similar characteristics to those of Ni 2p that can be deconvoluted into six distinct peaks. The observed peaks at 777.3 eV and 780.9 eV can be attributed to the Co 2p_3/2 state, while those at 793.6 eV and 795.2 eV correspond to the Co 2p_1/2 state. Hence, Co was predominantly present in a metallic state.^33,35 Two distinct peaks were detected at energy levels of 785 eV and 802.2 eV, respectively. The appearance of peaks at 777.1 eV and 792.1 eV can be assigned to the bonding between P and Co atoms.^22,25,34 The P 2p spectrum of NiCoP-600 in Fig. 3d exhibits three distinct peaks, indicative of metal phosphide's characteristic metal–P bonds. The P 2p_3/2 and P 2p_1/2 depicted binding energies of respectively 131.6 eV and 132.5 eV, with an extra peak around 137.1 eV corresponding to the bond between P and Ni/Co.^26–28 All these data confirmed the successful synthesis of NiCoP-600 catalysts.

Fig. 3 (a) Full spectrum of NiCoP-600 nanofibers. XPS spectra of (b) Ni 2p and (c) Co 2p of NiCo₂O₄ and NiCoP-600 nanofibers. (d) XPS spectrum of P 2p of NiCoP-600 nanofibers.

The electrocatalytic performances of NiCoP-600 nanofibers toward the OER were assessed by linear sweep voltammetry (LSV) in 1.0 M potassium hydroxide solution in comparison with NiO, Co₃O₄, NiCo₂O₄, Ni₂P/Ni₅P₄, and CoP₂ nanofibers under the same conditions. The recorded LSV curves of NiO, Co₃O₄, NiCo₂O₄, Ni₂P/Ni₅P₄, CoP₂, NiCoP-600 nanofibers, and RuO₂ are given in Fig. 4a. Before each catalytic test, repeated CV scans were performed at 5 mV⁻¹ and 1.0–1.65 V as pre-activation scans in the potential range of RHE. The required overpotential (η) at a specific current density was used to evaluate the performance of each electrocatalyst. Meanwhile, the overpotential of each catalyst was determined by analyzing the LSV curve at a current density of 50 mA cm⁻². Obviously, the overpotential of NiCoP-600 was significantly lower than those of NiO, Co₃O₄, NiCo₂O₄, Ni₂P/Ni₅P₄, CoP₂, NiCoP-600 nanofibers, and RuO₂. Under the same current density (j = 50 mA cm⁻²), the overpotential of NiCoP-600 was the lowest at 325 mV, a value very close to that of RuO₂ (313 mV) while surpassing those of NiO (460 mV), Co₃O₄ (439 mV), NiCo₂O₄ (400 mV), Ni₂P/Ni₅P₄ (384 mV), and CoP₂ nanofibers (348 mV). Further analysis of LSV curves revealed that NiCoP-600 preserved its superior overpotential (355 mV) under a current density of 100 mA cm⁻² when compared to Co₃O₄ (472 mV), NiCo₂O₄ (440 mV), Ni₂P/Ni₅P₄ (427 mV), and CoP₂ nanofibers (378 mV) (Fig. 4b), as well as outperforming some reported OER catalysts (Fig. 4d). In this paper, the OER characteristics exhibited by bimetallic phosphides surpassed those of monometallic phosphides, which, in turn were superior to those of bimetallic oxides and monometallic oxides. Also, the OER performance of NiCoP-600 nanofibers surpassed that of NiCoP-600 powders (Fig. S4†). Thus, the advantages of one-dimensional nanostructures over nanofibers with irregular shapes were substantial in terms of the OER.²⁹ Overall, the catalytic activity of NiCoP-600 nanofibers toward the OER looked remarkable when compared to other catalysts, demonstrating good performances.

Fig. 4 Electrocatalytic OER performances of NiO nanofibers, Co₃O₄ nanofibers, NiCo₂O₄ nanofibers, Ni₂P/Ni₅P₄ nanofibers, CoP₂ nanofibers, and NiCoP-600 nanofibers. (a) LSV curves of the prepared electrodes. (b) Corresponding overpotential of the synthesized catalysts at current densities of 50 mA cm⁻² and 100 mA cm⁻² used for comparison. (c) Comparison of Tafel plots of synthesized catalysts. (d) Comparison with reported OER catalysts.

The optimal experimental conditions in terms of nickel–cobalt ratio and corresponding calcination temperature were determined by comparing the OER activities of NiCo₂O₄ nanofiber catalysts with different Ni : Co molar ratios. The LSV curves of NCO-1 : 1, NCO-1 : 2, and NCO-2 : 1 are summarized in Fig. S5a.† Subsequently, a sequence of phosphating procedures was conducted at temperatures ranging from 400 °C to 800 °C, and structural and catalytic characterization studies were conducted to investigate the influence of calcination temperature. As shown in the SEM image (Fig. S6†), the calcination at 500 °C and 600 °C resulted in the formation of clear fiber shapes (Fig. S6a, b, e and f†). By comparison, calcination at 700 °C resulted in a semi-melt state (Fig. S6c and g†) while 800 °C caused complete melting and loss of the distinct fiber shape (Fig. S6d and h†). The corresponding LSV curves in Fig. S5b† indicate 600 °C as a suitable phosphating temperature for enhancing the OER activity of NiCoP-600 (Fig. S5b†). The samples calcined at 600 °C exhibited the best OER performances, while those calcined at 800 °C showed the worst OER performances. Therefore, the most favorable temperature for the fabrication of NiCoP nanofibers was identified as 600 °C.

The crystal structures of nanofibers with different compositions, namely NiCoP-400, NiCoP-500, NiCoP-600, NiCoP-700, and NiCoP-800 were determined by XRD analysis (Fig. S7†). At 400 °C, 500 °C, and 600 °C, minor amounts of NiP and CoP₃ were also detected. However, only NiCoP was present at 700 °C and 800 °C. The combination of this observation with the OER performance suggested superior OER performance of the non-pure phase of NiCoP when compared to its pure phase counterpart. Thus, the presence of other valence compounds synergistically enhanced the catalytic activity of NiCoP in promoting the oxygen evolution reaction. In sum, NiCoP-600 samples demonstrated the highest OER performances.

The Tafel slopes can be used to reveal the reaction mechanism and determine the inherent activity of catalysts through analysis of the reaction kinetic rate. Thus, Tafel analysis was conducted on all TMP “pre-catalysts” to analyse their OER kinetics. The results suggested a decrease in Tafel slopes in a similar order to the apparent OER activities. The Tafel slope was obtained by fitting the graph to the Tafel equation (η = b log j + a), where j denotes the current density and b represents the slope of the Tafel curve. As shown in Fig. 4c, NiCoP-600 presented the lowest Tafel slope at 96.4 mV dec⁻¹, a value lower than those of NiO (158.9 mV dec⁻¹), Co₃O₄ (99.2 mV dec⁻¹), NiCo₂O₄ (119.2 mV dec⁻¹), Ni₂P/Ni₅P₄ (113.3 mV dec⁻¹), and CoP₂ (99.1 mV dec⁻¹). As a result, the surface of the CoNiP catalyst exhibited a mechanism of rapid adsorption, implying enhanced electron transfer kinetics and an elevated rate of electron transfer.²⁶ The prepared NiCoP-600 nanofibers exhibited excellent properties and their one-dimensional nanostructure enhanced the regional dynamics of the active material, resulting in significant oxygen generation capacity of NiCoP-600.

As displayed in Fig. 5a and S8b,† the charge transfer resistance (R_ct) of NiCoP-600 was estimated to be 1.8 Ω, a value lower than those of NiO (122.1 Ω), Co₃O₄ (27.1 Ω), NiCo₂O₄ (22.3 Ω), Ni₂P/Ni₅P₄(7.6 Ω), and CoP₂ (3.8 Ω). Thus, NiCoP-600 exhibited faster electron transfer capability. Additionally, the one-dimensional carbon nanofiber structures effectively reduced the electrical contact resistance between adjacent NiCoP-600 nanoparticles, facilitating efficient electrical contact and rapid mass transfer of reactants and products during the reaction.

Fig. 5 (a) EIS patterns of NiO nanofibers, Co₃O₄ nanofibers, NiCo₂O₄ nanofibers, Ni₂P/Ni₅P₄ nanofibers, CoP₂ nanofibers, and NiCoP-600 nanofibers. (b) Variation in ΔJ/2 as a function of the scanning rate. (c) LSV curves before and after 5000 cycles. (d) Long-term stability of NiCoP-600 nanofibers at the current density of 10 mA cm⁻² for 50 h.

The electrochemically active surface area (ECSA) was evaluated using the capacitance of the electrochemical double layer (C_dl). The cyclic voltammetry (CV) activation process demonstrated that the electrode current can be segregated into two components. The first had to do with non-Faraday currents, encompassing charge and discharge currents, and the second dealt with Faraday currents associated with electrochemical reactions.^29,30 The difference in current density between charge–discharge reactions would linearly vary with the CV scanning speed. In the region with nonapplicable Faraday potential, the electrochemically accessible surface area (ECSA) of each catalyst was estimated by determining the electrochemical double-layer capacitance (C_dl) by the CV method to provide insights into the number of active sites. C_dl was obtained by performing a linear fit between the current density and scanning rate of cyclic voltammetry (CV) curves, where the slope represents the C_dl value. As shown in Fig. 5b, the C_dl of NiCoP-600 toward the OER reaction under different scanning rates was estimated to be 827 mF cm⁻², a value significantly higher than those of NiO (21 mF cm⁻²), Co₃O₄ (23 mF cm⁻²), NiCo₂O₄ (65 mF cm⁻²), Ni₂P/Ni₅P₄ (414 mF cm⁻²), and CoP₂ (550 mF cm⁻²). In the scanning potential range from 1.24 to 1.34 V, the ECSA of NiCoP-600 was the largest, indicating the presence of more active sites for the OER reaction when compared to other materials mentioned above. As displayed in Fig. 5b and S9a,† Cobalt-nickel bimetals contributed to increasing the C_dl values. For example, the C_dl value of NiCoP-600 (827 mF cm⁻²) was significantly higher than those of Ni₂P/Ni₅P₄ (414 mF cm⁻²) and CoP₂ (550 mF cm⁻²). The increased ECSA of bimetallic cobalt-nickel phosphating nanofibers resulted in a larger contact area between the catalyst and electrolyte, promoting the kinetics of the OER reaction.^32,33

During real-time manufacturing procedures, the durability of the electrocatalyst would also play a pivotal role in assessing the efficacy of catalyst materials since the catalyst material would facilitate the liberation of oxygen from the electrode surface during O₂ release. Thus, the long-term stability of NiCoP-600 was investigated in a potassium hydroxide (1.0 M) solution. As demonstrated in Fig. 5c, the current of the NiCoP-600 catalyst remained relatively stable (347 mV) even after 5000 cycles, suggesting excellent electrochemical durability in an alkaline environment. Additionally, long-term electrochemical stability testing at 50 mA cm⁻² for 50 h revealed good durability for the NiCoP-600 catalyst (Fig. 5d). The formation of P–M bonds can effectively prevent hydrogen bonding between oxygen atoms bonded to phosphorus and water molecules, thereby significantly enhancing stability. The structural stability of NiCoP-600 during electrocatalytic OER was verified by SEM imaging and EDS mapping before and after electrocatalytic OER. In Fig. S10,† the fibers did not break but the surface roughness increased after the OER. In Fig. S2a and b,† the EDS mapping results indicated a decrease in only the peak value of P, suggesting high structural stability for NiCoP-600.

The anodic process in the electrolytic cell involved a four-electron mechanism OER. By comparison to the hydrogen evolution reaction (HER), the OER would have slower kinetics with three generated adsorption intermediates (M–OOH_ads, M–O_ads, and M–OH_ads).³¹ In alkaline electrolytes, the OER was initiated by hydrolysis or OH coordination, followed by consecutive oxidation steps from M–OH_ads to M–O_ads, then to M–OOH_ads, ultimately producing O₂. All suggested mechanisms involved stages of hydroxyl coordination in alkaline environments, with M representing the active site and ads indicating the adsorbed form on the catalyst surface. Accordingly, the appropriate OER mechanism in alkaline electrolytes can be summarized as follows:

NiCoP + 2OH⁻ → NiCoP(OH)_{2 ads} + 2e⁻ (1)

NiCoP(OH)_{2 ads} + 2OH⁻ → NiCoPO_{2 ads} + 2H₂O + 2e⁻ (2)

NiCoPO_{2 ads} → NiCoP + 2O₂ (3)

or

NiCoPO_{2 ads} + 2OH⁻ → NiCoP(OOH)_{2 ads} + e⁻ (4)

NiCoP(OOH)_{2 ads} + 2OH⁻ → NiCoP + 2O₂ + 2H₂O + 2e⁻ (5)

Two primary pathways would influence oxygen production through distinct basic steps. The initial route entailed the direct amalgamation of two M–O_ads intermediates, proceeding through steps (1) → (2) → (3). In the following sequential progression of (1) → (2) → (4) → (5), the intermediate M–OOH_ads initially transformed into M–O_adsvia hydroxide, subsequently merging with another hydroxide to generate O₂. Note that reaction (3) always exhibited a higher thermodynamic barrier than reactions (4) and (5). The principal approaches for oxygen generation involved metal phosphide and sulfide following the second pathway.

4. Conclusions

Novel NiCoP-600 bimetallic phosphides were successfully prepared by electrostatic spinning technology, thermal annealing, and phosphatization processes. The catalysts demonstrated exceptional OER efficiency, with only 325 mV overpotential needed to attain a current density of 50 mA cm⁻² in 1 M potassium hydroxide solution. After 5000 cycles of OER, the catalysts also exhibited a remarkable overpotential of 347 mV, surpassing the performance of numerous reported transition metal phosphides. The one-dimensional nanostructure of the cross-linked network promoted charge transport, as well as facilitating electrolyte penetration and oxygen bubble release, thereby accelerating the catalytic kinetics of the OER. The introduction of bimetallic components also modulated the central electronic structure of monometallic phosphides. The NiCoP-600 catalysts outperformed their monometallic counterparts (Ni₂P/Ni₅P₄ and CoP₂ nanofibers) in all aspects. Overall, the proposed cost-effective nanocomposite catalysts look promising for future reversible electrochemical energy applications.

Author contributions

Yue Wang: conceptualization, data curation, writing – original draft. Xin Chang: software, formal analysis. Zexing Huang: software. Jiahui Fan: investigation. Lu Li: writing – review & editing. Mingyi Zhang: conceptualization, supervision, funding acquisition.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors are grateful to be supported in part by the National Natural Science Foundation of China (no. 52102228).

References

1. L. Rößner, H. Schwarz, I. Veremchuk, R. Zerdoumi, T. Seyller and M. Armbrüster, ACS Appl. Mater. Interfaces, 2021, 13, 23616–23626.
2. G. A. Gebreslase, M. V. Martínez-Huerta and M. J. Lázaro, J. Energy Chem., 2022, 67, 101–137.
3. Y.-C. Zhang, C. Han, J. Gao, L. Pan, J. Wu, X.-D. Zhu and J.-J. Zou, ACS Catal., 2021, 20, 12485–12509.
4. Q. Pan and L. Wang, J. Energy Chem., 2021, 485, 229335.
5. X. Li, Q. Hu, H. Yang, T. Ma, X. Chai and C. He, Chin. Chem. Lett., 2022, 33, 3657–3671.
6. F. Aftab, H. Duran, K. Kirchhoff, M. Zaheer, B. Iqbal, M. Saleem and S. N. Arshad, ChemCatChem, 2020, 12, 932–943.
7. Y. Li, R. Tong, W. Zhang and S. Peng, J. Catal., 2022, 410, 22–30.
8. G. M. S. Salvador, A. L. Silva, L. P. C. Silva, F. B. Passos and N. M. F. Carvalho, Int. J. Hydrogen Energy, 2021, 46, 26976–26988.
9. Sk. Riyajuddin, S. K. Tarik Aziz, S. Kumar, G. D. Nessim and K. Ghosh, ChemCatChem, 2019, 12, 1394–1402.
10. W. Yu, Y. Gao, Z. Chen, Y. Zhao, Z. Wu and W. Li, Chin. J. Catal., 2021, 42, 1876–1902.
11. L. Yu, J. Zhang, Y. Dang, J. He, Z. Tobin, P. Kerns, Y. Dou, Y. Jiang, Y. He and S. L. Suib, ACS Catal., 2019, 9, 6919–6928.
12. X. Li, J. Zhou, C. Liu, L. Xu, C. Lu, J. Yang, H. Pang, W. Hou and W. Hou, Appl. Catal., B, 2021, 298, 120578.
13. M. Ledendecker, S. K. Calderón, C. Papp, H.-P. Steinrück, M. Antonietti and M. Shalom, Angew. Chem., Int. Ed., 2015, 54, 12361–12365.
14. C. Lv, J. Liu, P. Lou, X. Wang, L. Gao, S. Wang and Z. Huang, Nanoscale, 2022, 14, 5430–5438.
15. Z. Gao, Z. Zeng, X. Xu, R. Mao, R. Jia and S. Han, Int. J. Hydrogen Energy DOI:10.1016/j.ijhydene.2023.07.179.
16. J. Bi, H. Ying, H. Xu, X. Zhao, X. Du, J. Hao and Z. Li, Chem. Commun., 2022, 58, 7817–7820.
17. P. P. Gao, M. Gao, T. Lei, Z. Ren, J. Luo, Z. Huang and A. Wu, Inorg. Chem. Commun., 2023, 156, 111143.
18. W. Yang, J. Tian, L. Hou, B. Deng, S. Wang, R. Li, F. Yang and Y. Li, ChemSusChem, 2019, 12, 4662–4670.
19. J. Zheng, X. Peng, Z. Xu, J. Gong and Z. Wang, ACS Catal., 2022, 12, 10245–10254.
20. C. Zhang, J. Wang, Y. Liu, W. Li, Y. Wang, G. Qin and Z. Lv, Chem.–Asian J., 2022, 17, e202200377.
21. P. M. Bodhankar, D. S. Dhawale, S. Giddey, R. Kumar and P. B. Sarawade, Sustainable Energy Fuels, 2022, 6, 5491–5502.
22. T. Wu, J. Fan, Q. Li, P. Shi, Q. Xu and Y. Min, Adv. Energy Mater., 2018, 8, 1701799.
23. M. A. Ahsan, T. He, K. Eid, A. M Abdullah, M. F. Sanad, A. Aldalbahi, B. Alvarado-Tenorio, A. Du and A. R. P. Santiago, ACS Appl. Mater. Interfaces, 2022, 14, 3919–3929.
24. A. R. P. Santiago, T. He, O. Eraso, M. A. Ahsan, A. N. Nair, V. S. N. Chava, T. Zheng, S. Pilla, O. Fernandez-Delgado, A. Du, S. T. Sreenivasan and L. Echegoyen, J. Am. Chem. Soc., 2020, 142, 17923–17927.
25. H. Wang, C. Li, J. An, Y. Zhuang and S. Tao, J. Mater. Chem. A, 2021, 9, 18421–18430.
26. I. Amorim, J. Xu, N. Zhang, D. Xiong, S. M. Thalluri, R. Thomas, J. P. S. Sousa, A. Araújo, H. Li and L. Liu, Catal. Today, 2020, 358, 196–202.
27. J. Yin, J. Jin, M. Lu, B. Huang, H. Zhang, Y. Peng, P. Xi and C.-H. Yan, J. Am. Chem. Soc., 2020, 142, 18378–18386.
28. X. Li, L. Xiao, L. Zhou, Q. Xu, J. Weng, J. Xu and B. Liu, Angew Chem. Int. Ed. Engl., 2020, 59, 21106–21113.
29. B. Yang, X. Chang, X. Ding, X. Ma and M. Zhang, J. Colloid Interface Sci., 2022, 623, 196–204.
30. S. Haschke, D. Pankin, Y. Petrov, S. Bochmann, A. Manshina and J. Bachmann, ChemSusChem, 2017, 10, 3644–3651.
31. H. Shi, H. Liang, F. Ming and Z. Wang, Angew. Chem., 2017, 129, 588–592.
32. X. Qiao, H. Kang, Y. Li, K. Cui, X. Jia, X. Wu and W. Qin, Appl. Catal., B, 2022, 305, 121034.
33. W. Jiang, Y. Tao, J. Ma, X. Liu, S. Zhu, W. Hao, Q. Xu and J. Fan, ACS Sustainable Chem. Eng., 2023, 11, 6629–6640.
34. M. A. Ahsan, T. He, K. Eid, A. M. Abdullah, M. L. Curry, A. Du, A. R. P. Santiago, L. Echegoyen and J. C. Noveron, J. Am. Chem. Soc., 2021, 143, 1203–1215.
35. M. Yu, X. Guo, X. Chang, X. Ma and M. Zhang, Sustainable Energy Fuels, 2022, 6, 5000–5007.
36. H. Liang, A. N. Gandi, D. H. Anjum, X. Wang, U. Schwingenschlögl and H. N. Alshareef, Nano Lett., 2016, 16, 7718–7725.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3se01412f

---