ZIF-L/PBA-Derived Self-Supporting Ni-Doped CoFeP Electrocatalysts for Bifunctional Water Splitting

Abstract

In recent years, transition metal-based catalytic materials have garnered considerable attention, particularly those exhibiting high catalytic efficiency toward both the oxygen evolution reaction (OER) and hydrogen evolution reaction (HER). In this work, a self-supporting ternary transition metal phosphide (CoFeNi_0.2P) with a hierarchical structure was synthesized using the Prussian blue analogue (PBA)/zeolitic imidazolate framework-L (ZIF-L) template. Benefiting from the hierarchical structure of the PBA/ZIF-L precursor and the electronic structure modulation induced by Ni doping, the resulting CoFeNi_0.2P demonstrates impressive bifunctional electrocatalytic activity. Specifically, in 1 M KOH electrolyte, the CoFeNi_0.2P catalyst requires an overpotential of only 88 mV to deliver 10 mA cm⁻² for the HER and 248 mV to achieve 50 mA cm⁻² for the OER. Moreover, it demonstrates satisfactory stability toward both the HER and OER. When integrated into a two-electrode electrolyzer, CoFeNi_0.2P enables a current density of 10 mA cm⁻² at a cell voltage of 1.59 V, maintaining robust performance for over 25 h. This study provides a feasible strategy for the rational design of hierarchical electrocatalysts for efficient overall water splitting.

1. Introduction

With the rapidly growing demand for fossil fuels, the global energy crisis has become an urgent challenge requiring immediate solutions [1,2,3]. Hydrogen, as a sustainable and green energy carrier, is considered a promising alternative to traditional fossil fuels, offering potential to alleviate both energy shortage crisis and environmental pollution [4]. However, the current water electrolysis technologies heavily rely on precious metal-based catalysts, such as Pt, Ru, and Ir, which are prohibitively expensive and scarce for large-scale applications [5]. Furthermore, while most hydrogen evolution reaction (HER) catalysts perform well in acidic media and oxygen evolution reaction (OER) catalysts are more effective in alkaline electrolytes, the design of bifunctional electrocatalysts that can catalyze both the HER and OER in the same electrolyte remains a significant challenge [6].

Considerable studies have been conducted on the possible alternatives to precious metal catalysts, such as transitional metal oxides [7], hydroxides [8], sulfides [9], and perovskites [10]. Furthermore, synergistic effects between various components in multimetallic catalysts have attracted significant attention recently due to their ability to enhance catalytic activity and optimize electronic structure [11]. To date, transition metal phosphides (TMPs) have attracted widespread attention from the scientific community due to their hydrogenase-like catalytic mechanism and high conductivity [11,12,13,14]. Furthermore, metal cation doping has been considered as a promising strategy to boost the intrinsic catalytic activity of the active sites, as it may not only enhance electron transfer during electrocatalysis but also modify the electronic structure of the catalysts that further optimize the reaction energy barrier for the HER or OER [15,16,17]. Nevertheless, achieving highly active and stable TMP-based bifunctional catalysts for both the HER and OER that are suitable for industrial applications remains elusive.

As is well known, metal−organic frameworks (MOFs) are highly regarded catalytic materials that have been extensively studied due to their tunable porous structures and abundant coordination sites [18,19]. Prussian blue analogues (PBAs), a subclass of MOFs, are known for their facile synthesis and cost-effectiveness. They typically conform to the general formula A_xM[M’(CN)₆]y·□_1−y·nH₂O (0 ≤ x ≤ 2; y < 1), where A denotes alkali metal ions, M and M’ correspond to transition metal ions, and □ represents the inherent vacancies in the M’(CN)₆ coordination units. In these crystalline frameworks, nitrogen-coordinated M cations are connected to carbon-coordinated M’ sites through cyanide (CN⁻) bridges [20]. PBAs feature highly tunable compositions, high specific surface areas, and abundant vacancies [21], offering great potential for electrocatalytic applications. However, their catalytic performance is often limited by their poor electrical conductivity and the limited accessibility of active metal centers buried within the bulk phase. Additionally, most reported PBAs exist as micro-sized powders, typically in cubic or irregular morphologies, which are prone to aggregation during the phosphidation process when used as templates for metal phosphide synthesis. In contrast, zeolitic imidazolate framework (ZIF) materials, another subclass of MOFs, have attracted extensive attention in recent years. Among them, ZIF-L, characterized by a two-dimensional layered structure with a leaf-like morphology, stands out as a promising precursor or template to construct binder-free hierarchical electrocatalysts, owing to its ability to grow directly on conductive substrates such as nickel foam (NF). Nevertheless, the high-temperature calcination of ZIF-L often results in structural collapse and the aggregation of the derived TMPs, thereby limiting their electrochemical performance [22]. Recently, the partial transformation of ZIF-L into other MOFs via postsynthetic means (such as ligand exchange reactions) has emerged as an effective strategy to construct heterogeneous catalysts [5,13,14].

Inspired by this, we propose a strategy employing Ni-doped ZIF-L as the starting material, which is subsequently converted into a CoFeNi_0.2-PBA material via a ligand exchange reaction, and ultimately transformed into a ternary transition metal phosphide, CoFeNi_0.2P. During the ligand exchange process, Fe(CN)₆³⁻ ions partially replace the imidazolate linkers, and coordinate with Co²⁺/Ni²⁺ to induce the nucleation of CoFeNi_x-PBA nanocrystals uniformly on the ZIF-L nanosheets. This surface-confined transformation preserves the two-dimensional morphology of ZIF-L while introducing abundant redox-active metal centers at the interface. As a result, the ZIF-L/PBA hybrid precursor enables the formation of a hierarchical ternary phosphide structure with uniformly distributed Co, Fe, and Ni species upon phosphidation. Furthermore, Ni doping modulates the electronic structure of adjacent Co and Fe elements, thereby enhancing intrinsic catalytic activity. This unique structural and compositional configuration endows the material with superior catalytic properties, offering a novel approach for the development of high-performance electrocatalysts.

2. Results and Discussion

2.1. Catalyst Synthesis and Characterization

Figure 1 schematically illustrates the three-step synthetic process. Firstly, CoNi_x-ZIF-L is directly grown on the surface of NF via a facile reaction between Co(NO₃)₂, Ni(NO₃)₂, and 2-Methylimidazole (2-MI) in an aqueous solution at room temperature (RT) [8]. Figure S1 shows the typical scanning electron microscope (SEM) images of the as-synthesized CoNi_0.2-ZIF-L, which exhibits a distinct plate-like morphology with a thickness around 200 nm. The synthesized CoNi_x-ZIF-L is then immersed in an aqueous solution of K₃[Fe(CN)₆] at RT and stirred continuously to facilitate the ligand exchange reaction, resulting in the formation of CoFeNi_x-PBA on the ZIF-L nanosheets. As presented in Figure S2, after the ligand exchange, the plate-like morphology is well preserved but the surface becomes rougher, suggesting the formation of PBA nanoparticles on the origin ZIF-L plates (Figure S2, Supplementary Materials). Afterwards, the phosphorization of CoFeNi_x-PBA is carried out by using NaH₂PO₂ as the phosphor source in an Ar atmosphere at 300 °C for 2h, yielding the ternary metal phosphide CoFeNi_0.2P.

Given that Ni doping content may influence the catalytic performance, the Co/Ni feeding ratio for catalyst synthesis was optimized. CoFeNi_xP (where x denotes the molar ratio of Ni to Co during ZIF-L synthesis, thus the corresponding sample names are designated as CoNi_x-ZIF-L, CoFeNi_x-PBA, and CoFeNi_xP) is used as a self-supporting electrocatalyst electrode for the OER, with CoFeP, CoFeNi_0.06P, and CoFeNi_0.33P serving as the control samples.

Figure 2a,b present the surface morphology of the typical CoFeNi_0.2P sample. Clearly, the plate-like architecture inherited from the ZIF-L and PBA templates is well retained, while the surface roughness further increases after phosphorization. Compared with undoped CoFeP (Figure 2c), the Ni-incorporated CoFeNi_xP exhibits thinner nanosheet structures (Figure S3, Supplementary Materials), with CoFeNi_0.2P displaying the thinnest nanosheets, thereby offering the highest specific surface area. Moreover, excessive Ni doping leads to the aggregation of the CoFeNi_xP plates (Figure S3e,f, Supplementary Materials).

The transmission electron microscope (TEM) image shows that the plate-like structure of CoFeNi_0.2-PBA is well maintained, with evenly interconnected ultrafine Co₂P, Fe₂P, and Ni₂P nanoparticles immobilized on the N and P dual-doped carbon porous nanosheets (Figure 2d,e) [5], and the carbon nanosheets derived from the pyrolysis of PBA frameworks serve as substrates, effectively anchoring and restricting the growth of phosphide nanoparticles during synthesis. The high-resolution TEM (HR-TEM) image distinctly reveals the coexistence of lattice fringes corresponding to Co₂P, Fe₂P, and Ni₂P nanoparticles, confirming the formation of three-phase heterostructures (Figure 2e). Specifically, the lattice fringes of 0.205, 0.293, and 0.251 nm match well with the (130) plane of Co₂P, (110) plane of Fe₂P, and (200) plane of Ni₂P, respectively [23,24]. Moreover, the selected area diffraction (SAED) pattern of CoFeNi_0.2P (Figure 2f) confirms the polycrystalline nature of the sample. Furthermore, the energy-dispersive X-ray spectroscopy (EDS) elemental mapping demonstrates the uniform distribution of Co, Fe, Ni, and P elements throughout the CoFeNi_0.2P material, as shown in Figure 2g–k.

X-ray diffraction (XRD) was used to characterize the crystal structure of the as-synthesized catalysts. As shown in Figure 3a, the introduction of Ni into ZIF-L does not alter its crystal structure, as evidenced by the matching diffraction patterns of Ni-doped ZIF-L and the simulated patterns, confirming the successful synthesis of ZIF-L. This conclusion is further supported by the Fourier-transform infrared (FT-IR) spectra in Figure S4. Furthermore, with the increasing Ni doping content, the diffraction peaks shift toward lower angles, indicating an expansion in the interplanar spacing of CoNi_x-ZIF-L (Figure 3b) [25]. After the ligand exchange process, no distinct diffraction peaks are observed in the XRD spectra (Figure S5), suggesting a significant reduction in crystallinity. Therefore, the FT-IR analysis was carried out to investigate the structural evolution during the ligand exchange process. As shown in Figure 3c and Figure S6, for CoNi_x-ZIF-L, the characteristic bands at 1143 and 760 cm⁻¹ are attributed to the C-H vibrations, while the absorption peaks at 420, 1384, and 1566 cm⁻¹ correspond to the Co-N bond vibration and the stretching vibrations of C-C and C=N, respectively [13]. After the ligand exchange reaction, the characteristic peaks of the original ZIF-L become weaker but remain detectable (760 cm⁻¹, 1343 cm⁻¹, 1384 cm⁻¹, and 1566 cm⁻¹), accompanying the emergence of several new peaks. In particular, the strong peak at 2080 cm⁻¹ is assigned to the stretching of the -C≡N- groups [26], while the bands observed at 596 and 458 cm⁻¹ are associated with the bending vibration of Fe-CN and Ni-CN/Co-CN, respectively [27]. These FT-IR results collectively confirm the successful formation of CoFeNi_x-PBA on the surface of ZIF-L. Moreover, the ZIF-L structure is retained and has not completely disappeared, indicating that the synthetic material is a kind of MOF-on-MOF structure. Figure 3d presents the XRD patterns of the CoFeNi_xP and CoFeP. Specifically, the diffraction peaks of the final materials match well with those of Co₂P (JCPDS card no. 32-0306), Fe₂P (JCPDS card no. 27-1171), and Ni₂P (JCPDS card no. 74-1385) phases, confirming the successful formation of ternary metal phosphides.

In addition, X-ray photoelectron spectroscopy (XPS) analysis was conducted to analyze the chemical composition and oxidation states of CoFeNi_0.2P. The survey spectrum in Figure S7 confirms the presence of Co, Fe, Ni, P, O, and C elements. As shown in Figure 4a, the high-resolution Co 2p spectrum exhibits two main spin-orbit peaks at 781.5 eV (Co 2p_3/2) and 797.4 eV (Co 2p_1/2), characteristic of Co²⁺ species [28], along with the additional peaks at 778.65 and 793.6 eV, which are attributed to the Co-P bond [29]. Similarly, the Fe 2p spectrum (Figure 4b) displays a doublet at 711.9 eV (Fe 2p_3/2) and 725.0 eV (Fe 2p_1/2), indicative of Fe²⁺ species, while the minor peaks at 706.8 and 720.3 eV correspond to the Fe-P bond [30,31]. Compared with CoFeP, the Co-P bond in CoFeNi_0.2P shows a positive shift about 0.3 eV. A similar shift is also observed for the Fe-P bond, indicating that the introduction of Ni atoms induces an electron withdrawal from Co and Fe atoms. This electron redistribution modulates the local chemical environment and electronic structure of CoFeP nanosheet [5,32]. As shown in Figure 4c, the high-resolution XPS spectrum of Ni 2p reveals two main peaks at 856.4 eV and 874.1 eV, corresponding to the Ni 2p3/2 and 2p1/2 spin-orbital components, respectively, confirming the presence of Ni²⁺ species [33]. In addition, the peaks at 852.8 eV and 869.8 eV are assigned to the Ni-P bond [29]. The P 2p spectrum in Figure 4d displays two peaks at 129.2 eV and 130.1 eV, which can be attributed to the P 2p_3/2 and P 2p_1/2 of the metal–P bond, respectively, along with a peak at 134.1 eV associated with the oxidized phosphorus species [29]. Notably, a negatively shift in P 2p peaks can be observed upon Ni incorporation, suggesting a change in the local electronic environment caused by electron transfer from Ni [5]. As shown in Figure 4e, the O 1s spectrum can be split into three peaks located at 531.4 eV, 532.2 eV, and 532.8 eV, corresponding to the surface hydroxide, C=O, and adsorbed water molecules, respectively [34]. In addition, the C 1s spectrum (Figure 4f) reveals three peaks at 284.8, 285.85, and 288.45 eV, which are attributed to the C-C, C-N, and O-C=O bonds, respectively [35].

2.2. Hydrogen Evolution Reaction Performance

To evaluate the electrocatalytic performance toward the HER, the as-synthesized catalysts were tested in 1 M KOH electrolyte using a three-electrode system. As shown in Figure 5a, the linear sweep voltammetry (LSV) curves illustrate the HER activity of all samples. Notably, CoFeNi_0.2P requires a low overpotential of merely 88 mV to achieve a current density of 10 mA cm⁻², outperforming CoFeP (126 mV), CoFeNi_0.06P (107 mV), and CoFeNi_0.33P (96 mV), as summarized in Figure 5b. The Tafel slope displayed in Figure 5c is a key parameter for evaluating HER kinetics. Under alkaline conditions, the HER typically proceeds via three primary steps:

Volmer step: H₂O + e⁻ → H_ads + OH⁻

(1)

Tafel step: H_ads + H_ads → H₂

(2)

Heyrovsky step: H_ads + H₂O + e⁻ → H₂ + OH⁻

(3)

The Tafel slope is a fundamental parameter for assessing the HER kinetics. To interpret the observed Tafel response of the synthesized catalysts, the associated reaction mechanisms are systematically discussed. Typically, a Tafel slope of ~120 mV dec⁻¹ is indicative of a Volmer-limited pathway, in which the initial step is rate-determining. In contrast, slopes in the range of 30–40 mV dec⁻¹ suggest a mixed Heyrovsky/Tafel-controlled mechanism, where subsequent electrochemical desorption and recombination steps dominate [36]. In our case, CoFeNi_0.2P demonstrates a significantly lower Tafel slope (71.73 mV dec⁻¹) compared to CoFeP (81.43 mV dec⁻¹), CoFeNi_0.06P (78.09 mV dec⁻¹), and CoFeNi_0.33P (75.16 mV dec⁻¹), revealing its superior HER kinetics governed by the Volmer–Heyrovsky mechanism.

Figure 5d shows the Nyquist impedance plots of the various catalysts, offering insights into their charge transfer kinetics during the HER process. Notably, CoFeNi_0.2P exhibits the smallest semicircular arc, suggesting a lower charge transfer resistance (R_ct) than its other counterparts (Table S1, Supplementary Materials). Interestingly, all CoFeNi_xP catalysts display smaller R_ct values than CoFeP, revealing the positive effect of Ni doping on charge transfer kinetics. As the Ni doping content increases, the R_ct first decreases and then increases, reaching a minimum for CoFeNi_0.2P. This suggests that the doping level in CoFeNi_0.2P is optimal for facilitating charge transfer during the HER process, consistent with the Tafel analysis. In addition, the double-layer capacitance (C_dl) was measured by CV scans at different scan rates within the non-Faradaic potential region of 0.175–0.275 V vs. RHE (Figure S8, Supplementary Materials), and the corresponding electrochemically active surface areas (ECSAs) were calculated. As illustrated in Figure 5e, CoFeNi_0.2P demonstrates the highest C_dl value (54.98 mF cm⁻²), suggesting a larger ECSA. To elucidate the effect of Ni doping on the intrinsic activity of the catalysts, the LSV curves were normalized for the ECSA. As depicted in Figure S9, the CoFeNi_0.2P catalyst still exhibits the best HER catalytic performance among all samples, further confirming the enhancement of its intrinsic catalytic activity [29].

Stability is also one of the essential factors for evaluating electrocatalysts. The durability of CoFeNi_0.2P was tested using chronoamperometry at a current density of 10 mA cm⁻². Notably, CoFeNi_0.2P shows excellent long-term stability, with the overpotential increasing by only ~10 mV after 50 h of continuous operation (Figure 5f). The leaf-like plate morphology of CoFeNi_0.2P is well preserved after the HER stability test, as confirmed by the SEM examination (Figure S10, Supplementary Materials). The XPS analysis of the post-HER catalyst (Figure S11) reveals negligible changes in the high-resolution spectra of Ni 2p, Co 2p, Fe 2p, and P 2p compared to their original states, suggesting the good structural stability of trimetallic phosphide.

2.3. Oxygen Evolution Reaction Performance

As a crucial half-reaction in water splitting, the oxygen evolution reaction (OER) is inherently constrained by its multi-step four-electron transfer mechanism. This intrinsic complexity leads to sluggish kinetics and high overpotentials, posing a major bottleneck for efficient energy conversion. The OER performance of the catalysts was then evaluated in 1 M KOH electrolyte. The backward scan curves were used for analysis to avoid the influence of oxidative peaks of Ni. Figure 6a presents the LSV curves of CoFeNi_xP and CoFeP catalysts at a scan rate of 5 mV/s. Clearly, CoFeNi_0.2P exhibits the lowest overpotential of 248 mV at a current density of 50 mA cm⁻², outperforming CoFeP (260 mV), CoFeNi_0.06P (251 mV), and CoFeNi_0.33P (257 mV). Notably, the redox peak observed at approximately 1.28 V (vs. RHE) in the LSV curves is ascribed to the anodic reduction of Ni³⁺ to Ni²⁺. The direct comparison in Figure 6b highlights the superior catalytic efficiency of CoFeNi_0.2P. It exhibits the lowest overpotential at current densities of 50, 100, and 250 mA cm⁻², suggesting that it requires a smaller voltage to sustain the same current density, thereby validating its exceptional OER activity. This performance positions CoFeNi_0.2P among the top-performing OER catalysts reported so far (Table S3, Supplementary Materials), outperforming many non-noble OER electrocatalysts such as Mn−(Co,Ni)2P/CoP/(N,S)−C (η₅₀ = 300 mV) [37], F–Fe–CoP (η₅₀ = 292 mV) [38], MXene/MIL Fe-53/ZIF-67 (η₅₀ = 292 mV) [39], 5-cycle Co₃O₄ (η₅₀ = 311 mV) [40], Co₃O₄/CeO₂ (η₅₀ = 259 mV) [41], Fe-Ni₃S₂@NF (η₅₀ = 297 mV) [42], and 5-cycle NiFe LDH (η₅₀= 259 mV) [43].

Furthermore, CoFeNi_0.2P exhibits a Tafel slope of 56.94 mV dec⁻¹ (Figure 6c), which is lower than those of CoFeP (59.28 mV dec⁻¹), CoFeNi_0.06P (57.61 mV dec⁻¹), and CoFeNi_0.33P (62.48 mV dec⁻¹), indicating its superior reaction kinetic in the OER process (Figure 6d). In addition, electrochemical impedance spectroscopy (EIS) was employed to investigate the charge transfer kinetics at the electrode/electrolyte interface of the prepared catalysts. As shown in Figure 6d, all CoFeNi_xP catalysts display smaller R_ct than CoFeP, revealing that Ni doping effectively promotes the charge transfer rate. Moreover, with the increase in Ni content, R_ct initially decreases and then increases, reaching its minimum for CoFeNi_0.2P (Table S2, Supplementary Materials). This trend suggests that an optimal Ni doping level accelerates charge transfer kinetics. The C_dl was then determined by conducting CV scans at various scan rates within the non-Faradaic potential window of 1.175–1.275 V vs. RHE (Figure S12, Supplementary Materials). As shown in Figure 6e, CoFeNi_0.2P exhibits the highest C_dl value (42.25 mF cm⁻²), surpassing those of CoFeP (37.81 mF cm⁻²), CoFeNi_0.06P (41.85 mF cm⁻²), and CoFeNi_0.33P (39.65 mF cm⁻²). The higher C_dl value of CoFeNi_0.2P indicates a larger ECSA and enhanced exposure of catalytic sites. The ECSA-normalized LSV curves further reveal the superior OER performance of CoFeNi_0.2P compared to the other catalysts (Figure S13), highlighting its enhanced intrinsic catalytic activity for the OER.

Additionally, the chronopotentiometry test at 10 mA cm⁻² reveals negligible potential elevation over 70 h, demonstrating the exceptional OER durability of the CoFeNi_0.2P catalyst (Figure 6f). The SEM observation confirms its well-preserved plate-like structure, with smaller nanoflakes derived on the surface, indicating the good structural stability of the catalyst (Figure S14, Supplementary Materials). To gain more insights into the surface chemical composition and active species involved in the OER, the XPS characterization was further carried out on the CoFeNi_0.2P catalyst after the OER test (Figure S11, Supplementary Materials). Obviously, after the OER reaction, most Co and Fe atoms are oxidized to their trivalent states. Meanwhile, the peak for Ni shifts positively by about 0.2 eV (Figure S11c), indicating an increase in its valence state after the reaction [44,45]. Moreover, after the stability test, the metal–P bonds almost disappear, leaving only oxidized species. In addition, the increase in M-O content observed in the O 1s spectrum further supports the formation of hydroxides. Based on the aforementioned XPS analyses, it can be concluded that metal oxides/(oxy)hydroxides form on the surface of the phosphide catalyst under highly oxidative conditions during the OER, which are considered the real active species for the reaction [44,45,46,47].

2.4. Overall Water Splitting Performance

As shown in Figure 7a,b, the bifunctional capability of CoFeNi_0.2P was demonstrated in a symmetric two-electrode alkaline electrolyzer, achieving a cell voltage of 1.59 V at 10 mA cm⁻², which is lower than that of the Pt/C||IrO₂ couple (1.72 V). This demonstrates the high-efficiency OER activity as an anode and the superior HER performance of the CoFeNi_0.2P, underscoring its potential as an integrated water-splitting catalyst. This performance is highly competitive compared to other electrocatalysts reported in the literature (Table S4, Supplementary Materials), such as CeOx@NiCo₂O₄/NF (1.66 V) [48], CoS_1.25Se_0.75@NC (1.67 V) [49], γ-FeOOH/Ni₃S₂ (1.66 V) [50], mCo_0.5Fe_0.5P/rGO (1.66 V) [51], CoMoS₄/NF (1.65 V) [52], Co/Mo₂C@NC-800 (1.67 V) [53], CoP@FeCoP/NC YSMPs (1.68 V) [54], and Te/FeNiOOH-NC (1.65 V) [55].

Additionally, long-term stability tests were also conducted on the two-electrode overall water splitting device. As shown in Figure 7c, when the current density is maintained at 10 mA cm⁻², the voltage remains stable at 1.59 V with minor fluctuations. The superior performance can be attributed to the hierarchical porous morphology of the catalyst, which stems from the structural advantages inherited from the ZIF-L/PBA template. To further evaluate the practical applicability of CoFeNi_0.2P in water electrolysis systems, its electron utilization efficiency was investigated. The Faraday efficiency (FE) graph in Figure 7d shows that the amount of collected H₂ is approximately double that of collected O₂, which aligns closely with the theoretical prediction for overall water splitting. This suggests that CoFeNi_0.2P can effectively split water into oxygen and hydrogen. With 98% Faradaic efficiency and a non-noble metal composition, CoFeNi_0.2P emerges as a cost-effective bifunctional catalyst, making it a promising alternative to noble metal-based systems.

3. Experimental Section

3.1. Materials and Chemicals

Nickel (II) nitrate hexahydrate (Ni(NO₃)₂·6H₂O), cobalt(II) nitrate hexahydrate (Co(NO₃)₂·6H₂O), 2-Methylimidazole, potassium(III) ferricyanide, and potassium hydroxide were purchased from Adamas Reagent Co., Ltd. (Shanghai, China). Sodium hypophosphite was purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. (Shanghai, China). Nickel foam (thickness: 1.6 mm) was obtained from Lizhiyuan Technology Co., Ltd. (Taiyuan, China). Commercial Pt/C catalyst (20 wt%) were bought from Johnson Matthey Company (Shanghai, China). IrO₂ and nafion (5 wt%) were obtained from Afar Aesar (Shanghai, China). All chemicals were used without further purification.

3.2. Sample Synthesis

3.2.1. Synthesis of Ni-Doped Co-MOF Precursor (CoNi_x-ZIF-L)

In a typical synthesis, Co(NO₃)₂·6H₂O (1.5 mmol) and Ni(NO₃)₂·6H₂O (0.3 mmol) were dissolved in 18.75 mL of deionized water to form a pink solution. Then, 18.75 mL of an aqueous solution of 2-methylimidazole (C₄H₆N₂, 12 mmol) was quickly added to the mixture. A pretreated piece of nickel foam (NF, 3 × 1.5 cm²) was immersed in the mixture and maintained at room temperature for 1 h. Afterward, the electrode was rinsed three times with distilled water and ethanol individually and then dried overnight. CoNi_0.06-ZIF-L, CoNi_0.33-ZIF-L, and Co-ZIF-L were synthesized using the same method, with the only difference being the amounts of nickel added: 0.1 mmol, 0.5 mmol, and 0 mmol, respectively.

3.2.2. Synthesis of CoFeNi_x-PBA

The CoFeNi_0.2-ZIF-L was immersed in 25 mL of aqueous solution containing 5 mmol of K₃[Fe(CN)₆] at RT for 1 h under stirring to facilitate rapid ligand exchange, resulting in CoFeNi_0.2-PBA nanosheet arrays. The electrodes were cleaned three times with distilled water and ethanol and dried overnight.

3.2.3. Synthesis of CoFeNi_xP

CoFeNi_x-PBA and NaH₂PO₂ (0.5 g) were placed in separate quartz boats with NaH₂PO₂ positioned at the upstream side. Then, phosphorization was carried out by heating the precursor samples to 300 °C under an Ar atmosphere at a rate of 2 °C min⁻¹, and this temperature was maintained for 2 h. After cooling to room temperature, the resulting CoFeNi_0.2P was directly used as a self-supporting electrode, eliminating the need for binders or further electrode preparation steps. The loading mass of active materials on the NF is about 4 mg cm⁻².

3.3. Material Characterization

The morphologies of the samples were characterized with field-emission scanning electron microscopy (FE-SEM, Quanta FEG450, FEI, Hillsboro, OR, USA) and transmission electron microscopy (TEM, TECNAI F30, FEI, USA). Elemental composition and distribution were analyzed with energy-dispersive X-ray spectroscopy (EDX, JEOL 2010, Tokyo, Japan) equipped with the FE-SEM. X-ray diffraction (XRD) was performed using a Bruker D8-Advance X-ray diffractometer (Bruker, Germany) with Cu K_α radiation to analyze the phase structure. Fourier-transform infrared (FT-IR) spectra were collected in the range of 4000–450 cm⁻¹ with a JASCO FT/IR-460 spectrophotometer. X-ray photoelectron spectroscopy (XPS, ESCALAB250Xi, Thermo Fisher, Waltham, MA, USA) was employed to analyze the chemical composition and valence states of the elements in the samples.

3.4. Electrochemical Measurement

All electrochemical tests were performed on the Gamry electrochemical workstation (INTERFACE 1010E, Warminster, PA, USA) in a standard three-electrode system. The electrochemical cell had a total volume of 30 mL, into which 20 mL of electrolyte was added, leaving approximately 1 cm of headspace above the solution surface to accommodate gas evolution during the electrochemical reactions. The as-prepared catalysts on the NF served as the working electrode, a Pt sheet was used as the counter electrode, and Hg/HgO acted as the reference electrode. A 1 M KOH solution was employed as the electrolyte solution. For the HER, the linear sweep voltammograms (LSVs) were measured over a potential range of −0.576 to 0.024 V vs. RHE at a scan rate of 5 mV s⁻¹. For the OER, the LSVs were recorded from 1.024 to 2.024 V vs. RHE at the same scan rate. Prior to all electrochemical measurements, the catalyst was electrochemically activated by performing LSV scans within the corresponding potential range at a scan rate of 50 mV s⁻¹ for 20 cycles. Electrochemical impedance spectroscopy (EIS) was measured in the frequency range of 0.1 Hz to 100 kHz with a 5 mV AC dither for the HER and OER.

CVs were tested at different scan rates, which are employed to estimate the double-layer capacitances (C_dl) of the catalysts. The C_dl value was further converted into the electrochemically active surface area (ECSA) by using a specific capacitance value that corresponds to a standard flat surface with a real surface area of 1 cm². In our calculation, 0.04 mF cm⁻² was used. Then, the ECSA was calculated using the following equation:

$$ECSA={\displaystyle \frac{C_{dl}-catalyst\hspace{0.33em}(\mathrm{mF}\hspace{0.17em}{\mathrm{cm}}^{-2})}{0.04\hspace{0.33em}(\mathrm{mF}\hspace{0.17em}{\mathrm{cm}}^{-2})\hspace{0.33em}per\hspace{0.33em}ECSA\hspace{0.33em}{\mathrm{cm}}^2}}$$

Chronoamperometry measurements were conducted at a constant current density of 10 mA cm⁻². Overall water splitting performance was evaluated in a two-electrode system using CoFeNi_0.2P as the catalyst for both the HER and the OER in 1 M KOH. The polarization curves were recorded at a scan rate of 5 mV s⁻¹. The Faraday efficiency (FE) theoretical precipitation is calculated as follows: n(O₂) = Q/nF (Q = it), where n(O₂) denotes the theoretical molar amount of O₂ precipitated, Q denotes the transferred charge, n denotes the number of electrons transferred in the OER/HER reaction (n = 4/2), F denotes the Faraday’s constant (F = 96,485 C/mol), i denotes the value of the electric current applied in water electrolysis, and t denotes the time of water electrolysis. In this work, the water displacement method was used to quantify the amount of oxygen evolved in a fixed time period [56]. The electrolysis was conducted at a constant current density of 50 mA cm⁻² for a duration of 1 h. All potentials reported in this paper were converted from vs. Hg/HgO to vs. RHE by adding a value of 0.098 + 0.059 × pH and were corrected for ohmic losses.

4. Conclusions

To sum up, a trimetallic phosphide has been successfully synthesized through a facile three-step method. Ni doping effectively modulates the electronic structure of CoFeNi_0.2P, while its hierarchical architecture ensures sufficient exposure of active surface area. Benefitting from this synergetic effect, CoFeNi_0.2P exhibits impressive electrocatalytic performance, achieving 50 mA cm⁻² at an overpotential of 248 mV for the OER and 50 mA cm⁻² at an overpotential of 139 mV for the HER. When employed as both a cathode and an anode catalyst, the assembled electrolyzer may drive a current density of 10 mA cm⁻² at a low cell voltage of 1.59 V, maintaining excellent durability over 25 h. This work opens a new avenue for improving the activity of transition metal MOF-derived catalysts and elucidates the tunability of the catalytic performance through electronic structure modulation.