Ni-Doped Co-Based Metal–Organic Framework with Its Derived Material as an Efficient Electrocatalyst for Overall Water Splitting

Abstract

Composite catalysts combining a metal–organic framework (MOF) with its derivatives have attracted significant attention in electrocatalysis due to their unique properties. In this study, we report the synthesis of a Ni-doped Co-1,4-benzenedicarboxylate (defined as Co₃Ni₁BDC) metal–organic framework via a straightforward solvothermal method, aiming to enhance oxygen evolution reaction (OER) activity. The introduction of Ni modulated the electronic structure, yielding high catalytic activity with an overpotential (η₁₀₀) of 300 mV and excellent stability for the OER. The Co₃Ni₁BDC material was further encapsulated with Co₂P nanoparticles via a controlled phosphating annealing process, forming a hybrid electrocatalyst (Co₃Ni₁BDC@Co₂P) to boost hydrogen evolution reaction (HER) performance. The Co₃Ni₁BDC@ Co₂P catalysts exhibited superior HER performance with low overpotentials of η₁₀ = 20 mV and η₁₀₀ = 127 mV, outperforming the Co₃Ni₁BDC precursor. An alkaline electrolyzer assembled with Co₃Ni₁BDC//Co₃Ni₁BDC@Co₂P achieved a cell voltage of 1.70 V at a current density of 20 mA cm⁻². This work provides a valuable idea for designing efficient electrocatalysts for overall water splitting.

1. Introduction

As global energy demands escalate alongside mounting environmental concerns over fossil fuel consumption, hydrogen has emerged as a pivotal clean energy vector due to its exceptional gravimetric energy density (142 MJ/kg) and carbon-neutral combustion profile [1]. Electrochemical water splitting plays a pivotal role in addressing the global energy crisis. The overall water electrolysis process comprises two critical half-reactions: the hydrogen evolution reaction (HER) at the cathode and the oxygen evolution reaction (OER) at the anode [2,3,4]. However, OER involves a complex four-electron transfer process, which suffers from sluggish kinetics and high energy barriers, significantly limiting hydrogen production efficiency [5,6]. This kinetic bottleneck fundamentally governs the energy conversion efficiency of water electrolyzers, with the OER overpotential accounting for approximately 80% of total voltage losses in commercial alkaline electrolysis systems. Although Pt and RuO₂ exhibit exceptional electrochemical activity for the hydrogen evolution reaction, their high cost impedes large-scale commercial application [7,8,9,10]. Therefore, the development of efficient and cost-effective electrocatalysts is crucial.

Recent studies have shown that the efficiency of hydrogen production largely depends on the structure and composition of the catalysts. Metal–organic frameworks (MOFs), characterized by their crystalline hybrid structures with periodically arranged metal clusters interconnected by multifunctional organic linkers (i.e., 1,4-benzenedicarboxylate, 2-methylimidazole), offer unique advantages in this context [11,12,13,14,15]. Strategic post-synthetic modifications of MOFs, including controlled pyrolysis, heteroatom doping, and in situ heterojunction construction, have emerged as powerful methodologies to engineer catalytic architectures. For example, Chen’s group developed self-supported 2D Fe-doped Ni-MOF nanosheets (NiFe-MOF) with exceptional performance and ultralow potential [16]. Similarly, Pan’s group introduced phosphide nanostructures at the interfaces of Zr-MOF-derived electrocatalysts, achieving high efficiency for HER [17]. Despite these advancements, the development of efficient bifunctional non-noble metal electrocatalysts remains a critical challenge.

In this work, we synthesized Ni-doped CoBDC via a straightforward solvothermal method. The as-synthesized Co₃Ni₁BDC electrocatalysts exhibited high catalytic activity with an overpotential (η₁₀₀) of 300 mV and excellent stability for OER. Furthermore, Co₃Ni₁BDC was encapsulated in Co₂P (denoted as Co₃Ni₁BDC@Co₂P) via a phosphating annealing method. The synergistic effect of highly dispersed Co₂P nanoparticles enabled Co₃Ni₁BDC@Co₂P to achieve exceptional HER catalytic activity with a low overpotential of 127 mV at a current density of 100 mA cm⁻². This work opens new avenues for the rational design of MOF-derived electrocatalysts for overall water splitting in energy materials.

2. Results and Discussion

2.1. Catalyst Synthesis and Characterization

Scheme 1 illustrates the preparation route for Co₃Ni₁BDC/NF (nickel foam) and Co₃Ni₁BDC@Co₂P/NF. The Co₃Ni₁BDC arrays were synthesized on nickel foam via a solvothermal method using Co²⁺ and Ni²⁺ as metal centers and 1,4-dicarboxybenzene as the organic ligand in a binary solvent system (15 mL DI water: 15 mL DMF) [18]. Subsequently, the obtained Co₃Ni₁BDC precursor was transferred into a tube furnace for phosphating under an Ar atmosphere, leading to the formation of the Co₃Ni₁BDC @Co₂P heterostructure on the surface of the partially decomposed Co₃Ni₁BDC. To confirm the successful synthesis before and after the Co₃Ni₁BDC treatment, X-ray diffraction (XRD) and Raman spectroscopy were performed. As shown in Figure 1a, the XRD patterns of Co₃Ni₁BDC and CoBDC matched the desired simulated structure (CCDC no. 985792). The corresponding Raman spectra are presented in Figure 1b. Peaks at 328 cm⁻¹ and 386 cm⁻¹ were attributed to the coordination bonds of Ni-O and Co-O. The peaks at 636 cm⁻¹, 815 cm⁻¹, and 866 cm⁻¹ corresponded to aromatic C-H bonds, while C-C bonds appeared at 1138 cm⁻¹ and 1184 cm⁻¹. The peaks at 1423 cm⁻¹, 1564 cm⁻¹, and 1616 cm⁻¹ for Co₃Ni₁BDC and CoBDC were associated with symmetric (νs) and asymmetric (νas) (-COO⁻) vibrations [19]. The similar peaks observed in Co₃Ni₁BDC indicated that Ni doping did not affect the crystalline structure. Figure 1c shows typical SEM images of Co₃Ni₁BDC, revealing a layered nanoarray structure. Compared to CoBDC, the Ni-doped treatment significantly affected the alignment structure. With an increased Ni/Co ratio, Co₂Ni₂BDC directly formed a nanosheet structure (Figure S1). Figure 1d shows a typical TEM image of the clear nanosheet layer structure. The corresponding energy-dispersive X-ray spectroscopy (EDX) mapping images indicate that the as-prepared Co₃Ni₁BDC exhibited a unique layered nanosheet structure with Ni, Co, C, and O elements homogeneously distributed throughout the nanostructure (Figure 1e₁–e₄).

After annealing, the powder XRD patterns at 40.8° and 42.1° were well indexed to Co₂P (JCPDS 32-0306) (Figure 2a), and nanosheets with a rougher surface are observed in Figure 2b. Small nanoparticles were observed on the surface of the Co₃Ni₁BDC@Co₂P nanosheets, forming a core–shell structure (Figure 2c). Figure 2d shows a high-resolution TEM (HRTEM) image of Co₃Ni₁BDC@Co₂P, revealing lattice fringes with a spacing of 0.22 nm, corresponding to the (201) plane. Furthermore, the selected area electron diffraction (SAED) pattern of Co₃Ni₁BDC@Co₂P (Figure 2e) reveals that the shell structure corresponds to the (201) plane of Co₂P. In addition, the SEM-EDS mappings in Figure 2f₁–f₅ demonstrate the homogeneous distribution of C, O, Ni, P, and Co elements.

The elemental composition and chemical states of CoBDC and Co₃Ni₁BDC were systematically investigated by X-ray photoelectron spectroscopy (XPS). As shown in the survey spectra (Figure 3a), both materials contained Ni, Co, O, and C elements. High-resolution analysis of the Co 2p region revealed distinct chemical environments. For CoBDC (Figure 3b), the deconvoluted peaks at 781.48 eV (Co 2p_3/2) and 797.43 eV (Co 2p_1/2) were characteristic of Co²⁺ species in Co-O coordination. In contrast, Co₃Ni₁BDC exhibited Co²⁺ signatures at slightly higher binding energies of 782.04 eV (Co 2p_3/2) and 798.62 eV (Co 2p_1/2), accompanied by satellite peaks at 785.30 eV and 803.66 eV, consistent with reported Co-O configurations in metal–organic frameworks [20]. Notably, the Co 2p_3/2 peak in Co₃Ni₁BDC showed a positive shift of 0.56 eV compared to CoBDC, suggesting enhanced electron density redistribution induced by Ni doping. This electronic modulation facilitated reactant adsorption/desorption during OER, thereby reducing activation energy barriers and improving electrocatalytic activity. The Ni 2p spectrum further confirmed the successful incorporation of Ni, with dominant peaks at 856.70 eV (Ni 2p_3/2) and 874.47 eV (Ni 2p_1/2), corresponding to Ni²⁺ oxidation states [21,22]. In the spectra of O 1s, three peaks could be clearly observed (Figure 3d). More specifically, the peaks at 530.5 eV corresponded to M-O, while the peaks at 531.82 eV could be attributed to -COOH and those at 533.15 eV to H₂O.

2.2. Oxygen Evolution Reaction Performance

The electrocatalytic behavior of Co₃Ni₁BDC is further confirmed by the linear sweep voltammetry (LSV) curves shown in Figure 4a. The overpotential at a current density of 100 mA cm⁻² for Co₃Ni₁BDC (300 mV) was significantly lower than that of pristine CoBDC (335 mV). Figure 4b shows that Co₃Ni₁BDC exhibited a smaller charge transfer resistance (R_ct) compared to IrO₂, a noble metal benchmark catalyst, further promoting electron transfer. Co₃Ni₁BDC exhibited the lowest Tafel slope of 88.33 mV dec⁻¹, which was smaller than those of Co₂Ni₂BDC (105.56 mV dec⁻¹), CoBDC (96.44 mV dec⁻¹), and IrO₂ (124.35 mV dec⁻¹) (Figure 4c). Additionally, the Ni-doped CoBDC demonstrated excellent OER stability, with no noticeable performance degradation even after 15 h of operation (Figure 4d). Cyclic voltammetry (CV) curves (Figure S2a,b) were recorded in the non-Faradaic potential region at different scan rates to calculate the double-layer capacitance (C_dl) and estimate the electrochemical surface area. As shown in Figure S2c, Co₃Ni₁BDC demonstrated a higher C_dl of 0.35 mF cm⁻² compared to CoBDC (0.27 mF cm⁻²), revealing that Ni doping significantly enhanced the electrochemically active surface area. The ECSA-normalized LSV curves provided insights into the intrinsic activities of the Co₃Ni₁BDC and CoBDC catalysts (Figure S2d).

To gain deeper insights into surface evolution during electrocatalysis, we performed an XPS analysis on Co₃Ni₁BDC before and after the OER test. As evidenced by the comparative spectra in Figure S3, the post-OER samples exhibited newly emerged Co³⁺ and Ni³⁺ signatures in their respective 2p regions. This oxidation state elevation aligns with the spontaneous formation of M-OOH intermediates (M = Co/Ni), widely recognized as catalytically active species in OER [23,24]. Notably, the intensified M-O peak area further corroborated the in situ generation of M-OOH phases. Crucially, despite such chemical transformations, the SEM images (Figure S4) confirmed that Co₃Ni₁BDC maintained its original nanosheet architecture even after prolonged operation, demonstrating exceptional structural robustness essential for sustained catalytic performance.

2.3. Hydrogen Evolution Reaction Performance

For hydrogen production, the iR-corrected LSV curves of HER demonstrated that Co₃Ni₁BDC@Co₂P exhibited the best HER performance in 1 M KOH. Figure 5a presents the LSV data of Co₃Ni₁BDC and Co₃Ni₁BDC@Co₂P to evaluate their HER performance. It is evident that the Co₃Ni₁BDC@Co₂P sample had the best HER performance, exhibiting the lowest overpotential (127 mV) at 100 mA cm⁻². The annealed samples exhibited a smaller charge transfer resistance (R_ct) compared to Co₃Ni₁BDC, facilitating electron transfer and further enhancing the electrode reaction kinetics (Figure 5b). Figure 5c shows that the annealed samples (77.65 mV dec⁻¹) exhibited slightly lower Tafel slopes than Co₃Ni₁BDC (105.78 mV dec⁻¹). Additionally, the annealed samples demonstrated excellent HER stability, with no significant performance degradation after 15 h of operation (Figure 5d). Figure S5a,b show the electrochemical double-layer currents obtained from the CV data of Co₃Ni₁BDC and Co₃Ni₁BDC@Co₂P at scan rates of 20–100 mV s⁻¹. The annealed samples exhibited an increased electrochemical surface area, leading to a higher C_dl. Co₃Ni₁BDC@Co₂P achieved a larger C_dl value of 10.26 mF cm⁻², approximately 10 times that of Co₃Ni₁BDC (Figure S5c). Figure S5d shows the ECSA-normalized HER polarization curves of Co₃Ni₁BDC@Co₂P.

The XPS survey in Figure S6 indicates the presence of P and Co signals in the sample of Co₃Ni₁BDC@Co₂P. Meanwhile, the Co-P structure of Co₃Ni₁BDC@Co₂P was consistent with the peak at 781.94 eV, which confirmed that P atoms were introduced into the as-prepared electrocatalyst via annealed treatment. The hierarchical structure derived from Co₃Ni₁BDC@Co₂P resulted in a large active surface area and abundant pores, which could promote the exposure of catalytically active sites and edge defects and reduce the diffusion pathways for charge and mass transfer, thereby enhancing HER performance. The peaks of P 2p at binding energies of 129.95 and 131.5 eV can be indexed to P 2p_3/2 and P 2p_1/2 [25,26]. Binding energies around 135.1 eV correspond to P-O as a result of surface oxidation. After stability testing, the P 2p orbit changed, and the corresponding peaks of P 2p_3/2 and 2p_1/2 disappeared due to the low content. There was no significant change in Co 2p, proving its excellent stability. Figure S7 shows the SEM images of Co₃Ni₁BDC@Co₂P after cycling, the catalyst still maintaining a good nanosheet structure.

2.4. Overall Water Splitting Performance

Encouraged by the efficient catalytic activity of Co₃Ni₁BDC and Co₃Ni₁BDC@Co₂P for both OER and HER, an alkaline electrolyzer was built using the Co₃Ni₁BDC//Co₃Ni₁BDC@Co₂P catalyst as both the cathode and the anode for water splitting (Figure 6a). A current density of 20 mA cm⁻² was achieved with a cell voltage of 1.70 V, which was not only better than that of IrO₂//Pt/C (1.75 V) but also better than other catalysts of the same type, compared in Table S1 [27,28,29,30,31,32,33,34]. The stability of Co₃Ni₁BDC//Co₃Ni₁BDC@Co₂P was examined by chronoamperometry (Figure 6b). It was found that the cell voltage hardly decayed after 16 h. Figure 6c shows that experimental value matched well with theoretical values, and the volume of the generated H₂ was nearly two times as large as that of the generated O₂.

3. Experimental Section

3.1. Materials

For this study, 1,4-dicarboxybenzene (H₂BDC, 98%), Co(NO₃)₂·6H₂O, Ni(NO₃)₂·6H₂O, and KOH (≥90%) were purchased from Shanghai Titan Technology Co., Ltd (China). Meanwhile, N, N-Dimethylformamide (DMF, 99.5%, AR) was purchased from Sinopharm Chemical (China) and nick foam (1.7 mm thick) was purchased from Kunshan Longshengbao Electronic Materials Co., Ltd (China). All chemicals were used as received without further purification.

3.2. Preparation of Co₃Ni₁BDC and Co₃Ni₁BDC@Co₂P

In a standard preparation procedure, nickel foam (NF) substrates (1.0 cm × 2.0 cm) underwent sequential cleaning with 3.0 M HCl, acetone, and deionized water, followed by drying at 60 °C for 2 h. The Co₃Ni₁MOF/NF composite was prepared through a solvothermal approach by combining stoichiometric amounts of Co(NO₃)₂·6H₂O, Ni(NO₃)₂·6H₂O, and 1,4-dicarboxybenzene (H₂BDC). Specifically, a precursor solution containing 336.75 mg cobalt nitrate hexahydrate, 112.25 mg nickel nitrate hexahydrate, and 332 mg organic linker was formulated in a binary solvent system (15 mL DI water: 15 mL DMF). This homogeneous mixture was sealed in a 50 mL Teflon-lined autoclave with pretreated NF substrates and maintained at 120 °C for 12 h. Subsequent cooling to ambient temperature enabled the collection of Co₃Ni₁BDC crystals through centrifugal separation, followed by triple ethanol rinsing and vacuum-drying at 30 °C. Analogous synthetic routes with adjusted metal salt ratios were employed to fabricate Co₂Ni₂BDC/NF and CoBDC/NF derivatives.

A phosphorus precursor (0.5 g sodium hypophosphite) was positioned upstream in the tubular reactor, while the as-prepared Co₃Ni₁BDC/NF was loaded downstream. The assembly underwent thermal annealing under argon protection with a programmed temperature protocol: heating at 5 °C·min⁻¹ to 300 °C followed by 2 h isothermal treatment. This phosphorization process yielded the hierarchical composite Co₃Ni₁BDC@Co₂P/NF as the final product.

3.3. Characterization

Crystalline phase analysis was conducted on a X’Pert PRO MPD diffractometer (Panalytical, Netherlands) with Cu Kα radiation (λ = 1.5406 Å). Structural validation was achieved through complementary spectroscopic techniques: Raman measurements (Horiba Labram HR800 Evo, Longjumeau, France, 532 nm laser excitation) coupled with FT-IR characterization (SPECTRUM 100 system). Morphological evolution and elemental distribution were investigated via field emission scanning electron microscopy (Quanta 450, FEI) integrated with energy-dispersive X-ray spectroscopy. For nanoscale structural interrogation, TEM observations were performed on a TECNAI F30 microscope (300 kV accelerating voltage), with specimens prepared by ultrasonic dispersion of catalyst nanosheets in anhydrous ethanol (30 min) followed by drop-casting onto copper grids. Surface chemical states were determined using an ESCALAB250Xi XPS system (Thermo Scientific) with Mg Kα irradiation (1253.6 eV), where binding energy calibration utilized adventitious carbon (C 1s = 284.6 eV) as the internal reference.

3.4. Electrochemical Measurements

Electrochemical measurements were conducted using a Gamry INTERFACE 1010E workstation (Warminster, PA, USA) with a conventional three-electrode configuration. The working electrode comprised catalyst-coated nickel foam (NF), paired with a platinum sheet counter electrode and an Hg/HgO reference electrode. All tests utilized 1 M KOH as the electrolyte. LSV for HER was performed from −0.5 to 0.1 V vs. RHE, while OER LSV scans spanned 0.8–2.0 V vs. RHE, both at 5 mV s⁻¹. Electrochemical impedance spectroscopy (EIS) analysis covered 0.1 Hz–100 kHz with a 5 mV AC perturbation. Double-layer capacitance values were derived from cyclic voltammetry (CV) scans at varying rates (10–100 mV s⁻¹).

Stability assessments employed chronopotentiometry at a fixed current density of 100 mA cm⁻². For overall water splitting evaluation, a two-electrode configuration was implemented using Co₃Ni₁BDC@Co₂P/NF (HER) and Co₃Ni₁BDC (OER) in 1 M KOH, with polarization curves acquired at 2 mV s⁻¹. Faraday efficiency calculations followed:

n(O₂) = Q/nF (Q = it)

where Q represents the charge transfer, n denotes the electron transfer number (4 for OER, 2 for HER), and F is Faraday’s constant (96,485 C/mol). Experimental validation used a drainage method with 50 mA current over 1 h electrolysis. the Measured potentials were converted to RHE scale via E_RHE = E_Hg/HgO + 0.098 + 0.059 × pH, with iR compensation applied.

4. Conclusions

In conclusion, this work demonstrates a rational design of hierarchical Ni-doped CoBDC electrocatalysts with interconnected nanosheet architectures grown on nickel foam. The strategic Ni incorporation induces favorable electronic modulation within the Co-based MOF, endowing the catalyst with exceptional OER activity (η₁₀₀ = 300 mV) and robust stability through optimized charge transfer kinetics. Through a precisely controlled partial phosphating process, we successfully constructed a core–shell Co₃Ni₁BDC@Co₂P heterostructure, where the Co₂P protective layer not only enhanced electrical conductivity but also created abundant active interfaces for the HER. Remarkably, the derived catalyst achieved ultralow HER overpotentials of η₁₀ = 20 mV and η₁₀₀ = 127 mV, representing a significant improvement over the parent Co₃Ni₁BDC material. The synergistic integration of these optimized components in a Co₃Ni₁BDC//Co₃Ni₁BDC@Co₂P alkaline electrolyzer enabled exceptional full water splitting performance, requiring only 1.70 V to drive 20 mA cm⁻²—a substantial advancement compared to conventional noble metal-based systems. This study highlights the effectiveness of combining elemental doping with interface engineering in MOF-derived catalysts, providing new insights into the development of cost-effective, durable electrocatalysts for sustainable hydrogen production. The proposed multi-step synthesis strategy, encompassing structural design, composition optimization, and surface modification, establishes a versatile paradigm for tailoring MOF-based materials toward efficient energy conversion applications.