The energy crisis and environmental pollution are serious issues caused by the continuous consumption of fossil fuels [1]. Developing renewable and clean sources of energy to replace traditional fossil fuels has become increasingly necessary [2-4]. Hydrogen production via electrochemical water splitting is considered to be an ideal method to solve these issues. The oxygen evolution reaction (OER) is a bottleneck for overall water splitting due to its sluggish reaction kinetics [5-7]. Therefore, efficient electrocatalysts should be developed for the OER to decrease the overpotential and increase the reaction rate. Although some state-of-the-art noble metal catalysts (e.g., IrO₂ or RuO₂) exhibit high electrocatalytic performance, their high cost and scarcity prohibit large-scale application [2, 8, 9]. Therefore, significant efforts have been devoted to developing low cost and earth-abundant electrocatalysts, including non-noble transition metal-based oxides [10], (oxy)hydroxides [11, 12], phosphides [13-15], nitrides [16], sulfides [17-20], and selenides [21-23]. Among them, Co-based materials have exhibited promising performance for the OER because oxidized cobalt centers can promote the formation of OOH species and subsequent deprotonation to form O₂ [24]. In addition, when Fe is doped into Co-based catalysts, the OER performance can be further improved [25, 26].

Metal-organic frameworks (MOFs), also known as porous coordination polymers, have been widely applied to medicine delivery [27], gas storage and separation [28, 29], catalysis [30], and sensing [31] due to their large specific surface areas, tunable pore sizes, periodic structures, and diversity of metal centers and organic linkers. For electrochemical water splitting, MOFs have been used as precursors and templates for synthesizing oxide/carbon composites with high OER activity [32-35]. However, tedious fabrication processes and high pyrolysis temperatures are generally required, which is unsuitable for practical applications. Consequently, the direct use of MOFs as electrocatalysts toward OER is desirable.

Herein, we report a facile ultrasonic method for fabricating Co/Fe-based bimetal-organic frameworks as highly efficient water oxidation catalysts in alkaline solution. The OER performance of the Co_xFe-MOFs followed a volcano-type pattern with increasing Fe content. The optimized Co_xFe-MOFs exhibited high catalytic activity with an overpotential of 280 mV at 10 mA cm^-2 and a Tafel slope of 44.7 mV dec^-1. This performance can be ascribed to the large specific surface area, excellent conductivity, and synergistic effects of Co and Fe.

CoCl₂·6H₂O was purchased from Tianjin Guangfu Technology Development Co., Ltd. FeCl₃·6H₂O, trimesic acid (H₃BTC), and RuO₂ were purchased from Aladdin. Triethylamine ((C₂H₅)₃N) and N, N-dimethylformamide (DMF) were purchased from Tianjin Fuyu Fine Chemical Co. Nafion (5%) was obtained from Sigma Aldrich. All reagents were used without further purification, and deionized water was used throughout the experiments.

The Co_xFe-MOFs were synthesized according to a previously reported method with some modifications [36]. In a typical synthesis, H₃BTC (0.75 mmol), CoCl₂·6H₂O (0.5 mmol), and FeCl₃·6H₂O (0.25 mmol) were dissolved in a mixed solution of DMF (32 mL), ethanol (2 mL), and deionized water (2 mL) under vigorous stirring. Subsequently, triethylamine (0.8 mL) was added rapidly to the above solution. The mixture was stirred for 5 min and continuously ultrasonicated for 8 h at room temperature. Finally, the products were collected via centrifugation, washed with ethanol three times, and dried at 60 ℃ overnight. A similar procedure was adopted for the preparation of the Co-, Co₃Fe-, Co₁Fe-, and Fe-MOFs using different molar ratios of Co/Fe chloride salts (1:0, 3:1, 1:1, 0:1) at the beginning with the total amount of metal salt held constant at 0.75 mmol.

Field-emission scanning electron microscopy (FESEM) was performed using a NOVA NanoSEM 450 instrument at 10 kV. Transmission electron microscopy (TEM), HRTEM, and EDS were performed using a JEM2100F TEM instrument at an acceleration voltage of 200 kV. X-ray diffraction (XRD) was performed using a D/max-2400 diffractometer with Cu K_α radiation (λ = 154.1 nm). X-ray photoelectron spectroscopy (XPS) was recorded on an ESCALAB 250Xi spectrometer with a monochromated Al K_α radiation source. Fourier transform infrared spectra (FT-IR) were collected using a NEXUS EURO instrument. Inductively coupled plasma-atomic emission spectrometry (ICP-AES) was performed with an Optima2000DV instrument from PerkinElmer company. The BET specific surface area was determined using an AUTO SORB-1-MP apparatus.

To prepare the working electrode, 5 mg of the prepared catalysts were added into a mixed solution containing 480 μL of water, 480 μL of ethanol, and 40 μL of a 5% Nafion solution. Subsequently, the mixed solution was sonicated for 30 min to obtain a homogeneous catalyst ink. Finally, 5 μL of the catalyst ink was immobilized on glassy carbon by drop-casting and dried at room temperature. All electrochemical measurements were performed using a standard three-electrode system (CHI 760E electrochemical station) at room temperature. The electrolyte was a 1 mol L^-1 KOH solution. A glass carbon electrode was utilized as the working electrode, and Hg/HgO and Pt wire were used as the reference and counter electrodes, respectively. All potentials were calibrated to the reversible hydrogen electrode (RHE) using the following equation: E(RHE) = E(Hg/HgO) + 0.098 V + 0.059PH. Linear sweep voltammetry (LSV) was recorded at a scan rate of 5 mV s^-1 with 95% iR-compensation. Electrochemical impedance spectroscopy (EIS) was performed at a frequency ranging from 10 kHz to 0.1 Hz at 1.53 V vs. RHE. The durability test was conducted by cyclic voltammetry (CV) from 1.1 to 1.7 V vs. RHE at a scan rate of 100 mV s^-1 for 500 cycles. The chronopotentiometry was measured at a current density of 10 mA cm^-2. To estimate the electrochemical active surface area (ECSA) of the catalyst, the electrochemical double layer capacitance (Cdl) was measured via CV in the non-Faradaic potential range of 0.1 to 0.2 V vs. Hg/HgO with scan rates of 10–50 mV s^-1. The turnover frequency (TOF) was calculated using the following formula: TOF = jA/4nF, where j is the current density at an overpotential of 280 mV, A is the geometric area of the GC electrode, 4 is the number of electrons transferred to produce one molecule of oxygen, n is the number of moles of active sites, and F is Faraday constant (96485 C mol^-1).

The fabrication process of the Co_xFe-MOFs is illustrated in Scheme 1. Triethylamine was used to accelerate the deprotonation of the organic ligands under ultrasonic conditions [37]. The Co_xFe-MOFs were formed by the coordination reaction of Co²⁺/Fe³⁺ with deprotonated trimesic acid (H₃BTC). The structural information of the as-prepared Co-, Co_xFe-, and Fe-MOFs was obtained by XRD, as shown in Fig. 1a. With increasing Fe content, the crystal structure of the Co_xFe-MOFs gradually changed from Co-MOF to Fe-MOF. FT-IR spectroscopy was performed to characterize the functional groups in the Co₂Fe-MOF (Fig. 1b). The free trimesic acid ligand was also analyzed by FT-IR for comparison. In the Co₂Fe-MOF, the absorption band at 3426 cm^-1 was attributed to the stretching vibration of the O–H motifs. The bands between 1300 and 1650 cm^-1 arose from the stretching vibrations of C=O and C–C bonds, respectively, on the benzene rings [38]. In particular, the bands at 1622 and 1371 cm^-1 can be ascribed to the respective asymmetric and symmetric stretching vibration bands of the carboxyl groups bound to metal ions. For comparison, the free trimesic acid ligand showed a stretching vibration of the carboxyl group at 1721 cm^-1. The difference in asymmetric vibration indicates that no free carboxylic groups were present in Co₂Fe-MOF [39].

The morphology and microstructure of Co₂Fe-MOF were characterized by SEM and TEM. The SEM images (Fig. 2a and 2b) indicate that Co₂Fe-MOF is composed of nanoparticles with an average size of approximately 50–100 nm. These nanoparticles were loosely packed into a porous structure, which is favorable for exposing more active sites for the OER. In the high-resolution TEM image of Co₂Fe-MOF (Fig. 2d), no obvious lattice fringes were observed due to the easily damaged MOF structure under high energy electron beam irradiation [40]. This result is consistent with the observation of weak circle rings shown at the top-right selected area electron diffraction (SAED) pattern. As shown by high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM; Fig. 2e), the EDS element mapping revealed a uniform distribution of Co, Fe, C, and O throughout the entire catalyst without impurities. EDS elemental analysis confirmed a molar ratio of 2:1 for Co: Fe (Fig. S1), consistent with the inductively coupled plasma (ICP) analysis (Table S1).

XPS measurements were obtained to investigate the surface composition and chemical valance states of the as-obtained Co₂Fe-MOF. The XPS survey spectra verified the coexistence of Co, Fe, C, and O (Fig. S2). In the Co 2p region (Fig. 3a), the binding energies of 781.9 and 797.7 eV can be ascribed to Co 2p_3/2 and 2p_1/2 peaks, respectively. In addition, two satellite peaks corresponding to Co 2p_3/2 and Co 2p_1/2 were observed at 786.5 and 803.2 eV, respectively. These results suggest the presence of Co²⁺ bound to oxygen [41]. The Fe 2p region shown in Fig. 3b exhibited two peaks at 711.5 and 725.3 eV, which were assigned to Fe 2p_3/2 and Fe 2p_1/2, respectively. Moreover, a satellite peak at 716.2 eV was observed, indicating the +3 valance state for iron [42]. The C 1s spectrum was deconvoluted into three peaks (Fig. 3c). The peak at 284.8 eV was attributed to C=C bond of the benzoic ring, the 288.4 eV peak arose from the carboxyl group (O=C–O) of BTC^3-, and the 286.0 eV peak was assigned to C–O [39, 43]. In the high-resolution XPS spectrum of O 1s (Fig. 3d), three peaks at 530.8, 531.6 and 532.7 eV were observed, arising from the coordinated oxygen, carbonyl group, and absorbed water, respectively [39]. Compared to those of the Co-MOF, the binding energies of the Co 2p_3/2 and Co 2p_1/2 spectra of Co₂Fe-MOF were positively shifted by 0.3 eV, indicating decreasing electron density at the Co centers. Furthermore, the Fe 2p_3/2 and Fe 2p_1/2 peaks were negatively shifted by 0.3 eV compared to those of Fe-MOF, suggesting an electron donation effect from Co to Fe (Fig. S3). The above results indicate that Fe incorporation can regulate the electronic environment of the Co centers. N₂ adsorption-desorption isotherms were recorded to calculate the specific surface area of Co₂Fe-MOF (Fig. S4). A IV-type isotherm with H3-type hysteresis loop indicated the existence of mesopores in Co₂Fe-MOF [44]. The specific surface area of Co₂Fe-MOF was determined to be 36.6 m² g^-1 based on the Brunauer-Emmett-Teller (BET) method. Large surface areas and mesoporous structures are favorable for the improved OER activity.

The electrocatalytic activities of the Co_xFe-MOFs toward the OER were evaluated by LSV in a typical three-electrode electrochemical cell with 1 mol L^-1 KOH as the electrolyte. A GC electrode with drop-casted catalysts was used as the working electrode and Hg/HgO and Pt wire were used as the reference and counter electrodes, respectively. Unless otherwise stated, the ohmic potential drop (iR) arising from solution resistance compensation was applied to all experimental data and all potentials were calibrated against the reversible hydrogen electrode (RHE). For comparison, bare GC and commercial RuO₂ electrodes were tested under the same conditions. As shown in Fig. 4a, the bare GC and Fe-MOF exhibited negligible OER activity. In contrast, the bimetal MOFs showed a significantly enhanced catalytic activity, unlike single metal materials. For the optimized Co₂Fe-MOF activity, an overpotential of 280 mV is required to reach a current density of 10 mA cm^-2, which is lower than those required for Co₁Fe-MOF (297 mV), RuO₂ (307 mV), and Co₃Fe-MOF (314 mV). This value is also lower than most recently reported Co/Fe-based OER electrocatalysts (Table S2). Meanwhile, Co₂Fe-MOF exhibited a lower Tafel slope of 44.7 mV dec^-1 than those of the Co₁Fe-MOF (48.9 mV dec^-1), Co₃Fe-MOF (55.6 mV dec^-1), Co-MOF (69.9 mV dec^-1), RuO₂ (82.0 mV dec^-1), and Fe-MOF (176.2 mV dec^-1) samples, reflecting the favorable kinetics of Co₂Fe-MOF toward the OER.

EIS analysis was conducted to investigate the electrochemical behavior of the prepared electrodes. The Co₂Fe-MOF exhibited a smallest semicircle diameter among these catalysts, as shown in the Nyquist plots (Fig. 4c), indicating that it possessed the lowest charge transfer resistance (R_ct). Additionally, the ECSAs were evaluated by calculating Cdl based on CV in the non-Faradaic region (Figs. S5 and S6). The Co₂Fe-MOF exhibited a lower Cdl value (2.33 mF cm^-2) compared to those of the Co-MOF (6.28 mF cm^-2) and Co₃Fe-MOF (3.96 mF cm^-2), indicating a superior intrinsic OER activity for the Co₂Fe-MOF. We further calculate the TOF to evaluate the intrinsic activity of the catalysts. Assuming all Co and Fe atoms are involved in the OER, the TOF value of Co₂Fe-MOF was 0.032 s^-1, higher than those of Co₁Fe-MOF (0.011 s^-1), Co₃Fe-MOF (0.008 s^-1), Co-MOF (0.002 s^-1), and Fe-MOF (0.0007 s^-1).

Stability is another important parameter for the practical application of electrocatalysts. Fig. 5a shows the LSV curves of the Co₂Fe-MOF before and after 500 CV scans at a scan rate of 100 mV s^-1. Only slight current density attenuation was observed for the polarization curves after 500 cycles, indicating the superior stability of Co₂Fe-MOF for the OER. From the chronopotentiometric curve in Fig. 5b, only a slight increase in the applied potential was required for the Co₂Fe-MOF to maintain a current density of 10 mA cm^-2 for a duration of 12000 s. The I-t curve of the Co₂Fe-MOF at an applied potential of 1.58 V vs. RHE is provided in Fig. S7. The observed slight decrease in catalytic activity was attributed to the oxygen bubble that was adsorbed on the electrode, preventing the intimate contact between the catalyst and electrolyte and detachment of the catalyst from the working electrode surface.

We synthesized a series of Co_xFe-MOFs with different Co/Fe molar ratios, using a simple ultrasonic method, as efficient electrocatalysts for the OER. The Co₂Fe-MOF showed the best catalytic performance among the prepared catalysts with a low overpotential of 280 mV at 10 mA cm^-2 and a Tafel slope of 44.7 mV dec^-1. Moreover, the Co₂Fe-MOF exhibited long-term durability for at least 12000 s. Its high OER performance can be attributed to the synergistic effects of Co²⁺ and Fe³⁺, mesoporous structure, and enhanced charge transfer rate. This study provides a promising method for examining highly active MOF materials for water splitting applications.