The depletion of fossil fuels and the associated serious environmental problems are the motivations for the intense research on new energy sources, including hydrogen [1-3]. As one of the most promising alternatives to fossil fuels, hydrogen exhibits a high energy density and is environmentally friendly [4-6]. Electrocatalytic water splitting, involving the hydrogen (2H⁺ + 2e^– → H₂) and oxygen (2H₂O → O₂ + 4H⁺ + 4e^–) evolution reactions, is one of the most appealing approaches for extracting molecular hydrogen from water [7, 8]. In these two electrochemical reactions, the ideal free energy change, ∆G₀, for the conversion of one H₂O molecule into H₂ and 1/2 O₂ under standard temperature and pressure is 237.2 kJ mol^–1, according to the Nernst equation, which corresponds to a reversible electrochemical cell voltage of 1.23 V [9]. However, to overcome the kinetic limitations of the actual electrochemical reaction, an overpotential is required to drive water splitting in an electrolyzer. Electrocatalysts can be used to promote the kinetics of the water electrolysis reaction and lower the required overpotential. For example, Ru and Ir oxides, such as RuO₂ and IrO₂, are well-known traditional oxygen evolution reaction (OER) catalysts, and Pt is considered as the state-of-the-art electrocatalyst for the hydrogen evolution reaction (HER) [10-12]. However, the high costs and limited abundances of these noble-metal-based materials seriously restrict their wide application in the industry. It is therefore desirable and urgent to design new efficient electrocatalysts based on materials with abundant resources. To date, a variety of alternative materials, including 3d transition metal (hydro)oxides (TMOs) [13], sulfides [14], and nanocomposites [15], have been reported for the overall water splitting that exhibit good electrocatalytic activities.

In the past decade, diverse nanostructures of Fe, Co, and Ni oxides, such as Co₃O₄ nanosheets, NiO nanosheets, and FeCo₂O₄, have been reported to be good electrocatalyst choices for both the HER and OER owing to the advantages of low toxicity, low cost, rich redox properties, and easy production [16-18]. However, their catalytic performances in terms of the overpotentials required for hydrogen and/or oxygen evolution and the long-term stabilities still need to be improved for these materials to compete with the noble metal electrocatalysts. The morphologies, chemical compositions, and electronic properties of TMO nanostructures should be elaborately tuned to obtain more efficient catalysts that display the desired catalytic performances. Recently, some studies have focused on non-noble-metal composites with minimal noble metal contents for achieving a high electrocatalytic activity. For example, Dong and coworkers [19] synthesized Ag-doped NiCoO mesoporous nanorods as a catalyst for the OER that reveal a high current density (140 mA cm^–2) at a low overpotential (370 mV). In addition, Zhu et al. [20] reported that Au/Ni//Fe LDH hybrid OER catalysts can display an ultrahigh current density of 500 mA cm^–2 at an overpotential of only 270 mV in alkaline medium. These reports suggested that the incorporation of Au, Ag, and other noble metals can modulate the TMO electronic structures and produce synergetic chemical coupling effects between the TMO catalysts and the dopant atoms, thus improving their catalytic performances. In addition, in situ growth of TMO nanowire arrays on unique 3D open metal substrates such as Ni and Cu foams results in a high adhesive strength of each of the active units participating in the HER and/or OER process [21, 22].

Here, Co_xNi_1-xO nanowire arrays with different Ir contents were fabricated on Ni foam through a combination of hydrothermal synthesis and subsequent calcination treatment. During the synthesis, the earth-abundant and freestanding Co_xNi_1-xO nanowire arrays were used as structural frameworks for diluting the incorporation of Ir (0.2 wt%–0.68 wt%). A Ni foam with a huge supporting area and a high electrical conductivity served as both a 3D substrate and Ni source for in situ growth of Co_xNi_1-xO nanowire arrays on the foam. When used as bifunctional electrocatalysts in the overall water splitting, the performance of the Ir-Co_xNi_1-xO/NF electrodes was found to depend strongly on the Ir content of Co_xNi_1-xO. The optimized Ir-Co_xNi_1-xO nanowire array electrocatalysts with an Ir content of only 0.57 wt% (denoted as Ir(0.57 wt%)-Co_xNi_1-xO) displayed the highest OER and HER activities, with low overpotentials of ~260 and ~53 mV, respectively, and a current density of 10 mA cm^–2 in an alkaline solution. Density functional theory (DFT) calculations were performed to determine why these Ir-incorporated Co_xNi_1-xO nanowire arrays exhibited enhanced electrocatalytic activities. The present work revealed the importance of precisely tuning TMOs through noble metal doping to obtain low-cost high-performance electrocatalysts.

Cobalt(II) nitrate hexahydrate (Co(NO₃)₂·6H₂O, AR, ≥99%), ammonium fluoride (NH₄F, AR, ≥96%), and urea (CO(NH₂)₂, AR, ≥99%) were purchased from Chemical Supply Co., Ltd. (Kent, UK). Ir chloride hydrate (IrCl₃·xH₂O, powder, Ir > 52%) was obtained from Shanghai Macklin Reagent Co., Ltd. (Shanghai, China). These chemicals were used as-received without further purification. All the experiments used ultrapure deionized water (18.2 MΩ·cm).

Ir-Co_xNi_1-xO nanowire arrays were synthesized on Ni foam by hydrothermal method, which was followed by a calcination treatment. Before the synthesis, the Ni foam (1 cm × 3 cm × 0.1 cm) was pretreated with acetone and concentrated HCl (3 mol L^–1) for 10 min each and washed thoroughly with deionized water to obtain clean surfaces. In a typical synthesis, 0.262 g Co(NO₃)₂·6H₂O was dissolved in 30 mL distilled water and stirred for 5 min to ensure homogeneity. Then, 0.1 g NH₄F and 0.162 g urea were added with stirring. Subsequently, a diluted aqueous IrCl₃ solution (0.002 mol L^–1) was slowly injected by using a pipetting gun. To obtain Ir-Co_xNi_1-xO with different Ir contents, the IrCl₃ solution was added in volumes of 500, 1000, and 1500 μL. Subsequently, the mixed solution and a piece of the cleaned Ni foam were transferred to a 50 mL Teflon-lined stainless steel autoclave, which was sealed and maintained at 120 ℃ for 18 h. After cooling to room temperature, the Ni foam was removed, rinsed with deionized water, and dried at 60 ℃ overnight. Finally, the resulting products were calcined in Ar atmosphere at 400 ℃ for 2 h. Co_xNi_1-xO control samples, which contained no Ir, were also prepared by the same procedure in the absence of IrCl₃ solution.

The synthesized products were first characterized by scanning electron microscopy (SEM; FEI Co., Hillsboro, OR, USA). Powder X-ray diffraction (XRD) analyses were conducted with a Rigaku Dmax 2500 X-ray diffractometer equipped (Rigaku Corp, Tokyo, Japan) with Cu K_α radiation (λ = 1.541874 Å). Transmission electron microscopy, high-resolution transmission electron microscopy (TEM and HR-TEM, respectively), and energy dispersive X-ray elemental mapping analyses (EDX) were performed on an FEI F20 (FEI Corp.) at 200 kV. The Ir contents of the samples were determined by inductively coupled plasma-mass spectrometry (ICP-MS). X-ray photoelectron spectroscopy (XPS) was performed on a PHI-5000 X-ray photoelectron spectrometer (Physical Electronics, Inc., Chanhassen, MN, USA) using Al K_α X-rays as the excitation source.

All electrochemical measurements were performed using the standard three-electrode system that was connected to a CHI760E electrochemical workstation (CHI Instruments, Inc., Austin, TX, USA), with 1 mol L^–1 KOH (pH 13.7) serving as the electrolyte. Graphite rods were used as the counter electrode and 3 mol L^–1 KCl saturated Ag/AgCl (Tianjin Aida, China) was employed as the reference electrode. The samples on Ni foam were directly used as the working electrodes. Before the tests, the electrolyte was bubbled with high-purity N₂ (99.999%) for 30 min and the bubbling was maintained throughout the experiment. Linear sweep voltammetry (LSV) polarization curves were recorded at a scan rate of 2 mV s^–1 with 85% iR drop compensation. All the measured potentials (vs. Ag/AgCl) were converted to the reversible hydrogen electrode (RHE) scale, in accordance with the literature [23]. The Tafel plots were fitted using the Tafel equation (η = blog(j) + a), in which η is the overpotential, j the current density, and b the Tafel slope. Electrochemical impedance spectroscopy (EIS) measurements were also carried out in the frequency range 0.01 Hz to 100 kHz with an applied potential at –10 and 10 mA cm^–2 for the HER and OER, respectively. Chronopotentiometric measurements were used to evaluate the long-term stabilities of the electrode materials. The overall water splitting process was carried out by using the Ir-Co_xNi_1-xO/NF electrode as both the anode and cathode in a two-electrode configuration to examine the resulting LSV polarization curve at 1 mV s^–1. The control Co_xNi_1-xO/NF electrodes (without any Ir incorporation) were also tested for the purpose of comparison.

DFT+U calculations were carried out by employing the Vienna ab initio simulation package [24]. Generalized gradient approximation with Perdew-Burke-Ernzerh functional [25] was used to calculate the exchange-correlation energy. The cutoff energy for plane-wave expansion was set at 400 eV and the Brillouin zone sampled using 5 × 5 × 3 gamma-centered Monkhorst-Pack grids. For structural optimization, the crystal structures were fully relaxed until the residual force was < 0.01 eV Å^–1. The electronic properties of the materials were studied by considering a Co_xNi_1-xO crystal structure with one Ir atom substituting one Co atom to examine the effects of Ir doping. In order to better describe the 3d electrons of the Co and Ni atoms, the effective U values of 3.7 and 6.45 eV were used, respectively. For calculation of the Gibbs free energy [26, 27], the (001) surface of Ir-Co_xNi_1-xO was used as an example, as it is a common surface of crystalline CoO, in which the Co atoms in the plane were selectively substituted with Ni and Ir atoms to form an alloy surface. The adsorption energies on different sites on the (001) plane were calculated using ∆G_X* = E_X* + E_ZPE – TS_X* [28, 29], where X^* represents the adsorbed intermediates formed in the HER and OER, such as *H, *O, *OH, and *OOH, and E_X*, E_ZPE, and TS denote the binding energy, zero-point energy, and entropic contributions, respectively.

The crystal structures of the as-obtained products were first characterized by XRD, which revealed peaks at 36.48°, 42.38°, and 61.48° for all the samples; this corresponded to the rock salt structure of cubic CoO (space group P63/mmc, a = b = c = 3.25 Å, α = 90°, β = 90°, γ = 120°; Fig. 1). Notably, the incorporation of Ir did not alter the crystal structure and all the specimens still exhibited the cubic structure. A slight shift to higher angles was observed for the diffraction peaks of the Co_xNi_1-xO sample, in comparison with those of stoichiometric CoO (PDF no. 78-0431), which was attributed to the lattice contraction caused by partial Co ion substitution with the smaller Ni ions [30, 31]. The three strong peaks at 44.5°, 51.8°, and 76.4° in the XRD patterns originated from the Ni substrate (PDF no. 87-0712). The morphology and microstructure of the Ir-Co_xNi_1-xO nanowire arrays on Ni foam were examined by SEM and TEM. The SEM images of the Ir-Co_xNi_1-xO products recorded at low and high magnifications showed needle-like, high-density, and vertically aligned nanowire arrays that uniformly covered the entire Ni foam surface (Figs. 2a and 2b). The average nanowire length was estimated to be ∼5 μm. No obvious morphological differences were observed between the Ir-Co_xNi_1-xO products upon altering the amount of IrCl₃ solution added during the synthesis (Figs. S1a–S1d). The typical TEM images of Ir-Co_xNi_1-xO products clearly revealed highly porous nanowires ~100 nm in diameter (Figs. 2c and 2d). The pores observed in the wires are believed to be generated by the release of H₂O and other gaseous molecules during the calcination process, an effect that is commonly seen in TMO nanostructures obtained from hydroxide precursors [32]. The HR-TEM images indicated that the grains in the nanowires were well-crystallized (Fig. 2e). The resolved lattice spacing of 0.246 nm corresponded to the (111) crystal plane distance in the cubic phase Co_xNi_1-xO [19]. Elemental mapping of Ir-Co_xNi_1-xO revealed a homogeneous distribution of Co, Ni, and O in the composite and quite uniform Ir doping throughout the nanowires (Figs. 2f–2j). The atomic ratio of Co/Ni in the products was close to 0.71/0.29. The real Ir concentrations in the prepared Ir-Co_xNi_1-xO composites were determined by ICP-MS analysis, and found to be 0.2 wt%, 0.57 wt%, and 0.68 wt%, respectively (Table S1).

The surface chemical states of the products were further investigated by XPS. The full XPS spectra collected from the Ir-Co_xNi_1-xO products revealed the presence of Co, Ni, O, and Ir in the composites, which suggested successful synthesis of the Ir-Co_xNi_1-xO samples (Fig. S2). The high-resolution Co 2p and Ni 2p XPS spectra of the Co_xNi_1-xO products with or without Ir doping showed that both were composed of four deconvoluted peaks (Figs. 3a and 3b). The Co 2p core-level spectra exhibited two peaks at 780.52 and 796.16 eV, along with satellite peaks at 786.39 and 802.68 eV, which were assigned to the Co 2p_3/2 and Co 2p_1/2 orbitals, respectively. The spin-orbit split between the two main Co 2p peaks was ~16 eV, which implied that Co existed in the Co²⁺ state [33, 34]. Similarly, the two peaks with the binding energies of 855.09 (Ni 2p_3/2) and 872.88 eV (Ni 2p_1/2) that were observed in the high-resolution Ni 2p spectrum originated from Ni²⁺ [35-37]. In addition, the indiscernible doublet peak feature in the Ir 4f XPS profile suggested low Ir contents of the Co_xNi_1-xO products, which is in accordance with the ICP results (Fig. 3c). Notably, the broad peak at 67.70 eV on the high-energy side of Ir 4f corresponds to the Ni 3s orbital [38]. The XPS spectrum of the O 1s region was fitted with two O peaks (Fig. 3d). The peak at 529.50 eV was ascribed to lattice O, and the peak at 531.10 eV to the surface hydroxyl groups [39, 40]. Further, the XPS spectra of the Ir-containing materials showed slight negative shifts in the peaks of the Co and Ni species, in comparison to those of bare Co_xNi_1-xO (control), which suggested that the electronic structure of Co_xNi_1-xO had been modified by Ir doping [41-43]. Notably, the XPS peak shifts of the doped samples implied changes in the electronic configurations and possibly stronger electronic interactions in Ir-Co_xNi_1-xO, which might have played an important role in improving the electrocatalytic activity [44-46].

The OER catalytic activities of the obtained Ir-Co_xNi_1-xO/NF electrodes were first evaluated in 1 mol L^–1 KOH solution by using the typical three-electrode system. The superiority of the minimal effect of Ir doping was verified by examining the weight proportion of Ir in the Co_xNi_1-xO nanowires, which was found to be controlled at less than ~1 wt%; at the same time, the bare Co_xNi_1-xO/NF electrocatalysts were tested for comparison. The LSV curves of all the Ir-Co_xNi_1-xO/NF electrodes exhibited sharper onset potentials at ~1.48 V versus RHE and displayed significantly enhanced anodic currents, compared with that of bare Co_xNi_1-xO/NF (Fig. 4a). More interestingly, when tuning the minimal Ir content of the Co_xNi_1-xO nanowire arrays, Ir(0.57 wt%)-Co_xNi_1-xO/NF revealed the lowest overpotential of ~260 mV at 10 mA cm^–2 current density, which exceeded those of Ir(0.68 wt%)-Co_xNi_1-xO/NF and Ir(0.2 wt%)-Co_xNi_1-xO/NF (~270 and 280 mV, respectively). In contrast, the bare Co_xNi_1-xO/NF control electrodes displayed the highest overpotential of ~305 mV at 10 mA cm^–2 current density. The Ir mass-related OER activities of Co_xNi_1-xO/NF were also determined from the related Tafel slopes. The 0.57 wt% Ir-doped Co_xNi_1-xO/NF exhibited the smallest Tafel slope of 48 mV dec^–1, which indicated a more rapid OER on Ir(0.57 wt%)-Co_xNi_1-xO/NF electrode than on the others (Fig. 4b) [47]. The EIS results revealed that Ir(0.57 wt%)-Co_xNi_1-xO/NF exhibited a much smaller charge-transfer resistance (R_ct) of 0.59 Ω than Ir(0.2 wt%)-Co_xNi_1-xO/NF, Ir(0.68 wt%)-Co_xNi_1-xO/NF, and bare Co_xNi_1-xO/NF (0.97, 0.66, and 1.33 Ω, respectively) (Fig. 4c). This allowed identification of the highest OER rate, which corresponded to the smallest Tafel slope [48]. The durability of an electrocatalyst is another important parameter that must be considered for practical applications. Long-term chronopotentiometric measurements showed that the Ir(0.57 wt%)-Co_xNi_1-xO/NF electrodes were inherently stable in alkaline media during the OER process (Figs. 4d and S3). The double-layer capacitances (C_dl) were measured to determine the electrochemical active surface areas of the different samples. As shown in Fig. S4, the C_dl of Ir(0.57 wt%)-Co_xNi_1-xO/NF electrode was estimated to be ~80 mF cm^–2, which is much larger than those of the Ir(0.68 wt%)-Co_xNi_1-xO and Ir(0.2 wt%)-Co_xNi_1-xO samples. The results revealed that Ir(0.57 wt%)-Co_xNi_1-xO nanowire arrays on Ni foam can provide a larger catalytically active surface area and a higher interfacial contact area with the electrolyte, which suggest a higher electrocatalytic activity of the Ir(0.57 wt%)-Co_xNi_1-xO/NF electrode.

The HER performances of the different catalysts were characterized in the same electrolyte (Fig. 5a). Similar to the results for the OER, the Ir-Co_xNi_1-xO electrodes with 0.57 wt% Ir exhibited the best HER activity, with an overpotential of only ~53 mV at a current density of 10 mA cm^–2, which was lower than those of the Ir(0.68 wt%)-Co_xNi_1-xO/NF, Ir (0.2 wt%)-Co_xNi_1-xO/NF, and bare Co_xNi_1-xO/NF electrodes (83, 127, and 148 mV, respectively). Furthermore, as with the case of the OER, the Tafel slope was used to explore the kinetics of the HER process [49]. Again, the Ir(0.57 wt%)-Co_xNi_1-xO/NF electrodes exhibited the smallest Tafel slope of 70 mV dec^–1 among the four catalysts, which corresponded to the fastest HER (Fig. 5b). Similarly, EIS measurements were used to further investigate the HER kinetics at the electrode surfaces. The semicircle observed in the high-frequency range of the Nyquist plot indicated that the Ir(0.57 wt%)-Co_xNi_1-xO/NF electrodes displayed a much lower charge transfer resistance than the other samples, which suggested higher charge and mass transfer efficiencies after the doping of an appropriate amount of Ir in the Co_xNi_1-xO nanowire frameworks (Fig. 5c). The superior robustness of the Ir(0.57 wt%)-Co_xNi_1-xO/NF electrodes for the HER was confirmed from the chronopotentiometric curves (Fig. 5d). After 50 h of continuous operation at a constant current density of 10 mA cm^–2, the potential of Ir-Co_xNi_1-xO/NF remained at almost its initial value, which indicated superior durability of the Ir-Co_xNi_1-xO electrocatalyst.

DFT+U calculations were performed to evaluate the effects of Ir doping of Co_xNi_1-xO catalysts on the HER and OER. Co_xNi_1-xO displays the same cubic structure as rock salt CoO (Figs. 6a and S5). Figs. 6b–d show the calculated PDOS and TDOS distributions of CoO, Co_xNi_1-xO, and Ir-Co_xNi_1-xO, respectively. Pure CoO displays a considerably large bandgap of 1.3 eV, which is similar to the results reported by Qiao et al. [50]. Compared to that of pure CoO, the bandgap of Co_xNi_1-xO is slightly decreased to ~0.84 eV. Both CoO and Co_xNi_1-xO exhibit semiconductor characteristics with low electrical conductivities. However, after doping Co_xNi_1-xO with Ir, Ir-Co_xNi_1-xO already exhibits a half-metal-like conductivity, as evidenced by the impurity states that are found near the Fermi energy level, which confirm that the electrical conductivity of Co_xNi_1-xO is significantly enhanced by Ir doping. The enhanced conductivity of Co_xNi_1-xO achieved through Ir doping can directly result in increased turnover frequency and hydrogen generation current [50], which play significant roles in continuous hydrogen production. To further understand the changes in the hydrogen evolution ability of Co_xNi_1-xO doped with Ir, we systematically calculated the Gibbs free energies (ΔG_H*) of H adsorption on pure CoO, Co_xNi_1-xO, and Ir-Co_xNi_1-xO surfaces. After fully optimizing the adsorption positions of H on the surfaces of the three catalysts, we found that, for the pure CoO structure, the total energy of the H atoms adsorbed on the O atoms is lower than that of the H adsorbed on the Co atoms, whereas, for the CoNiO and Ir-CoNiO structures, the total energy of the H atoms adsorbed on the Co atoms is lower than that of the H atoms adsorbed on the other atoms (Table. S2). The H adsorption energies are shown in Fig. 7; the ΔG_H* of H^* on Ir-Co_xNi_1-xO (0.523 eV) was much less than those of H^* on CoO and Co_xNi_1-xO (0.882 and 0.823 eV, respectively). Clearly, these results allowed us to conclude that the Ir incorporation optimized the free energy of H adsorption on Co_xNi_1-xO for the HER.

Since the OER activities of electrocatalysts have proved to be closely related to the changes in the Gibbs free energies [51], we also calculated the catalytic activities of pure CoO, Co_xNi_1-xO, and Co_xNi_1-xO doped with Ir atoms for the OER. Based on the calculated results, shown in Fig. 8 and Table S3, the free energy ΔG > 0 of each step indicates that the transfer step of every pair of protons and electrons in the OER is endothermic. In pure CoO, the Gibbs free energy is ∆G₂ = 2.12 eV, and the corresponding overpotential is calculated to be η_OER = 0.89 eV. For Co_xNi_1-xO catalysts, the free energy of the second step is almost unchanged, and the overpotential is only reduced by 0.02 eV, which suggests that the substitution of Ni atoms in the CoO crystal has not greatly improved the catalytic activity of CoO. For Co_xNi_1-xO doped with Ir atoms, the potential-limiting step is dramatically decreased in free energy to ∆G₂ = 1.77 eV, and the overpotential is reduced to η_OER = 0.54 eV, which indicates that the Ir doping can greatly improve the oxygen evolution performances of the Co_xNi_1-xO electrocatalysts.

Considering the outstanding OER and HER activities of these Ir-Co_xNi_1-xO catalysts, a two-electrode system using Ir(0.57 wt%)-Co_xNi_1-xO/NF as both the cathode and anode was constructed for water electrolysis. The polarization curves of Ir(0.57 wt%)-Co_xNi_1-xO/NF‖Ir(0.57 wt%)-Co_xNi_1-xO/NF electrolyzer and its Co_xNi_1-xO/NF‖Co_xNi_1-xO/NF counterpart in 1.0 mol L^–1 KOH showed that a cell voltage of only 1.55 V, without iR-correction in the Ir(0.57 wt%)-Co_xNi_1-xO/NF-based electrolyzer, was needed to achieve a current density of 10 mA cm^–2; thus, this electrolyzer outperformed the bare Co_xNi_1-xO-based-electrolyzer (1.64 V, Fig. 9a). It was notable that bubbles due to hydrogen and oxygen evolutions at the cathode and anode, respectively [52-54], were clearly observed (Fig. 9a, inset). The superior OER and HER activities, as well as the higher water splitting activities, of the Ir(0.57 wt%)-Co_xNi_1-xO catalyst are comparable and even better than those of many previously reported TMO electrocatalysts (Tables S4–S6). The durability of the fabricated Ir(0.57 wt%)-Co_xNi_1-xO/NF‖Ir(0.57 wt%)-Co_xNi_1-xO/NF electrolyzer was further evaluated through long-term galvanostatic tests (Figs. 9b and S6). Impressively, an almost constant cell voltage of ~1.55 V was well maintained for over 50 h at the set current density of 10 mA cm^–2. Overall, the designed Ir-Co_xNi_1-xO electrocatalyst exhibited efficient water splitting with good stability.

Minimal Ir-incorporated Co_xNi_1-xO nanowire arrays on Ni foam were successfully prepared through a facile hydrothermal-annealing combined method. When used as electrocatalysts, Ir-Co_xNi_1-xO/NF significantly lowered the overpotentials for both the OER and HER. Among all the samples, Ir(0.57 wt%)-Co_xNi_1-xO/NF exhibited the highest catalytic activity in terms of low overpotentials of 260 and 53 mV for the OER and HER, respectively, at a current density of 10 mA cm^–2. Through XPS measurements, EIS analysis, and DFT calculations, this enhancement was proposed to result from the reduced charge-transfer resistance and modified electronic structure of the Co_xNi_1-xO catalysts after proper Ir doping. In addition, a water-splitting electrolyzer using Ir(0.57 wt%)-Co_xNi_1-xO/NF as both the anode and cathode afforded a cell voltage of only 1.55 V to obtain 10 mA cm^–2; it also exhibited superior capacity retention. This study provided a rational means of using TMO nanostructures as base materials for adding small amounts of noble metals to obtain high-performance electrocatalysts.

This work was financially supported by the National Natural Science Foundation of China (51772255), the Hunan Provincial Innovation Foundation For Postgraduate (CX2017B274), the National Basic Research Program of China (2015CB921103), and the Program for Changjiang Scholars and Innovative Research Team in University (IRT13093).