Zhiqun
Bai
a,
Ruoxuan
Guo
a,
Jingwang
Kuang
a,
Huifang
Chen
a,
WeiHao
Sha
a,
Ao
Xie
a,
Jingchao
Liu
*b,
Pingyu
Wan
a and
Yang
Tang
*a
aInstitute of Electrochemistry, Beijing University of Chemical Technology, Beijing 100029, China
bSchool of Computer Science and Engineering, Beihang University, Beijing 100191, China. E-mail: liujingchao@buaa.edu.cn
First published on 7th July 2025
Water-splitting is a critical technology for the conversion and storage of renewable energy. The slow anodic process dynamics with high overpotential greatly limit the commercialization of electrolytic hydrogen production. Herein, we proposed an electrochemical in situ etching method to achieve molybdenum component-controlled dissolution of ultrathin NiFeMo layered double hydroxides (LDHs) nanosheet arrays on nickel–iron foam (NFF) to create in situ reconstructed oxygen evolution reaction (OER) active sites in LDHs. The Mo-etched material exhibited significantly enhanced catalytic performance and demonstrated exceptional OER activity in alkaline media, achieving a low overpotential of 288 mV at 10 mA cm−2 and 784 mV at 1000 mA cm−2, along with a desired Tafel slope of 43.52 mV dec−1. The enhanced reaction kinetics was contributed by Mo doping-induced electronic structure optimization and interfacial stabilization. Furthermore, an assembled alkaline electrolyzer with NiFeMo/NFF maintained a stable electrolysis voltage (1.695 ± 0.022 V) under simulated industrial fluctuating conditions (30 wt% KOH, 85 °C, 3000 A m−2) for 210 h, exhibiting a super-low voltage decay rate (0.1 mV h−1). The superior performance-stability properties of the new catalyst displayed new material design strategies for efficient industrial hydrogen-production systems.
Nickel–iron (NiFe)-based transition metal oxides/hydroxides have garnered significant attention due to their abundance and cost-effectiveness. In 2014, Lu and colleagues found that nickel–iron LDH nanosheet arrays exhibited excellent catalytic activity and stability against the OER. NiFe LDH nanosheet arrays were prepared on three-dimensional nickel foam substrates after hydrothermal treatment at 120 °C for 12 h. This strategy could provide abundant active sites, promote gas precipitation, and have small charge transfer resistance, which is conducive to the electron transfer process.6 Studies have shown that activity can be significantly improved by elemental doping (e.g., Mo, Co), defect engineering or nanostructuring. Ding Lei et al. used a simple and rapid one-step electrodeposition process to prepare W-doped NiFe (NiFeW)-layered dihydroxide with ultrathin nanosheet characteristics at room temperature and pressure as a bifunctional catalyst for water electrolysis.7 Chen and colleagues prepared a FeCo2O4/MoS2 binary composite catalyst by a hydrothermal method and calcination method. It was found that MoS2 had high activity and good stability in the HER, and was more promising than precious metal catalysts. In oxidative dynamics, its rich redox potential enabled it to adsorb ions (Co2+/Co3+, Fe2+/Fe3+) and enhance the activity of the OER.8 Doping (e.g., Mo, Co), defect engineering, and nanostructuring have enhanced their activity,9 but two fundamental issues remain unresolved: (1) insufficient active site density leading to high overpotentials; (2) structural degradation (peeling/oxidation) under harsh operational conditions (e.g., elevated temperature/current density).10 Although third-metal incorporation can optimize electronic structures for activity, conventional chemical doping often causes segregation of active species11 and fails to address interfacial stability. In addition, the lack of precise control of nanomorphology in wet chemical synthesis limits the further improvement of performance. Hence, we used an electrochemical deposition method to synthesize Mo-doped stable catalysts more conveniently.
We developed a highly efficient and stable NiFeMo/NFF electrocatalyst, which exhibited OER performance in alkaline media. Electrochemical measurements revealed that the NiFeMo/NFF catalyst achieved an overpotential of 288 mV at 10 mA cm−2 and 784 mV at 1000 mA cm−2 in 1 M KOH electrolyte, demonstrating superior electrocatalytic activity. Kinetic analyses further confirmed its optimized reaction kinetics, as evidenced by a Tafel slope of 43.52 mV dec−1. To evaluate practical applicability, a prototype alkaline electrolyzer assembled with NiFeMo/NFF was subjected to simulated industrial fluctuating conditions (30 wt% KOH electrolyte, 85 °C, and 3000 A m−2) for 210 h of continuous operation, during which the electrolysis voltage remained stable at 1.695 ± 0.022 V with a negligible voltage decay rate of 0.1 mV h−1, showcasing exceptional dynamic response capability and long-term durability. The synergistic enhancement in performance and stability originated from dual mechanisms: Mo doping-induced optimization of electronic structure and interfacial stabilization. These mechanisms provide new material design strategies for developing highly active and durable OER catalysts “tailored” for industrial hydrogen-production systems.
This two-stage cycle was repeated 20 times to complete the etching process, yielding electrode A. Subsequently, electrode A was rinsed with ultrapure water and dried in a vacuum oven at 70 °C. After drying, electrode A was thermally annealed in a muffle furnace to obtain the final NiFeMo/NFF composite electrode.
For comparative analyses, the electrochemical performance of NiFeMo/NFF, NiFe/NFF, and NFF electrodes was evaluated. Linear sweep voltammetry (LSV) was employed to measure the relationship between overpotential and current density, with the voltage scan range set from 0 to 1 V at a scan rate of 5 mV s−1. The Tafel slope was calculated using the equation η = a + blog
J, where η is the overpotential, b is the Tafel slope, and J is the current density. Chronoamperometry was performed at various current densities (10, 100, 300, 500, 800, and 1000 mA cm−2) to monitor the potential decay over time. Cyclic voltammetry (CV) was conducted in the non-faradaic region with scan rates ranging from 10 to 100 mV s−1. The double-layer capacitance (Cdl) was determined by fitting the CV curves, and the effective electrode surface area (ECSA) was calculated using the formula ESCA = A Cdl/Cs, where A = 1 cm2 (electrode area) and Cs = 0.04 mF cm−2 (standard value for ECSA determination).
Finally, electrochemical impedance spectroscopy (EIS) was performed in potentiostatic mode at a constant current density of 10 mA cm−2, with frequency ranging from 105 Hz to 10−2 Hz and an alternating signal amplitude of 3 mV. Long-term stability testing under industrial alkaline electrolysis conditions was carried out using a Blue & Battery test system (D350A) to evaluate the durability of the electrode in real-world applications.
As shown in Fig. 2a–c, the Mo-doped NiFeMo/NFF electrode exhibited aggregated nanoparticle structures (∼486.9 nm, as observed in Fig. S1†), indicating a large specific surface area and abundant electrochemical active sites. Notably, Mo doping acted as an effective corrosion inhibitor: the undoped NiFe/NFF electrode exhibited severe surface cracking (Fig. 2d–f), whereas the NiFeMo/NFF electrode maintained a uniform and intact surface morphology. This structural integrity contributed directly to enhanced catalytic activity and long-term stability.
The corresponding energy-dispersive X-ray (EDS) mapping confirmed uniform elemental distribution of Ni, Fe, Mo, and O across the NiFeMo/NFF electrode (Fig. 2g). The atomic percentage was quantified as Ni (22.06%), Fe (19.72%), Mo (2.99%), and O (53.32%) (Table S1†).
X-ray diffraction (XRD) analyses (Fig. S2†) indicated that the NiFeMo/NFF composite material exhibited typical Ni crystal structure characteristics. The diffraction peaks observed at 44.5°, 51.85°, and 76.4° perfectly matched the (111), (200), and (220) crystal planes of standard Ni (PDF#: 04-0850). Notably, the XRD spectrum of NiFeMo/NFF did not show significant differences compared with NiFe/NFF, which may have been due to the strong diffraction signal from the substrate material masking other possible crystal phase features, such as NiFe-LDH and NiFe2O4. Subsequent high-resolution transmission electron microscopy (HRTEM) further confirmed the microstructure and revealed the presence of localized crystallization regions.
HRTEM of the morphology and lattice fringes of the electrode material is presented in Fig. 3. Initial observations of the surface morphology revealed that wrinkled nanosheets formed a “nanoflower”-like structure. Lattice fringe measurements identified spacings of 0.197 nm, 0.23 nm, 0.25 nm, and 0.24 nm, corresponding to the (018), (015), (012) crystal planes of NiFe-LDH and the (311) crystal plane of NiFe2O4, respectively. These data further confirmed the presence of both NiFe-LDH and NiFe2O4 components on the surface, indicating that the nanosheets were composed of NiFe oxides and NiFe-LDH. The strong diffraction peaks from the NFF substrate may have overshadowed other component peaks. Additionally, the poor crystallinity of NiFeMo/NFF (predominantly amorphous structure) likely explained the absence of distinct Mo elemental peaks in the XRD pattern.
X-ray photoelectron spectroscopy (XPS) was employed to investigate the chemical composition and electronic states of NiFeMo/NFF and NiFe/NFF, as shown in Fig. 4a–e. The NiFe/NFF surface consisted primarily of Ni, Fe, and O elements, while the NiFeMo/NFF spectrum confirmed the coexistence of Ni, Fe, Mo, and O, consistent with EDS results, thereby verifying Mo doping.
![]() | ||
Fig. 4 (a) XPS wide spectrum of NiFe/NFF and NiFeMo/NFF. (b) Deconvolved Ni 2p, (c) Fe 2p, (d) O 1s, and (e) Mo 3d peak-fitting spectra. |
As depicted in Fig. 4b, the high-resolution Ni 2p XPS spectrum of NiFeMo/NFF exhibited peaks at 855.43 eV and 873.25 eV, corresponding to the Ni2+ 2p3/2 and 2p1/2 transitions, respectively. Additional peaks at 856.90 eV and 875.06 eV were attributed to Ni3+ 2p orbital splitting. Notably, satellite peaks characteristic of Ni–O bonds were observed at 862.03 eV and 880.25 eV. Quantitative peak fitting revealed that the area of the Ni3+ 2p3/2 peak in the Mo-doped catalyst was 121.5% greater than that of the undoped sample, indicating that Mo6+ modified the electronic structure of Ni2+ through electron transfer and d-orbital coupling, rendering it more prone to lose electrons and oxidize into Ni3+. Moreover, the crystal-field distortion induced by Mo6+ enhanced the asymmetry of the local coordination field, optimizing the oxidation potential energy of Ni2+. Meanwhile, it could further facilitate the conversion of Ni2+ to Ni3+ by modulating oxygen activity or adsorption behavior, ultimately increasing the Ni3+ content significantly.
Further analyses suggested that Mo doping modulated the local electronic structure of Ni sites, enhancing the electron-capture capability of adjacent O atoms and thereby optimizing the charge transfer efficiency. This mechanism enabled Ni2+ to lose electrons readily and form high-valent Ni3+ species, which are crucial for efficient OER catalysis.
Fig. 4c presents the high-resolution Fe 2p XPS spectrum of NiFeMo/NFF. The peaks at 711.04 eV and 724.05 eV corresponded to Fe2+ 2p3/2 and 2p1/2 orbitals, respectively. Meanwhile, peaks at 713.53 eV and 725.91 eV were attributed to Fe3+ 2p orbital transitions. Notably, Mo doping induced a leftward shift in the Fe2+ 2p3/2 binding energy by 0.31 eV and the Fe3+ 2p3/2 peak by 0.25 eV. This chemical-shift phenomenon indicated that Mo altered the local electronic environment of Fe atoms significantly, potentially through electronic interactions or coordination restructuring between Mo and Fe (e.g., Mo acting as an electron donor/acceptor). Additionally, the satellite peaks observed at 718.06 eV and 731.75 eV further confirmed the complex chemical states of Fe, suggesting non-crystalline matrix environments and possible surface oxidation effects. The 3d6 configuration of Fe2+ inherently exhibited high oxidation inertness in strongly alkaline environments (compared with Ni2+), and the electron cloud distribution of its d orbitals showed a weak response to electron transfer from Mo6+. As shown in Fig. 4c, although the Fe2+ 2p peak shifted leftwards, the proportion of Fe3+ did not increase significantly, indicating that the electron transfer failed to drive the oxidation of Fe2+ to Fe3+.12,13
Fig. 4d shows the high-resolution O 1s XPS spectrum, revealing significant contributions from multiple oxygen species. Peak-fitting analysis identified two primary components: 529.94 eV (lattice oxygen bound to metal centers, M–O) and 530.58 eV (stable coordination structures). Notably, a shoulder peak at 531.82 eV was attributed to surface hydroxyl oxygen species (O–H) or chemisorbed oxygen near defect sites, directly linked to the presence of surface oxygen vacancies. Quantitative analyses revealed that the area of the O–H–H related shoulder peak in Mo-doped samples increased by 115.6% compared with the undoped counterpart, confirming that transition metal doping effectively promoted the formation of surface hydroxyl oxide active phases.
Based on the high-resolution analysis of Mo 3d XPS (Fig. 4e), incorporation of high-valent Mo (Mo6+) into the catalyst system was confirmed. Peak fitting revealed that the characteristic peaks at 232.68 eV (Mo 3d5/2) and 235.79 eV (Mo 3d3/2) corresponded to the typical binding energies of Mo6+, indicating that Mo existed stably in the +6 oxidation state within the composite.
Introduction of this high-valent dopant induced significant reconstruction of electronic structure, demonstrating that Mo doping redistributed the electron density at Ni/Fe sites and promoted the transformation of Ni2+ to highly active Ni3+ species. Combined with O 1s spectrum analysis, the strong electronegativity of Mo6+ optimized the coordination environment of metal centers, but also enhanced the activation of lattice oxygen through the Jahn–Teller effect, thereby further improving catalytic performance.
To assess OER performance quantitatively, Fig. 5c summarizes the overpotentials (η10 and η1000) at current densities of 10 and 1000 mA cm−2 (without IR compensation). As shown in Table S2,† NiFeMo/NFF demonstrated the lowest overpotentials across all current densities: η10 = 288 mV and η1000 = 784 mV. Post-IR compensation (65%), the LSV curves (Fig. S3†) revealed further improvements (η100 = 322 mV and η1000 = 331 mV), confirming its exceptional electrocatalytic activity under industrial-relevant conditions.
Systematic evaluation of OER performance was further conducted via chronoamperometry. Fig. 5d shows the current–time responses at stepwise current densities (10–1000 mA cm−2, each maintained for 300 s). All samples exhibited structural stability, with NiFeMo/NFF demonstrating the most pronounced OER activity, characterized by the lowest polarization curve. This result highlighted the synergistic effect of Mo doping in simultaneously enhancing both activity and durability.
Fig. 5e shows the double-layer capacitance fitting curves of four materials. The NiFeMo/NFF material exhibited the highest double-layer capacitance at 14.28 mF cm−2, significantly surpassing that of NiFe/NFF (Cdl = 3.36 mF cm−2) and NFF (0.94 mF cm−2) by 4.25-times and 15.19-times, respectively (Table S3†). These data suggested that NiFeMo/NFF had the largest electrochemical specific surface area and more effective active components on NFF, aligning with the results of SEM characterization. The doping of Mo and Fe led to increased electrochemical active area due to the nanosheet and nanoparticle structures loaded on the surface.
Moreover, the results of the alternating current impedance test showed that the solution resistance RS was 0.71 Ω, the charge transfer resistance Rct of NFF was 219.7 Ω, and the NiFeMo/NFF composite material doped with iron and molybdenum reduced it to 6.88 Ω, as shown in Table S4.† Post-calcination, NiFeMo/NFF exhibited the lowest Rct compared with that of NiFe/NFF. These findings demonstrated that the introduction of Mo effectively reduced the electron transfer resistance of the catalyst, establishing a more efficient charge transfer pathway. Combined with the results of the Tafel test, NiFeMo/NFF displayed superior intrinsic catalytic activity towards the OER.
We employed a “customized” NiFeMo/NFF composite anode paired with a team-developed Ni-based cathode to construct an alkaline electrolysis system with an effective electrode area of 20 cm2. Construction of the electrolytic cell is depicted in Fig. 7. Under simulated industrial fluctuating conditions (30 wt% KOH electrolyte, 85 °C operation temperature, and current density of 3000 A m−2), the system underwent continuous electrolysis for 210 h. As shown in Fig. 6c, the voltage–time curve revealed stable operation at 1.695 V with a voltage fluctuation range of ±1.1%, demonstrating exceptional electrochemical stability and activity under harsh operational parameters. Notably, the NiFeMo/NFF anode maintained consistent voltage output even at the industrially relevant current density of 3000 A m−2, confirming its capability to meet the technical requirements of large-scale “green” hydrogen-production systems. This performance highlighted the potential for practical implementation of the material in next-generation electrolyzers.
Furthermore, we elucidated the dynamic evolution mechanisms of transition metal catalysts under industrially relevant current densities through coordinated elemental doping and structural engineering. These findings provide theoretical insights and practical hints for designing highly active and durable non-precious metal electrodes for large-scale green hydrogen production.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5nr01666e |
This journal is © The Royal Society of Chemistry 2025 |