Unraveling the role of Mo/Ni synergy in β-Ni(OH)₂ for high-performance urea oxidation catalysis

Abstract

Advancing electrocatalytic urea oxidation requires the precise engineering of active sites to enhance efficiency and durability. In this study, the atomic-level synergy between molybdenum (Mo) single atoms and β-Ni(OH)₂ is explored to develop a highly efficient electrocatalyst for the urea oxidation reaction (UOR). By anchoring Mo atoms onto β-Ni(OH)₂, electronic modulation at the Ni sites is induced, optimizing their oxidation state and fostering robust Ni–O–Mo interactions that stabilize catalytic activity. This tailored electronic structure facilitates rapid charge transfer, suppresses undesired phase transitions, and enhances reaction kinetics. The Mo/β-Ni(OH)₂ demonstrates exceptional UOR performance, achieving a current density of 10 mA cm⁻² at a significantly reduced potential of 1.37 V, which is 220 mV lower than that of the oxygen evolution reaction. Additionally, it exhibits a remarkably low Tafel slope of 24.4 mV dec⁻¹ and retains 98.6% of its activity after 40 hours of continuous operation. In situ spectroscopic analysis confirms the dynamic structural evolution of Mo/β-Ni(OH)₂ during the UOR, revealing the stabilization of high-valence Ni species and the suppression of NiOOH formation. Furthermore, it elucidates the electronic reconfiguration at Mo and Ni sites, which plays a crucial role in enhancing catalytic performance. The modulation of coordination environments and electronic states significantly lowers the energy barriers along the UOR pathway, as supported by density functional theory calculations. These findings provide fundamental insights into metal-site interactions, offering guidance for the design of electrocatalysts with enhanced performance and stability.

---

1 Introduction

A promising technological pathway to mitigate both the energy crisis and environmental challenges involves the use of renewable energy for hydrogen production via water electrolysis or electrochemical CO₂ reduction.^1–4 Considerable efforts have been devoted to advancing catalyst development, understanding reaction mechanisms and designing efficient electrocatalysts for these cathodic processes.^5,6 However, the energy efficiency of these electrolysis systems is severely constrained by the thermodynamically unfavorable oxygen evolution reaction (OER) at the anode, which results in substantial energy loss.^7,8 For instance, thermodynamic analyses of CO₂ reduction systems indicate that 92.8% of the total energy consumption is attributed to the OER.⁹ To enhance energy utilization, integrating cathodic reactions with alternative anodic reactions that consume less energy is a critical energy-saving strategy.^10–12 Given the high equilibrium potential of OER (1.23 V), identifying anodic reactions with more favorable oxidation potentials is a key breakthrough in overcoming these limitations. The urea oxidation reaction (UOR) presents a significantly lower thermodynamic potential (0.37 V vs. SHE) compared to OER, which not only drastically reduces the energy demands of electrolysis systems but also provides the additional benefit of treating urea-rich wastewater.^13,14 This dual function makes the UOR an attractive solution for both sustainable hydrogen production and wastewater treatment, offering considerable potential for practical applications. Nevertheless, despite its lower theoretical oxidation potential, the UOR suffers from inherently sluggish kinetics due to the complex 6-electron transfer process and the formation of multiple reaction intermediates.^15,16 While noble metal-based catalysts can accelerate this process, their high cost and scarcity impede widespread application.^17,18 Therefore, the development of cost-effective, highly active, and durable UOR electrocatalysts is essential for the large-scale commercialization of this energy conversion technology.

In recent years, various design and synthesis strategies for electrocatalysts have been developed, with particular focus on nickel-based catalysts, which exhibit superior catalytic activity for the UOR compared to other transition metals.^19–23 This enhanced activity is attributed to nickel's ability to readily bond with urea, facilitated by the interaction of unpaired d-electrons or empty d-orbitals with the catalytically active Ni³⁺ state.²⁴ Numerous approaches have been employed to further enhance anodic catalytic performance by modifying the catalyst's surface properties, including heteroatom doping,^19,21 defect engineering and interface engineering.^20,25–27 For nanoscale nickel-based catalysts, reducing the particle size increases the specific surface area, thereby exposing more active sites. Doping with other metal elements can further refine the local electronic structure surrounding Ni centers, reducing d-orbital electron density and creating vacant orbitals, thereby enhancing the adsorption of reaction intermediates and promoting catalytic activity. Additionally, nickel-based composite catalysts can establish internal electric fields, and their heterostructures provide synergistic effects that not only increase the effectiveness of active sites but also alter the surface charge distribution of the catalyst. This modification enhances interactions with reaction intermediates, ultimately boosting catalytic efficiency. However, most research in this area has predominantly focused on catalyst synthesis and performance, with limited investigation into the underlying synergistic mechanisms of nickel-based catalysts. Another challenge is that nickel-based catalysts undergo structural reconstruction during anodic oxidation, forming both β-NiOOH (Ni in the +3 oxidation state) and γ-NiOOH (Ni in ∼+3.6 oxidation state).²⁸ The distorted structure of γ-NiOOH introduces stacking defects in the lattice, which impede OH⁻ adsorption and hinder charge transfer within the catalyst.^29,30 Moreover, studies suggest that approximately half of the Ni³⁺ in β-NiOOH disproportionates into Ni²⁺ and Ni⁴⁺, with Ni²⁺ not being an effective active site.^31,32 Density functional theory (DFT) calculations have further revealed that the desorption of COO* from NiOOH is a rate-limiting step, leading to the deactivation of NiOOH surfaces over time and reducing the overall efficiency of urea electrolysis.³³

To overcome the structural instability and phase transformation issues in nickel-based electrocatalysts, a novel strategy was developed by anchoring single-atom molybdenum (Mo) onto β-Ni(OH)₂. This rationally engineered electrocatalyst leverages the high valence state of Mo to induce profound electronic restructuring within the Ni lattice, fostering the formation of robust Ni–O–Mo coordination bonds. These bonds not only enhance charge delocalization but also stabilize Ni active sites in elevated oxidation states, thereby optimizing their catalytic efficacy. The electronic coupling between Mo and Ni strengthens the catalyst's resistance to unwanted phase transitions, particularly suppressing the formation of NiOOH species that typically lead to deactivation. By employing a multifaceted approach integrating in situ spectroscopic analysis and DFT simulations, this study unveils the intricate electronic interactions that govern Mo/Ni synergy, providing new insight into the design principles of nickel-based catalysts with stability and activity for urea oxidation reactions.

2 Experimental section

2.1. Preparation of β-Ni(OH)₂

To synthesize β-Ni(OH)₂, 10 mmol of NiCl₂·6H₂O was dissolved in 100 mL of deionized (DI) water to prepare solution A. In parallel, 12 mmol of NaOH was dissolved in 100 mL of DI water to form solution B. Solution A was heated to 80 °C, and solution B was gradually added dropwise under continuous stirring. After 30 minutes of reaction, stirring was stopped, and a green precipitate formed. The precipitate was then collected by centrifugation and thoroughly washed with DI water five times, yielding the final β-Ni(OH)₂ product.

2.2. Preparation of Mo/β-Ni(OH)₂

Five separate dispersions were prepared by suspending 50 mg of β-Ni(OH)₂ in 100 mL of pH 10 aqueous solution, adjusted with NaOH. To each dispersion, 1, 2, 3, 4 and 5 mg of MoCl₅ were added, respectively. The mixtures were stirred at 30 °C for 24 hours. After the reaction, the precipitates were collected by centrifugation and washed with deionized water five times, yielding β-Ni(OH)₂ samples with varying Mo loadings.

2.3. Characterization

Crystalline structure analysis was conducted using X-ray diffraction (XRD, Bruker D2 Phaser) with Cu Kα radiation (λ = 1.54 Å). X-ray photoelectron spectroscopy (XPS) measurements were carried out with a PHI 5000 Versa Probe (ULVAC-PHI, Japan) system using a monochromatic Al Kα X-ray source. Field emission scanning electron microscopy (FE-SEM) images were captured with a Hitachi SU8220 system. Elemental distribution analysis was performed using energy-dispersive X-ray spectroscopy (EDS) integrated with high-resolution transmission electron microscopy (HRTEM) in scanning transmission electron microscopy (STEM) mode. The morphology of the prepared samples was characterized using spherical-aberration-corrected field emission transmission electron microscopy (ULTRA-HRTEM, JEOL JEM-ARM200FTH, operated at 80 and 200 kV). Metal content was quantified using inductively coupled plasma optical emission spectroscopy (ICP-OES, Agilent 720). X-ray absorption spectroscopy (XAS) for the Mo and Ni K-edges was conducted at SPring-8 (Japan), on the 12B2 Taiwan beamline of the National Synchrotron Radiation Research Center (NSRRC), operated at 8.0 GeV with a constant current of approximately 100 mA. In situ Raman spectroscopy was performed using a UniDRON system (CL Technology) with a 50× objective lens and a 633 nm laser.

2.4. Electrochemical measurement

Electrochemical measurements were carried out at room temperature using an electrochemical workstation (Autolab PGSTAT302N) with a standard three-electrode system in 1 M KOH containing 0.33 M urea. For all measurements, 5 mg of the sample was dispersed in a mixture of 1 mL ethanol and 20 μL of a 5 wt% Nafion solution. The resulting catalyst ink was loaded onto a 1 cm² nickel foam (NF) electrode and dried. A platinum wire was used as the counter electrode, and an Ag/AgCl electrode in a 3 M KCl solution served as the reference electrode. Linear sweep voltammetry (LSV) was employed to assess the electrocatalytic activity of the prepared catalyst towards urea oxidation. The double-layer capacitance (C_dl) was determined using cyclic voltammetry (CV) in the non-faradaic region over a small potential range at different scan rates.

2.5. Atomic scale modelling and simulations

The DFT calculations were carried out by the projector-augmented wave (PAW) method as implemented in the Vienna Ab initio Simulation Package provided by Medea (Medea-VASP).^34–37 The interactions were described using the Generalized Gradient Approximation with the Perdew–Burke–Ernzerhof (GGA-PBE) exchange–correlation functional.³⁸ The atomic orbitals are defined using plane-wave basis sets with a cut-off energy of 400 eV. Real space projection operators were used and the convergence criterion for self-consistent field calculations was set to 10⁻⁷ eV between consecutive steps using the blocked Davidson Algorithm. A Monkhorst mesh of 1 × 2 × 1 was considered in K-sampling which corresponds to actual k-spacings of 0.349 × 0.262 × 0.349 per Å. To calculate the Zero Point Energies (ZPEs) and contributions of vibrational modes of the atoms in the system, the Phonon module in the Medea platform was used over the full Brillouin zone in an interaction range of 10 Å where asymmetric atoms were displaced by ±0.02 Å.³⁹

The adsorption energy of urea on Mo/β-Ni(OH)₂ was calculated as follows

E_f = E_total − E_Mo/Ni(OH)2 − E_urea (1)

whereE_total is the total energy of the whole system, E_Mo/Ni(OH)2 is the energy of the β-Ni(OH)₂ or Mo/β-Ni(OH)₂ layer, and E_urea is the energy of the isolated urea molecule. According to this equation, a negative E_f indicates exothermic behavior while a positive E_f indicates endothermic behavior.

The computational hydrogen electrode model was used to calculate the Gibbs free energy changes (ΔG) of the reaction intermediates.⁴⁰

ΔG = ΔE + ΔZPE − TΔS (2)

The electronic energy change is denoted as ΔE, the change in zero-point energy (ZPE) is ΔZPE, the temperature of 0 K is T, and the change in vibrational entropy is ΔS.

3 Results and discussion

A single-atom Mo/β-Ni(OH)₂ catalyst was prepared by anchoring Mo atoms onto β-Ni(OH)₂ through the immersion of a molybdenum precursor, MoCl₅, into a dispersed β-Ni(OH)₂ solution. The β-Ni(OH)₂ was synthesized via co-precipitation at 80 °C. To introduce Mo atoms onto the prepared β-Ni(OH)₂, MoCl₅ solution was added to the β-Ni(OH)₂ aqueous solution, which had been adjusted to pH 10 using NaOH. The solution was then stirred at 30 °C for 24 h, allowing sufficient time for the Mo atoms to anchor onto the β-Ni(OH)₂ surface (Fig. 1a). EDS images revealed well-distributed Mo on β-Ni(OH)₂ (Fig. 1b and S1†). Furthermore, ICP-OES analysis quantified the Mo content in Mo/β-Ni(OH)₂ to be 2.19 wt% (note: the optimal condition for the UOR, Table S1†). The morphology of Mo/β-Ni(OH)₂ exhibited a nanosheet structure closely resembling that of pure β-Ni(OH)₂ (Fig. S2†), suggesting that the introduction of Mo did not disrupt the sheet-like morphology. Nitrogen adsorption–desorption isotherms and BET analysis (Fig. S3†) revealed that Mo incorporation slightly increases the surface area from 146.5 to 148.8 m² g⁻¹, with minimal impact on the layered structure. Both β-Ni(OH)₂ and Mo/β-Ni(OH)₂ exhibit type-IV isotherms and mesoporous features, with BJH pore sizes centered at 10–50 nm. The average pore width decreases slightly (from 11.4 to 10.8 nm), and the pore volume shows a marginal reduction upon Mo addition, likely due to partial pore filling. To gain deeper insights into the structural characteristics, TEM was employed to image Mo/β-Ni(OH)₂ (Fig. S4†). The TEM images revealed well-defined sheet-like structures, free from small particles, indicating that Mo nanoparticles or clusters did not form during the synthesis process. This suggests that the Mo atoms were likely dispersed at an atomic level, avoiding aggregation. The introduction of Mo did not alter the thickness of the nanosheets, which consistently remained around 2 nm, highlighting the stability of the β-Ni(OH)₂ framework upon Mo incorporation (Fig. S5†). Notably, HR-TEM images showed that the (101) interplanar spacing in both β-Ni(OH)₂ and Mo/β-Ni(OH)₂ was 2.3 Å (Fig. S6†). In addition, the diffraction pattern of Mo/β-Ni(OH)₂ displayed several diffraction peaks at 19.2°, 33.0°, 38.3°, 51.9°, 58.9° and 62.6°, corresponding to the (001), (100), (101), (102), (110) and (111) planes of β-Ni(OH)₂, respectively, as indexed by the standard JCPDS card number 14-0117 (Fig. S7†). This finding confirms that Mo/β-Ni(OH)₂ retains an isostructural relationship with β-Ni(OH)₂, indicating that the introduction of Mo did not alter the primary crystal structure of the host material. The absence of additional peaks in the XRD pattern of Mo/β-Ni(OH)₂ is significant, implying that Mo atoms are likely well-dispersed on the β-Ni(OH)₂ surface at an atomic level. Raman spectroscopy offered deeper insights into the structural integrity of β-Ni(OH)₂ following Mo incorporation. The characteristic bands of β-Ni(OH)₂ were observed at 309 cm⁻¹ and 445 cm⁻¹, corresponding to the E-type vibration of Ni–OH (γ_Ni–OH) and the A_1g stretching of Ni–O (ν_Ni–O), respectively (Fig. S8†).⁴¹ Additionally, a sharp peak at 3578 cm⁻¹ was attributed to the symmetric stretching of hydroxyl groups (ν_OH) in β-Ni(OH)₂. Upon the introduction of Mo into the β-Ni(OH)₂, a weak additional band corresponding to the Mo–O stretching vibration (ν_Mo–O) appeared, along with a broadening of the γ_Ni–OH and ν_Ni–O peaks.⁴² This broadening suggests strong interactions between the Mo and Ni atoms, indicating that Mo atoms are effectively integrated into β-Ni(OH)₂. AC-HAADF-STEM images revealed numerous dispersed points with significant contrast, highlighted by red circles, which are attributed to the strong scattering of Mo atoms (Fig. 1c). These high-contrast points indicate that Mo atoms are successfully anchored onto the support material at an atomic scale. Moreover, the intensity distribution analysis in the X–Y direction showed that the spacing between Mo atoms was approximately 2.3 nm with the diameter of a Mo atom being 0.30 nm (Fig. 1d and e). This spacing exceeds the effective atomic radius of Mo, providing strong evidence that the Mo atoms are atomically dispersed rather than forming dimers or clusters.

Fig. 1 (a) Schematic illustration of the preparation of Mo/β-Ni(OH)₂. (b) HAADF-STEM-EDS elemental mapping of Mo/β-Ni(OH)₂. (c) Aberration-corrected (AC) HAADF-STEM image of Mo/β-Ni(OH)₂. Mo atoms are highlighted by red circles. The intensity profiles obtained in areas labeled (d) 1 and (e) 2 of Mo/β-Ni(OH)₂.

The chemical composition and electronic interactions between Mo single atoms and the β-Ni(OH)₂ support were investigated using XPS. The XPS survey scan of Mo/β-Ni(OH)₂ confirmed the presence of Ni, O and Mo elements as expected (Fig. S9†). The Ni 2p XPS spectrum of Mo/β-Ni(OH)₂ showed two prominent spin–orbit peaks corresponding to Ni²⁺ 2p_1/2 (873.06 eV) and Ni²⁺ 2p_3/2 (855.34 eV), along with two satellite peaks (Fig. 2a). These peaks exhibited a positive shift of 0.5 eV compared to those in pristine β-Ni(OH)₂, indicating a strong synergistic electronic interaction between Ni and Mo atoms. This shift suggests an alteration in the electronic environment around the Ni atoms, likely due to electron transfer from Ni to Mo. The relatively weak intensity of the Mo 3d signal was attributed to the low Mo content in the sample, consistent with the single-atom dispersion of Mo (Fig. 2b). The Mo 3d region revealed two pairs of peaks at 231.90/234.68 eV and 233.32/236.31 eV, corresponding to Mo⁵⁺ and Mo⁶⁺ oxidation states, respectively, with Mo⁵⁺ being the predominant species. O 1s XPS spectra of β-Ni(OH)₂ and Mo/β-Ni(OH)₂ were deconvoluted into three peaks at 530.4, 531.2, and 532.7 eV, corresponding to M–O, M–OH, and physisorbed H₂O, respectively (Fig. S10†). Compared to β-Ni(OH)₂, Mo/β-Ni(OH)₂ shows a decreased M–O and an increased M–OH signal, suggesting that Mo incorporation disrupts lattice oxygen bonding and promotes hydroxyl-rich environments. Further insights into the coordination environment and valence states of Mo and Ni atoms in Mo/β-Ni(OH)₂ were obtained through XAS. The shift in the absorption edge position of the Ni K-edge for Mo/β-Ni(OH)₂ to higher energy compared to both β-Ni(OH)₂ and the reference Ni(OH)₂ sample provided clear evidence of an elevated average oxidation state for the Ni sites, exceeding +2 (Fig. 2c). This shift suggested a significant alteration in the local electronic structure around Ni atoms, attributable to the incorporation of Mo atoms. This finding aligns with XPS results, which also confirm an increase in the Ni oxidation state upon Mo doping. The introduction of Mo into β-Ni(OH)₂ facilitates electron transfer from the Ni sites to the Mo sites through the formation of Ni–O–Mo bonds. The Fourier transform extended X-ray absorption fine structure (EXAFS) spectra at the Ni K-edge showed two significant peaks at approximately 1.5 Å and 2.7 Å for both β-Ni(OH)₂ and Mo/β-Ni(OH)₂, corresponding to the scattering paths of the closest oxygen atoms (Ni–O) and the second neighboring metal atoms (Ni–Ni) surrounding the Ni atoms (Fig. 2d). These features are consistent with those observed in Ni(OH)₂ but differ from those of Ni foil, indicating that the coordination environment of Ni sites in Mo/β-Ni(OH)₂ is similar to that in Ni(OH)₂. The X-ray absorption near edge structure (XANES) of Mo K-edge for Mo/β-Ni(OH)₂ shows that the absorption edge is situated between those of Mo foil and MoO₃, indicating that the oxidation state of Mo sites in Mo/β-Ni(OH)₂ is slightly below +6 (Fig. 2e). This suggests that the Mo species are likely present in a mixed oxidation state, such as Mo⁵⁺ and Mo⁶⁺. The partial reduction of Mo in Mo/β-Ni(OH)₂ may result from electronic interactions between Mo and Ni within the Ni–O–Mo bridge. The EXAFS of Mo/β-Ni(OH)₂ reveals a dominant Mo–O scattering feature at 1.2 Å,⁴³ which is significantly different from the Mo–Mo scattering paths observed at 2.4 Å in Mo foil and 3.2 Å in MoO₃ (Fig. 2f), confirming that Mo is not present in its metallic Mo form or as MoO₃. Moreover, the lack of a Mo–Cl path demonstrated that the precursor MoCl₅ was fully hydrolyzed,⁴⁴ forming hydroxyl complexes that subsequently anchored to the surface of β-Ni(OH)₂ through a dehydration reaction. The EXAFS fitting results further substantiated this mechanism, showing that each Mo atom in Mo/β-Ni(OH)₂ was coordinated with approximately four oxygen atoms, one of which was shared with a Ni atom (Fig. S11 and Table S2†). This specific coordination environment confirmed the formation of a Mo–O–Ni bridge, indicating that Mo was anchored to the surface of β-Ni(OH)₂ rather than being incorporated into its lattice. Wavelet transform (WT) EXAFS analysis provided further macroscopic evidence that Mo exists as isolated atoms in Mo/β-Ni(OH)₂ (Fig. 2g). The WT contour plots revealed a single intensity maximum around 8.155 Å⁻¹, corresponding to Mo–O coordination. Crucially, no intensity maxima related to Mo–Mo signals were observed, in stark contrast to the WT plots of Mo foil and MoO₃, which exhibited clear Mo–Mo coordination peaks. This absence of Mo–Mo coordination strongly supports the conclusion that the anchored Mo was present as atomically dispersed species and exhibited strong electronic interactions with the β-Ni(OH)₂ support.

Fig. 2 High-resolution XPS spectra of (a) Ni 2p and (b) Mo 3d. (c) Ni K-edge XANES spectra and (d) Ni K-edge k³-weighted Fourier transform (FT) of EXAFS spectra of Mo/β-Ni(OH)₂ and β-Ni(OH)₂, where Ni foil and Ni(OH)₂ were used as reference materials. (e) Mo K-edge XANES spectra and (f) Mo K-edge EXAFS spectra of Mo/β-Ni(OH)₂ and β-Ni(OH)₂, where Mo foil and MoO₃ were used as reference materials. (g) Wavelet transform for the k³-weighted EXAFS spectra of Mo foil, MoO₃ and Mo/β-Ni(OH)₂.

To validate the potential application of Mo/β-Ni(OH)₂ in the UOR, its performance was systematically evaluated in a typical three-electrode setup using an electrolyte composed of 1.0 M KOH and 0.33 M urea. Before electrochemical testing, the Mo/β-Ni(OH)₂ catalyst was carefully deposited onto commercial Ni foam and washed with ethanol to ensure a clean and active surface. Initial optimization experiments involved varying the concentration of MoCl₅ to synthesize different Mo/β-Ni(OH)₂ samples. The current density of Mo/β-Ni(OH)₂ increased with the Mo content, peaking at 2.19 wt% Mo before plateauing or slightly decreasing (Fig. S12†).

The electrochemical performance of the Mo/β-Ni(OH)₂ electrocatalyst for the UOR was further assessed and compared with the OER using LSV polarization curves in an alkaline medium (Fig. 3a). A potential of 1.37 V was required to reach a current density of 10 mA cm⁻² in the presence of urea, which is 220 mV lower than that needed for the OER. This significant potential difference underscores the superior UOR performance of Mo/β-Ni(OH)₂. Moreover, Mo/β-Ni(OH)₂ performance far surpassed that of commercial RuO₂ used for the OER. At higher current densities, this advantage became even more pronounced, suggesting that the UOR could be an ideal alternative to the conventional anodic OER. Besides, at a current density of 50 mA cm⁻², the UOR potential of Mo/β-Ni(OH)₂ was significantly lower than that of β-Ni(OH)₂, MoO₃, MoO₃/β-Ni(OH)₂ and Ni foam (Fig. 3b and c). Tafel slope analysis, performed to evaluate reaction kinetics, revealed a slope of 24.4 mV dec⁻¹ for Mo/β-Ni(OH)₂, which is substantially lower than that of the control catalysts, underscoring its highly favorable UOR kinetics and confirming the superior reaction rate of Mo/β-Ni(OH)₂ (Fig. 3d). Moreover, the negative scan LSV curves of Mo/β-Ni(OH)₂ and β-Ni(OH)₂ were recorded to avoid the overlapping of the oxidative transformation of Ni²⁺ to Ni³⁺ and the onset of urea oxidation in the positive scan direction. As shown in Fig. S13,† both Mo/β-Ni(OH)₂ and β-Ni(OH)₂ exhibit clear cathodic peaks around 1.3 V (vs. RHE), corresponding to the reduction of Ni³⁺ to Ni²⁺. Notably, the peak area for Mo/β-Ni(OH)₂ is larger than that of β-Ni(OH)₂, indicating a higher proportion of Ni³⁺ in Mo/β-Ni(OH)₂. Additionally, the observation that MoO₃/β-Ni(OH)₂ slightly outperformed β-Ni(OH)₂ in the UOR suggests that a minimal fraction of MoO₃ nanoparticles in contact with β-Ni(OH)₂ may act similarly to atomically dispersed Mo. To gain deeper insight into the intrinsic activity of Mo/β-Ni(OH)₂, LSV polarization curves were normalized by the electrochemical active surface area (ECSA), isolating the contribution of each active site from the overall performance. The ECSA was estimated via electrochemical double-layer capacitance (C_dl) measurements for Mo/β-Ni(OH)₂, β-Ni(OH)₂, MoO₃ and MoO₃/β-Ni(OH)₂ using CV in the non-faradaic region (Fig. S114†). The C_dl for Mo/β-Ni(OH)₂ was 11.6 mF cm⁻², comparable to that of β-Ni(OH)₂ (10.3 mF cm⁻²), indicating that the anchoring of Mo atoms did not significantly alter the surface roughness (Fig. 3e). After normalization by ECSA, Mo/β-Ni(OH)₂ exhibited markedly superior intrinsic catalytic activity relative to β-Ni(OH)₂, highlighting the enhanced catalytic role of single atom Mo incorporation (Fig. S15†). In addition, due to the smaller number of connections between Mo sites and Ni sites, the electrocatalytic activity of MoO₃/β-Ni(OH)₂ is slightly enhanced compared with β-Ni(OH)₂. This observation emphasized the importance of single atom Mo in achieving synergistic effects with Ni sites. Furthermore, multistep chronoamperometry demonstrated that Mo/β-Ni(OH)₂ responded rapidly to increasing applied potentials, reflecting favorable mass transport properties, such as efficient electrolyte diffusion and effective release of gaseous products (Fig. S16†). These characteristics indicate that Mo anchoring not only enhances the intrinsic catalytic activity but also optimizes the electrochemical interface, contributing to high reaction rates under operational conditions. Notably, Mo/β-Ni(OH)₂ exhibits superior performance compared to most reported transition metal-based UOR catalysts (Fig. 3f).^45–61 Furthermore, chronoamperometry (v–t) measurement was conducted to evaluate the durability of Mo/β-Ni(OH)₂. As shown in Fig. 3g, the Mo/β-Ni(OH)₂ catalyst exhibited only a 1.4% decay after 40 hours at a constant current density of 10 mA cm⁻², indicating excellent long-term stability. The polarization curves recorded before and after the stability test (Fig. S17†) reveal that Mo/β-Ni(OH)₂ exhibits a minimal positive shift of approximately 3 mV at 100 mA cm⁻², whereas β-Ni(OH)₂ shows a considerably larger shift of about 11 mV. Additionally, the onset potential of Mo/β-Ni(OH)₂ remains relatively unchanged after prolonged operation, indicating that the Mo-anchored β-Ni(OH)₂ effectively maintains its electrochemical activity during extended UOR testing. Comparison of the morphology of the as-prepared catalyst and postcatalyst showed that there was no significant change in the surface morphology of Mo/β-Ni(OH)₂. On the other hand, the flake-like structure of β-Ni(OH)₂ was partially reduced, likely due to phase transformation (Fig. 3h and S18†). To further elucidate the origins of the morphological changes, TEM characterization was performed. Pristine β-Ni(OH)₂ exhibited a well-defined nanosheet morphology prior to testing (Fig. S19†). However, after 40 hours of stability testing, the nanosheet features became indistinct and were replaced by aggregated structures. This aggregation phenomenon likely arises from the phase transformation of β-Ni(OH)₂. In contrast, Mo/β-Ni(OH)₂ after the long-term stability test retained a discernible nanosheet morphology, despite the presence of some localized aggregation. Raman spectra support the morphological observations, showing that β-Ni(OH)₂ undergoes a notable phase transition after stability testing, with new peaks at 472 and 552 cm⁻¹ corresponding to NiOOH (Fig. S20†). In contrast, Mo/β-Ni(OH)₂ retains β-Ni(OH)₂ features alongside emerging NiOOH signals, indicating that Mo anchoring inhibits full-scale transformation and preserves structural integrity. XPS analysis further reveals a larger Ni 2p binding energy shift (∼0.7 eV) for Mo/β-Ni(OH)₂ compared to β-Ni(OH)₂ (∼0.4–0.5 eV), suggesting more pronounced Ni oxidation facilitated by Mo (Fig. S21†). Notably, the Mo 3d spectra remain unchanged, indicating that Mo likely acts as a stable electronic promoter rather than an active site. Additionally, after 1000 cycles of LSV, the overpotential at 50 mA cm⁻² increased by only 8 mV. These findings suggest that atomically dispersed Mo on β-Ni(OH)₂ not only enhances UOR kinetics and intrinsic activity but also improves structural stability, making it a highly promising catalyst for UOR applications.

Fig. 3 (a) LSV curves of Mo/β-Ni(OH)₂ in 1.0 M KOH with and without 0.33 M urea. (b) UOR polarization curves of β-Ni(OH)₂, MoO₃/β-Ni(OH)₂, MoO₃ and Mo/β-Ni(OH)₂. (c) Their corresponding potentials at 10, 50 and 100 mA cm⁻². (d) The Tafel plots for Mo/β-Ni(OH)₂ and the corresponding reference materials. (e) C_dl of Mo/β-Ni(OH)₂ and the corresponding reference materials derived from current density versus scan rate. (f) Comparison of the Tafel slope and potential at 100 mA cm⁻² for various Ni-based UOR electrocatalysts. (g) Long-term chronopotentiometric stability determined at a current density of 10 mA cm⁻², with the inset showing the UOR polarization curves from the 10th and 1000th scans. (h) SEM image of Mo/β-Ni(OH)₂ after 40 h of stability test.

To further elucidate the interaction between Mo sites and the β-Ni(OH)₂ support in the Mo/β-Ni(OH)₂ catalyst, particularly in relation to UOR activity and stability, in situ XAS was employed. This technique enabled an independent examination of the structural and oxidation states of each metal under electrochemical conditions.^62–64 The XANES analysis of Mo/β-Ni(OH)₂ revealed that the Mo K-edge absorption edge remained stable over a potential range of 1.3 to 1.45 V (Fig. 4a), indicating that the local structure around Mo sites is preserved during the UOR process. This observation is further supported by the first-order derivatives of the XANES spectra, which showed no discernible shifts (Fig. S22†). In contrast, the Ni sites in both electrocatalysts exhibited a significant shift in the absorption edge to higher energy with increased anodic potential, suggesting the formation of higher oxidation states of Ni beyond Ni²⁺, likely due to the generation of reactive Ni³⁺/Ni⁴⁺ species (Fig. 4b and c). Notably, the Ni oxidation state in Mo/β-Ni(OH)₂ increased more substantially during the UOR process than in β-Ni(OH)₂, suggesting that Mo facilitates the oxidation of Ni sites, closely linked to the high intrinsic catalytic activity of Mo/β-Ni(OH)₂. This trend in Ni oxidation states also aligns with the UOR polarization curves. To further investigate changes in the local atomic environment of the electrocatalysts during the UOR, in situ Mo and Ni K-edge EXAFS spectra were analyzed, and the extracted structural parameters are summarized in Tables S3–S5.† Under applied anodic potentials, bond distances and coordination numbers for the Mo–O and Mo–Ni scattering paths in Mo/β-Ni(OH)₂ showed no observable changes, with consistent structural features and no evidence of metal oxide formation (Fig. S23†). This suggests that the local structure surrounding Mo remains stable, with no evidence of reconstruction occurring during the UOR process. The lack of clear variation in the oxidation state, bond length and coordination number at the Mo site throughout the reaction further implies that Mo does not serve as the active site in the UOR. However, the anchored Mo atoms could promote the oxidation state of Ni sites, thereby enhancing the catalytic activity of the Ni sites. This interaction between Mo and Ni played a crucial role in modulating the electronic environment, thus indirectly boosting the overall electrocatalytic performance.

Fig. 4 In situ XANES spectra of the (a) Mo K-edge, (b) Ni K-edge for Mo/β-Ni(OH)₂ and (c) Ni K-edge for β-Ni(OH)₂ during the UOR. (d) The dynamic oxidation state of Mo and Ni sites for Mo/β-Ni(OH)₂ and β-Ni(OH)₂. (e) In situ Ni K-edge EXAFS fitting curves in R space of Mo/β-Ni(OH)₂. (f) The dynamic bond length of Ni–O and Ni–Ni for Mo/β-Ni(OH)₂ and β-Ni(OH)₂. In situ Raman spectra of (g) Mo/β-Ni(OH)₂ and (h) β-Ni(OH)₂. (i) The dynamic ratio of I_γ/I_ν for Mo/β-Ni(OH)₂ and β-Ni(OH)₂.

In Mo/β-Ni(OH)₂, the Ni–O bond distance remained stable at 1.57 Å from OCV to 1.4 V (Fig. 4e and f). However, at 1.45 V, this bond length shortened by approximately 0.05 Å, while the Ni–Ni bond length increased slightly by around 0.06 Å. These adjustments suggest the formation of mixed Ni³⁺/Ni⁴⁺ states, indicating that the Ni sites in Mo/β-Ni(OH)₂ are the primary active centers for UOR electrocatalysis. In contrast, for β-Ni(OH)₂ without Mo, the Ni–O bond distance decreases significantly from 2.01 Å at 1.35 V to 1.90 Å at 1.45 V, along with a pronounced increase in the Ni–Ni bond distance (Fig. S24†). This pattern aligns with a phase transition from Ni(OH)₂ to NiOOH, indicating surface rearrangement or reconstruction during the UOR in the absence of Mo. The relatively longer Ni–O bond distance in Mo/β-Ni(OH)₂ suggests enhanced structural resilience against bond contraction and deformation due to Ni site oxidation during the anodic UOR process. This stabilizing effect may suppress potential-induced phase transitions, underscoring the role of anchored Mo atoms in preserving the structural integrity of Ni sites, which could enhance the durability and efficiency of the electrocatalyst under UOR conditions.

In situ Raman spectroscopy further confirmed the structural stability of Mo/β-Ni(OH)₂. During anodic polarization, the γ_Ni–OH and ν_Ni–O bands remained stable, indicating the preserved structural integrity of Mo/β-Ni(OH)₂ (Fig. 4g–i). As the anodic potential increased, the intensity ratio of γ_Ni–OH to ν_Ni–O (I_γ/I_ν) increased from 0.7 to 1.0. This effect is primarily due to strong interactions between urea and Mo/β-Ni(OH)₂, which restrict the stretching vibration of the Ni–O bond. This strong interaction also indicates that high-valence Ni sites promote urea adsorption, thereby enhancing catalytic activity. In contrast, β-Ni(OH)₂ underwent a distinct structural transition during the UOR process, as shown by the decreasing intensity of the γ_Ni–OH and ν_Ni–O peaks with increasing anodic potential. Additionally, two peaks observed near 460 cm⁻¹ and 530 cm⁻¹ at potentials above 1.4 V, corresponding to the bending and stretching vibrational modes of Ni–O in NiOOH, confirm the phase transition from β-Ni(OH)₂ to NiOOH during the UOR. This observation is consistent with the in situ XAS results. These findings suggest that Mo atoms on the β-Ni(OH)₂ surface stabilize the structure, especially in areas rich in high-valence dangling bonds, preventing surface reconstruction.

In addition, ex situ characterization was performed to reinforce the supporting evidence. Ex situ TEM (Fig. S25†) reveals that pristine β-Ni(OH)₂ undergoes severe nanosheet collapse and aggregation at higher potentials, indicative of phase reconstruction into NiOOH, whereas Mo/β-Ni(OH)₂ largely retains its morphology with only mild aggregation. Ex situ XRD (Fig. S26†) further shows that β-Ni(OH)₂ maintains its phase up to 1.40 V, but a weak shoulder at 18.4° appears at 1.45 V, assigned to the (001) plane of NiOOH. In contrast, Mo/β-Ni(OH)₂ shows no new diffraction features across the voltage range, suggesting that Mo anchoring effectively suppresses phase transition. Ex situ XPS analyses of Ni 2p reveal a positive binding energy shift from A.P. to OCV for β-Ni(OH)₂, which remains stable up to 1.35 V (Fig. S27†). At higher potentials (1.40–1.45 V), the binding energy increases further. In comparison, Mo/β-Ni(OH)₂ exhibits earlier and more substantial Ni oxidation across the potential range, consistent with the in situ XANES results, indicating that Mo facilitates the formation of catalytically active high-valence Ni species. Meanwhile, the Mo 3d spectra remain largely unchanged, demonstrating that Mo sites remain chemically stable (Fig. S28†). Overall, ex situ TEM, XRD and XPS characterization studies provide solid evidence that Mo incorporation enhances the structural and chemical stability of β-Ni(OH)₂ during the UOR, reinforcing the conclusions drawn from in situ spectroscopic analyses.

Based on the experimental results, comprehensive DFT calculations were conducted to elucidate the underlying mechanisms responsible for the high catalytic activity of Mo/β-Ni(OH)₂ in the UOR. As previously discussed, Mo atoms do not integrate into the β-Ni(OH)₂ lattice but instead adsorb onto its surface. The DFT results confirmed that the formation energy for Mo on the β-Ni(OH)₂ surface is significantly lower than for Mo doped within the β-Ni(OH)₂ lattice (Fig. 5a). Furthermore, differential charge density analyses revealed that charges from the original Ni sites in Mo/β-Ni(OH)₂ are transferred to Mo sites once Mo atoms anchor onto the surface (Fig. 5b and c). This charge transfer phenomenon was validated by XAS results, which indicated that the Ni sites in Mo/β-Ni(OH)₂ exhibit a higher oxidation state compared to those in β-Ni(OH)₂, demonstrating a synergistic interaction between the Mo and Ni sites. The high oxidation state of Ni sites not only enhances catalytic activity during the UOR but also promotes the adsorption of urea molecules during the electrocatalytic process, a finding confirmed by in situ Raman spectroscopy. DFT analysis further clarified that the adsorption of urea molecules at the Ni sites in Mo/β-Ni(OH)₂ is highly exothermic as compared to adsorption at the Ni sites in β-Ni(OH)₂ (Fig. 5d). The improved adsorption energetics at the Mo-modified Ni sites highlights the critical role of Mo in enhancing both the electronic structure and catalytic performance for UOR. Additionally, the adsorption energy of urea molecules on the Mo atom is found to be higher than that on the Ni site in Mo/β-Ni(OH)₂. This indicates that the Ni site plays a more significant role as the active site for catalytic activity in Mo/β-Ni(OH)₂, as the lower adsorption energy at the Ni site facilitates more favorable interactions with urea molecules. Fig. 5e illustrates the free energy changes (ΔG) along the reaction coordinate for UOR on both catalysts. The evolution of reaction intermediates and their corresponding configurations on the catalyst surface are depicted for key steps, offering a comparative understanding of the catalytic mechanisms. The incorporation of Mo atoms into β-Ni(OH)₂ lowers the energy barriers for the UOR as the calculated ΔG is highly negative in comparison to ΔG of the UOR on Ni(OH)₂ surface for all reaction steps. The Ni–O–Mo linkages formed in Mo/β-Ni(OH)₂ enhance the adsorption and stabilization of critical intermediates. This stabilization promotes efficient electron transfer and facilitates the oxidative conversion of urea. The higher oxidation state of Ni achieved in Mo/β-Ni(OH)₂ is critical for accelerating the rate-determining step, as evidenced by the significantly negative ΔG for the step involving the formation of high-valence Ni species.

Fig. 5 (a) The formation energy of the Mo-surface configuration and doped-Mo at the Ni site. (b) Difference charge densities of Mo adsorbed on β-Ni(OH)₂vs. β-Ni(OH)₂. (c) Average difference charge density along the Y-axis in the Mo/β-Ni(OH)₂ layer in the (001) plane. (d) Adsorption energy of urea on different sites in Mo/β-Ni(OH)₂ and β-Ni(OH)₂ in the (001) plane. (e) Reaction free energy profiles of the UOR in Mo/β-Ni(OH)₂ and β-Ni(OH)₂.

4 Conclusion

In summary, this study successfully developed and characterized a Mo atom-dispersed β-Ni(OH)₂ catalyst, demonstrating its exceptional electrocatalytic performance for the UOR. Through comprehensive XPS, XAS and in situ analyses, it was revealed that Mo atoms are anchored on the β-Ni(OH)₂ surface, forming robust Ni–O–Mo linkages that induce significant electronic modifications at the Ni sites. This interaction increases the oxidation state of the Ni sites, resulting in enhanced urea adsorption and catalytic efficiency. DFT calculations further confirmed that Mo anchoring facilitates electron transfer from Ni to Mo, thereby lowering the energy barriers for urea adsorption at Ni sites and contributing to the highly intrinsic activity and stability of Mo/β-Ni(OH)₂. Notably, the structural stability of the catalyst under operational conditions was verified, with the catalyst maintaining its morphology and electrocatalytic performance after prolonged testing. These findings emphasize the crucial role of Mo–Ni synergistic interactions in enhancing the electronic properties and structural durability of Ni-based catalysts for the UOR. The Mo/β-Ni(OH)₂ catalyst thus offers a promising approach for efficient urea oxidation, providing valuable insights into the design of advanced catalysts for energy-related applications.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its ESI.†

Author contributions

Hsiao-Chien Chen: writing – review & editing, resources, project administration, funding acquisition, supervision, conceptualization. Abdul Shabir: software, investigation. Kun-Hua Tu: investigation.

Conflicts of interest

The authors declare no competing financial interest.

Acknowledgements

The authors acknowledge support from the National Science and Technology Council (Contract No. NSTC 113-2221-E-182-060-), Chang Gung University (URRPD2Q0021) and Chang Gung Memorial Hospital (CMRPD1L0221). We would also like to thank the Chang Gung Memorial Hospital Microscopy Core Laboratory for their assistance with the SEM techniques.

References

1. Y. C. Chu, K. H. Chen, C. W. Tung, H. C. Chen, J. Wang, T. R. Kuo, C. S. Hsu, K. H. Lin, L. D. Tsai and H. M. Chen, Adv. Mater., 2024, 36, 2400640.
2. L. Luo, Z. Huang and Q. Tang, J. Mater. Chem. A, 2024, 12, 5145–515525.
3. M. Wang, C. Feng, W. Mi, M. Guo, Z. Guan, M. Li, H. C. Chen, Y. Liu and Y. Pan, Adv. Funct. Mater., 2024, 34, 2309474.
4. Y. Feng, Y. Guan, L. Wen and Y. Liu, J. Mater. Chem. A, 2024, 12, 20300–20306.
5. Y. Zhu, K. Fan, C. S. Hsu, G. Chen, C. Chen, T. Liu, Z. Lin, S. She, L. Li, H. Zhou, Y. Zhu, H. M. Chen and H. Huang, Adv. Mater., 2023, 35, 2301133.
6. C. J. Chang, Y. Zhu, J. Wang, H. C. Chen, C. W. Tung, Y. C. Chu and H. M. Chen, J. Mater. Chem. A, 2020, 8, 19079–19112.
7. X. Lin, L. Zeng and M. Wu, J. Mater. Chem. A, 2024, 12, 17793–17816.
8. S. Chu and A. Majumda, Nature, 2012, 488, 294–303.
9. S. Verma, S. Lu and P. J. A. Kenis, Nat. Energy, 2019, 4, 466–474.
10. X. Xu, H. C. Chen, L. Li, M. Humayun, X. Zhang, H. Sun, D. P. Debecker, W. Zhang, L. Dai and C. Wang, ACS Nano, 2023, 17, 10906–10917.
11. Q. Fang, J. Wen, H. Wang, X. Wei, L. Jiao, X. Luo, M. Sha, Y. Qin, M. Liu, L. Zheng, W. Gu, H. Zhong, L. Hu and C. Zhu, Nano Energy, 2024, 131, 110217.
12. X. Xu, H. C. Chen, L. Li, M. Humayun, X. Zhang, H. Sun, J. Jia, C. Xu, M. Bououdina, L. Sun, X. Wang and C. Wang, Adv. Funct. Mater., 2024, 34, 2408823.
13. Y. Li, X. Chen, Y. Yu, K. Zhang, Y. Cheng, W. He, Q. Luo and S. Gao, Appl. Catal., B, 2024, 354, 124150.
14. Z. H. Zhang, Z. R. Yu, Y. Zhang, A. Barras, A. Addad, P. Roussel, L. C. Tang, M. Naushad, S. Szunerits and R. Boukherroub, J. Mater. Chem. A, 2024, 12, 19321–19330.
15. E. T. Sayed, T. Eisa, H. O. Mohamed, M. A. Abdelkareem, A. Allagui, H. Alawadhi and K. J. Chae, J. Power Sources, 2019, 417, 159–175.
16. Z. Y. Yu, C. C. Lang, M. R. Gao, Y. Chen, Q. Q. Fu, Y. Duan and S. H. Yu, Energy Environ. Sci., 2018, 11, 1890–1897.
17. J. Li, S. Wang, S. Sun, X. Wu, B. Zhang and L. Feng, J. Mater. Chem. A, 2022, 10, 9308–9326.
18. Y. Wang, C. Wang, H. Shang, M. Yuan, Z. Wu, J. Li and Y. Du, J. Colloid Interface Sci., 2022, 605, 779–789.
19. G. Zhan, L. Hu, H. Li, J. Dai, L. Zhao, Q. Zheng, X. Zou, Y. Shi, J. Wang, W. Hou, Y. Yao and L. Zhang, Nat. Commun., 2024, 15, 5918.
20. W. Ge, L. Lin, S. Q. Wang, Y. Wang, X. Ma, Q. An and L. Zhao, J. Mater. Chem. A, 2023, 11, 15100–1512130.
21. H. Xie, Y. Feng, X. He, Y. Zhu, Z. Li, H. Liu, S. Zeng, Q. Qian and G. Zhang, Small, 2023, 19, 2207425.
22. X. Zheng, J. Yang, P. Li, Z. Jiang, P. Zhu, Q. Wang, J. Wu, E. Zhang, W. Sun, S. Dou, D. Wang and Y. Li, Angew. Chem., Int. Ed., 2023, 62, e202217449.
23. Y. Zhu, C. Liu, S. Cui, Z. Lu, J. Ye, Y. Wen, W. Shi, X. Huang, L. Xue, J. Bian, Y. Li, Y. Xu and B. Zhang, Adv. Mater., 2023, 35, 2301549.
24. A. S. Rasal, H. M. Chen and W. Y. Yu, Nano Energy, 2024, 121, 109183.
25. X. Zhang, J. Zhang, Z. Ma, L. Wang, K. Yu, Z. Wang, J. Wang and B. Zhao, J. Colloid Interface Sci., 2024, 665, 313–322.
26. B. Song, Y. Feng, J. Su, L. Li, Q. Shao and J. Lu, ACS Appl. Nano Mater., 2023, 6, 23142–23149.
27. S. Xu, X. Ruan, M. Ganesan, J. Wu, S. K. Ravi and X. Cui, Adv. Funct. Mater., 2024, 34, 2313309.
28. D. K. Bediako, B. Lassalle-Kaiser, Y. Surendranath, J. Yano, V. K. Yachandra and D. G. Nocera, J. Am. Chem. Soc., 2012, 134, 6801–6809.
29. A. Mavrič, M. Fanetti, Y. Lin, M. Valant and C. Cui, ACS Catal., 2020, 10, 9451–9457.
30. S. Y. Lee, I. S. Kim, H. S. Cho, C. H. Kim and Y. K. Lee, Appl. Catal., B, 2021, 284, 120986.
31. Y. F. Li and A. Selloni, J. Phys. Chem. Lett., 2014, 5, 3981–3985.
32. F. Dionigi and P. Strasser, Adv. Energy Mater., 2016, 6, 1600621.
33. L. Wang, Y. Zhu, Y. Wen, S. Li, C. Cui, F. Ni, Y. Liu, H. Lin, Y. Li, H. Peng and B. Zhang, Angew. Chem., Int. Ed., 2021, 60, 10577–10582.
34. MedeA, Version 3.0, MedeA is a Registered Trademark of Materials Design, Inc., San Diego, USA.
35. G. Kresse and J. Furthmuller, Phys. Rev. B:Condens. Matter Mater. Phys., 1996, 54, 11169–11186.
36. G. Kresse and J. Furthmuller, Comput. Mater. Sci., 1996, 6, 15–50.
37. G. Kresse and D. Joubert, Phys. Rev. B:Condens. Matter Mater. Phys., 1999, 59, 1758.
38. J. P. Perdew, B. Kieron and E. Matthias, Phys. Rev. Lett., 1996, 77, 3865.
39. K. Parlinski, Z. Q. Li and Y. Kawazoe, Phys. Rev. Lett., 1997, 78, 4063–4066.
40. J. K. Nørskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, J. R. Kitchin, T. B. Pedersen and H. Jonsson, J. Phys. Chem. B, 2004, 108, 17886–17892.
41. R. Baddour-Hadjean, F. Fillaux and J. Tomkinson, Phys. B, 1995, 213–214, 637–639.
42. L. O. Alemán-Vázquez, E. Torres-García, J. R. Villagómez-Ibarra and J. L. Cano-Domínguez, Catal. Lett., 2005, 100, 219–226.
43. C. C. Cheng, Y. X. Yeh, Y. C. Ting, S. H. Lin, K. Sasaki, Y. M. Choi and S. Y. Lu, J. Mater. Chem. A, 2022, 10, 8784–8797.
44. S. Borg, W. Liu, B. Etschmann, Y. Tian and J. Brugger, Geochim. Cosmochim. Acta, 2012, 92, 292–307.
45. N. Chen, Y. X. Du, G. Zhang, W. T. Lu and F. F. Cao, Nano Energy, 2021, 81, 105605.
46. H. Du, H. Hu, X. Wang, N. Ran, W. Chen, H. Zhu, Y. Zhou, M. Yang, J. Wang and J. Liu, Small, 2024, 20, 2401053.
47. B. Sang, C. Xue, S. Guo, X. Wang, S. Xie, C. Wang and R. Q. Li, Chem. Commun., 2024, 60, 6643–6646.
48. M. Cai, Q. Zhu, X. Wang, Z. Shao, L. Yao, H. Zeng, X. Wu, J. Chen, K. Huang and S. Feng, Adv. Mater., 2023, 35, 2209338.
49. X. Ding, L. Pei, Y. Huang, D. Chen and Z. Xie, Small, 2022, 18, 2205547.
50. S. Xu, D. Jiao, X. Ruan, Z. Jin, Y. Qiu, Z. Feng, L. Zheng, J. Fan, W. Zheng and X. Cui, Adv. Funct. Mater., 2024, 34, 2401265.
51. T. X. Nguyen, Z. Y. Wei, T. M. Zheng, Y. H. Su, K. S. Chuang and J. M. Ting, Chem. Eng. J., 2024, 496, 153841.
52. Z. Wang, W. Liu, J. Bao, Y. Song, X. She, Y. Hua, G. Lv, J. Yuan, H. Li and H. Xu, Chem. Eng. J., 2022, 430, 133100.
53. Z. Y. Yu, C. C. Lang, M. R. Gao, Y. Chen, Q. Q. Fu, Y. Duan and S. H. Yu, Energy Environ. Sci., 2018, 11, 1890–1897.
54. Y. Chen, J. Ge, Y. Wang, X. Zhao, F. Zhang and X. Lei, ACS Appl. Nano Mater., 2024, 7, 12091–12100.
55. M. M. Meshesha, D. Chanda and B. L. Yang, Appl. Catal., B, 2024, 344, 123635.
56. V. Maheskumar, A. Min, C. J. Moon, R. A. Senthil and M. Y. Choi, Small, 2023, 4, 2300212.
57. C. Chen, S. He, K. Dastafkan, Z. Zou, Q. Wang and C. Zhao, Chin. J. Catal., 2022, 43, 1267–1276.
58. Y. Qiu, X. Dai, Y. Wang, X. Ji, Z. Ma and S. Liu, J. Colloid Interface Sci., 2023, 629, 297–309.
59. Y. Zheng, P. Tang, X. Xu and X. Sang, J. Solid State Chem., 2020, 292, 121644.
60. L. Wang, W. He, D. Yin, Y. Xie, H. Zhang, Q. Ma, W. Yu, Y. Yang and X. Dong, Chem. Eng. J., 2023, 462, 142254.
61. H. Roh, C. Lim, D. Kim, T. Park and K. Yong, Appl. Surf. Sci., 2023, 610, 155520.
62. M. Barawi, C. A. Mesa, L. Collado, I. J. Villar-García, F. Oropeza, V. A. de la Peña O'Shea and M. García-Tecedor, J. Mater. Chem. A, 2024, 12, 23125–23146.
63. Y. P. Huang, C. W. Tung, T. L. Chen, C. S. Hsu, M. Y. Liao, H. C. Chen and H. M. Chen, Nanoscale, 2022, 14, 8944–8950.
64. J. Wang, H. C. Chen, H. Y. Tan, C. M. Tan, Y. Zhu and H. M. Chen, ACS Appl. Mater. Interfaces, 2022, 14, 22681–22696.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta00935a

---