Surface Activation of NiMo Oxyfluoride/Fe‐Oxyhydroxide Heterostructures via Defect Engineering for Enhanced Photoelectrochemical Water Oxidation

ABSTRACT

Efficient photoelectrochemical water oxidation critically depends on bulk conductivity and effective charge separation at the semiconductor electrolyte interface. Herein, we report a hybrid heterostructure design strategy using nickel molybdenum oxyfluoride (NiMoOF) with iron oxyhydroxide (FeOOH), synthesized through a hydrothermal synthesis route combined with subsequent spray pyrolysis. The hybridization of NiMoOF and FeOOH in a NiMoOF/FeOOH hybrid heterostructure significantly increases the surface‐active area and promotes interfacial charge transfer by passivating surface defects, compared to pristine NiMoOF. As a result, the NiMoOF/FeOOH hybrid heterostructured photoanode exhibits a markedly enhanced photocurrent density of 2.18 mA cm⁻² at 1.23 V vs. RHE, substantially outperforming pristine NiMoOF. The photoanode also demonstrates excellent operational stability maintained over 20 h under 1.5 illumination. Furthermore, quantitative analysis reveals significant improvements in both bulk (η_bulk = 28.91%) and surface (ηs_urface = 90.38%) charge separation efficiencies. This work demonstrates a scalable and effective strategy for designing high‐performance photoanodes through synergistic compositional and interfacial engineering, offering valuable insights into advanced photoelectrochemical water‐splitting systems.

A hybrid NiMoOF/FeOOH photoanode with engineered oxygen vacancies and cationic defects was successfully fabricated, enhancing electronic structure and photoelectrochemical performance.

1. Introduction

Photocatalytic water splitting holds significant promise for addressing critical challenges related to energy consumption and environmental pollution [1]. Photoelectrochemical (PEC) water splitting that mimics natural photosynthesis enables direct solar to hydrogen conversion, offering a sustainable and environmentally friendly energy solution. However, the practical implementation of PEC systems remains challenging due to the high overpotential and sluggish kinetics of the water oxidation half‐reaction, which significantly reduce photocatalyst efficiency [2, 3]. As such, the overall PEC performance critically depends on the optoelectronic and interfacial properties of semiconductor photoelectrodes. Since the discovery of TiO₂ as a photoanode for water decomposition, researchers have explored a wide range of alternative materials, including modified TiO₂ [4, 5]_, WO₃ [6, 7]_, BiVO₄ [8, 9]_, and α‐Fe₂O₃ [10, 11, 12]_. Despite their widespread investigation, these materials typically exhibit low photoconversion efficiencies due to limitations such as poor charge carrier mobility and slow interfacial oxygen evolution kinetics [13]. To enhance PEC performance, several strategies have been explored, including the construction of homo‐ and heterojunctions, surface modification with cocatalyst, and elemental doping. Nevertheless, single‐component metal oxide photoanodes continue to suffer from high rates of bulk and interfacial charge recombination, which significantly limit their PEC activity [14, 15]. To improve charge separation and transport, the fabrication of heterostructured photoelectrodes by integrating multiple semiconducting components has emerged as an effective solution. These heterostructures facilitate interfacial charge transfer and reduce recombination, thereby enhancing overall PEC performance.

Among the various heterostructured photoanodes, nickel molybdenum oxide (NiMoO₄, referred to as NMO) has recently attracted attention due to its synergistic combination of nickel oxide (NiO) and molybdenum oxide (MoO₃), which provides suitable energy band positions and high chemical stability [16, 17, 18]. Its monoclinic crystal structure, comprising tetrahedral Mo and octahedral Ni sites, facilitates efficient charge transport and separation, making it a strong candidate for solar‐driven water oxidation [19]. For instance, Gao et al. [8] reported that NiMoO₄/BiVO₄ heterostructured exhibit a synergistic enhancement in PEC activity, while Tian et al. [20] demonstrated that integrating NMO with twinned Cd_0.5Zn_0.5S improves electrocatalytic behavior. To further augment the PEC performance of NMO, various strategies have been employed, including heterostructure engineering, defect engineering, and metal doping. These approaches aim to overcome the intrinsic limitations of single‐component systems and have shown promising enhancements in both charge separation efficiency and catalytic activity [21, 22, 23, 24, 25, 26, 27]. Among these, defect engineering has been extensively utilized in photocatalysis to enhance visible light absorption and suppress the recombination of photogenerated charge carriers. Additionally, the introduction of secondary semiconductors or conductive interlayers is known to enhance interfacial charge transport. One such promising cocatalyst is FeOOH, which has demonstrated excellent activity in PEC and oxygen evolution reaction (OER) applications due to its favorable molecular structure and reaction kinetics [28, 29]. When integrated with NiMo‐based semiconductors, FeOOH can function as an electron mediator, enhancing charge transfer dynamics and surface reaction kinetics [30, 31, 32]. However, achieving simultaneous improvements in surface catalytic activity and photogenerated charge separation remains a significant challenge. Addressing these limitations requires a deeper understanding of the interplay between component materials and the underlying charge transfer processes.

Herein, we report a photoanodic treatment‐assisted strategy for the novel hybrid heterostructure design using nickel molybdenum oxyfluoride (NiMoOF, referred to as NMF) and NiMoOF/FeOOH (referred to as NMFF) hybrid heterostructure photoanode film was successfully fabricated via the spray pyrolysis technique. Experimental results reveal that the NMFF heterostructure exhibits significant structural and compositional advantages, including a relatively high photoactive surface area attributed to abundant defect sites and a strong synergistic interaction between NiMoOF and FeOOH. The presence of oxyfluoride‐induced defects effectively modifies the surface‐active sites of the NMF photoanode, thereby enhancing charge transfer efficiency. Under 1‐sun illumination, the optimized NMFF photoanode shows a photocurrent density of 2.18 mA cm⁻² at 1.23 V vs. RHE, representing a notable improvement over the pristine NMF photoanode. To evaluate its photocatalytic performance, the long‐term operational stability of the NMFF photoanode was also investigated. A comprehensive analysis of charge transfer processes and structural evolution during operation provides critical insights into the mechanisms underlying the enhanced photoelectrochemical performance and durability of the heterostructure.

2. Experimental Section

2.1. Materials

Sodium fluoride (NaF), nickel(II) nitrate hexahydrate (Ni(NO₃)₂·6H₂O), ammonium molybdate tetrahydrate ((NH₄)₆Mo₇O₂₄·4H₂O), and iron(III) nitrate hexahydrate (Fe(NO₃)₃·6H₂O) were all purchased from Sigma‐Aldrich and used without further purification.

2.2. Synthesis of NiMoO, NiMoOF, and Hybrid Heterostructured NiMoOF/FeOOH

NMO nanorods were produced via a modest hydrothermal process followed by spray pyrolysis, as shown in Scheme 1. A solution was prepared by dissolving 0.1 M nickel nitrate, 0.1 M ammonium molybdate, and 0.3 M urea in 40 mL of DI water under constant stirring to form a clear precursor solution. This solution was then transferred to a Teflon‐lined stainless‐steel autoclave and heated at 180°C for 15 h. The resulting NMO product was collected via centrifugation, washed numerous times with ethanol and deionized water, and dried in a vacuum oven at 60°C. NMF was synthesized using a similar hydrothermal method. The precursor solution was prepared by dissolving 0.1 M nickel nitrate, 0.1 M ammonium molybdate, and 0.3 M sodium fluoride in 40 mL of deionized water under stirring. The mixture was autoclaved at 180°C for 15 h. The resulting product was recovered by centrifugation, washed several times with ethanol and water, and dried under vacuum at 60°C.

SCHEME 1.

Schematic illustration of the synthesis process for NMO, NMF, and heterostructured NMFF electrodes via the hydrothermal method followed by spray pyrolysis.

Thin films of NMO and NMF were then fabricated onto indium tin oxide (ITO) substrates via spray pyrolysis. In each case, 20 mg of powders (either NMO or NMF) were dissolved in 30 mL of deionized water and stirred for 2 h to obtain homogeneous suspensions. The resulting solutions were sprayed onto ITO substrates preheated to 150°C. To synthesize the NMFF heterostructure, 20 mg of NMF was dispersed in 30 mL of deionized water under continuous stirring. Subsequently, varying concentrations of iron nitrate (5, 10, 15, 20, and 25 mM) were added to the mixture, which was stirred for an additional 2 h to form a uniform solution. This solution was then sprayed onto an ITO substrate preheated to 150°C, as illustrated in Figure S1. The resulting NMFF thin films were subsequently used for characterization and PEC measurements. The material characterization methods and photoelectrochemical analysis in detail are described in the Supporting Information.

3. Results and Discussion

3.1. Structural and Morphological Properties

The crystal structures of the synthesized samples were analyzed using X‐ray diffraction (XRD), as shown in Figure 1a. The NMO sample exhibits distinct diffraction peaks at 10.92°, 27.38°, 29.98°, 33.31°, 37.68° which correspond to the (110), (112), (220), (310), (202) crystal planes of NiMoO·xH₂O (JCPDS No. 13–0128). In the NMF and NMFF samples, new peaks appeared at 25.71° and 29.15°, which are attributed to the NaNiF₃ phase (JCPDS No. 00‐010‐0158) [33, 34, 35]. Additionally, reflections at 12.22°, 23.73°, 27.27°, and 30.50° are assigned to the MoO_2.8F_0.2 phase (JCPDS No. 00‐025‐1103). The presence of these additional diffraction peaks in the NMF and NMFF samples confirms the successful formation of polycrystalline NiMo‐oxyfluoride structures, demonstrating the incorporation of fluoride species and the evolution of a new crystal phase beyond the original NiMoO·xH₂O framework [35, 36]. Notably, no diffraction peaks corresponding to FeOOH were observed, suggesting either low content or amorphous nature of FeOOH in NMFF [30, 31, 32, 37]. The reflections marked with a plus sign (+) at 35.2°, 50.5°, 60.1°, and 43.92° are attributed to the underlying ITO substrate. These XRD results confirm the formation of a heterostructured NMFF material. Further structural analysis was carried out using Raman spectroscopy (Figure 1b). The Mo═O stretching vibration band at ∼975.74 cm⁻¹ in NMO shifts to ∼953.77 cm⁻¹ in NMF and ∼950.86 cm⁻¹ in NMFF, indicating bond environment alterations induced by oxyfluoride incorporation [38]. The bands at ∼822.71 cm⁻¹ (NMO), ∼831.45 cm⁻¹ (NMF), and ∼840.72 cm⁻¹ (NMFF) correspond to asymmetric stretching modes of MoO₄ tetrahedra, showing increased intensity with higher Mo content, which suggests Mo enrichment [39]. A vibrational band around 353.56 cm⁻¹ is attributed to the symmetric/asymmetric bending of Ni─O bonds [40]. The overall shifts in the Mo═O modes and emergence of new bands confirm local structural distortions introduced by F and FeOOH incorporation, which may enhance electronic conductivity and catalytic activity.

FIGURE 1.

(a) X‐ray diffraction patterns of NMO, NMF, and NMFF recorded in the 2θ range of 10°–70°, and (b) Raman spectra of NMO, NMF, and NMFF catalyst electrodes.

SEM images of the NMFF electrode (Figure 2a; Figure S2a) show densely packed, nanorod‐like structures. In contrast, NMO exhibits nano‐needle‐like morphologies (Figure S3a,b) while NMF maintains uniform nanorod arrangements (Figure S4a,b). Elemental mapping images and EDAX spectra of the NMFF sample are provided in Figure S2d–f. Similarly, the elemental mapping and EDAX spectra for NMO and NMF are presented in Figures S3c,d and S4c,d, respectively. The presence of nanorod‐like structures in NMFF was further confirmed by transmission electron microscopy (TEM), as shown in Figure 2b; Figure S2b,c. The high‐resolution TEM (HR‐TEM) image (Figure 2c) reveals well‐resolved lattice fringes. In the magnified images (Figure 2d,e), interplanar spacings of 0.29 and 0.30 nm are observed, corresponding to the (130) and (201) plane, respectively, as confirmed by JCPDS cards 025–1103 and 010–0158. The selected area fast Fourier transform (FFT) pattern (Figure 2f) further confirms the crystalline nature of the NMFF nanorods. Elemental mapping images (Figure 2g) demonstrate the uniform distribution of Ni (cyan), Mo (pink), O (red), F (orange), and Fe (green) across the nanorods, indicating successful formation of a well‐integrated heterostructure. The uniform distribution of Fe observed in the FE‐SEM elemental mapping and HR‐TEM EDX analysis (Figure S2e,f), together with the XPS results, confirms the successful incorporation of Fe species within the NMFF heterostructure despite the absence of distinct FeOOH diffraction peaks in the XRD patterns, which may be attributed to the low loading amount and/or amorphous nature of FeOOH. This well‐defined nanostructured is expected to enhance photoelectrocatalytic activity by increasing the electrochemically active surface area and promoting efficient charge separation and transfer.

FIGURE 2.

Morphological analysis of the best‐performing NMFF catalyst electrode: (a) FE‐SEM image showing nanorod‐like morphology, (b) TEM image of a single NMFF nanorod, consistent with the SEM results, (c) HRTEM image, (d–e) lattice fringes with interplanar spacings of 0.29 and 0.30 nm, (f) FFT image, and (g) HADF‐STEM elemental mapping of the NMFF nanorod showing the presence of Ni (cyan), Mo (pink), O (red), F (orange), and Fe (green).

X‐ray photoelectron spectroscopy (XPS) was used to investigate the surface elemental composition and valence states of the samples. Consistent with the EDS elemental mapping results, the XPS survey spectra (Figure S5) reveals distinct signals corresponding to Ni, Mo, O, F, and Fe. In the Ni 2p core‐level spectrum (Figure 3a), The Ni 2p XPS spectrum exhibits four main peaks located at 855.69 eV (Ni²⁺ 2p_3/2) and 873.41 eV (Ni²⁺ 2p_1/2), corresponding to a spin–orbit splitting of 17.72 eV. Additional peaks observed at 857.05 eV (Ni³⁺ 2p_3/2) and 875.10 eV (Ni³⁺ 2p_1/2), along with their associated shake‐up satellite features, indicate the coexistence of both Ni²⁺/Ni³⁺ oxidation states [41, 42]. The Mo 3d spectrum (Figure 3b) displays peaks at 232.37 and 235.56 eV, consistent to Mo 3d_5/2 and Mo 3d_3/2, respectively. The energy separation of 2.93 eV confirms the presence of Mo⁶⁺ species [41, 43]. A distinct peak at 684.67 eV in the F 1s spectrum (Figure 3c) indicates the incorporation of fluorine into the crystal lattice in both NMF and NMFF, suggesting the successful formation of metal–fluoride bonds [44, 45]. In the Fe 2p spectrum of NMFF (Figure 3d), the peaks located at approximately 710.6 eV (Fe^{2 +} 2p_3/2) and 723.49 eV (Fe^{2 +} 2p_1/2) confirm the presence of Fe²⁺ species. Additional peaks observed at 713.05 eV (Fe^{3 +} 2p_3/2) and 727.02 eV (Fe³⁺ 2p_1/2) are attributed to Fe³⁺ species. Deconvolution of the Fe 2p spectrum further confirms the coexistence of mixed Fe²⁺/Fe³⁺ oxidation states in NMFF [30, 31, 32]. The O 1s spectra of all three samples (Figure 3e) exhibit two distinct peaks at 530.08 and 528.94 eV. These are attributed to oxygen in hydroxyl groups and oxygen involved in metal oxygen bonds, particularly in the NMO sample. A distinct peak at 531.6 and 531.9 eV, observed prominently in NMF and NMFF, is associated with oxygen coordination defects, confirming the presence of oxygen vacancies [46, 47, 48]. Moreover, XPS also provided insight into the electronic structure modifications during surface treatment. After oxyfluoride incorporation, a shift of the core‐level peaks toward higher binding energies was observed, indicating electron withdrawal from the metal centers and a decrease in local electron density, which can be attributed to the strong electronegativity of fluorine and the formation of oxyfluoride bonds. Following FeOOH deposition, the peaks shifted toward higher binding energies, suggesting a decrease in electron density. These sequential shifts confirm the successful chemical modification of the surface, further validating the composition and the formation of oxygen vacancies and structural defects on the NMO surface factors that are essential for enhancing charge transfer and photoelectrochemical performance. Additionally, Fourier‐transform infrared (FT‐IR) spectra (Figure S6) reveal a new vibrational band at 1556 cm⁻¹ in NMF and NMFF, assigned to the asymmetric stretching mode of HCOO– groups, confirming the presence of defect‐related species [46, 47, 49]. To further probe the defects, electron spin resonance (ESR) spectroscopy was conducted (Figure 3f). All samples show a symmetrical ESR signal, but the NMFF sample exhibits significantly higher signal intensity, indicating a greater concentration of oxygen vacancies (V_O) induced by the incorporation of FeOOH [50].

FIGURE 3.

XPS analysis of all three samples: (a) core‐level Ni 2p, (b) core‐level Mo 3d, (c) core‐level F 1s, (d) core‐level Fe 2p, (e) core‐level O 1s, and (f) EPR spectra of NMO, NMF, and NMFF.

3.2. Optical and Electronic Properties

The UV‐visible absorption spectra (Figure 4a) of the NMO photoanodes reveal notable changes in optical behavior following the incorporation of oxygen vacancies and FeOOH. The pristine NMO sample displays an absorption edge at 314 nm, indicating primary absorption in the ultraviolet region. Upon introducing oxygen vacancies in NMF, the absorption edge redshifts to 341 nm, extending into the visible range suggesting improved light‐harvesting capability due to defect‐induced band structure modulation [32, 51]. Further incorporation of FeOOH into NMF to form NMFF induces a more pronounced redshift, with an absorption edge at 421 nm. Tauc plot analysis (Figure 4b) shows that the optical bandgap decreases from 2.41 eV (NMO) to 2.23 eV (NMF), and further to 1.87 eV (NMFF), consistent with improved visible‐light absorption and bandgap tuning due to defect engineering and heterostructure formation. Photoluminescence (PL) spectra (Figure 4c) provide insight into charge recombination dynamics. The NMO sample shows the highest PL intensity, indicating a high recombination rate of photogenerated charge carriers. In contrast, NMF exhibits reduced PL intensity, suggesting slower recombination due to defect‐induced charge trapping [52]. Notably, both NMF and NMFF display a distinct shoulder peak in the PL spectra, which can be attributed to the presence of oxygen defect states [51]. The NMFF sample demonstrates the lowest PL intensity, confirming a significant reduction in recombination centers and enhanced charge separation efficiency [52, 53].

FIGURE 4.

Optical and electronic properties: (a) UV–vis spectra, (b) Tauc Plot, (c) PL spectra at an excitation wavelength of 380 nm, (d) photocurrent density potential plot under 1‐sun illumination, (e) Photocurrent density at 1.23 V vs. RHE for NMFF and compared with recent literature reported photoanodes. (f) Tafel plot (g) photocurrent–time plot at 1.23 V vs. RHE, (h) ABPE, and (i) IPCE spectra at 1.23 V vs. RHE.

3.3. Pec Water Oxidation Performance

The photoelectrochemical performance of the photoanodes was evaluated in 1 M KOH (pH 13.4) under AM 1.5 G illumination (100 mW cm⁻²). The Linear sweep voltammetry (LSV) curves obtained under chopped light illumination are shown in Figure 4d and Figures S7 and S8. The pristine NMO photoanode exhibited a photocurrent density of 0.51 mA cm⁻² at 1.23 V vs. RHE, with an onset potential of 1.01 V vs. RHE. The introduction of oxyfluoride defects in the NMF photoanode led to enhanced performance, increasing the photocurrent of 1.27 mA cm⁻² and slightly lowering the onset potential to 0.98 V. Notably, the NMFF heterostructured photoanode achieved a significantly enhanced photocurrent density of 2.18 mA cm⁻² at 1.23 V vs. RHE, along with a reduced onset potential of 0.96 V. To verify that the enhanced photocurrent is associated with the synergistic interaction between NiMoOF and FeOOH, control experiments were conducted using different concentrations of FeOOH (5, 10, 15, 20, and 25 mM) as well as by directly depositing FeOOH onto ITO. As shown in Figure S8, the FeOOH/ITO electrode exhibited negligible photocurrent, confirming that the improved photoelectrochemical performance originates from the synergistic effect between NiMoOF and FeOOH. Among these, the 20 mM FeOOH–NiMoOF heterostructure exhibited the highest photocurrent density, outperforming all other concentrations. Figure 4e and Table S1 provides a summary of key electrochemical parameters, benchmarking the performance of these materials against other reported systems and confirming the beneficial role of oxygen vacancies and FeOOH incorporation. The Tafel analysis (Figure 4f) further provides insight into the intrinsic OER kinetics of the photoanodes. NMO, NMF, and NMFF exhibit Tafel slopes of 137.7, 125.2, and 113.9 mV dec^{− 1}, respectively, indicating a progressive improvement in reaction kinetics. The significantly lower Tafel slope of NMFF suggests faster charge‐transfer processes and more favorable OER kinetics, consistent with its enhanced carrier separation efficiency and superior photoelectrochemical performance.

A comparison of photocurrent densities at 1.23 V vs. RHE for all samples is presented in Figure 4g. NMFF exhibits the highest average photocurrent, reflecting improved charge separation and water oxidation efficiency [53]. The applied bias photon‐to‐current conversion efficiency (ABPE), estimated from the LSV data, is shown in Figure 4h NMFF achieved an ABPE of 0.1%, outperforming NMF (0.05%) and NMO (0.02%). This enhancement is attributed to the synergistic effects of surface defect passivation and improved charge transfer dynamics via the formation of NMFF heterostructures. Figure 4i displays the incident photon‐to‐current conversion efficiency (IPCE) results, where NMFF reached 25.81% at 420 nm, surpassing NMF (23.07% at 340 nm) and NMO (16.75% at 310 nm), indicating more effective charge carrier generation and collection. In addition, Figure S9 shows that NMFF achieved an absorbed photon‐to‐current efficiency of 28.68% at 420 nm, compared to 26.04% for NMF and 18.62% for NMO. These results highlight the contribution of FeOOH to enhanced light harvesting, charge separation, and catalytic performance for OER.

To further evaluate charge separation, bulk and surface efficiencies were analyzed using photocurrent density measurements in 1 M KOH and 0.5 M Na₂SO₃ electrolytes (Figure S10) [32, 54]. As shown in Figure 5a, the bulk charge separation efficiency (η_bulk) was 12.49% for NMO, increased to 17.68% for NMF, and further improved to 22.15% for NMFF at 1.23 V vs. RHE. This progressive enhancement aligns with the increasing density of oxygen vacancies and the formation of a heterostructure. Supporting this, the PL spectra (Figure 4c) exhibit reduced emission intensity for NMFF, consistent with suppressed charge recombination. Next, surface charge separation efficiencies (η_surface) were also analyzed (Figure 5b). NMFF achieved 58.63%, outperforming NMF (41.34%) and NMO (20.01%), suggesting that FeOOH effectively passivates surface trap states and facilitates charge extraction. ECSA values were determined via cyclic voltammetry (CV) in the non‐Faradaic potential window of 0.9–1.0 V at scan rates of 10–100 mV/s. The measurements were conducted under both dark and light conditions, as shown in Figure S11a–c (dark); Figure S11d–f (light). As shown in Figure 5c, NMFF exhibits the highest double‐layer capacitance (Cdl) of 3.13 µF, surpassing that of NMF (2.12 µF) and NMO (1.54 µF). The corresponding ECSA values (Table S2) follow the same trend, with NMFF reaching 78.25 cm². Under illuminated conditions, the Cdl of NMFF further increases from 3.13 to 3.51 µF, accompanied by an enhancement in ECSA to 87.75 cm². This increase under light irradiation indicates the generation of additional electrochemically accessible active sites and a larger effective surface area. Moreover, the TOF_ECSA (Figure 5d) further highlights the superior intrinsic activity of NMFF, with a value of 0.39 s⁻¹, nearly double that of pristine NMO (0.19 s⁻¹) and substantially higher than NMF (0.28 s⁻¹). Owing to its distinct structure and composition, these results indicate that illumination significantly enhances the intrinsic activity of NMFF. The improved TOF_ECSA further confirms that NMFF possesses a higher density of catalytically active sites and superior turnover thereby facilitating improved OER activity [55, 56].

FIGURE 5.

Charge separation efficiencies: (a) η_bulk, (b) η_surface, (c) C_dl plot, (d) TOF bar graph under light and dark condition, (e) long‐term stability at 1.23 V vs. RHE, and (f) hydrogen and oxygen evolution under 1‐sun illumination. (g) Mott–Schottky plots recorded under dark conditions (h) Carrier lifetime calculated from the open‐circuit potential (OCP) decay curves. (i) TRPL decay curves.

When evaluating a photoanode for practical applications, both conversion efficiency and long‐term stability are critical factors. As shown in Figure 5e, the NMFF photoanode maintained a stable photocurrent over 20 h at 1.23 V RHE, indicating excellent operational durability. This long‐term stability is likely due to the efficient consumption of photogenerated holes, which suppresses charge recombination. Water‐splitting performance was further evaluated by gas evolution measurements (Figure 5f). After 12 h of PEC operation, the NMFF photoanode produced 388 µmol of H₂ and 194 µmol of O₂, while retaining 94.8% of its initial photocurrent. The Faradaic efficiency of the photoanode for O₂ evolution was approximately 95%, indicating efficient charge utilization toward the oxygen evolution reaction.

Post‐stability characterization confirmed the robustness of NMFF photoanode. LSV curves before and after testing remained nearly unchanged (Figure S12), and SEM images (Figure S13) revealed no significant morphological degradation. XRD patterns (Figure S14) were recorded to examine the structural changes after prolonged stability testing. The diffraction pattern shows that the sample retains its original crystalline structure, with characteristic peaks at 12.17°, 21.30°, 29.26°, 30.21°, and 37.75°, which are indexed to NaNiF₃ and MoO_2.8F_0.2, consistent with the as‐prepared NMFF sample. Slight changes in peak intensity are observed after the stability test, while no new peaks appear in the pattern. The XPS spectra of Ni 2p (Figure S15) shows a shift toward higher binding energies along with an increased number of Ni³⁺ oxidation states, while other XPS spectra remain unchanged. This confirms the formation of NiOOH active phase during initial cycling of the OER activity and lasting over the studied stability duration. This phase transition was further corroborated by Raman spectroscopy (Figure S16), which exhibited a characteristic peak at 553.19 cm⁻¹ corresponding to NiOOH [57, 58]. The M─O bonds observed in Raman spectroscopy are associated with the surface conversion of Ni species into NiOOH phase which is highly active phase for OER. Additionally, Mo dissolving was observed in both the EDX spectra and XPS (Figures S13e and S15c) as well as in the Raman analysis (Figure S16) [59], indicating partial dissolution of Mo during the reaction, consistent with expected behavior after prolonged alkaline operation. The Mo leaching is further supported with the ICP analysis (Table S3), providing quantitative confirmation of Mo dissolution after the stability test which may be the possible reason for activity degradation. In particular, the XPS spectra shows a shift toward lower binding energy, suggesting an increase in donor density and enhanced electron density in the remaining framework. The O 1s core‐level spectra (Figure S15d) exhibited a small shift and reduced intensity of the defect‐related peak, suggesting moderate surface restructuring. A corresponding downshift in the F 1s region (Figure S15e) suggests the oxidation of M─F bonds to M─O bonds, likely caused by partial Mo leaching during prolonged OER operation [60, 61]. Additionally, a 0.31 eV shift in the Fe 2p peak (Figure S15f) is attributed to surface reconstruction and likely indicates an increase in electroactive surface area; this shift further suggests an elevated oxidation state with increased Fe³⁺ content, which actively contributes to the OER process [58]. Crucially, despite these surface‐level chemical changes, the overall crystal structure and morphology remained intact, as confirmed by XRD and SEM. These results underscore the excellent structural integrity and long‐term durability of the NMFF photoanode during PEC water splitting.

3.4. Investigations Into Charge Carrier Behaviors

To gain a comprehensive understanding of the photoanodes, the light‐harvesting efficiency (LHE) (Figure S17) and the theoretical maximum photocurrent density (J_abs) (Figure S18) were investigated. Among all samples, NMFF exhibited the highest LHE and J_abs, indicating enhanced light absorption and charge carrier separation. This improvement is attributed to the synergistic effect of oxyfluoride and FeOOH incorporation, which collectively enhance conductivity, charge mobility, and suppress recombination losses. In contrast, NMO showed the lowest values due to insufficient light absorption and high recombination. Mott–Schottky analysis confirmed all samples as p‐type semiconductors. NMFF showed the highest donor density (64.87), followed by NMF (60.28) and NMO (39.22) (Figure 5g, as calculated using Equation S4), suggesting greater charge carrier availability in NMFF [62]. Open‐circuit potential (▵OCP) (Figure S19) measurements were conducted to evaluate the photovoltage and charge carrier dynamics of the photoanodes. NMO, NMF, and NMFF exhibited OCP values of 65, 76, and 103 mV, respectively, with NMFF showing the highest photovoltage, indicative of more efficient charge separation and suppressed recombination. Analysis of the OCP decay curves further revealed that NMFF possesses a longer photogenerated carrier lifetime compared to NMO and NMF, confirming more effective electron‐hole separation (Figure 5h). These observations were corroborated by time‐resolved photoluminescence (TRPL) analysis, where NMFF exhibited the longest fluorescence lifetime (1.12 ns) relative to NMF (0.88 ns) and NMO (0.52 ns) (Figure 5i), further evidencing improved charge separation and reduced recombination in NMFF.

In addition to charge carrier dynamics, surface wettability plays a critical role in facilitating interfacial charge transfer. Contact angle measurements (Figure 6a–c) revealed that NMO, NMF, and NMFF exhibit angles of 47.5°, 41.0°, and 31.6°, respectively. The significantly lower contact angle of NMFF indicates enhanced hydrophilicity, which promotes improved electrolyte infiltration and interfacial contact. This favorable wettability complements the longer photogenerated carrier lifetime observed from OCP and TRPL analyses, collectively contributing to more efficient charge separation and enhanced oxygen evolution reaction (OER) performance.

FIGURE 6.

(a‐c) Contact angle measurements, (d‐f) Bode plot and (g‐i) PEIS plot at 1.0‐1.5 V vs. RHE, (j) K_ct, (k) K_rec, and (l) Φ_ct at potential range from 1.0‐1.5 V vs. RHE.

To probe the reaction kinetics and mechanism of the photoanodes, Bode plots (Figure 6d–f; Figure S20) were analyzed in the potential range of 1.0–1.5 V vs. RHE. For NMO and NMF, increasing the applied potential leads to a gradual decrease in phase angle and a rightward shift of the peak, indicating OER activity [63]. In contrast, NMFF exhibits a pronounced decrease in the low‐frequency phase angle even at moderate potentials, reaching a minimal value at 1.5 V vs. RHE, indicative of an extremely rapid OER process. The peak shift toward higher frequencies and the shorter characteristic time constants further demonstrate faster charge carrier dynamics and more efficient separation of photogenerated electron‐hole pairs under illumination, confirming the superior intrinsic activity of NMFF. Photoelectrochemical impedance spectroscopy (PEIS) was further performed to evaluate the interfacial charge transport properties of all photoanodes in the potential range of 1.0–1.5 V vs. RHE (Figure 6g–i). An equivalent circuit model (Figure S20a) was employed to fit the impedance spectra. The circuit consists of a series resistance (Rs), bulk/charge‐transfer resistance (Rbulk and Rct), and bulk/surface capacitances (Cbulk and Csurf) [64, 65]. As shown in the Nyquist plots, the semicircle diameter gradually decreases with increasing applied potential, reflecting enhanced charge transfer kinetics. Compared to NMO and NMF, NMFF exhibits a smaller semicircle diameter, demonstrating more efficient interfacial charge transfer under illumination. Measurements at 1.23 V vs. RHE under dark and light conditions (Figure S20b,c) reveal that illumination leads to the smallest semicircle, further confirming superior charge transport. The PEIS analysis provides critical insights into the resistive and capacitive behaviors of NMO, NMF, and NMFF photoanodes, elucidating their charge transport and recombination characteristics. To assess the role of surface states in charge transfer across the solid electrolyte interface (SEI), the density of surface states (DOSS) was calculated from the C_surf values using Equation S13. As shown in Figure S21, NMFF exhibited the highest DOSS, with its peak shifting toward more positive potential. This shift suggests that the increased density of interface‐related surface states facilitates more efficient interfacial charge transfer. In contrast, NMO and NMF showed lower DOSS values and less favorable potential shifts, indicating a limited role of surface states in those materials [65, 66]. As shown in Figure S22a, NMFF exhibited the lowest R_bulk, indicative of improved internal charge mobility. The R_ct at the SEI, shown in Figure S22b, was also the lowest for NMFF, confirming that FeOOH modification and surface passivation substantially enhance interfacial charge transfer by reducing recombination. Furthermore, NMFF exhibited the highest C_bulk (Figure S22c), reflecting improved charge storage capabilities within the photoanode material. The C_surf (Figure S22d) was also maximized for NMFF, implying a greater capacity for surface charge accumulation and enhanced surface passivation.

To evaluate the combined effects of heterostructure formation and defect passivation on surface kinetics, the rate constants for surface charge transfer (K_ct) and surface recombination (K_rec) were estimated across a range of applied potentials. As shown in Figure 6j, the NMFF photoanode exhibited a K_ct value of 20 s–¹ at 1.23 V vs. RHE, significantly surpassing those of NMF and NMO. This improvement indicates accelerated charge injection into the electrolyte, enabled by enhanced charge carrier transport across the heterostructure. In contrast, the K_rec values (Figure 6k) for NMFF were the lowest among all samples at potentials above 1.23 vs. RHE. The marked suppression of recombination is attributed to the effective passivation of surface trap states by FeOOH, as well as its catalytic role in facilitating the OER [53]. The presence of oxyfluoride species in NMF further contributes to surface defect passivation, facilitating charge carrier migration toward the electrolyte [36]. At lower applied potentials, however, NMO exhibited a K_rec value exceeding K_ct, indicating significant surface charge recombination, which is consistent with the increased PL emission intensity observed for this sample. This trend confirms the presence of substantial recombination losses in NMO, which limit its PEC performance. Figure 6l shows the calculated charge transfer efficiency (Φ_CT) as a function of applied potential. NMFF consistently exhibited the highest Φ_CT values across the tested voltage range, demonstrating that the synergy between internal electric field‐driven separation and surface trap passivation leads to more efficient transport of photogenerated charge carriers. These findings highlight the effectiveness of integrating heterostructures with surface engineering strategies to optimize PEC water‐splitting performance.

3.5. Proposed Photoelectrochemical Mechanism

The enhanced PEC performance of the NMFF photoanode can be attributed to the synergistic effects of defect passivation and charge transfer invigoration via oxyfluoridization of NiMoOF and incorporation of FeOOH in NiMoOF/FeOOH hybridized heterostructures. Upon light absorption, the internal electric field at the NMFF heterostructure facilitates spatial separation of photogenerated electron‐hole pairs. Electrons migrate toward the conductive substrate, while holes are driven to the photoanode surface, reducing bulk recombination and enhancing interfacial charge transfer (Figure 7) [67]. The oxyfluoride component plays a crucial role in suppressing charge recombination by passivating surface and bulk defect states and providing more favorable electronic pathways for charge migration. In addition, it facilitates efficient hole transport toward the FeOOH layer, thereby improving hole extraction and accelerating surface oxygen evolution kinetics. FeOOH incorporation not only boosts donor density but also improves bulk conductivity and charge mobility. More importantly, FeOOH passivates surface defects, reducing trap‐state recombination and promoting hole transfer from the space‐charge region to the electrolyte (Figure 7a). This effect is validated by the reduced k_rec values observed for NMFF/FeOOH, which also functions as an active OER catalyst, enhancing interfacial charge transfer and enabling efficient hole injections into the electrolyte [54, 58]. Meanwhile, the oxyfluoride component in NMF assists in surface passivation and facilitates smoother charge‐transport pathways (Figure 7b). In contrast, NMO shows substantial recombination at lower potentials, with K_rec exceeding K_ct (Figure 7c), indicating poor charge extraction. NMFF consistently exhibits higher Φ_CT, confirming that the synergistic combination of defect passivation and charge transfer invigoration in NMFF minimizes recombination losses and promotes efficient charge movement through surface states at the solid electrolyte interface. Furthermore, oxyfluoride species formed through NiMoOF incorporation create abundant active sites, which facilitate water adsorption and improve OER activity. This highlights the importance of compositional and structural tuning for enhanced PEC performance. This mechanism underscores the vital role of compositional and structural tuning in achieving superior PEC water oxidation performance.

FIGURE 7.

Schematic illustration of the possible charge transfer mechanism under 1‐sun illumination for (a) NMFF, (b) NMF, and (c) NMO. The blue arrow indicates hole transfer; black and green arrows represent bulk and surface recombination, respectively. The pink region denotes the oxyfluoride layer on NMO.

4. Conclusions

A hybrid heterostructure NiMoOF/FeOOH photoanode was successfully fabricated using a bulk‐surface phase coupling strategy combined with defect engineering to enhance PEC water‐splitting performance. The introduction of oxygen vacancies and lattice defects in the NiMoOF layer significantly improved its electrical conductivity and charge transfer efficiency. Additionally, the FeOOH surface layer functioned both as a trap‐state passivation agent and a photo‐corrosion barrier, leading to reduced charge recombination and improved photoanode durability. Band structure analysis clarified the contribution of each deposition stage to the improved optoelectronic properties of the composite. The optimized NiMoOF/FeOOH photoanode achieved a high photocurrent density of approximately 2.18 mA cm⁻² at 1.23 V vs. RHE and demonstrated excellent long‐term stability, retaining over 94% of its initial performance after 20 h of continuous illumination. These results highlight the effectiveness of heterostructure engineering, combined with surface defect passivation and enhanced charge transfer dynamics, in significantly improving PEC performance. The proposed design strategy offers a promising pathway for the development of highly efficient and durable photoanodes for next‐generation solar‐driven water‐splitting systems.

Author Contributions

Amol S. Salunke: validation, conceptualization, methodology, formal analysis, data curation, writing – original draft; Nabeen K. Shrestha: writing – review & editing, investigation, formal analysis; Anshika Gupta, Jyoti V. Patil, Sawanta S. Mali, Ramesh J. Deokate, Chang Kook Hong: formal analysis, data curation; Sejoon Lee: writing – review & editing; Sangeun Cho: supervision, investigation, writing – review & editing; Akbar I. Inamdar: investigation, supervision, data curation, writing – review & editing.

Conflicts of Interest

The authors declare no competing financial interest.

Supporting information

Supporting File: smll74166‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data supporting the findings of this study are available upon reasonable request from the corresponding author.