Revealing Local Coordination–Modulated Oxygen Evolution Reactivity in High‐Entropy Layered Double Hydroxides

ABSTRACT

Layered double hydroxides (LDHs) are promising electrocatalysts for the oxygen evolution reaction (OER), but their sluggish kinetics and structural instability under industrial operating conditions severely limit their practical applications. Here, we demonstrate that high‐entropy NiCoZnFeCr‐LDHs (HE‐LDHs), in which the Fe³⁺: Cr³⁺ ratio is precisely controlled, can modulate redox behavior, local coordination, and catalytic activity. NiCoZnFe_0.8Cr_0.2‐LDH achieves superior OER performance with an overpotential of 199 mV at 10 mA cm⁻² with a Tafel slope of 37 mV dec⁻¹, while maintaining outstanding durability at 1 A cm⁻² over 1200 h, among the best reported for non‐noble LDH‐based catalysts. Operando spectroscopic analyses reveal that compositional variations tune the oxidation behavior of metals and the local phase evolution, including the reconstruction of Ni species into γ‐NiOOH during OER, together with a particularly significant alteration in Zn local coordination from octahedral to tetrahedral. Density functional theory calculations support that the tetrahedral Zn induces electron localization on adjacent oxygen sites, enabling internal hydrogen bonding that substantially lowers the energy barriers of the potential‐determining steps and stabilizes reaction intermediates in the lattice oxygen‐mediated mechanism (LOM). This work provides fundamental insights into the structure–activity relationships by composition tuning in multicomponent OER catalysts.

Precise Fe³⁺:Cr³⁺ compositional control in high‐entropy NiCoZnFeCr layered double hydroxides drives an octahedral‐to‐tetrahedral Zn coordination transformation that activates lattice oxygen participation by inducing electron localization on adjacent lattice oxygen sites and enabling hydrogen bonding with intermediates. This optimized catalyst achieves exceptional oxygen evolution performance with low overpotential, fast kinetics, and long‐term durability.

1. Introduction

Water electrolysis has been widely recognized as a sustainable and efficient approach for green hydrogen production, enabling large‐scale storage of renewable energy [1, 2]. However, its overall energy conversion efficiency is still restricted by the sluggish oxygen evolution reaction (OER), which requires a complex four‐electron transfer and a relatively large overpotential compared to the hydrogen evolution reaction (HER) [3]. Despite extensive efforts to improve OER performance, many state‐of‐the‐art OER catalysts still rely on Ir‐ and Ru‐based materials owing to their excellent intrinsic activity and benchmark OER performance [4, 5, 6]. However, their scarcity and high cost hinder large‐scale implementation. Therefore, the development of scalable and cost‐effective electrocatalysts that exhibit both high activity and long‐term durability under industrially relevant conditions remains a critical challenge.

Among various non‐noble catalysts, layered double hydroxides (LDHs) comprising transition metals such as Ni, Co, and Fe have emerged as particularly promising candidates for the OER in alkaline media [7, 8, 9]. This structure provides a high density of active sites and facilitates the incorporation of multiple metal cations, which can work synergistically to modulate redox properties and improve electron transfer during OER [10]. Ni, Co, and Fe are particularly effective because they undergo reversible oxidation‐state changes (e.g., Ni²⁺/Ni³⁺, Co²⁺/Co³⁺) under anodic conditions, which are closely associated with OER activity [11, 12, 13]. In addition, interlayer anions help to balance the positive charge of the metal hydroxide layers, stabilizing the lamellar structure, and can influence the local electronic environment through interlayer interactions and anion regulation effects [14, 15, 16]. However, conventional binary or ternary LDHs exhibit inherent limitations in compositional and structural flexibility, while their layered structure is susceptible to highly oxidative OER conditions [17, 18]. This induces detrimental effects, including phase transformation, cation dissolution, and unwanted surface reconstruction, which collectively undermine both catalytic activity and durability [19, 20, 21]. Overcoming these limitations necessitates the development of advanced LDH architectures with superior structural robustness and compositional tunability.

Recently, high‐entropy materials (HEMs), which incorporate multiple principal elements into a single crystalline lattice, have emerged as promising catalysts. The high configurational entropy in these systems stabilizes multicomponent phases and promotes synergistic interactions among metal cations [22, 23, 24, 25]. When applied to LDH structures, the high‐entropy approach provides a broad compositional space for regulating electronic structures, coordination environments, and defect chemistry, thereby offering opportunities to optimize catalytic activity and durability in alkaline OER [26, 27, 28, 29, 30, 31]. Despite these advantages, the underlying structure–property relationships governing redox dynamics and active site evolution in high‐entropy LDHs (HE‐LDHs) remain elusive. Particularly, the M²⁺/M³⁺ cation ratio is a critical compositional parameter that strongly influences LDH phase formation and catalytic performance. Imbalances in this ratio can induce defective structures and phase segregation, altering the local chemical environment, potentially affecting the distribution and activity of active sites [30, 31]. However, systematic studies examining how the relative ratio of trivalent metals modulates structure, composition, and catalytic behavior in HE‐LDHs are lacking, highlighting a significant gap in the rational design of these materials.

Herein, we report an in‐depth study on NiCoZnFeCr HE‐LDHs, focusing on how trivalent metal (Fe³⁺ and Cr³⁺) composition affects structural, electronic, and OER mechanistic properties. By varying the Fe³⁺/Cr³⁺ ratio across a series of compositions (0.8:0.2, 0.5:0.5, and 0.2:0.8) at constant M²⁺/M³⁺ precursor ratio, we tuned the local coordination environment and electronic structure within the LDH framework. The optimized NiCoZnFe_0.8Cr_0.2‐LDH catalyst achieves exceptional OER performance with an overpotential of 199 mV at 10 mA cm⁻², a Tafel slope of 37.4 mV dec⁻¹, as well as remarkable stability over 1200 h at 1 A cm⁻², surpassing all other compositions examined within the HE‐LDH series. In situ extended X‐ray absorption fine structure (EXAFS) analysis indicated significant modifications in Zn coordination with a tetrahedral geometry, while in situ X‐ray absorption and Raman spectroscopy revealed that the oxidation states and phase evolution of Ni, Co, and Cr vary with cation ratio. Mechanistic studies using ¹⁸O‐isotope differential electrochemical mass spectrometry (DEMS) demonstrated that these high‐performing HE‐LDH catalysts operate via a lattice oxygen‐mediated mechanism (LOM), with the extent of LOM participation dependent on cation composition. Density functional theory (DFT) calculations corroborated that the stabilization of the Zn‐induced structural modifications within NiCoZnFe_0.8Cr_0.2‐LDH thermodynamically favors the LOM pathway, directly linking cation composition to local structure and enhanced OER activity. Collectively, these findings establish compositional engineering as an effective strategy for modulating structure and designing high‐performance HE‐LDH electrocatalysts.

2. Result and Discussion

2.1. Material Characterization

HE‐LDHs were synthesized on a nickel foam via a hydrothermal process at 160°C for 12 h (Figure 1a). The selected metal nitrate precursors comprise 3d transition metals (Cr, Fe, Co, Ni, Zn), wherein Fe and Cr are present as trivalent (+3) cations, while Co, Ni, and Zn exist as divalent (+2) cations. A series of HE‐LDH structures was successfully prepared with precisely controlled Fe/Cr composition ratios (i.e., 0.8:0.2, 0.5:0.5, and 0.2:0.8) while maintaining a constant molar ratio of divalent‐to‐trivalent metal precursors (M²⁺/M³⁺ = 3:1) (Figure S1). Compositions in the synthesized samples were quantified using an inductively coupled plasma‐optical emission spectrometer (ICP‐OES, Table S1). First, X‐ray diffraction (XRD) analysis demonstrated that NiFe‐LDH, NiCoZnFe‐LDH, and HE‐LDHs with varying trivalent metal ratios retain a single‐phase structure of hexagonal iron nickel carbonate hydroxide hydrate structure (PDF#40‐0215) (Figures 1b and S2). The slight shift in the low‐angle reflections relative to the NiFe‐LDH reference pattern can be attributed to interlayer‐spacing changes induced by the incorporation of multiple metal species. Furthermore, the progressive broadening of the diffraction peaks with increasing elemental complexity suggests increased structural disorder within the LDH. Scanning electron microscopy (SEM) images confirmed uniform and homogeneous growth of well‐defined sheet structures across the entire sample (Figure S3). High‐resolution TEM analysis substantiated the characteristic sheet‐like morphology of LDHs, with individual sheets exhibiting ultrathin, nanometer‐scale thickness (Figure 1c). A lattice fringe analysis using inverse fast Fourier transform (IFFT) on two highlighted regions revealed interplanar spacings of 0.231 and 0.256 nm, corresponding to the (015) and (012) crystallographic planes of the LDH structure, respectively. Selected‐area electron diffraction (SAED) patterns consistently confirmed that the sample has a hexagonal iron‐nickel carbonate hydroxide hydrate structure (Figure 1d,e).

FIGURE 1.

Synthesis and material characterization. (a) Synthesis scheme of High‐Entropy LDH on nickel foam. (b) XRD patterns. (c) High‐Resolution TEM image of NiCoZnFe_0.8Cr_0.2‐LDH. (d) SAED pattern. (e) HRTEM image with the corresponding IFFT. (f) HAADF‐STEM image with STEM‐EDS elemental mappings.

All three HE‐LDH (NiCoZnFe _x Cr _1‐x ‐LDH: x = 0.2, 0.5, and 0.8) catalysts exhibit similar Fourier transform infrared (FTIR) spectra, confirming the preservation of interlayer chemistry and the characteristic LDH framework (Figure S4). Specifically, a broad absorption band centered at ∼3400 cm⁻¹ corresponds to the stretching vibration of hydroxyl groups, while the peak at ∼1650 cm⁻¹ is assigned to the bending mode of interlayer water molecules [32]. The signal observed at ∼1360 cm⁻¹ arises from the stretching vibrations of interlayer carbonate or nitrate species [33]. In the low‐wavenumber region, the bands located at ∼770 and ∼510 cm⁻¹ are attributed to metal–oxygen (M–O) lattice vibrations [34]. Scanning transmission electron microscopy‐energy dispersive X‐ray spectroscopy (STEM‐EDS) elemental mapping revealed a uniform distribution of all five incorporated elements throughout the sample matrix (Figures 1f and S5). Moreover, X‐ray photoelectron spectroscopy (XPS) analysis confirmed nearly identical oxidation states for Ni, Co, Fe, Zn, and Cr, with no notable shifts across the HE‐LDH series (Figures S6–S10, Table S2). The O 1s spectra of all samples can be deconvoluted into the same major oxygen components, including lattice oxygen (M‐O), hydroxyl groups (M‐OH), oxygen vacancies (V_o), and adsorbed water (H₂O_Ads), indicating that the overall types of surface oxygen species are maintained across the samples (Figure S11 and Table S3). Nevertheless, composition‐dependent differences in the relative fractions of the oxygen vacancies are observed, showing a gradual increase with increasing compositional complexity, from binary to high‐entropy compositions, which may be associated with local structural distortions induced by compositional complexity in multi‐metal LDHs. In addition, within the HE‐LDH samples, a certain degree of variation in the V_o fraction is detected upon tuning the Fe/Cr ratio, indicating that not only the overall elemental complexity but also the specific trivalent‐metal composition can modulate the local defect structure. This trend is generally consistent with previously reported behavior in multi‐metal oxide/hydroxide systems, where increasing the number of constituent elements enhances local lattice disorder and promotes defect‐related oxygen formation.

Overall, all HE‐LDH samples exhibit highly comparable physicochemical characteristics. Electron microscopy and diffraction analyses confirm uniform sheet‐like morphologies and preserved LDH crystallinity. Elemental mapping reveals homogeneous metal distributions, and XPS analysis shows nearly identical metal oxidation states and surface oxygen environments. These results demonstrate that the HE‐LDH samples share highly similar morphological, structural, and surface electronic characteristics prior to electrochemical evaluation.

2.2. OER Performances

Following a comprehensive analysis, electrochemical OER measurements were conducted for the HE‐LDHs and control samples, including NiFe‐LDH and NiCoZnFe‐LDH in 1 м KOH electrolyte. Polarization curves were corrected with 90% iR compensation (Figure 2a). Among all catalysts examined, NiCoZnFe_0.8Cr_0.2‐LDH demonstrated markedly superior OER performance, requiring overpotentials of only 199 mV at 10 mA cm⁻² and 236 mV at 100 mA cm⁻². NiCoZnFe_0.5Cr_0.5‐LDH also exhibited excellent activity with overpotentials of 219 and 267 mV at 10 and 100 mA cm⁻², respectively, although slightly less active (Figure 2b). After completing the electrochemical measurements, post‐electrochemical characterizations were performed to assess possible structural and compositional changes after OER testing. SEM and TEM analyses revealed no noticeable morphological changes, and EDS mapping consistently showed a uniform distribution of all constituent elements (Figures S12‒S14). Although the XRD patterns exhibited reduced peak intensities, all major diffraction peaks remained at identical positions, confirming that the crystal structure was preserved (Figure S15).

FIGURE 2.

Electrochemical OER performance. (a) Linear sweep voltammetry curves in the negative direction (scan rate of 1 mV s⁻¹) in 1 м KOH electrolyte. (b) Overpotential comparison at j =  10, 100, and 250 mA cm⁻². (c) Tafel slope plots of NiFe‐LDH, NiCoZnFe‐LDH, and HE‐LDHs, (d) Comparison of the overpotentials and Tafel slopes of this work (NiCoZnFe_0.8Cr_0.2‐LDH) with other high entropy reported in the literature. (e) Stability test of NiCoZnFe_0.8Cr_0.2‐LDH at 1.0 A cm⁻².

To further evaluate OER kinetics, Tafel slopes were extracted from the polarization curves (Figure 2c). Consistent with the overpotential trends, NiCoZnFe_0.8Cr_0.2‐LDH displayed the lowest Tafel slope of 37.4 mV dec⁻¹, indicating optimal reaction kinetics. Notably, when comparing catalysts containing Cr with control catalysts including NiCoZnFe‐LDH and NiFe‐LDH, a dramatic reduction in Tafel slope was observed upon Cr incorporation, specifically when all five metallic elements were present. Both the overpotential and Tafel slopes indicated that the OER performance of the NiCoZnFeCr‐LDH catalysts was largely enhanced compared to NiFe‐LDH or NiCoZnFe‐LDH. This enhancement is consistent with previous reports showing that Cr incorporation into LDH frameworks modulates the electronic structure of neighboring transition‐metal centers, enhancing OER kinetics [35, 36, 37, 38].

The electrochemical surface area (ECSA) was also compared by performing cyclic voltammetry at various scan rates within the non‐Faradaic region (Figure S16) and extracting double‐layer capacitance (C_dl) values (Figure S17). Accordingly, the differences in catalytic performance reflect contributions from intrinsic catalytic properties beyond variations in active surface area, as indicated by the ECSA‐normalized OER activities (Figure S18). Despite exhibiting larger ECSA values, NiFe‐LDH and NiCoZnFe‐LDH catalysts still showed inferior OER performance compared to the HE‐LDH catalysts (Table S4). The substantial enhancement in OER activity upon Cr incorporation suggests that Cr plays a role in promoting intrinsic activity beyond surface area contributions. The synergistic multi‐metal interactions, modified electronic structures, and the high‐entropy effect associated with configurational entropy in the HE‐LDH system can collectively drive the enhanced OER performance. Electrochemical impedance spectroscopy (EIS) was further employed to evaluate the intrinsic resistance and charge‐transfer behavior (Figure S19). Measurements were conducted at 1.47 V versus reversible hydrogen electrode (RHE), near the onset overpotential for OER. The Nyquist plots revealed that NiCoZnFe_0.8Cr_0.2‐LDH and NiCoZnFe_0.5Cr_0.5‐LDH exhibited much lower charge‐transfer resistance (R_ct) than NiFe‐LDH and NiCoZnFe‐LDH. Specifically, NiCoZnFe_0.8Cr_0.2‐LDH displayed the lowest charge‐transfer resistance of 3.6 Ω, followed by NiCoZnFe_0.5Cr_0.5‐LDH with 3.8 Ω. These results indicate reduced kinetic barriers for electron transfer at the catalyst‐electrolyte interface, consistent with the enhanced OER activity. Notably, the lower R_ct values of these catalysts are consistent with their higher current densities and lower overpotentials observed in LSV measurements, confirming that enhanced interfacial charge transfer contributes significantly to their improved OER performance.

Based on its exceptional performance, the best‐performing catalyst, NiCoZnFe_0.8Cr_0.2‐LDH, was benchmarked against other recently reported high‐entropy alloys and LDH‐based catalysts from the past five years (Figure 2d) [27, 34, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48]. Excluding noble metal‐containing catalysts, NiCoZnFe_0.8Cr_0.2‐LDH exhibited significantly lower overpotentials and Tafel slopes, demonstrating faster OER kinetics than most reported systems. Furthermore, the durability of the NiCoZnFe_0.8Cr_0.2‐LDH catalyst was first evaluated by a 100 h chronoamperometric test at 1 A cm⁻² (Figure S20). The catalyst exhibited negligible potential decay during this period, indicating excellent operational stability. Subsequently, the structural and compositional integrity of NiCoZnFe_0.8Cr_0.2‐LDH was confirmed by post‐stability characterization, which revealed no significant structural degradation, morphological changes, or compositional variations, demonstrating that the catalyst retains both structural and elemental stability even under high current density operation (Figures S21 and S22). To further verify that the observed current originated predominantly from oxygen evolution rather than side reactions, gas quantification was performed using a water displacement method during OER operation at 1 A cm⁻² (Figures S23 and S24). The experimentally measured gas volumes agreed well with the values theoretically expected from the passed charge for O₂ at the anode and H₂ at the cathode, respectively. This indicates that the Faradaic efficiencies remained close to 100% with negligible contributions from parasitic reactions such as peroxide formation. Furthermore, the catalyst demonstrated stable OER performance for more than 1,200 h at 1 A cm^−2, highlighting its promise as a highly durable electrocatalyst for practical applications (Figure 2e). Over 1200 h, the potential remained stable, showing remarkable stability even under high‐current operation. When compared with recently reported high‐entropy alloy and LDH catalysts in the last five years (Figure S25), NiCoZnFe_0.8Cr_0.2‐LDH exhibited outstanding stability at significantly higher current densities despite comprising solely non‐noble transition metals [27, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60]. This exceptional durability is attributed to the intrinsic structural robustness of the HE‐LDH framework. The incorporation of multiple metal species, Ni, Co, Zn, Fe, and Cr, leads to an entropy‐stabilized configuration, which thermodynamically favors structural stability and suppresses phase segregation under harsh OER conditions [61]. In addition, the coexistence of diverse metal–oxygen coordination environments can effectively redistribute local electronic and chemical stresses, thereby preventing localized degradation. Such a stabilized multi‐metal framework may also mitigate excessive surface reconstruction, contributing to the observed long‐term durability at high current densities.

2.3. Structural and Electronic Evolution Comparison From In Situ/Operando Spectroscopy Investigation

To investigate real‐time structural and electronic evolution during OER, in situ/operando X‐ray absorption spectroscopy (XAS) measurements were performed. The catalysts selected for this study were NiCoZnFe_0.8Cr_0.2‐LDH and NiCoZnFe_0.5Cr_0.5‐LDH, which exhibited the lowest and second‐lowest overpotentials, respectively. The in situ cell setup is presented in Figure S26. The characterization was performed using nanosheets grown on carbon paper (Figure S27). X‐ray absorption near‐edge structure (XANES) spectra were acquired for each element at open circuit potential (OCP), 1.25 V, 1.35 V, and 1.45 V versus RHE. Significant spectral evolution was observed at Ni K‐edge XANES from 1.35 V onward, consistent with oxidation of Ni species and NiOOH phase formation (Figure 3a) [62, 63]. These changes are consistent with the previous Ni LDH catalyst. At 1.45 V, the intensity of the main absorption peak decreased, and the absorption edge shifted positively by approximately 1.52 eV relative to OCP, confirming further Ni oxidation. Overall, Ni exhibited the largest potential‐dependent edge shift among all metals, indicating it undergoes the most significant oxidation during OER. Cr K‐edge spectra under identical potential conditions showed no appreciable changes in the main absorption peak intensity or oscillation patterns, indicating the absence of major phase transitions (Figure 3b). Nevertheless, a positive energy shift emerged at 1.35 V, similar to that observed for Ni, suggesting partial Cr oxidation near the OER onset potential. Similarly, the Co K‐edge exhibited potential‐dependent edge shifts. A positive shift of 0.64 eV was observed at 1.45 V relative to OCP, while the main absorption peak intensity and oscillation features remained largely unchanged, indicating minimal structural rearrangement of Co sites (Figure 3c). In contrast, nearly discernible changes in the Fe and Zn K‐edge spectra were not observed over the applied potential range, suggesting that these elements contribute primarily to structural or electronic‐stabilizing roles on the catalyst surface (Figures S28 and S29) [64].

FIGURE 3.

Operando spectral analysis. (a) In situ X‐ray absorption near edge structure (XANES) spectra for the Ni K‐edge, (b) Cr K‐edge, (c) Co K‐edge of NiCoZnFe_0.8Cr_0.2‐LDH. (d) In situ X‐ray absorption near edge structure (XANES) spectra for the Ni K‐edge, (e) Cr K‐edge, (f) Co K‐edge of NiCoZnFe_0.5Cr_0.5‐LDH. (g) In situ Fourier‐transformed k³‐weighted extended X‐ray absorption fine structure (EXAFS) spectra for Ni K‐edge of NiCoZnFe_0.8Cr_0.2‐LDH. (h) In situ spectra for Ni K‐edge of NiCoZnFe_0.5Cr_0.5‐LDH. (i) Heatmap of in situ Raman Spectra of NiCoZnFe_0.8Cr_0.2‐LDH with the potential range between OCP to 1.4 V. (j) Heat map of in situ Raman spectra of NiCoZnFe_0.5Cr_0.5‐LDH with the potential range between OCP to 1.4 V.

To further investigate potential‐dependent electronic evolution of the control catalyst, NiCoZnFe_0.5Cr_0.5‐LDH, in situ XAS measurements were conducted under OCP, 1.25 V, 1.35 V, and 1.45 V versus RHE. Notably, all three redox‐active metals (Ni, Cr, and Co) in NiCoZnFe_0.5Cr_0.5‐LDH showed consistently smaller potential‐induced shifts compared to NiCoZnFe_0.8Cr_0.2‐LDH, indicating attenuated electronic evolution with higher Cr content (Figure 3d–f). At the Ni K‐edge XANES, a positive shift was observed from 1.35 V onward, indicating Ni oxidation and progressive NiOOH phase formation. Upon increasing the potential to 1.45 V, the Ni K‐edge of the NiCoZnFe_0.5Cr_0.5‐LDH catalyst shifted by 1.25 eV relative to OCP. Compared to the NiCoZnFe_0.8Cr_0.2‐LDH catalyst, this shift was 0.27 eV smaller, suggesting that Ni oxidation is less favorably induced at a given potential. The Cr K‐edge spectra also exhibited gradual positive shifts with increasing potential, showing a shift of 0.76 eV from OCP to 1.45 V, which was 0.41 eV smaller than that of the NiCoZnFe_0.8Cr_0.2‐LDH catalyst. Similarly, the Co K‐edge exhibited positive shifts of 0.41 eV from OCP to 1.45 V, which was 0.23 eV smaller than the 0.64 eV shift in NiCoZnFe_0.8Cr_0.2‐LDH. Similar to the NiCoZnFe_0.8Cr_0.2‐LDH catalyst, both Fe and Zn K‐edge spectra remained unchanged across the applied potential range (Figures S28 and S29). While the Zn K‐edge also showed no potential‐dependent changes, the Zn spectra exhibited unanticipated structural differences as the composition ratio between NiCoZnFe_0.8Cr_0.2‐LDH and NiCoZnFe_0.5Cr_0.5‐LDH was varied, which are discussed later.

Collectively, the in situ XAS results demonstrate that the NiCoZnFe_0.5Cr_0.5‐LDH exhibits reduced oxidation responsiveness to the applied potential compared to the NiCoZnFe_0.8Cr_0.2‐LDH catalyst, particularly for Ni, Cr, and Co. The reduced edge shifts indicate that fewer high‐valent Ni and Co species are formed at a given potential, which may account for the comparatively lower OER activity. Additionally, although Cr has generally been regarded as not serving a direct active site, in situ XAS reveals modulation of the Cr edge, suggesting that Cr participates in the redox process through electronic interactions with neighboring metals. Then, it can tune the local electronic environment and indirectly influence the oxidation behavior of adjacent Ni and Co species [65, 66]. These observations underscore the importance of precise compositional tuning in HE‐LDHs to optimize the formation of redox‐active species and enhance OER performance.

To further probe the local atomic environment and bond distances, operando EXAFS measurements were conducted with particular focus on Ni K‐edge under applied potentials (Figures 3 g,h, S30, and S31, Table S6). For the main catalyst, NiCoZnFe_0.8Cr_0.2‐LDH, the Ni K‐edge EXAFS spectra at OCP and 1.25 V exhibited minimal structural changes, with the Ni–O and Ni–M peaks remaining at similar radial distances. Correspondingly, EXAFS fitting revealed only slight variations in the Ni–O and Ni–M bond distances from 2.044 and 3.048 Å at OCP to 2.028 and 3.082 Å at 1.25 V, respectively. Upon increasing the potential to 1.35 and 1.45 V, both Ni–O and Ni–M peaks shifted to lower radial distances, indicating bond contraction [67, 68]. The Ni–O bond distance decreased to 1.954 Å at 1.35 V and further to 1.923 Å at 1.45 V, while the Ni–M distance decreased to 2.835 and 2.829 Å, respectively. Meanwhile, the control catalyst NiCoZnFe_0.5Cr_0.5‐LDH exhibited a different structural evolution. Although the Ni–O and Ni–M bond distances decreased with increasing potential, the EXAFS spectrum at 1.35 V displayed two distinct coordination features beyond 2 Å, and satisfactory fitting required inclusion of an additional Ni(OH)₂ coordination environment. It suggests the coexistence of oxidized NiOOH species and incompletely oxidized Ni species. These results imply that the extent of Ni oxidation is influenced by the relative trivalent metal composition within the HE‐LDH framework. In NiCoZnFe_0.8Cr_0.2‐LDH, the altered local electronic environment appears to facilitate the formation of high‐valent Ni species under OER‐relevant potentials, which may contribute to its superior OER activity.

To gain deeper insight into the structural transformation behavior of HE‐LDHs during oxygen evolution and to pinpoint their catalytically active species, in situ Raman spectroscopy was employed under operating OER conditions (Figure 3i,j). As shown in the heatmap of NiCoZnFe_0.8Cr_0.2‐LDH, distinct vibrational features appear even at OCP, which can be attributed to the E_g bending δ(Ni^III–O) and A_1g stretching ν(Ni^III–O) modes of the α‐Ni(OH)₂ phases [69]. The potential was subsequently increased from 1.25 V in 0.01 V increments, and at 1.30 V, a clear enhancement of the δ(Ni^III–O) band relative to the ν(Ni^III–O) vibration was observed, indicating the onset of phase reconstruction toward γ ‐NiOOH (Figure S32).

Compared to the OCP spectrum, both Raman bands exhibited an evident red shift, with the δ(Ni^III–O) mode moving from 458.1 to 454.0 cm⁻¹, and the ν(Ni^III–O) vibration shifting from 533.8 to 528.3 cm⁻¹ as the potential reached 1.30 V. These positions correspond well to the characteristic γ‐NiOOH vibrations reported in previous studies [70, 71]. Importantly, the appearance of these features occurs near the OER onset potential, implying that the structural conversion of Ni species is directly linked to the activation of catalytic sites. Together with the Ni K‐edge XAS results, these operando Raman observations demonstrate that the reconstruction of Ni species into γ‐NiOOH occurs concomitantly with the onset of OER, directly linking Ni‐centered structural transformation to catalytic activation in HE‐LDHs.

In contrast, the NiCoZnFe_0.5Cr_0.5‐LDH catalyst exhibited a similar but slightly delayed structural evolution. The characteristic δ(Ni^III–O) and ν(Ni^III–O) peaks appeared at a higher potential of approximately 1.32 V, indicating that the oxidation of Ni(II) to Ni(III) and subsequent formation of the NiOOH phase occurred at a later stage compared with NiCoZnFe_0.8Cr_0.2‐LDH. These results revealed that the local electronic structure and metal–oxygen covalency around Ni sites were altered by the trivalent Fe³⁺/Cr³⁺ ratio. The resulting redistribution of charge density may stabilize the Ni(II)‐dominated hydroxide lattice, thereby increasing the energy barrier for the Ni(II) to Ni(III) oxidation and subsequent phase transformation. Such a trend aligns well with the electrochemical performance data, where the higher activation potential correlates with slightly reduced OER activity, reinforcing the link between Ni oxidation dynamics and the catalytic efficiency of HE‐LDHs.

2.4. Structural Modification of Zn Sites of NiCoZnFe_0.8Cr_0.2‐LDH

The variation in trivalent metal composition exerted a pronounced influence on the redox behavior of multiple cations within the HE‐LDH lattice. In situ XAS measurements revealed that the oxidation of Ni, Co, and Cr proceeded more extensively in NiCoZnFe_0.8Cr_0.2‐LDH than in NiCoZnFe_0.5Cr_0.5‐LDH, indicating that subtle adjustments in the trivalent‐metal ratio can collectively modulate the electronic structure and redox dynamics of the entire multi‐cation network. Remarkably, this compositional tuning also led to an unanticipated structural response at the Zn sites. Although the two catalysts exhibited nearly identical physicochemical characteristics as determined by XRD and electron microscope, their Zn K‐edge XAS spectra showed distinctly different spectral features (Figure 4a,b). When compared with reference materials, the XANES profile of NiCoZnFe_0.5Cr_0.5‐LDH closely matched that of the reference with Zn occupying the octahedral site (ZnAl‐LDH), whereas NiCoZnFe_0.8Cr_0.2‐LDH showed spectral characteristics similar to Zn with tetrahedral geometry (zinc ferrite). The crystal structures of both reference compounds were validated by XRD (Figures S33 and S34) and further confirmed through their respective EXAFS fittings (Figure S35, Table S5). Using these verified references as benchmarks, EXAFS fitting of the samples confirms that Zn in NiCoZnFe_0.5Cr_0.5‐LDH predominantly adopts an octahedral geometry, while Zn in NiCoZnFe_0.8Cr_0.2‐LDH exhibits a tetrahedral coordination (Figure S36). Additionally, wavelet transformation of EXAFS (WT‐EXAFS) analysis of ZnAl‐LDH, zinc ferrite (Figure S37), alongside NiCoZnFe_0.5Cr_0.5‐LDH and NiCoZnFe_0.8Cr_0.2‐LDH samples (Figure S38), revealed clear structural similarity between NiCoZnFe_0.5Cr_0.5‐LDH and ZnAl‐LDH, as well as between NiCoZnFe_0.8Cr_0.2‐LDH and zinc ferrite. Furthermore, the second coordination shell peaks enabled discrimination between the two distinct coordination environments, as the M–M distances differ substantially between octahedral Zn (∼ 2.80 Å) and tetrahedral Zn (∼3.39 Å).

FIGURE 4.

Structural investigation of NiCoZnFe_0.8Cr_0.2‐LDH. (a–b) XAS spectra of Zn for various samples. (c–d) Atomic‐resolution STEM image of NiCoZnFe_0.8Cr_0.2‐LDH overlaid with model LDH structure and its enlarged image, respectively. (e) Intensity profile of the STEM image. (f) Simulated STEM image of the model LDH. (g) Intensity profile of the simulated STEM image as a function of distance. (h) Supercell structure of NiCoZnFe_0.8Cr_0.2‐LDH (top view). (i) Schematic illustration of the NiCoZnFe_0.8Cr_0.2‐LDH structure (side view, blue circles indicate transition metals). (j) Model structure of NiCoZnFe_0.8Cr_0.2‐LDH (side view).

To understand the local geometry of Zn in NiCoZnFe_0.8Cr_0.2‐LDH, the zinc ferrite, which resembles the XANES spectra of NiCoZnFe_0.8Cr_0.2‐LDH, is schematically shown in Figure S39 (cleaved along (111) facet, side view, and top view, respectively). Red octahedra represent FeO_6, whereas gray tetrahedra represent ZnO₄. The side view shows that a plane consisting of Fe octahedra is formed, and two tetrahedra are placed on top and bottom of the plane. From the top view, the structure is also considered as two zinc tetrahedra placed on an empty octahedron with octahedral vacancy V_Fe'''. An empty octahedron and two tetrahedra are connected via face‐sharing. From the well‐known fact that LDHs form layers that consist of octahedra, it is considered that NiCoZnFe_0.8Cr_0.2‐LDH also mainly forms layers that consist of M(OH)₆ octahedra while simultaneously having some amount of octahedral metal vacancies to have Zn tetrahedra on both sides. Previous reports of metal hydroxide structures that contain Zn(OH)₄ tetrahedra on octahedral matrices, such as zinc hydroxide nitrate [71] and namuwites [72]_, further support the viability of the existence of tetrahedral Zn in such structures.

To investigate how the layers are stacked in NiCoZnFe_0.8Cr_0.2‐LDH and to establish a plausible structural model, an atomic‐resolution STEM image was analyzed and compared with the simulated STEM image (Figure 4c–g). Five hydroxide layers were used in the model structure for the simulation, and a Debye‐Waller factor of 0.87, extracted from the experimental XRD pattern (Figure S40), was used as input to account for thermal vibrations. In an atomic‐resolution STEM image (Figure 4c) overlapped with the model structure of LDH, it can be seen that the two alternating layers of NiCoZnFe_0.8Cr_0.2‐LDH are stacked in the same manner [73]. Note that AA‐stacked LDH would yield a triangular arrangement. For further investigation, STEM image simulation of the model LDH structure was conducted and compared with the experimental STEM image. From the honeycomb‐like placements of atoms shown both in Figure 4d,f, it is again confirmed that NiCoZnFe_0.8Cr_0.2‐LDH consists of two alternating layers. The similarities in the intensity profiles of the real STEM image (Figure 4e) and the simulated STEM image (Figure 4g) validate the STEM image simulation.

Next, the representative cell of NiCoZnFe_0.8Cr_0.2‐LDH was constructed, as shown in Figure 4h–j, which presents a schematic illustration of the structure. After structure relaxation with DFT, the extended supercell of NiCoZnFe_0.8Cr_0.2‐LDH with tetrahedral Zn coordination is obtained. Figure 4h represents the top view of the structure, whereas Figure 4i,j represent the side view of the structure. Two layers facing opposite directions appear to alternate, with carbonate ions intercalated between them. To reflect its high‐entropy nature, metals on the octahedral site were sampled while conserving the metal compositions using special quasi‐random structures (SQSs) simulation [74]. The supercell structure reflects the metal compositions and includes CO₃ ²⁻ ions, in agreement with experimental FT‐IR spectroscopy results (Figure S4). On the other hand, the DFT‐relaxed supercell structure of NiCoZnFe_0.5Cr_0.5‐LDH is shown in Figure S41, which reflects the ideal stoichiometry and intercalated anions. Moreover, to validate the existence of tetrahedral Zn in our model structures, the formation enthalpies of tetrahedral and octahedral Zn configurations were compared. The results confirmed that Zn preferentially exists in tetrahedral form rather than in octahedra, and the difference in formation enthalpies is larger in NiCoZnFe_0.8Cr_0.2‐LDH than in NiCoZnFe_0.5Cr_0.5‐LDH (Figure S42). Finally, XANES simulation was performed based on the structure we constructed for NiCoZnFe_0.8Cr_0.2‐LDH (Figure S43). The simulated XANES profile exhibited the same features as the experimentally obtained spectra, further validating the constructed structure and confirming that Zn exists in a tetrahedral coordination environment (Supplementary Note 3).

Importantly, prior to discussing the mechanistic implications, we performed EXAFS analysis of the Ni, Co, and Fe sites in NiCoZnFe_0.5Cr_0.5‐LDH and NiCoZnFe_0.8Cr_0.2‐LDH to assess whether the change in Zn‐site coordination induces broader structural rearrangements within the LDH framework. Their coordination numbers and bond distances were found to remain largely unchanged despite the variation in the Zn coordination environment (Figures S44–S46, Table S5), indicating that the structural perturbation is localized around the Zn–O bonds. To estimate whether the activity enhancement arises solely from Zn‐site modulation, we further characterize the Zn states of the control NiCoZnFe‐LDH catalyst by XANES analysis (Figure S47). This control sample exhibits XANES features similar to those of NiCoZnFe_0.8Cr_0.2‐LDH, indicating that Zn adopts a comparable tetrahedral‐like coordination in both catalysts. Thus, Zn‐site coordination alone cannot fully account for the catalytic enhancement, particularly the distinct activity difference between NiCoZnFe‐based and high‐entropy LDH systems in LSV measurements (Figure 2a). While the local Zn geometry may contribute to the catalytic behavior, it is not the unique descriptor of the overall activity trend. These results imply that Fe/Cr compositional tuning and multi‐metal interactions within the NiCoZnFeCr‐LDH framework play a key role in OER improvement, as elucidated in the following mechanistic section.

2.5. Origin of the Enhanced OER Activity With Composition Modulation: Mechanistic Investigation

To understand the origin of the enhanced OER activity with NiCoZnFe_0.8Cr_0.2‐LDH, the reaction mechanisms for NiCoZnFe_0.8Cr_0.2‐LDH and NiCoZnFe_0.5Cr_0.5‐LDH were investigated with the structures in Figure 4. The mechanism of OER can be divided into two categories: the adsorbate evolution mechanism (AEM) and the lattice oxygen‐mediated mechanism (LOM) [75]. Because the tetramethyl ammonium hydroxide (TMAOH) is known to bind with O₂ ²⁻ intermediates, we have conducted LSV measurements with an electrolyte containing TMAOH and compared the curves with the as‐obtained LSV curves with bare 1 м KOH media (Figure 5a–c) [27, 76]. The NiFe‐LDH displayed almost indistinguishable LSV curves in KOH and TMAOH electrolytes, showing that both the onset potential and the slope of the anodic sweep were conserved. This insensitivity to cation substitution suggests that the reaction pathway on the NiFe‐LDH catalyst is dominated by surface‐adsorbed intermediates, with almost negligible contribution from lattice oxygen.

FIGURE 5.

OER Mechanism Study. (a) Linear sweep voltammetry curves measured in 1 м KOH and 1 м TMAOH for NiFe‐LDH, (b) NiCoZnFe_0.8Cr_0.2‐LDH, (c) NiCoZnFe0.5Cr0.5‐LDH, (d) Raman spectra of the corresponding LDH catalysts after electrochemical test under TMAOH, (e) DEMS quantification of ³⁴O₂/³⁶O₂ ratios over three cycles of NiCoZnFe_0.8Cr_0.2‐LDH and NiCoZnFe_0.5Cr_0.5‐LDH.

In stark contrast, NiCoZnFe_0.8Cr_0.2‐LDH exhibited a noticeable decrease in OER activity upon replacing KOH with TMAOH. The onset potential shifted to more positive values, and the overall current density was suppressed throughout the anodic sweep. Such behavior is consistent with catalysts whose OER performance is partially sustained by lattice oxygen participation; where oxygen exchange pathways are blocked or inhibited by TMA⁺, the kinetics slow down accordingly. The same trend was observed for NiCoZnFe_0.5Cr_0.5‐LDH, which likewise exhibited substantial current suppression and a notably flattened OER profile in TMAOH.

Additionally, Raman measurements were performed to confirm the presence of TMAOH species on the catalyst surfaces for the three catalysts (Figure 5d). For NiFe‐LDH, no noticeable TMA⁺‐related vibrational features were detected, indicating little to no interaction between the electrolyte cation and the catalyst surface. In contrast, both high‐entropy LDH catalysts exhibited distinct high‐frequency peaks attributable to TMA⁺ species at 751 cm⁻¹ and 949cm⁻¹, demonstrating that TMA⁺ was retained at or near their surface after electrochemical operation [77]. The selective presence of these TMA⁺ features on the HE‐LDHs supports that their surface oxygen environments are more dynamically involved during OER, allowing cation accumulation or association, which does not occur on NiFe‐LDH. Mechanistically, this behavior aligns with catalysts capable of lattice‐oxygen redox, whereas the absence of such features in NiFe‐LDH is consistent with an adsorbate‐mediated pathway.

As TMAOH poisoning measurements and Raman spectroscopy mainly provide qualitative evidence for lattice oxygen‐related OER characteristics, ¹⁸O isotope differential electrochemical mass spectrometry (DEMS) analysis was further performed to more directly assess the extent of lattice oxygen participation during OER. (Figures 5e and S48). NiCoZnFe_0.8Cr_0.2‐LDH exhibited a pronounced ³⁴O₂ signal throughout all three cycles, resulting in a ³⁴O₂/³⁶O₂ ratio of approximately 0.25. The strong and reproducible formation of ³⁴O₂ indicates significant exchange between lattice oxygen and the electrolyte‐derived oxygen atoms, consistent with a dominant LOM [78, 79, 80]. Meanwhile, NiCoZnFe_0.5Cr_0.5‐LDH showed a lower level of ³⁴O₂ evolution, giving a ³⁴O₂/³⁶O₂ ratio of approximately 0.10 across repeated cycles. While reduced relative to NiCoZnFe_0.8Cr_0.2‐LDH, the detectable formation of ³⁴O₂ clearly indicates that lattice‐oxygen participation still occurs, albeit to a lesser extent. These results suggest a mixed catalytic pathway in which lattice oxygen contributes but does not dominate the entire OER.

Based on reaction mechanism studies, DFT simulations were employed to elucidate the structure‐property relationship between the two structures. The LOM‐active oxyhydroxide phase was deduced from the as‐constructed initial LDH structures (Figures S49 and S50, Supplementary Note 1). One notable feature of the structure is that the dangling oxygen atom bound to the inner tetrahedral Zn participates in a hydrogen‐bond network with water molecules, contributing to the stability of the oxyhydroxide phase. Using the as‐constructed structure models, reaction pathways for OER on two structures are displayed (Figures 6a and S51). The reaction site is consistent with the experimental results from the XANES analysis (Supplementary Note 2).

FIGURE 6.

Computational study on OER mechanism (a) Proposed reaction mechanism of NiCoZnFe_0.8Cr_0.2‐LDH. (b–c) Reaction scheme drawn for NiCoZnFe_0.8Cr_0.2‐LDH and NiCoZnFe_0.5Cr_0.5‐LDH. (d) Relaxed structure of *OH‐adsorbed NiCoZnFe_0.8Cr_0.2‐LDH with the M‐O bond lengths shown for octahedra adjacent to Zn tetrahedra. (e) Relaxed structure of *OH‐adsorbed NiCoZnFe_0.8Cr_0.2‐LDH with Mulliken charge.

To investigate the detailed mechanisms among various possible LOM reaction schemes, common OER adsorbates (M‐OH and M‐OOH) shared by multiple reaction mechanisms are examined on the surface [81]. Surprisingly, the dangling oxygen bound to the outer tetrahedral Zn in NiCoZnFe_0.8Cr_0.2‐OOH was sufficiently negative to take a proton from an oxygen atom located more than 3 Å away (Movie S1). This proton abstraction also occurred for *OH (Movie S2), indicating that the first *OH adsorption proceeds via a highly spontaneous dissociative mechanism (Figure S52). Following dissociative *OH adsorption, oxygen evolution to the gaseous form is favored over deprotonation since O₂ evolution is a purely chemical step with a low barrier. The subsequent *OH adsorption fills the V_O ^•• followed by deprotonation and another *OH adsorption step that forms the O‐O bond. Finally, deprotonation from the *OOH moiety restores the original structure, completing the catalytic cycle. In contrast, NiCoZnFe_0.5Cr_0.5‐OOH was modeled using the typical oxygen‐vacancy site mechanism (OVSM), as no evidence of ³²O₂ was observed in the DEMS results [82].

Based on these steps, energy diagrams for the two reaction pathways were constructed using DFT calculations (Figure 6b,c). Consistent with previous studies, NiCoZnFe_0.5Cr_0.5‐OOH exhibited a potential‐determining step (PDS) at the O‐O bond formation with ΔG_PDS of 0.80 eV. Notably, the PDS for NiCoZnFe_0.8Cr_0.2‐OOH was found to be the deprotonation from *OOH, which occurs after O‐O bond formation with smaller ΔG_PDS of 0.60 eV. The smaller ΔG_PDS is consistent with experimental results exhibiting the faster OER kinetics with NiCoZnFe_0.8Cr_0.2‐OOH.

To elucidate the underlying reason for this difference, a part of the DFT‐relaxed structure with *OH adsorption is shown, along with its bond lengths (Figure 6d). The O‐H distance of 1.88 Å clearly corresponds to a hydrogen bond, which is primarily responsible for lowering the activation energy of the O‐O bond formation step. This reduction in barrier height consequently shifts the potential‐determining step (PDS) from O‐O bond formation to the subsequent deprotonation step. The formation of hydrogen bond is further analyzed using Mulliken charge analysis (Figure 6e) [83]. The lattice oxygen participating in the hydrogen bond exhibits a charge of ‐0.49 due to the influence of the hydrogen, whereas other lattice oxygen atoms have Mulliken charges ranging from ‐0.77 to ‐1.09, demonstrating the distinct behavior of this oxygen site. Furthermore, since oxygen atoms bound to tetrahedral Zn possess the most negative charges, each adjacent corner‐sharing octahedron (Ni and Co in this case) is forced to form two strong M‐O bonds with short bond lengths. This effectively weakens the bond strength for the rest of the M‐O bonds, leading to electron localization on this oxygen atom [84]. The charge density difference plot showing the effective difference in charge due to the adsorption of *OH on lattice oxygen (Figure S53) supports the fact that an oxygen with such electron localization is available to form an internal hydrogen bond. This localization facilitates hydrogen‐bond formation, in contrast to that in the NiCoZnFe_0.5Cr_0.5‐LDH (Figure S54). Therefore, the hydrogen bond between the adsorbate's H and lattice oxygen stabilizes the adsorbate and lowers the energy barrier for the O‐O bond formation step. Additionally, the projected density of states (PDOS) analysis reveals that the d‐band center of NiCoZnFe_0.8Cr_0.2‐OOH lies closer to the Fermi level than that of NiCoZnFe_0.5Cr_0.5‐OOH, supporting the enhanced interaction between OER adsorbates and NiCoZnFe_0.8Cr_0.2‐OOH (Figures S55 and S56).

To summarize the mechanism investigation, the DEMS investigation showed that both NiCoZnFe_0.8Cr_0.2‐LDH and NiCoZnFe_0.5Cr_0.5‐LDH structures undergo the LOM mechanism, and OER mechanisms based on the LOM reaction scheme were studied using the γ‐phase oxyhydroxide structures via DFT calculations. Consistent with experimental results, the reaction barrier of NiCoZnFe_0.8Cr_0.2‐OOH (0.6 eV) was substantially lower than that of NiCoZnFe_0.5Cr_0.5‐OOH (0.8 eV). The differences in reaction barriers and PDS are attributed to structural distinctions between the two catalysts. The oxygen atoms bound to outer tetrahedral Zn were found to carry a significantly negative charge compared to other oxygen atoms, resulting in strong M‐O bonds in corner‐sharing octahedra and the localization of electrons at the active oxygen sites, supported by Mulliken charge and bond length analyses. This demonstrates that compositionally‐induced structural modification of the LDH framework is a viable strategy for both lowering energy barriers and tuning the PDS of OER.

3. Discussion

The superior OER performance of NiCoZnFe_0.8Cr_0.2‐LDH can be rationalized by a synergistic coupling of electronic, geometric, and thermodynamic effects enabled by its high‐entropy composition. In conventional NiFe‐LDH and NiFeCo‐based LDH systems, enhanced OER activity has primarily been reported to originate from electronic modulation, including charge redistribution, stabilization of high‐valent Ni species, and partial promotion of lattice oxygen participation [19, 20, 21]. The incorporation of Cr further extends this electronic tuning by modifying the redox behavior of neighboring transition‐metal centers; however, these effects remain largely confined to electronic structure engineering within a limited compositional space [37, 38]. In contrast, the present work reveals that Zn coordination modulation, governed by the Fe/Cr ratio, induces a local structural reconfiguration that enables specific interactions with OER intermediates. This unique structural feature directly stabilizes key adsorbates and lowers the energy barrier for O–O bond formation, thereby providing a mechanistic pathway beyond conventional electronic effects. This promoting role of Zn in enhancing OER activity is notably evident in the high‐entropy composition of NiCoZnFeCr‐LDH matrix. Furthermore, the increased structural perturbation associated with the five‐component composition thermodynamically stabilizes the active phase and suppresses phase segregation under prolonged OER operation, directly accounting for the long‐term durability at high current density. Taken together, the high‐entropy design strategy enables a synergistic convergence of electronic‐structure optimization and geometric perturbation, demonstrating multi‐component engineering as a powerful approach to advancing OER electrocatalyst performance.

4. Conclusion

In conclusion, this study demonstrates that precise compositional tuning of trivalent metals in HE‐LDH is an effective strategy to induce structural evolution, facile redox behavior, and enhanced OER activity under alkaline conditions. Systematic variation of the Fe³⁺/Cr³⁺ ratio reveals that the trivalent metal composition strongly influences metal oxidation dynamics and local coordination environments, with a particularly important role in the structural modification of tetrahedrally coordinated Zn formation. Operando spectroscopic analyses further reveal that OER activation is accompanied by the reconstruction of Ni species into γ‐NiOOH, while isotope‐labeling experiments indicate that the associated structural evolution correlates with variations in lattice oxygen participation during OER. DFT calculations further confirm that Zn‐induced coordination changes lower the energetic barriers of key steps in the LOM. Overall, this work establishes an explicit structure–composition–mechanism relationship in HE‐LDHs and highlights compositional engineering as a powerful design strategy for developing stable and efficient OER electrocatalysts suitable for industrially relevant operation.

5. Experimental Section

5.1. Materials

Nickel foam (TMAX‐Foam3) was purchased from Xiamen Tmax Battery Equipment Limited with a thickness of 1.6 mm. Hydrochloric acid (ACS reagent, 37%), nickel(II) nitrate hexahydrate (Ni(NO₃)₂·6H₂O, 99.999% trace metal basis), cobalt(II) nitrate hexahydrate (Co(NO₃)₂·6H₂O, 99.999% trace metal basis), zinc(II) nitrate hexahydrate (Zn(NO₃)₂·6H₂O, purum p.a., crystalized, ≥99.0%), iron(III) nitrate nonahydrate (Fe(NO₃)₃·9H₂O, ≥99.999% trace metal basis), chromium(III) nitrate nonahydrate (Cr(NO₃)₃·9H₂O, ≥99.99% trace metal basis), ammonium fluoride (NH₄F, ACS reagent, ≥98.0%), urea (CH₄N₂O, ACS reagent, 99.0‐100.5%), potassium hydroxide (KOH, semiconductor grade, pellets, 99.99% trace metals basis (Purity excludes sodium content.)), tetramethylammonium hydroxide solution (1.0 м ± 0.02 м in H₂O, solution, ACS reagent grade), multielement standard solution 6 for ICP (TraceCERT, 23 elements, 100 mg L⁻¹ each in nitric acid and hydrofluoric acid) were purchased from Sigma‐Aldrich. Carbon paper (AvCarb MGL 190) was purchased from Fuel Cell Store. Deionized water (18.2 MΩ·cm) was used for all experiments.

5.2. Material Synthesis

Before synthesis, nickel foam substrates (1 × 3.5 cm²) were thoroughly cleaned using ultrasonication in 2 м HCl and acetone for 15 min each. Subsequently, the substrates were rinsed multiple times with ethanol and deionized water. High‐entropy LDH samples were prepared using Ni(NO₃)₂·6H₂O, Co(NO₃)₂·6H₂O, Zn(NO₃)₂·6H₂O, Fe(NO₃)₃·9H₂O, and Cr(NO₃)₃·9H₂O as metal precursors. To achieve the desired molar ratio of M^{2 +}: M^{3 +} = 3:1, divalent metal precursors were used at 0.25 mmol each, while trivalent metal precursors were employed at 0.125 mmol each. These precursors were dissolved in 5 mL of deionized water. Concurrently, 4.675 mmol of NH₄F and 7.5 mmol of urea were dissolved in a separate 5 mL portion of deionized water. After each solution was fully vortexed, the solutions were combined and stirred continuously for 30 min. The pre‐cleaned nickel foam substrate and the precursor solution were then transferred into a 25 mL Teflon‐lined hydrothermal autoclave and heated at 160°C for 12 h. Finally, the resulting HE‐LDH was thoroughly washed with deionized water and ethanol, then dried in a vacuum oven. Control catalysts, NiFe‐LDH and NiCoZnFe‐LDH, were prepared using identical procedures, with a total cation concentration and a molar ratio (M^{2 +}: M^{3 +} = 3:1) of metal precursors (Figure S1).

5.3. Material Characterization

X‐ray diffraction (XRD) patterns were recorded with a Bruker D2 Phaser diffractometer equipped with Cu Kα radiation (λ = 1.5406 Å) over a 2θ range of 20° to 70° with a step size of 0.01°. The metal ions in the synthesized catalysts were quantified using an ICP‐OES, PerkinElmer Avio 220MAX. Calibration was conducted with commercial ICP standard solutions diluted to 0.1, 0.2, 1, and 2 ppm. High‐resolution transmission electron microscopy (HR‐TEM), high‐angle annular dark‐field scanning transmission electron microscopy (HAADF‐STEM), and STEM‐EDS mapping were performed using a JEM‐F200 (TFEG) instrument operating at an accelerating voltage of 200 kV. Atomic‐resolution imaging was conducted using a JEM‐ARM200f (Cs‐corrected STEM) microscope at an accelerating voltage of 200 kV. Synchrotron X‐ray absorption spectroscopy (XAS) measurements were carried out at the 7D and 1D beamlines of Pohang Accelerator Laboratory (PAL, Republic of Korea). In situ experiments were primarily conducted at the 1D beamline in fluorescence mode using a custom‐designed polytetrafluoroethylene (PTFE) electrochemical cell. Fourier‐transform infrared (FT‐IR) spectra were recorded using a Spectrum Two (PerkinElmer) spectrometer over the range of 300–4000 cm^{− 1}. X‐ray photoelectron spectroscopy (XPS) measurements were performed using an AXIS SUPRA+ (Kratos Analytical) system equipped with a monochromated Al Kα X‐ray source (1487 eV) and small‐spot focusing capability.

5.4. Electrochemical Measurements

Electrochemical measurements were conducted in a one‐pot three‐electrode configuration using a potentiostat (VSP‐3e, BioLogic). The prepared catalysts (NiFe‐LDH, NiCoZnFe‐LDH, NiCoZnFe_0.8Cr_0.2‐LDH, NiCoZnFe_0.5Cr_0.5‐LDH, and NiCoZnFe_0.2Cr_0.8‐LDH) on the nickel foam were used as the working electrodes. A platinum mesh and an Hg/HgO (1.0 м KOH, 0.098 V vs. SHE) electrode served as the counter and reference electrodes, respectively. Prior to all electrochemical tests, the electrolyte (1.0 м KOH) was thoroughly purged with high‐purity N₂ gas for at least 30 min to remove dissolved oxygen. Continuous N₂ purging was maintained throughout all measurements to ensure an oxygen‐free environment. All measured potentials were converted to the reversible hydrogen electrode (RHE) scale using the following equation:

$$E\left(\mathrm{RHE}\right)=E\left(\mathrm{Hg}/\mathrm{HgO}\right)+0.05916\times \mathrm{pH}+0.098\mathrm{V}$$

Before recording the polarization curves, all catalysts were electrochemically activated by repeated cyclic voltammetry (CV) scans in 1.0 м KOH until stable profiles were obtained. Linear sweep voltammetry (LSV) measurements were then performed from high to low potential in the range of 0.093–1.093 V versus SHE (equivalently, 1.0 – 2.0 V vs. RHE) at a sweep rate of 1 mV·s⁻¹. The obtained polarization data were corrected for ohmic resistance losses, with a 90% iR‐compensation automatically applied.

Tafel slopes were derived from the LSV curves according to the Tafel equation:

$$\eta =a+b\log \left|j\right|$$

where 𝜂 is the overpotential, 𝑗 is the current density, and 𝑏 represents the Tafel slope. The linear region of the 𝜂‐log𝑗 plot was used to extract the slope value for each catalyst.

To evaluate the electrochemical surface area (ECSA), CV measurements were recorded within a non‐Faradaic potential region at scan rates of 20, 40, 60, 80, 100, and 120 mV·s⁻¹. The double‐layer capacitance (C_dl) was obtained by plotting the difference in anodic and cathodic current densities (j_a‐j_c ​) as a function of the scan rate, and the slope of the linear fit was divided by two to obtain C_dl.

Then, ECSA was calculated according to:

$$\mathrm{ECSA}=\frac{{\mathrm{C}}_{\mathrm{dl}}}{{\mathrm{C}}_{\mathrm{s}}}$$

where Cs is the specific capacitance and was assumed to be 0.04 mF cm⁻² for all LDH samples.

The ECSA‐normalized current density was then calculated using:

$$j_{\mathrm{ECSA}}=\frac{j_{\mathrm{geo}}}{\mathrm{ECSA}}$$

where j _geo is the geometric current density and j _ECSA is the current density normalized by the electrochemically active surface area.

Electrochemical impedance spectroscopy (EIS) measurements were performed at 0.550 V versus Hg/HgO over a frequency range from 200 kHz to 1 mHz with an AC amplitude of 10 mV.

Long‐term stability tests were carried out at a constant current density of 1.0 A·cm⁻² for 1200 h in a continuous‐flow system, with the electrolyte level maintained by continuously circulating 1.0 м KOH to prevent depletion and ensure consistent reaction conditions.

5.5. In Situ Raman Spectroscopy Measurement

Raman Spectra were obtained using an InVia Qontor confocal Raman microscope (RENISHAW) with a 532 nm laser. Exposure time was 20 s with 3 accumulation conditions. In situ Raman spectroscopy was performed using a custom‐made electrochemical cell through a 63x water‐immersion objective lens. Under the three‐electrode system, NiFe‐LDH, NiCoZnFe_0.8Cr_0.2‐LDH, and NiCoZnFe_0.5Cr_0.5‐LDH were used as a working electrode, and the Pt wire and Hg/HgO (1 м KOH) were utilized as counter and reference electrodes, respectively, in 1 м KOH electrolyte. To obtain Raman spectra under steady‐state conditions at various applied potentials, multi‐step chronoamperometry (CA) was conducted at each potential during both the anodic sweep (0 to 1.45 V vs. RHE) and the cathodic sweep (1.45 to 0 V vs. RHE).

5.6. DEMS Oxygen‐18 Isotope Experiment

The DEMS experiments were conducted using a thin‐layer flow cell coupled to a Hiden HPR40 mass spectrometer and operated under potentiostatic control. The catalyst film was preconditioned in 1.0 м KOH solution, and a 1.0 м KOH/isotopically labeled H₂ ¹⁸O was used as the electrolyte. The electrolyte flow rate was kept at 85 µL min^{− 1} and circulated through a peristaltic pump. The catalyst film was prepared on carbon paper and integrated into the DEMS system through a minimal amount of conductive binder, which was confirmed to have no electrochemical influence. For optimal gas collection efficiency, the working electrode was positioned approximately 0.2 mm above a gas‐permeable Teflon membrane. During cyclic voltammetry sweeps from 1.0 to 1.65 V versus RHE (without iR compensation) at a scan rate of 2 mV s^{− 1}, ion currents corresponding to m/z = 32 (^{1 6}O₂), m/z = 34 (^{1 6}O^{1 8}O), and m/z = 36 (^{1 8}O₂) were continuously recorded. Prior to data acquisition, the mass spectrometer filament was activated for 30 min to stabilize the background signal.

5.7. Density Functional Theory Calculation

To represent three respective compositions, NiCoZnFe_0.8Cr_0.2(OH)₈, NiCoZnFe_0.5Cr_0.5(OH)_8, and NiCoZnFe_0.2Cr_0.8(OH)₈, supercells having the same atomic composition were generated. Each supercell contained 100, 80, and 100 atoms, respectively. For a supercell of NiCoZnFe_0.8Cr_0.2(OH)₈ with tetrahedral Zn, one carbonate ion was inserted between two layers. In order to represent the randomness of high‐entropy LDH in a finite cell, special quasirandom structure (SQS) calculations were first conducted for each cell. The weights of the first coordination shell were intentionally set to zero to ignore the metal‐oxygen bond. Afterward, the generalized gradient approximation (GGA) scheme was adopted to describe the exchange‐correlation interaction in DFT calculations, in conjunction with the Perdew‐Burke‐Ernzerhof (PBE) functional [85]. A plane‐wave basis set with 450 eV energy cutoff and a projector‐augmented wave (PAW) method were used to describe the behavior of core electrons [86]. Structural optimization was conducted with a self‐consistent field (SCF) convergence threshold of 1.0 × 10⁻⁶ eV and a force convergence threshold of 5.0 × 10⁻² eV Å⁻¹. A 2 × 2 × 1 Monkhorst‐Pack grid was used to sample the Brillouin zone. A vacuum of 12 Å was introduced along the c‐axis in order to avoid interactions along the c‐axis, resulting in cell sizes of 8.475  ×  8.475  ×  26.3 ų for NiCoZnFe_0.8Cr_0.2(OH)₈ with tetrahedral Zn. To account for the effect of magnetism in 3d transition metals, Hubbard U parameters were introduced, consistent with the reported values from Ceder's group [87]. Changes in Gibbs free energies with applied potential were calculated using the following equation:

$$\mathrm{\Delta }G=\mathrm{\Delta }E+\mathrm{\Delta }ZPE-T\mathrm{\Delta }S-neU$$

where ΔG denotes the Gibbs free energy of adsorption, E is the energy of the given species, ΔZPE is the zero‐point energy difference between species, and neU is a term for representing the applied potential in the computational hydrogen electrode model [88].

5.8. STEM Image Simulation

For simulating a STEM image, the DrProbe software [89] was used. In detail, a simple layered double hydroxide structure was constructed based on the abovementioned procedure with different k‐points sampling (3 × 3 × 3). The projection axis for viewing the simulated image was set to be [001], whereas the projection orientation was set to be [1‐20]. From the XRD pattern, the Debye–Waller factor was calculated and used as input to the thermal vibration factor (Biso), with a value of 0.87. Multiple slices along the z‐axis were used for the simulation, whereas corresponding absorption potentials were generated during the slice generation procedure. The acceleration voltage was set to 200 kV for the STEM image, and the probe was adjusted to account for Cs spherical aberration.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File 1: smll74576‐sup‐0001‐SuppMat.docx.

Supporting File 2: smll74576‐sup‐0002‐MovieS1.mp4.

Supporting File 3: smll74576‐sup‐0003‐MovieS2.mp4.

Data Availability Statement

Source code and necessary information for running SQS calculations of high‐entropy LDHs, followed by various simulations (e.g., DFT, STEM image simulation) used in the literature, are accessible at the following link: https://github.com/jihoonkim2000/HE_LDH