Lanthanum‐Induced Gradient Fields in Asymmetric Heterointerface Catalysts for Enhanced Oxygen Electrocatalysis

Abstract

Metal‐nitrogen‐carbon (M‐N‐C) catalysts display considerable potential as cost‐effective alternatives to noble metals in oxygen electrocatalysis. However, uncontrolled atomic migration and random structural rearrangement during pyrolysis often lead to disordered coordination environments and sparse active sites, fundamentally limiting their intrinsic catalytic activities and long‐term durability. Herein, a novel strategy is reported for use in directionally regulating atomic migration pathways via the incorporation of a foreign metal (La). By exploiting the differences in the atomic migration priorities via the Kirkendall effect, directional control of atomic diffusion is achieved to fabricate a well‐defined asymmetric multiphase heterointerface catalyst (LaN/LaFe‐NC). The presence of high‐density, structurally well‐defined active sites – along with a built‐in directional electric field across the heterointerface – substantially enhances the efficiency of interfacial charge transport, thus improving the intrinsic activity and stability in oxygen electrocatalysis. When incorporated into a rechargeable Zn‐air battery, LaN/LaFe‐NC delivers a high power density of 211 mW cm⁻², with an exceptional cycling stability of >240 h. This study establishes a generalizable atomic‐level design strategy for use in engineering robust heterointerface catalysts and offers valuable insights for application in advancing next‐generation renewable energy conversion and storage systems.

Uncontrolled atomic migration during the pyrolysis of metal‐N‐C catalysts often leads to disordered coordination environments, active site loss, and poor oxygen electrocatalytic activities. Herein, a La‐mediated atomic migration strategy enables the fabrication of a well‐defined asymmetric heterostructure (LaN/LaFe‐NC) featuring a built‐in electric field. This architecture stabilizes the Fe active sites, enhances interfacial charge transport, and delivers an outstanding oxygen electrocatalysis performance.

1. Introduction

Metal‐nitrogen‐carbon (M‐N‐C) catalysts, owing to their atomically dispersed active sites and exceptional metal utilization efficiencies, are widely regarded as promising alternatives to noble metal‐based electrocatalysts for use in the oxygen reduction reaction (ORR) and oxygen evolution reaction (OER).^{[ 1 − 4 ]} Currently, synthesizing M‐N‐C catalysts typically involves treatment at high temperatures ranging from 900 to 1100 °C,^{[ 5 ]} but strong interactions between their metal species and carbon matrices due to the uncontrollability of pyrolysis can induce severe carbon migration and structural rearrangement. This results in the outward diffusion of the metal atoms through the graphitic layers, ultimately leading to their irreversible volatilization from the surface.^{[ 6 − 8 ]} This diffusion‐driven loss leads to low densities of active metal sites (typically <3 wt.%) and highly disordered metal coordination environments.^{[ 9 − 11 ]} Additionally, the directly exposed graphitic layers are prone to corrosion or delamination under ORR conditions, further compromising long‐term durability.^{[ 12 ]} Certain metals, such as Ni, exhibit distinct migration behaviors during pyrolysis, generally migrating toward the catalyst surface as atomic clusters and partially assembling into crystalline domains.^{[ 13} , ^{14 ]} Although this behavior may aid in retaining metal content, the resulting random surface aggregation lacks the high activity of a well‐defined M‐N‐C coordination structure, and it is prone to metal leaching under electrocatalytic conditions.^{[ 15 ]}

Although atomic migration is conventionally regarded as a primary cause of structural degradation and active site loss, it also presents an underexplored opportunity for structural engineering – using atomic migration as a driving force to fabricate multiphase heterointerface architectures. One promising strategy involves introducing a foreign element to precisely regulate the migration behavior by leveraging the disparities in the diffusion kinetics and migration priorities between the foreign and base metals during pyrolysis.^{[ 16} , ¹⁷ , ^{18 ]} Under the influence of the Kirkendall effect, the foreign metal can preferentially migrate toward the surface, forming a stabilizing protective layer, whereas the base metal remains embedded within the carbon matrix to generate high‐density M‐N‐C active sites. This strategy based on directional atomic migration not only mitigates metal loss but also enables the fabrication of heterostructures with interfacial electric fields and potential gradients – thus enhancing the tunability of their electronic structures, accelerating their charge transfer kinetics, and ultimately improving their electrocatalytic activities and durability. Rare‐earth elements hold great potential as foreign metals. On one hand, their unique lanthanide contraction effect and relatively large ionic radii can induce lattice distortion and structural reconstruction of the base metal upon incorporation, leading to the formation of local unsaturated sites and defect structures. These features create favorable conditions for subsequent metal migration and the generation of active sites. On the other hand, the partially filled 4f orbitals of rare‐earth elements can enable gradient orbital coupling within heterostructures, modulate interfacial electronic interactions, and further enhance electrocatalytic activity.^{[ 19} , ²⁰ , ²¹ , ^{22 ]}

Herein, we report a rationally designed asymmetric heterogeneous catalyst (LaN/LaFe‐NC), wherein the incorporation of a foreign metal (La) mediates a unique atomic migration pathway and mechanism of structural rearrangement during pyrolysis. Leveraging the Kirkendall effect, La and Fe exhibit distinctly different migration priorities: La preferentially migrates toward the catalyst surface, forming a uniform, atomically thin LaN shell. This external LaN layer not only enhances the structural stability during electrocatalysis but also induces significant interfacial charge redistribution by forming an asymmetric LaN/graphene/M‐N‐C heterointerface. The intermediate graphene layer acts as an electron bridge, facilitating the formation of a built‐in electric field with a directional gradient across the interface, thus accelerating the overall ORR/OER kinetics. Meanwhile, the differences in the migration priorities effectively suppress Fe volatilization, enabling high Fe retention (16.5 wt.%) within the LaFe‐NC core and promoting the formation of well‐defined, high‐density active sites. As a result, LaN/LaFe‐NC exhibits an exceptional bifunctional ORR/OER performance, realizing a respective high ORR half‐wave potential (E _1/2) of 0.917 V and low OER overpotential of 260 mV at 10 mA cm⁻². This outstanding bifunctionality is further validated in a rechargeable Zn‐air battery (ZAB), delivering a respective peak power density and cycling stability of 211 mW cm⁻² and >240 h. Our study presents a novel strategy for use in regulating directional atomic migration, enabling atomic‐level structural engineering and the fabrication of multiphase heterointerface catalysts. These findings offer critical insights into the rational design of next‐generation high‐performance electrocatalysts and lay a solid foundation for their application in advanced renewable energy conversion and storage technologies.

2. Results and Discussion

2.1. Design and Morphological Characterization of the LaN/LaFe‐NC Catalyst

To synthesize the Fe‐N‐C catalyst, we employed Fe Prussian blue analogues (FePBA) as the precursor and La as the foreign metal due to its distinct structural and thermodynamic advantages.^{[ 23} , ^{24 ]} La can readily migrate through defect‐rich carburized layers during pyrolysis via polarization effects, and it can also promote carbon migration – leading to local carbon enrichment and facilitating the formation of cementite (Fe₃C).^{[ 25 ]} This behavior modulates the migration dynamics of the Fe species. As Fe₃C is a critical crystalline phase in Fe‐N‐C catalysts,^{[ 26} , ^{27 ]} we employed a Fe₃C‐based model in density functional theory (DFT) calculations, comparing the relaxation behaviors of the La atoms at the surface and in the bulk phase of the catalyst (Figure S1, Supporting Information). The total energy of La is 0.92 eV lower at the surface compared to that within the bulk, indicating a strong thermodynamic driving force for surface enrichment over lattice incorporation. Accordingly, as the Fe‐N‐C/graphitic structure gradually evolves during pyrolysis, La preferentially accumulates at the Fe‐N‐C/graphitic interface and migrates toward the catalyst surface, facilitating the formation of a multiphase interfacial architecture. Additionally, we compared the ORR and OER activities of dual‐metal Fe‐N‐C catalysts incorporating various foreign metal elements and single‐atom catalysts, as shown in Figure S2 (Supporting Information). The results demonstrate that La/Fe‐NC exhibits the highest activity for both ORR and OER, clearly indicating that La is more effective than the other metal elements in enhancing the electrocatalytic performance of Fe‐N‐C systems. Based on this design strategy, we first synthesized FePBA and defective FePBA (d‐FePBA) via hydrothermal acid treatment. Scanning electron microscopy (SEM) reveals that although d‐FePBA retains the characteristic cubic morphology of FePBA, distinct surface etching features are observed after acid treatment, indicating the introduction of structural defects and vacancies – favorable sites for subsequent La incorporation (Figure S3a,b, Supporting Information). We then employed wet chemical adsorption to incorporate La ions into the d‐FePBA framework (La@d‐FePBA), followed by pyrolysis to obtain the final catalyst. Different La loadings and pyrolysis temperatures were systematically investigated (Figure S4, Supporting Information), and the optimized synthesis conditions yielded LaN/LaFe‐NC as the final catalyst (see Experimental Section for details). Under identical conditions, FePBA is directly pyrolyzed to yield Fe‐NC as a control sample (Figure S3c, Supporting Information). Characterizing Fe‐NC reveals significant structural self‐compression during pyrolysis, with its particle size reduced to ≈44% of that of the original FePBA. High‐resolution transmission electron microscopy (HR‐TEM) indicates that Fe‐NC is encapsulated by thick graphene layers, forming a dense shell‐like structure that can severely impede interfacial electron transport and restrict access to the active sites (Figure S5a,b, Supporting Information). The internal metallic Fe core exhibits a complex crystal structure characterized by the coexistence of multiple crystalline phases – including metal and metal oxide, carbide, and nitride – indicating substantial structural heterogeneity (Figure S5c, Supporting Information). Inductively coupled plasma optical emission spectroscopy (ICP‐OES) reveals that the Fe content within Fe‐NC is only 0.67 wt.%, suggesting extensive Fe volatilization during pyrolysis. Such volatilization results in the substantial loss of active sites, thus compromising its electrocatalytic performance (Figure S5d and Table S1, Supporting Information).

SEM of LaN/LaFe‐NC reveals a rough surface morphology (Figure S6, Supporting Information), suggesting substantial structural reconstruction during pyrolysis. The HR‐TEM image shown in Figure 1a further depicts a well‐defined asymmetric heterointerface comprising three distinct regions. The intermediate domain comprises ≈10 layers of curved graphene with an interlayer spacing of 3.38 Å, corresponding to the (002) plane of graphene.^{[ 28 ]} Within each graphene layer, dislocated lattice fringes are observed, indicating structural defects. High‐angle annular dark‐field scanning transmission electron microscopy (HAADF‐STEM) was employed to resolve the local crystalline features of each region (Figure 1b). In the inner LaFe‐NC core (Region 1, Figure 1c), a distorted orthorhombic Fe₃C structure is identified, with atomic spacings of 2.06 and 2.37 Å, corresponding to the short and long Fe─C bond configurations, respectively. The core lattice remains uniform, with no evidence of isolated single‐atom Fe sites or aggregated metallic particles, further indicating a highly ordered coordination environment. In the outermost surface domain (Region 2, Figure 1d), lattice fringes with an interplanar spacing of 3.06 Å are assigned to the (111) plane of cubic LaN, confirming the surface migration of La and subsequent formation of a stable, atomically thin LaN shell. Inverse fast Fourier transform (IFFT) analysis of the interfacial domain (Region 3, Figure 1e) reveals pronounced lattice distortion induced by La accumulation at the LaFe‐NC/graphene boundary, resulting in atomic rearrangement and lattice expansion. Notably, the HAADF‐STEM images (Figure 1f; Figure S7, Supporting Information) reveal a distinct atomic migration pathway originating from the LaFe‐NC core, through the graphene interlayer, and terminating at the LaN shell. The atomic spacing along this migration pathway reaches 3.11 Å – larger than typical Fe─Fe or Fe─C bond lengths^{[ 29} , ^{30 ]} – strongly suggesting that migration is dominated by La rather than Fe. Geometric phase analysis was conducted to further quantify this effect, and the spatial tensile strain component was mapped along the ε _yy direction (Figure 1f). A well‐defined strain band aligned with the atomic migration pathway is observed, extending from the LaFe‐NC core into the graphene region, and the maximum compressive strain reaches 20%, indicating pronounced local lattice distortion and atomic rearrangement. Collectively, these findings provide compelling structural evidence of La‐mediated atomic migration and interfacial reconstruction during pyrolysis, which are central to the formation of the asymmetric heterostructure within LaN/LaFe‐NC.

Figure 1.

Morphological and interfacial characterizations of the LaN/LaFe‐NC catalyst. a) HR‐TEM image of LaN/LaFe‐NC. b) Aberration‐corrected HAADF‐STEM image with the three marked regions (1–3) used in atomic‐scale analysis. c–e) Magnified HAADF‐STEM images of regions 1–3, respectively, along with corresponding IFFT analyses, lattice spacing profiles, and 3D atomic Gaussian fitting maps. f) Aberration‐corrected HAADF‐STEM image and geometric phase analysis, revealing the atomic migration pathway and local strain distribution. g) STEM image of the heterointerface used in EDS line scanning. h) EDS line scan profiles of the elemental distributions across the heterointerface. i) Atomically resolved contour maps of the EELS line scans of the C K‐edge, N K‐edge, Fe L‐edge, and La M‐edge, highlighting the element‐specific spatial distributions across the LaN/graphene/LaFe‐NC interface.

Electron energy loss spectroscopy (EELS) reveals the distinct spatial distributions of La and Fe within LaN/LaFe‐NC (Figures S8 and S9, Supporting Information). As shown in Figure 1g–i, the C K‐edge spectra indicate that carbon is predominantly located within the graphene interlayers. Notably, the 1s → σ* resonance at 292.0 eV increases in intensity at the graphene domain compared to that at the LaFe‐NC core, suggesting a high density of structural defects within each graphene interlayer.^{[ 31} , ^{32 ]} The La M‐edge spectra indicate that the La signal (M₄ and M₅) gradually increases when moving from the LaFe‐NC core, reaching an initial peak at the LaFe‐NC/graphene interface, indicating La accumulation at the LaFe‐NC/graphene boundary. The second peak is observed at the LaN shell on the catalyst surface, confirming surface enrichment via directional La migration. The profile of the N K‐edge spectrum is similar, with a strong surface‐localized signal corresponding to the formation of the thermodynamically stable LaN shell. In sharp contrast, the Fe L₂ and L₃ edges are confined entirely within the LaFe‐NC core, and thus, the Fe atoms do not exhibit significant outward migration under La mediation but remain within the core as FeN_xC_y active structures. This contrasting behavior highlights the difference in the migration priority between La and Fe during pyrolysis and underscores the effectiveness of the strategy based on La‐mediated atomic migration‐driven structural rearrangement. We subsequently analyzed the Fe L_2,3‐edges and La M_4,5‐edges, where the L₃/L₂ and M₅/M₄ intensity ratios serve as sensitive indicators of the oxidation states of the respective metals.^{[ 33 ]} The edge intensities were normalized, and spatial variations in the L₃/L₂ and M₅/M₄ ratios were systematically examined across different regions. As shown in Figure S10 (Supporting Information), the La M₅/M₄ ratio shows minimal variation, indicating that the oxidation state of La remains largely stable. In contrast, the Fe L₃/L₂ ratio remains constant in the core region of LaFe‐NC but decreases significantly at the LaFe‐NC/graphene interface, suggesting a local reduction in the oxidation state of Fe. This change likely results from the migration of Fe atoms during pyrolysis, where Fe−N₄ active sites are formed at the graphene edge. These findings further demonstrate that preferential La migration during pyrolysis limits Fe diffusion, confining Fe largely within the inner core. Consequently, a stable interfacial synergy is established between Fe₃C and the edge‐located Fe−N₄ sites on the graphene. This synergistic structure promotes electron transport and optimizes the reaction pathway, ultimately enhancing the overall electrocatalytic performance. Energy‐dispersive X‐ray spectroscopy (EDS) mapping further confirms the distinct elemental distributions and highlights the preservation of the Fe active species (Figure S11, Supporting Information). Elemental quantification via selected‐area EDS reveals that the Fe content within the LaFe‐NC core is as high as 16.5 wt.% (Figure S12, Supporting Information). These results conclusively indicate that the proposed atomic migration mechanism not only suppresses Fe volatilization and aggregation but also enhances the retention of active metal sites, thus contributing to the superior ORR/OER performance of LaN/LaFe‐NC.

2.2. Phase Transformation During Pyrolysis

Introducing La enabled the successful fabrication of the asymmetric heterointerface catalyst by leveraging the differences in the atomic migration priorities. However, a critical question remains: how does this unique heterointerface structure evolve during pyrolysis? To address this question, we first performed thermogravimetric analysis (TGA) to monitor the key reaction stages during the thermal decomposition of the precursor. Notably, FePBA (the precursor to Fe‐NC) exhibits significant mass losses throughout pyrolysis, primarily due to Fe volatilization (Figure S13, Supporting Information). The derivative thermogravimetry (DTG) thermogram shown in Figure 2a delineates the decomposition pathway of La@d‐FePBA (the precursor to LaN/LaFe‐NC), and the initial mass loss at <200 °C is mainly due to water evaporation. At 267–1000 °C, multiple successive mass loss steps are observed, reflecting complex structural evolution and phase transitions. We conducted X‐ray diffraction (XRD) and Fourier transform infrared (FT‐IR) spectroscopy using La@d‐FePBA samples pyrolyzed at temperatures corresponding to key DTG transition points to elucidate the origins of these transitions. FePBA and d‐FePBA exhibit the typical PBA framework structure at room temperature (Figure S14a, Supporting Information). However, as shown in Figure 2b and Figure S14b (Supporting Information), La@d‐FePBA undergoes three major phase transitions upon heating: 1) At 267 °C, a broad diffraction peak is observed at 2θ ≈ 22°, which is typically associated with amorphous carbon, originating from the thermal decomposition of cyanide (CN⁻) ligands and initial rearrangement of the carbon framework.^{[ 34} , ³⁵ , ^{36 ]} 2) At >310 °C, diffraction peaks corresponding to hexagonal Fe₂O₃ emerge, coinciding with the DTG peak at 319 °C, indicating that oxide formation dominates the structural evolution in this temperature range. 3) At ≈650 °C, the peaks representing Fe₂O₃ are no longer observed, and the structure transitions into a composite phase comprising Fe₃C, Fe₃N, and LaN, signifying substantial atomic migration and structural rearrangement within LaN/LaFe‐NC. FT‐IR spectroscopy further elucidates the evolution of the carbon structures and functional groups. The FT‐IR spectrum of FePBA displays the characteristic vibrational modes of PBA, whereas that of d‐FePBA exhibits a pronounced shift in the ν(CN) peak position, indicating the disruption of the Fe–CN framework and successful defect introduction (Figure S15a,b, Supporting Information). These structural vacancies provide favorable sites for La incorporation. As shown in Figure 2c and Figure S15c,d (Supporting Information), the FT‐IR spectrum of La@d‐FePBA at <350 °C retains the strong ν(CN) and δ(HOH) peaks associated with the PBA framework. With increasing temperature, the intensities of these peaks gradually decrease, corresponding to the decomposition of the CN⁻ ligands. At 650 °C, new peaks are observed at ∼1480 and ∼1042 cm⁻¹, which are assigned to the ν(C═C) and ν(C─O─C) vibrations of graphene, respectively.^{[ 37} , ^{38 ]} These signals further confirm the formation of the graphene interlayers and development of the multiphase heterointerface structure.

Figure 2.

Phase transformation and chemical evolution during pyrolysis. a) DTG thermograms of La@d‐FePBA and FePBA. Contour maps of the b) XRD patterns and c) FT‐IR spectra of La@d‐FePBA collected at different pyrolysis temperatures. d) Raman spectra of LaN/LaFe‐NC synthesized at different temperatures (700–900 °C) compared to that of FePBA. High‐resolution e) N 1s and f) Fe 2p XPS spectra of LaN/LaFe‐NC, Fe‐NC, and FePBA, highlighting the differences in their chemical bonding states.

Based on this analysis, we elucidated the unique mechanism of atomic migration and structural rearrangement underlying the formation of LaN/LaFe‐NC. Driven by the distinct migration priorities of La and Fe, La atoms preferentially migrate across the graphene defects toward the outermost catalyst surface during pyrolysis. The La atoms then react with the nitrogen species released during thermal decomposition to form the thin thermodynamically stable LaN shell. In parallel, Fe atoms undergo intense carbothermal reduction in the carbon‐rich environment under La‐induced local lattice distortion, forming FeN_xC_y active structures within the LaFe‐NC core. Via this coupled migration‐rearrangement mechanism, the well‐defined asymmetric LaN/graphene/LaFe‐NC heterostructure is formed, integrating compositional, structural, and interfacial advantages to support high‐performance electrocatalysis.

Raman spectroscopy was employed to evaluate the formation and degree of carbon defects within LaN/LaFe‐NC during pyrolysis. As shown in Figure 2d and Figure S16 (Supporting Information), the sample pyrolyzed at 900 °C (LaN/LaFe‐NC) exhibits the highest I _D/I _G ratio (1.06). Moreover, the defect level in LaN/LaFe‐NC is significantly higher than that in Fe‐NC, indicating a greater density of defects and microporous structures within the graphene layers. Such a high defect density is advantageous for facilitating charge and mass transport between the LaFe‐NC core and the LaN surface, thereby accelerating charge transfer and catalytic reaction kinetics, and ultimately improving ORR/OER performance of LaN/LaFe‐NC. This is further supported by nitrogen adsorption‐desorption isotherm analysis, which reveals the high specific surface area of LaN/LaFe‐NC of 581.10 m² g⁻¹, which substantially exceeds those of FePBA (108.26 m² g⁻¹) and Fe‐NC (490.39 m² g⁻¹, Figure S17, Supporting Information). These findings confirm that atomic migration induces abundant defects and porosity within the carbon matrix, thus modulating the local electronic structure and enhancing electron transport, ultimately increasing the catalytic activity in oxygen electrocatalysis.

X‐ray photoelectron spectroscopy (XPS) analysis was conducted to gain further insights into the changes in oxidation states and electronic structures. As shown in Figure 2e and Table S2 (Supporting Information), the N 1s XPS spectrum of LaN/LaFe‐NC reveals a high concentration of pyridinic nitrogen (40.80% of the total nitrogen species),^{[ 39 ]} which is widely recognized as an effective active site in the ORR.^{[ 40 ]} The high‐resolution Fe 2p spectrum of Fe‐NC reveals a distinct new peak at 707.02 eV, corresponding to metallic Fe species, indicating atomically dispersed Fe or small clusters formed at elevated temperatures (Figure 2f). In contrast, Fe predominantly occurs within LaN/LaFe‐NC in the Fe³⁺ oxidation state (84.33%),^{[ 41 ]} suggesting that La‐mediated atomic migration stabilizes Fe in the form of FeN_xC_y active structures. Additionally, a notable positive shift in the peak representing Fe³⁺ is observed, indicating a modified electronic environment around the Fe center and further confirming the intrinsic modulation of the electronic structure of LaN/LaFe‐NC. The La 3d spectrum reveals the clear splitting of the La 3d_5/2 and La 3d_3/2 orbitals, and thus, La primarily occurs in the La³⁺ oxidation state (Figure S18, Supporting Information). Combined with the metal–N signature in the N 1s spectrum, these results further confirm that La–N coordination is the dominant chemical state of La within LaN/LaFe‐NC. We also performed depth‐profiling XPS measurements to investigate the distribution of chemical compositions from the surface to the inner phase of LaN/LaFe‐NC. As shown in Figure S19 (Supporting Information), both La and Fe exhibit distinct electronic structures across different regions of the heterostructure. At the interface between the defective graphene and the LaFe‐NC core, a high concentration of low‐valence Fe species (Fe⁰) was detected, indicative of abundant Fe–N₄ moieties coordinated within the Fe‐N‐C structure—widely recognized as key active sites for ORR. With increasing depth into the LaFe‐NC core, the Fe³⁺ signal becomes dominant, consistent with the XRD results showing that Fe predominantly exists in the form of Fe₃C and Fe₃N. While these inner phases may not serve as primary active sites like Fe–N₄, they provide crucial structural and electronic support that synergizes with the interfacial Fe–N₄ centers to enhance overall ORR/OER performance. The La 3d XPS spectra further reveal significant binding‐energy shifts across the heterostructure. From the outer LaN shell to the underlying defective graphene layer, the La 3d binding energy exhibits a slight negative shift of 0.15 eV. With increasing depth into the LaFe‐NC core, the La 3d peaks gradually shift toward higher binding energies, with a total shift of 0.7 eV. These variations indicate that La experiences distinct electronic structural changes within the heterostructure. Notably, the spin–orbit splitting between La 3d_5/2 and La 3d_3/2 decreases from 16.52 eV at the surface to 16.23 eV in the LaFe‐NC core, further confirming the progressive chemical‐state transition of La across the heterostructure. In summary, La plays a pivotal role in regulating the spatial distribution of Fe species. Preferential La migration suppresses excessive Fe diffusion, thereby promoting the formation of highly active Fe–N₄ sites at the graphene/LaFe‐NC interface. Meanwhile, the La‐doped Fe₃C/Fe₃N core provides structural stability and facilitates electronic interactions, collectively underpinning the enhanced bifunctional electrocatalytic performance.

2.3. Electrocatalytic performances of LaN/LaFe‐NC in the ORR and OER

The electrocatalytic performance of LaN/LaFe‐NC was evaluated in 0.1 m KOH using a rotating disk electrode (RDE) and benchmarked against those of Fe‐NC, FePBA, and commercial Pt/C. In an O₂‐saturated solution, the cyclic voltammogram of LaN/LaFe‐NC displays distinct peaks representing oxygen reduction, which are no longer observed in an N₂‐saturated solution, confirming its strong ORR catalytic activity (Figure S20, Supporting Information). The linear sweep voltammogram of LaN/LaFe‐NC measured at 1600 rpm reveals a higher diffusion‐limited current density and an excellent E _1/2 of ≈6 mA cm⁻² and 0.917 V, respectively, indicating efficient oxygen activation and rapid electron transfer (Figure 3a). At 0.85 V, LaN/LaFe‐NC exhibits the highest kinetic current density (J _k, 53.2 mA cm⁻²) and a low Tafel slope of 51 mV dec⁻¹ (Figure 3b; Figure S21, Supporting Information), highlighting the favorable reaction kinetics. Furthermore, its mass activity (MA) and specific activity (SA) are 1.9‐ and 1.3‐fold higher than those of Fe‐NC and 3.9‐ and 4.2‐fold higher than those of Pt/C, respectively (Figure 3b), confirming that LaN/LaFe‐NC displays lower reaction barriers and enhanced electron transfer properties. Critically, after 10 000 cycles of an accelerated durability study, LaN/LaFe‐NC exhibits only a slight decrease of ≈11 mV in its E _1/2 (Figure S22, Supporting Information), underscoring its remarkable long‐term stability in the ORR. Post‐stability characterizations using HR‐TEM, XRD, Raman, and XPS further confirm the morphological and structural robustness of LaN/LaFe‐NC during prolonged ORR operation (Figures S23 and S24, Supporting Information).

Figure 3.

Electrocatalytic performances of LaN/LaFe‐NC in the ORR and OER. a) ORR polarization curves of LaN/LaFe‐NC, Fe‐NC, FePBA, and commercial Pt/C, as measured in O₂‐saturated 0.1 m KOH at 1600 rpm. b) Half‐wave potentials (E _1/2), kinetic current densities (J _k), mass activities (MAs), and specific activities (SAs) of the respective catalysts at 0.85 V. c) OER polarization curves of LaN/LaFe‐NC, Fe‐NC, FePBA, and IrO₂ in O₂‐saturated 0.1 m KOH at 1600 rpm. d) Summary of the OER‐related parameters: overpotential, Tafel slope, R _ct, C _dl, and TOF. e) Long‐term stabilities of LaN/LaFe‐NC and IrO₂, as evaluated via chronopotentiometry at 10 mA cm⁻² and methanol tolerance studies. f) H₂O₂ yields (bottom) and electron transfer numbers (top) of LaN/LaFe‐NC and Pt/C, as measured using a RRDE in O₂‐saturated 0.1 m KOH. g) Comparison of the bifunctional catalytic performances, showing the E _1/2 values and overpotentials at 10 mA cm⁻² (E_{j 10}) for the ORR and OER, respectively. The dashed lines indicate the potential gaps (ΔE) between the ORR and OER.

LaN/LaFe‐NC also exhibits a remarkable OER catalytic performance under the same conditions. As shown in Figure 3c, it delivers a low overpotential of 260 mV at 10 mA cm⁻², accompanied by a small Tafel slope of 78 mV dec⁻¹, indicating its favorable OER kinetics. A comparative radar plot of five key OER parameters shown in Figure 3d reveals that LaN/LaFe‐NC exhibits the lowest charge transfer resistance (R _ct). This underscores the role of its optimized interfacial architecture in facilitating rapid electron transport and improving the overall catalytic efficiency. Furthermore, LaN/LaFe‐NC displays the highest double‐layer capacitance (C _dl, 9.73 mF cm⁻²) and electrochemically active surface area (ECSA, 47.76 m² g⁻¹), and also achieves a high oxygen turnover frequency (TOF) of 1.336 s⁻¹ (versus the reversible hydrogen electrode (RHE)). The ECSA and TOF of LaN/LaFe‐NC are 2.4‐ and 2.3‐fold higher than those of Fe‐NC and 1.6‐ and 3.2‐fold higher than those of IrO₂, respectively (Table S3, Figures S25 and S26, Supporting Information), further confirming its superior intrinsic activity and reaction kinetics in the OER. Notably, LaN/LaFe‐NC displays an exceptional operational stability, maintaining a constant current density of 10 mA cm⁻² for 250 h, with negligible performance decay (Figure 3e). To further investigate the stability of LaN/LaFe‐NC and elucidate the surface evolution of the LaN shell during OER operation, a series of post‐stability characterizations was performed, including XRD, ICP‐OES, electrochemical impedance spectroscopy (EIS), and time‐of‐flight secondary ion mass spectrometry (ToF‐SIMS). The ToF‐SIMS analysis (Figure S27, Supporting Information) reveals the presence of weak OH⁻ and O⁻ signals on the LaN/LaFe‐NC surface, suggesting the formation of a thin La (hydro)oxide layer. Post‐stability XRD patterns (Figure S28a, Supporting Information) confirm that the crystalline structure of LaN/LaFe‐NC remains largely intact, with no detectable phase transformation or structural degradation, indicating that the in situ‐formed passivation layer is of limited thickness and does not compromise the structural integrity. Consistently, EIS analysis (Figure S28b, Supporting Information) shows only a negligible increase in impedance after long‐term OER operation, implying that the (hydro)oxide layer is sufficiently thin and porous to allow efficient electron transport without introducing significant charge‐transfer resistance. Moreover, ICP‐OES measurements (Figure S28c, Supporting Information) reveal minimal leaching of metal species, particularly La, further corroborating the structural stability of LaN/LaFe‐NC under OER conditions. Collectively, these results demonstrate that the excellent OER durability of LaN/LaFe‐NC arises from a self‐passivation mechanism, wherein the LaN shell plays a critical role in mitigating corrosion and ensuring long‐term electrochemical stability. It also exhibits an excellent methanol tolerance – sustaining a stable current during chronoamperometric evaluation for >1000 s following methanol injection – highlighting its promise for practical application in metal‐air battery systems. Moreover, the ORR and OER performances of LaN/LaFe‐NC were confirmed through multiple independent repeated experiments, yielding low relative standard deviation (RSD) values and thus verifying its high reproducibility (Figure S29, Supporting Information).

Regarding the reaction pathways, Fe‐NC generates a relatively high yield of H₂O₂ (≈10%) as a byproduct during the ORR (Figure S30, Supporting Information). In contrast, LaN/LaFe‐NC follows a highly selective four‐electron ORR pathway, as confirmed via Koutecký–Levich (K‐L) analysis (Figure S31, Supporting Information), with an H₂O₂ yield of <4%, which is even lower than that of commercial Pt/C (≈6%, Figure 3f). The suppressed H₂O₂ production suggests that LaN/LaFe‐NC enables more effective O─O bond cleavage, thus minimizing the generation of undesirable byproducts and enhancing the overall catalytic efficiency. Overall, compared to those of previously reported M‐N‐C‐based catalysts, LaN/LaFe‐NC exhibits a superior bifunctional ORR/OER electrocatalytic activity, surpassing the performances of established benchmarks (Figure 3g; Table S4, Supporting Information). These results suggest that LaN/LaFe‐NC is a highly promising bifunctional catalyst for use in next‐generation rechargeable metal‐air batteries.

2.4. Mechanistic Insights into the Electrocatalytic Activity

To elucidate the intrinsic driving forces behind electron transfer and interfacial microstructure modulation, we employed ultraviolet photoelectron spectroscopy (UPS) and ultraviolet‐visible‐near‐infrared absorption spectroscopy to probe the work functions and electronic band structures of the catalysts (Figure  4a; Figure S32, Supporting Information). Compared to that of Fe‐NC (3.76 eV), LaN/LaFe‐NC exhibits a lower work function of 3.38 eV, indicating a reduced energy barrier for electron transfer to adsorbed oxygen species, thus facilitating the enhanced ORR activity. As shown in Figure 4b, Fe‐NC displays a valence band maximum (VBM) at −7.69 eV with a bandgap energy of 2.23 eV. Conversely, LaN/LaFe‐NC exhibits an upward‐shifted VBM at −6.18 eV and a narrower bandgap of 2.12 eV, positioning the electronic states closer to the Fermi level and thus improving charge accessibility. Mott‐Schottky analysis further supports these observations: whereas both catalysts are n‐type semiconductors, the Mott‐Schottky plot of LaN/LaFe‐NC exhibits a notably smaller slope, indicating its higher charge carrier density (Figure S33, Supporting Information). These band structure modifications suggest that the asymmetric multiphase heterostructure of LaN/LaFe‐NC induces significant electron redistribution across the interface, resulting in the formation of a built‐in directional electric field.^{[ 42} , ^{43 ]} The defect‐rich graphene interlayer plays a pivotal role as an electronic bridge, facilitating vertical charge transport between the LaFe‐NC core and LaN shell. This enhanced interfacial coupling not only accelerates electron mobility but also optimizes the charge transfer kinetics, collectively increasing the catalytic activities and stabilities in the ORR and OER. To further elucidate the electronic interactions at the asymmetric heterointerface of the LaN/LaFe‐NC and their influence on electrocatalytic performance, we carried out additional density functional theory (DFT) calculations. Bader charge analysis revealed a substantial electron transfer of ≈2.27 e from the LaFe‐NC core to the LaN shell in the model structure (Figure S34, Supporting Information), providing direct evidence for the formation of an intrinsic built‐in electric field within the heterostructure. This result is consistent with our UPS measurements, which showed a reduced work function and an upward shift of the valence band, further confirming modulation of the interfacial charge distribution by the internal electric field. Such interfacial electron transfer enables efficient electronic coupling between the LaN shell and the Fe‐N‐C active sites in the LaFe‐NC core via the graphene interlayer, while simultaneously optimizing the local electronic environment of the Fe sites. Moreover, the charge redistribution plays a critical role in enhancing the structural stability of Fe by modifying its local coordination environment and strengthening Coulombic interactions with surrounding anions. Notably, compared with Fe‐NC, incorporation of La increases the average negative charge on anions from −1.15 to −1.17 e, further stabilizing the Fe‐N‐C active centers. To further verify the role of the LaN/LaFe‐NC heterostructure in regulating reaction kinetics, we constructed free energy diagrams for both OER and ORR and compared them with those of Fe‐NC (Figure S35, Supporting Information). The LaN/LaFe‐NC system exhibits more balanced four‐step free energy profiles, with lower energy barriers for the rate‐determining steps—particularly for OER, where a markedly reduced overpotential is achieved. In summary, the asymmetric heterostructure of LaN/LaFe‐NC induces interfacial electron transfer and charge redistribution, which not only optimizes the adsorption of reaction intermediates and mitigates excessive electronic interactions between Fe sites and intermediates (relative to Fe‐NC) through the built‐in electric field, but also enhances the structural stability of Fe‐N‐C active sites. These synergistic effects collectively accelerate reaction kinetics and endow LaN/LaFe‐NC with superior bifunctional electrocatalytic performance toward both OER and ORR.

Figure 4.

Mechanistic origin of the enhanced electrocatalytic performance of LaN/LaFe‐NC. a) UPS spectra of LaN/LaFe‐NC and Fe‐NC, which reveal their respective work functions (Φ). b) Energy band diagrams of LaN/LaFe‐NC and Fe‐NC, indicating electronic structure modulation. c) In situ Raman spectra and the corresponding contour map of LaN/LaFe‐NC collected under applied potentials in the range 1.0–0 V. The dashed lines indicate the regions corresponding to the adsorbed O₂ ⁻ species and D, G, and Dʹ bands. d) Evolution of the FWHMs and peak positions of the D, G, and Dʹ bands in (c) as functions of the applied potential (1.0–0.5 V). e) Bode plots of LaN/LaFe‐NC and Fe‐NC under ORR conditions (1.1–0.4 V), highlighting the transition between the non‐ORR and ORR regions (indicated by the dashed lines).

In situ Raman spectroscopy was conducted to gain deeper insight into the reaction intermediates and catalytic mechanism during the ORR (Figure 4c). The band observed at ≈1035 cm⁻¹, with a gradually increasing intensity as the applied potential decreases from 1.0 to 0 V, is assigned to the O─O stretching vibration mode of adsorbed anionic superoxide species (O₂ ⁻).^{[ 44 ]} The persistence of this feature across the entire potential window suggests that the protonation of adsorbed O₂ ⁻ is likely the rate‐determining step of the ORR. Additionally, noticeable potential‐dependent shifts are observed in the D, G, and Dʹ bands, which are characteristic signatures of various types of carbon structures. The Raman spectra were deconvoluted to further understand the active site‐intermediate interactions, and the full width at half maximum (FWHM) of each band was analyzed (Figure 4d; Figure S36, Supporting Information). The FWHM of the D band – associated with edge defects–increases with decreasing potential and the peak shifts positively, indicating enhanced interactions between the edge sites and reactive intermediates. Remarkably, the FWHMs of the G and Dʹ bands, which are attributed to the in‐plane vibrations of the sp ² basal planes and their associated defects, respectively, also increase with decreasing potential. However, these peaks shift negatively, suggesting that the basal planes are also actively involved in the 4e⁻ ORR pathway.^{[ 45 ]} Collectively, these results demonstrate that the edge‐ and basal‐plane carbon sites within LaN/LaFe‐NC contribute to ORR catalysis. The defective graphene interlayer not only provides abundant active sites but is also crucial in activating the inner basal‐plane carbon, enhancing interfacial charge transfer, and significantly accelerating the ORR kinetics.

Dynamic EIS was employed to investigate the charge transfer kinetics and mass transport behavior of LaN/LaFe‐NC during the ORR. As shown in Figure 4e and Figure S37 (Supporting Information), the Bode contour maps recorded over the potential range 1.1–0.4 V reveal a distinct evolution in the electron transport characteristics. Compared to that in the Bode plot of Fe‐NC, the Bode plot of LaN/LaFe‐NC consistently exhibits a lower phase‐angle peak across the entire potential window, indicating a reduced interfacial R _ct and an enhanced electronic conductivity.^{[ 46 ]} Furthermore, as the potential decreases, the phase angle of LaN/LaFe‐NC initially shifts toward the high‐frequency region, followed by a continuous shift toward the low‐frequency region, suggesting a progressive reduction in the charge transfer impedance and an improved mass transport capacity. These features collectively confirm the superior electron transport properties and efficient electron‐oxygen coupling behavior of LaN/LaFe‐NC during the ORR. A thiocyanate (SCN⁻) poisoning study was conducted to further verify that La incorporation effectively stabilizes a higher number of catalytically active Fe sites (Figure S38, Supporting Information). As a well‐established inhibitor of Fe coordination, SCN⁻ interacts strongly with the Fe within the FeN_xC_y structures, deactivating the active sites and suppressing the ORR activity.^{[ 47 ]} Upon adding 0.01 M SCN⁻ to the electrolyte, LaN/LaFe‐NC exhibits a more pronounced decline in ORR performance than that of Fe‐NC, highlighting its higher density of Fe‐based active sites. In addition, a NO₂ ⁻ poisoning study was performed, wherein Fe‐based active sites formed stable adducts with NO₂ ⁻ anions, leading to catalyst deactivation, while metal‐free N/C catalyst remained unaffected.^{[ 48 ]} Cyclic voltammetry (CV) measurements before and after exposure to 0.125 mol dm⁻³ NO₂ ⁻ (Figure S39, Supporting Information) revealed significant changes in charge transfer and the appearance of reduction peaks for LaN/LaFe‐NC, whereas no such effects were observed for the metal‐free N/C control. This selective poisoning behavior confirms that Fe‐based moieties are the true catalytic centers. Therefore, Fe remains the dominant active center in LaN/LaFe‐NC, and La‐mediated structural modulation enhances the stabilities and levels of utilization of the Fe sites, thus significantly increasing the overall ORR catalytic activity. To gain deeper insight into the function of the LaN shell in enhancing ORR activity and stability, we employed an EDTA‐assisted leaching method to selectively remove the LaN layer from LaN/LaFe‐NC, yielding EDTA‐LaN/LaFe‐NC.^{[ 49 ]} XRD and Raman spectra (Figure S40, Supporting Information) of EDTA‐LaN/LaFe‐NC confirm the successful removal of the LaN shell, with negligible impact on the graphene layer and the LaFe‐NC core. The characteristic Fe₃C structural features remain intact, demonstrating robust structural stability. A slight increase in carbon defect density was observed, likely arising from localized mild etching effects during the EDTA treatment. As shown in Figure S41a,b (Supporting Information), although EDTA‐LaN/LaFe‐NC retains a certain degree of ORR activity, the reaction pathway shifts from a 4e⁻ process to a 2e⁻ process, accompanied by a pronounced decrease in ECSA (Figure S41c, Supporting Information). This degradation is mainly attributed to the removal of the mesoporous LaN layer, which reduces the active surface area. It further demonstrates that the heterostructure incorporating the LaN shell effectively regulates the ORR pathway and enhances selectivity toward the desired main product. Furthermore, the disruption of the asymmetric heterostructure alters the overall charge distribution and electron transfer pathways, as evidenced by the markedly increased charge transfer resistance (Figure S41d, Supporting Information). These results demonstrate that the presence of LaN not only improves electrocatalytic activity by constructing an asymmetric heterointerface and generating an internal electric field to facilitate charge transfer, but also suppresses H₂O₂ formation and modulates the ORR reaction pathway. Furthermore, after 10 000 ORR cycles, EDTA‐LaN/LaFe‐NC exhibits pronounced performance degradation and significantly reduced methanol tolerance (Figure S41e,f, Supporting Information). These further confirm that the presence of the LaN shell not only preserves the catalytic performance of the LaFe‐NC core, but also plays a crucial role in enhancing the long‐term ORR activity and stability of LaN/LaFe‐NC, thereby underscoring the importance of the La‐mediated heterostructure in overall performance enhancement.

2.5. Performance of the ZAB Based on LaN/LaFe‐NC

Given its excellent bifunctional catalytic activity and durability, a rechargeable aqueous ZAB was assembled under ambient air using LaN/LaFe‐NC as the air cathode. The resulting LaN/LaFe‐NC‐based ZAB delivers a superior open‐circuit potential of 1.49 V (Figure 5a), higher peak power density (210.92 ± 0.48 mW cm⁻², RSD = 0.23%, Figure 5b; Figure S42, Supporting Information), and larger specific capacity of 790.37 mAh g⁻¹ Zn (Figure 5c) and higher charging/discharging current densities (Figure S43, Supporting Information), outperforming the Pt/C + IrO₂‐based benchmark device (1.37 V, as shown in Figure S44 (Supporting Information), 94.08 mW cm⁻², and 718.69 mAh g⁻¹ Zn, respectively). This outstanding power density exceeds that of most recently reported M‐N‐C and transition metal‐based catalysts in ZABs (Table S5, Supporting Information), underscoring the superior device‐level performance and practical applicability of LaN/LaFe‐NC. To assess the long‐term durability and reversibility, galvanostatic cycling studies were conducted using LaN/LaFe‐NC‐, Fe‐NC‐, and Pt/C + IrO₂‐based ZABs. As shown in Figure 5d, the LaN/LaFe‐NC‐based ZAB operates continuously and stably for >200 h at 5 mA cm⁻², with negligible performance degradation. In contrast, the Pt/C + IrO₂‐based ZAB displays significant decay within 20 h, whereas the Fe‐NC‐based ZAB undergoes rapid deterioration after 100 h. Remarkably, even at elevated current densities of 10 and 20 mA cm⁻², the LaN/LaFe‐NC‐based ZAB maintains stable operation for >240 h, indicating its outstanding long‐term durability (Figure 5e; Figure S45, Supporting Information). Post‐cycling TEM and EDS further confirm that the LaN/LaFe‐NC air cathode retains its structural integrity and elemental distribution after extended cycling (Figure S46, Supporting Information), highlighting its robust architecture. As a practical demonstration, the fabricated LaN/LaFe‐NC‐based ZAB successfully powers a light‐emitting diode (LED) display (Figure 5f), underscoring the considerable potential of LaN/LaFe‐NC as a high‐performance electrocatalyst for use in next‐generation energy storage and conversion technologies.

Figure 5.

ZAB performance enabled by LaN/LaFe‐NC. a) Open‐circuit potential of the ZAB assembled with LaN/LaFe‐NC as the air cathode. Inset: optical image showing potential measurement using a multimeter. b) Discharge polarization curves and the corresponding power density plots of the ZABs based on LaN/LaFe‐NC, Fe‐NC, and Pt/C + IrO₂. c) Specific capacities normalized with respect to Zn mass for the LaN/LaFe‐NC‐ and Pt/C + IrO₂‐based ZABs at a constant current density of 10 mA cm⁻². d) Galvanostatic charge/discharge cycling performances of the ZABs assembled with LaN/LaFe‐NC, Fe‐NC, and Pt/C + IrO₂ at 5 mA cm⁻². e) Cycling stabilities of the LaN/LaFe‐NC‐based ZABs at elevated current densities of 10 and 20 mA cm⁻². f) Optical image of an LED display powered using a LaN/LaFe‐NC‐based ZAB, indicating its practical applicability.

Dynamic relaxation time (DRT) analysis was conducted during battery charging to gain deeper insight into the intrinsic mechanisms behind the enhanced performance and durability of the LaN/LaFe‐NC‐based ZAB, as shown in Figure 6 . LaN/LaFe‐NC exhibits four distinct regions in terms of temporal response during charging, each corresponding to key electrochemical processes within the metal‐air battery system.^{[ 50} , ⁵¹ , ^{52 ]} The τ₃ peak is primarily associated with oxygen adsorption and diffusion. With increasing charging potential, the intensity of the τ₃ peak decreases sharply, and it shifts markedly toward the high‐frequency region, indicating substantial enhancements in the kinetics of oxygen mass transfer. Notably, as the potential reaches 2.1 V, the intensity and position of the τ₃ peak stabilize, suggesting that LaN/LaFe‐NC even maintains an efficient oxygen transport capacity at a high current density, with no significant diffusion limitations. In parallel, the τ₄ peak corresponds to the zinc deposition kinetics and related interfacial charge transfer processes. As the charging current gradually increases, the τ₄ peak exhibits a similar behavior to that of the τ₃ peak, indicating a reduction in the interfacial impedance and a more uniform Zn deposition profile. This optimized deposition behavior effectively suppresses dendrite growth, thus enhancing the cycling stability of the battery and extending its operational lifespan. In conclusion, LaN/LaFe‐NC enables a high energy conversion efficiency and an excellent long‐term stability by simultaneously improving the dynamics of oxygen mass transport and zinc deposition. These findings highlight LaN/LaFe‐NC not only as a highly efficient bifunctional electrocatalyst for use in the ZAB system but also as a promising candidate for broader application in ion‐based energy storage technologies.

Figure 6.

Dynamic charge relaxation behavior of LaN/LaFe‐NC during battery operation. DRT spectra obtained via the deconvolution of the EIS data collected during charging. The dashed line denotes the applied charging potential of 2.1 V, highlighting the stabilization of the charge transfer and mass transport kinetics at a high potential.

3. Conclusion

In summary, we developed a highly efficient and durable heterointerface electrocatalyst (LaN/LaFe‐NC) by leveraging a novel foreign metal (La)‐mediated atomic migration strategy, guided by the difference in the mobilities of the metal atoms. This structure‐directed approach effectively suppressed the loss of active metal sites, significantly enhancing their density and stabilities, thus endowing the catalyst with an outstanding bifunctional ORR/OER activity and excellent performance in a ZAB. The superior performance of LaN/LaFe‐NC was attributed to its unique asymmetric multiphase heterostructure, wherein the defect‐rich graphene interlayer acted as an electronic bridge between the LaFe‐NC core and LaN shell. This architecture induced a built‐in directional electric field across the interface, which not only enhanced the interfacial electronic coupling and charge transport efficiency but also optimized the overall charge transfer kinetics. Combined, these features substantially improved the intrinsic electrocatalytic activity and long‐term operational stability. This study not only introduces a rational design strategy for use in atomic‐level structural control and multi‐metal heterostructure fabrication, but also provides valuable insights into the development of next‐generation high‐performance electrocatalysts for use in renewable energy technologies.

4. Experimental Section

Materials

All chemicals, including Fe(NO₃)₃·9H₂O (≥99.95%), K₄[Fe(CN)₆]·3H₂O (≥98.0%), NH₄Cl (≥99.5%), HCl (37%), La(NO₃)₃·6H₂O (99.9%), and KOH (≥85%), were purchased from Sigma–Aldrich (St. Louis, MO, USA) and used as received without further purification.

Material Synthesis

To synthesize FePBA, 0.5 mmol of Fe(NO₃)₃·9H₂O and 0.5 mmol of K₄[Fe(CN)₆]·3H₂O were separately dissolved in 20 mL of deionized (DI) water. The two solutions were mixed under continuous magnetic stirring and transferred to a Teflon‐lined autoclave, followed by reaction at 60 °C for 12 h. The resulting precipitate was collected via centrifugation, washed three times each with DI water and ethanol, and dried at 80 °C overnight to yield FePBA. To introduce structural defects, 10 mmol of NH₄Cl and 1 mmol of HCl were dissolved in 40 mL of DI water. The as‐prepared FePBA was dispersed into the solution under vigorous stirring and ultrasonicated for 10 min, and the mixture was then transferred to a Teflon‐lined autoclave and reacted at 150 °C for 2 h. The resulting product, d‐FePBA, was collected, washed with DI water and ethanol, and dried at 80 °C overnight. To incorporate La, 0.05 mmol of La(NO₃)₃·6H₂O was dissolved in 50 mL of ethanol. d‐FePBA (50 mg) was then dispersed in this solution, followed by ultrasonication for 1 h and vigorous stirring for 24 h to enable La³⁺ adsorption. The resulting product, La@d‐FePBA, was filtered, dried at 80 °C overnight, finely ground, and annealed at 900 °C for 1 h under an Ar atmosphere in a tube furnace to yield the final LaN/LaFe‐NC catalyst. For comparison, Fe‐NC was prepared by directly annealing the as‐prepared FePBA at 900 °C for 1 h under an Ar atmosphere.

Material Characterization

The morphologies of the synthesized catalysts were examined using a field‐emission scanning electron microscope (Regulus8230, Hitachi, Tokyo, Japan) equipped with an energy‐dispersive X‐ray spectrometer. HAADF‐STEM was conducted using a Titan 80‐300 TEM/STEM system (FEI, Hillsboro, OR, USA) equipped with a spherical aberration corrector and operated at 200 kV. XRD was performed using a high‐power diffractometer (D/MAX 2500V/PC, Rigaku, Tokyo, Japan) operated at 40 kV and 200 mA, with a scan rate of 2° min⁻¹ over the 2θ range 20°–80°. XPS (ESCALAB 250Xi, Thermo Fisher Scientific, Waltham, MA, USA) with monochromatic Cu Kα radiation was used to analyze the valence states and near‐surface chemical composition. TGA was performed using an SDT Q600 system (TA Instruments, New Castle, DE, USA), and FT‐IR spectroscopy was conducted using a Frontier spectrometer (PerkinElmer, Waltham, MA, USA) and KBr pellets. Raman spectroscopy was performed using a confocal Raman microscope (LabRAM ARAMIS IR², Horiba, Kyoto, Japan) to probe the carbon structure and defect characteristics.

Electrocatalytic Measurements

RDE and rotating ring‐disk electrode (RRDE) measurements were performed in a polyethylene three‐electrode electrochemical cell using a RRDE‐3A system (BAS, Tokyo, Japan) connected to a CHI 760E electrochemical workstation (CH Instruments, Austin, TX, USA). A graphite rod and Ag/AgCl (saturated KCl) electrode were used as the counter and reference electrodes, respectively. All measured potentials were converted to the RHE scale using the Nernst equation: E _RHE  =  E _Ag/AgCl + 1.023 V. The working electrode was prepared via drop casting. The catalyst ink was prepared by dispersing the catalyst in a mixed solvent containing 10, 120, and 360 µL of 5 wt.% Nafion solution, isopropanol, and DI water, respectively, followed by bath sonication. The resulting ink was drop‐cast onto a glassy carbon electrode (geometric area: 0.2827 cm²), yielding a catalyst loading of 0.17 mg cm⁻². Prior to measurement, the working electrode was activated using 30 cycles of cyclic voltammetry between 0.40 and 1.10 V versus RHE in O₂‐saturated 0.1 m KOH at a scan rate of 50 mV s⁻¹. The ORR polarization curves were recorded in O₂‐saturated 0.1 m KOH over the potential range 0.20–1.10 V versus RHE at respective scan and rotation rates of 5 mV s⁻¹ and 1600 rpm. The current density was normalized with respect to the geometric surface area of the electrode (0.2827 cm²). The ECSA was calculated from the C _dl, which was obtained by cyclic CV measurements in the non‐Faradaic potential region (0.8−0.9 V) at scan rates ranging from 20 to 100 mV s⁻¹. The ECSA was then determined using the equation ECSA = C _dl/C _s, where C _s denotes the specific capacitance of a smooth surface (0.04 mF cm⁻²). The durability was assessed via an accelerated durability study involving 10 000 continuous potential cycles between 0.60 and 0.95 V versus RHE in O₂‐saturated 0.1 m KOH. All electrochemical data are reported without compensating for ohmic resistance.

ZAB Measurements

A rechargeable aqueous ZAB system was assembled using polished zinc foil as the anode and an electrolyte comprising 6 m KOH mixed with 0.2 m zinc acetate. The as‐prepared catalyst or commercial Pt/C + IrO₂ was coated onto carbon paper (TGP‐H‐60, Toray Industries, Tokyo, Japan), which was pretreated via immersion in a mixed acid solution of H₂SO₄ and HNO₃ (3:1, v/v) for 12 h, and the coated carbon paper was then used as the air cathode. The catalyst loading was controlled at 1.0 mg cm⁻², and the charge/discharge polarization curves were recorded using a CHI 760E electrochemical workstation. The galvanostatic charge/discharge cycling studies were conducted at various current densities, with each cycle comprising 10 min of discharge and then 10 min of charge under ambient conditions. The specific capacity is calculated using the following equation:

$$\mathrm{Specific}\mathrm{capacity}=\frac{\mathrm{Current}\times \mathrm{Time}}{\mathrm{Mass}\mathrm{of}\mathrm{consumed}\mathrm{Zinc}}$$ (1)

In Situ Raman Spectroscopy

The LaN/LaFe‐NC was dispersed in a Nafion solution, sonicated in a bath, and drop‐cast onto carbon paper to prepare the working electrode. The prepared sample was then mounted in a custom‐designed electrochemical Raman cell equipped with a quartz window and configured in a three‐electrode setup. A graphite rod and saturated Ag/AgCl electrode were used as the counter and reference electrodes, respectively. Electrochemical control was maintained using a CHI 760E workstation (CH Instruments Inc.), and in situ Raman spectra were recorded using a confocal Raman microscope (LabRAM ARAMIS IR²).

Computational Methods

All DFT calculations were performed using the Materials Studio software package (BIOVIA, San Diego, CA, USA) and the Vienna Ab initio Simulation Package (VASP). The computational setup, including the exchange–correlation functionals, basis sets, and convergence criteria, is described in the Supporting Information.

Conflict of Interest

The authors declare no conflict of interest.

Supporting information

Supporting Information

Supplemental Video 1

Zhang Y., Jeong S., Park J., et al. “Lanthanum‐Induced Gradient Fields in Asymmetric Heterointerface Catalysts for Enhanced Oxygen Electrocatalysis.” Adv. Mater. 38, no. 1 (2026): e11117. 10.1002/adma.202511117

Data Availability Statement

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