Electronic Modulation of the 3D Architectured Ni/Fe Oxyhydroxide Anchored N-Doped Carbon Aerogel with Much Improved OER Activity

It remains a big challenge to develop non-precious metal catalysts for oxygen evolution reaction (OER) in energy storage and conversion systems. Herein, a facile and cost-effective strategy is employed to in situ prepare the Ni/Fe oxyhydroxide anchored on nitrogen-doped carbon aerogel (NiFeOx(OH)y@NCA) for OER electrocatalysis. The as-prepared electrocatalyst displays a typical aerogel porous structure composed of interconnected nanoparticles with a large BET specific surface area of 231.16 m2·g−1. In addition, the resulting NiFeOx(OH)y@NCA exhibits excellent OER performance with a low overpotential of 304 mV at 10 mA·cm−2, a small Tafel slope of 72 mV·dec−1, and excellent stability after 2000 CV cycles, which is superior to the commercial RuO2 catalyst. The much enhanced OER performance is mainly derived from the abundant active sites, the high electrical conductivity of the Ni/Fe oxyhydroxide, and the efficient electronic transfer of the NCA structure. Density functional theory (DFT) calculations reveal that the introduction of the NCA regulates the surface electronic structure of Ni/Fe oxyhydroxide and increases the binding energy of intermediates as indicated by the d-band center theory. This work provides a new method for the construction of advanced aerogel-based materials for energy conversion and storage.


Introduction
With the development of economic globalization and the growth of the population, the exploitation of low-cost and efficient energy conversion/storage technologies has attracted widespread attention [1]. The electrolysis of water into hydrogen and oxygen has been considered one of the most effective ways to alleviate energy consumption and develop a carbon-free economy [2]. The OER (4OH − →2H 2 O + O 2 + 4e − ) process is the anodic half-reaction of the water-splitting reaction, however, it is a four-proton-electron coupling transfer process, which results in a slow kinetic process and severely impedes the energy conversion efficiency [3,4].
Noble-metal-based oxides such as RuO 2 and IrO 2 are usually considered commercial OER electrocatalysts. However, the applications of noble metals are limited due to their scarcity and high cost, which has restricted their large-scale production and commercialization [5,6]. Therefore, researchers are committed to developing cost-effective electrocatalysts with excellent OER performance, such as transition metal oxides [7], spinel semiconductor oxides [8], layered double hydroxides [9], metal-based oxyhydroxides [10], perovskite oxides [11], etc. Among them, transition metal-based hydroxides with low kinetic energy barriers have made great progress in structural modification and electrocatalytic properties [12,13]. In particular, Ni and Fe-containing OER catalysts are versatile substitutes for precious metal species due to their earth-abundance, low-costing, environmental benignity, and theoretically high catalytic activity. Ni/Fe oxyhydroxides are The X-ray diffraction (XRD) patterns of the resulting samples showing the phases and compositions are exhibited in Figure 1a. It is mentioned that FeO x (OH)y@NCA, and NiO x (OH)y@NCA are controlling samples without introducing Ni, and Fe, respectively. It is found that the broad peaks at around 26.3 • can be ascribed to (002) crystal planes of graphite C (PDF No. , indicating the amorphous structure of the carbon aerogelbased electrocatalyst. The peaks at 2θ = 26.   phase can be observed with the increase in Ni amount. Meanwhile, the intensities of the peaks attributed to FeOOH and NiOOH become stronger as the Ni/Fe molar ratio increases. It is worth mentioning that the intensity of the Fe and Ni species is weak, which is due to the fact that the Ni/Fe oxyhydroxide shows low crystallinity and the porous carbon support occupies most of the electrocatalyst. That is to say that the Ni 7 FeO x (OH) y @NCA sample shows an amorphous structure without strong diffraction peaks. In addition, one of the most typical characteristics of aerogel is its porous and disordered structure, therefore, we have claimed that the diffraction results indicate the typical aerogel structures. However, the existence of the Fe/Ni oxyhydroxide can be verified by TEM and XPS analysis afterward.
NiFeOx(OH)y, the surface electronic structures of NiFeOx(OH)y are regulated, and the binding energy between the electrocatalyst surface and the intermediate is increased. Therefore, this work provides a simple strategy to construct low-cost and efficient electrocatalysts for applications in fuel cells, energy storage, and conversion, and other energy regeneration fields.

Chemical Composition and Structural Analysis
The X-ray diffraction (XRD) patterns of the resulting samples showing the phases and compositions are exhibited in Figure 1a. It is mentioned that FeOx(OH)y@NCA, and NiOx(OH)y@NCA are controlling samples without introducing Ni, and Fe, respectively. It is found that the broad peaks at around 26.3° can be ascribed to (002) crystal planes of graphite C (PDF No. , indicating the amorphous structure of the carbon aerogelbased electrocatalyst. The peaks at 2θ = 26. 5   phase can be observed with the increase in Ni amount. Meanwhile, the intensities of the peaks attributed to FeOOH and NiOOH become stronger as the Ni/Fe molar ratio increases. It is worth mentioning that the intensity of the Fe and Ni species is weak, which is due to the fact that the Ni/Fe oxyhydroxide shows low crystallinity and the porous carbon support occupies most of the electrocatalyst. That is to say that the Ni7FeOx(OH)y@NCA sample shows an amorphous structure without strong diffraction peaks. In addition, one of the most typical characteristics of aerogel is its porous and disordered structure, therefore, we have claimed that the diffraction results indicate the typical aerogel structures. However, the existence of the Fe/Ni oxyhydroxide can be verified by TEM and XPS analysis afterward.  The Raman spectrum provides intrinsic and rich structural information about the resulting samples ( Figure 1b). The resulting samples show the well-defined D-band and Gband at 1385 and 1587 cm −1 , respectively. The D-band is associated with defects and partial disorder structures on the catalyst surface, while the G-band represents the E2g vibrational mode in the carbon domain, which shows the degree of graphitization of the resulting samples [30,31]. Generally, the peak intensity ratio of the D and G bands (I D /I G ) has been considered an effective evaluation criterion for evaluating the disorder degree of carbonbased material. The estimated I D /I G values of the three samples are extremely similar (0.582 for Ni 7 FeO x (OH) y @NCA, 0.594 for FeO x (OH) y @NCA, and 0.581 for NiO x (OH) y @NCA, respectively). This is caused by the fact that the porous NCA support occupies most of all the three electrocatalysts, with a small amount of oxyhydroxide anchored on the porous supports. Therefore, the effect of metal/metal oxide content on the degree of graphitization is extremely weak. In addition, we have also tested the I D /I G value (0.798) of the metal-free NCA sample. It is found that as compared with the pristine metal-free NCA sample, the I D /I G value is much decreased, showing the higher order states of NCA after the introduction of the oxyhydroxide. It is inferred that the introduction of Ni/Fe oxyhydroxide improves the graphitization of NCA, therefore increasing the charge transfer rate of the electrons during the four elementary steps, which is responsible for the much enhanced OER activity [32]. Figure 1c,d show the nitrogen adsorption/desorption isotherms and BJH size distributions of the samples with different Ni/Fe molar ratios. According to Figure 1c, the curve of FeO x (OH) y @NCA exhibits type IV isotherm with a saturated adsorption plateau, which indicates that it is typically mesoporous. The other four curves all show mixed isotherms of II and IV with H 3 -type hysteresis loops, which show no saturated adsorption plateaus, therefore, indicating the co-presence of large pores and mesopores inside the pore structures. The characteristic capillary condensation regions of the curves for the resulting samples are all in the range of 0.9-1.0, indicating that it is a typical aerogel with a three-dimensional mesoporous network structure [33,34]. It is noteworthy that the Ni 12 FeO x (OH) y @NCA sample possesses the largest adsorbed volume, and therefore it has the largest BJH adsorption pore volume of 2.20 cm 3 ·g −1 . As for the BJH distribution curves, it is noted that single-peaked pore size distributions of the FeO x (OH) y @NCA and NiO x (OH) y @NCA have been displayed. It is noted that the FeO x (OH) y @NCA has a strong peak around 15 nm, showing its rather homogeneous mesoporous structure. This indicates that the introduction of Ni has slowed down the kinetic process of the gelation and reduced the average pore size of the resulting samples. Furthermore, the most probable diameter of the Ni 12 FeO x (OH) y @NCA is around 90 nm. However, there are no distinct peaks for the Ni 3 FeO x (OH) y @NCA and Ni 7 FO x (OH) y @NCA samples. This is caused by the fact that with a large amount of Ni in the solution during the sol-gel process, the hierarchical structures including the micropores, mesopores, and large pores have formed. It is found that the optimal sample (Ni 7 FO x (OH) y @NCA) has a medial specific surface area, an adsorption pore volume, and an average pore size among the six samples (Table 1). In addition, it is worth mentioning that although the Ni 7 FeO x (OH) y @NCA sample does not have the largest BET specific surface area among the samples, it possesses the largest electrochemically active surface area (ECSA) via the electrochemical test afterward, indicating the much enhanced OER activity. The BET specific surface areas of the resulting samples are 196.07, 238.64, 614.53, 496.48, and 611.76 m 2 ·g −1 , respectively, among which the Ni 12 FeO x (OH) y @NCA and NiO x (OH) y @NCA possess extremely large BET specific surface areas larger than 600 m 2 ·g −1 . The introduction of Ni has extended the mesopores to large pore regions as indicated by Figure 1c, which can be responsible for the much-improved BET specific surface areas. It is worth mentioning that the BET specific surface area of the optimal Ni 7 FeO x (OH) y @NCA (238.64 m 2 ·g −1 ) is much higher than those in the previously reported works [35][36][37], which facilitates the exposure of more active sites in the catalyst during the OER process. Meanwhile, the BJH adsorption pore volumes of the resulting samples are 0.05, 0.06, 0.14, 0.05, and 0.15 cm 3 ·g −1 , respectively. In addition, the average Gels 2023, 9,190 5 of 17 pore sizes are 18.05, 16.14, 32.10, 12.10, and 17.37 nm, respectively. It is noted that the optimal Ni 7 FeO x (OH) y @NCA possesses an appropriate pore size of 16.14 nm, within the middle range of the typical mesoporous aerogels, which is favorable to the enhancement of OER electrocatalytic activity.

Synthetic Route and Pore Morphology Analysis
The resulting NiFeO x (OH) y @NCA sample was obtained by in situ preparation of Ni/Fe oxyhydroxides on NCA via the sol-gel method ( Figure 2a). The NiFeO x (OH) y @NCA sample was prepared by gelation, supercritical drying, and high-temperature heat treatment of the RF solution containing N-doped Ni/Fe metal salts (Ni/Fe/NRF). The Scanning Electron Microscope (SEM) images (Figure 3a,b, S1 and S2) demonstrate that the resulting samples all show a 3D porous morphology composed of interconnected nanoparticles, indicating the typical aerogel structures. The FeO x (OH) y @NCA sample ( Figure S1) exhibits a mesoporous and homogeneous structure with smooth morphology, and the nanoparticle diameters are at around 10-20 nm, with pore diameters of 10-30 nm. The NiO x (OH) y @NCA sample ( Figure S2) has a rough surface with smaller nanoparticle diameters of 5-10 nm, and the pore sizes are located at 20-40 nm. which is consistent with the BJH pore size distribution analysis. In addition, it is noted that a small number of large pores and particle aggregation are exhibited in the as-prepared NiO x (OH) y @NCA sample due to the rich amount of Ni. As for the optimal Ni 7 FeO x (OH) y @NCA (Figure 2b,c), it is clear that the diameters of the nanoparticles are about 20-30 nm, which is between those of the FeO x (OH) y @NCA and NiO x (OH) y @NCA samples. In addition, the optimal sample shows many pores with diameters at several tens of nanometers, and it is noted that there are some large pores with diameters larger than 100 nm due to the introduction of Ni affecting grain growth. These hierarchical pore structures of the Ni 7 FeO x (OH) y @NCA aerogel provides more active sites for electron transfer, loading of the Ni/Fe oxyhydroxide, and diffusion channels for the intermediates. Therefore, the synergy effect between the N-doped carbon aerogel and Ni/Fe oxyhydroxide accelerates its OER activity. further verified that the Ni/Fe oxyhydroxide has been well anchored on the surface of the NCA, which is conducive to the further improvement of its OER performance.

X-ray Photoelectron Spectroscopy Analysis
XPS analysis is carried out to analyze the surface chemical composition and valence state of resulting electrocatalysts. The corresponding high-resolution C1s spectrum of the Ni 7 FeO x (OH) y @NCA sample (Figure 4a) shows the distinguishable C-C bond at 284.0 eV, the C-N bond at 285.1 eV, and the π-π* bond at 289.7 eV [38]. As shown in Figure 4b, the peaks at around 532.0 eV, 532.3 eV, and 534.0 eV can be ascribed to the metal-oxides, O-C-O, and C-OH, respectively [19,39]. As shown in Figure 4c, the peaks at around 398.1 eV, 400.6 eV, 401.2 eV, and 402.6 eV can be rationally assigned to the pyrazine N, pyridine N, graphitic N, and oxidized N, respectively [29,40]. The high-resolution XPS spectra of the Fe 2p and Ni 2p of the optimal Ni 7 FeO x (OH) y @NCA (Figure 4d,e) can be split into 2p3/2 and 2p1/2 doublets on account of the spin-orbit coupling of the electrons. The peaks at 710.8 eV and 723.9 eV can be assigned to the Fe 2p 3/2 and Fe 2p 1/2 of Fe 3+ . In addition, it could be observed that the peaks at 706.8 eV and 718.4 eV are caused by the Fe 2p 3/2 and Fe 2p 1/2 of Fe 0 [34]. In addition, the peaks at 715.3 eV and 726.9 eV belong to the satellite peaks of Fe. As depicted in Figure 4e, the high-resolution Ni 2p spectra are consistent with previous literature reports. The peaks at 854.7 eV and 872.7 eV result from the Ni 2p 3/2 and Ni 2p 1/2 of Ni 3+ , and the peaks at 851.8 eV and 869.2 eV can be ascribed to the Ni 2p 3/2 and Ni 2p 1/2 of Ni 0 . In addition, the peaks at 857.8 eV and 876.0 eV belong to the multiplet-split of Ni 3+ 2p 3/2 and Ni 3+ 2p 1/2 , and the peaks at 861.3 eV and 879.6 eV are caused by the satellite peaks of Ni [39,41], which keeps consistent with the results of the XRD, and it further confirms the successful fabrication of the NiFeO x (OH) y . The XPS spectra of C, N, and O ( Figure S3a-c) for FeO x (OH) y @NCA and NiO x (OH) y @NCA samples are almost identical to the optimal Ni 7 FeO x (OH) y @NCA sample. The results obtained by the Lorentzian-Gaussian function with different contributions are concluded in Table S1. As compared with the controlling FeO x (OH) y @NCA sample, it is mentioned that the binding energy of the Fe 0 and Fe 3+ for the optimal sample shifts negatively, as well as the oxidized N, graphitic N, and pyrazine N peaks, which is caused by the electrons obtained from the adjacent Ni. As compared with the NiO x (OH) y @NCA sample, the binding energy of Ni 0 and Ni 3+ of the optimal sample both shifts negatively, which may be also caused by the introduced Fe providing rich electrons in the resulting aerogels.

OER Performance Analysis
The linear sweep voltammetry (LSV) curves of NiFeOx(OH)y@NCA are shown in Figure 5a. When the current density reaches 10 mA·cm −2 , the overpotential of the

OER Performance Analysis
The linear sweep voltammetry (LSV) curves of NiFeO x (OH) y @NCA are shown in Figure 5a. When the current density reaches 10 mA·cm −2 , the overpotential of the Ni 7 FeO x (OH) y @NCA, FeO x (OH) y @NCA, NiO x (OH) y @NCA, Ni 7 FeO x (OH) y @CA, commercial RuO 2 , and Ni 7 FeO x (OH) y are 304, 404, 547, 430, 356, and 532 mV, respectively. The optimal Ni 7 FeO x (OH) y @NCA shows the lowest overpotential, which is even superior to the commercial electrocatalyst RuO 2 , indicating that a proper Ni/Fe molar ratio promotes the OER property. According to the LSV curve, the performance of OER decreases sharply when there is no Fe element, therefore, we speculate that the Fe atom is likely to be the active site. Additionally, the kinetic process of the OER can be evaluated from the Tafel slope plots calculated from the polarization curves. As shown in Figure 5b, the Ni 7 FeO x (OH) y @NCA exhibits the smallest Tafel slope of 72 mV·dec −1 , which is much better than those of the FeO x (OH) y @NCA (145 mV·dec −1 ), NiO x (OH) y @NCA (210 mV·dec −1 ), Ni 7 FeO x (OH) y @CA (112 mV·dec −1 ), RuO 2 (102 mV·dec −1 ), and Ni 7 FeO x (OH) y (126 mV·dec −1 ) that the optimal Ni 7 FeO x (OH) y @NCA has the fastest OER kinetics. To further investigate the electrocatalytic performance of NiFeO x (OH) y @NCA, we have performed cyclic voltammetric (CV) measurements to determine the double-layer capacitances (C dl ), which is considered an effective evaluation method of the electronic catalytic surface area (ECSA) [42]. The CV curves of the resulting electrocatalysts at varying scan rates are shown in Figure S4. Furthermore, as shown in Figure 5c, the C dl of optimal Ni 7 FeO x (OH) y @NCA is 8.62 mF·cm −2 , which is larger than those of the FeO x (OH)y@NCA (6.02 mF·cm −2 ), NiFeO x (OH)y@NCA (4.69 mF·cm −2 ), NiO x (OH) y @NCA (2.78 mF·cm −2 ), and the commercial RuO 2 (6.95 mF·cm −2 ). It is demonstrated that by rationally designing the structure of NiFe grown on NCA, larger active sites can be exposed. A larger ECSA favors water molecule adsorption and close contact with the electrolyte, as well as abundant active sites for electrocatalytic reactions. Figure 5d shows the overpotentials of the resulting samples, among which the optimal Ni 7 FeO x (OH) y @NCA shows excellent OER activity among the controlling samples. In addition, the effects of resorcinol (R)/Fe molar ratios on the OER electrocatalytic performance are investigated ( Figure S5). It is worth noting that the R/formaldehyde (F) and the Ni/Fe molar ratio are fixed, and only various R/Fe molar ratios are carried out for the OER test. It is found that when the molar ratio of R/Fe is 80, the Ni 7 FeO x (OH) y @NCA sample shows the best OER activity among the five samples. This indicates that the OER activity is dependent on the moderate content of the Ni/Fe oxyhydroxide loaded on the NCA substrate. The effect of heat-treatment temperature on OER performance is also investigated and it is found that the electrocatalyst obtained by heat-treatment at 900 • C shows the best OER activity among the three samples ( Figure S6).
Stability is a crucial factor in assessing the feasibility of OER electrocatalytic performance. The polarization curve of the optimal Ni 7 FeO x (OH) y @NCA only slightly drifts after extended scanning of 2000 cycles (Figure 6a). The time-dependent potential profile of the Ni 7 FeO x (OH) y @NCA is also evaluated at 10 mA·cm −2 over 8 h (Figure 6a, inset), which further verifies that the optimal Ni 7 FeO x (OH) y @NCA is stable during the OER process. Furthermore, the SEM images of the resulting Ni 7 FeO x (OH) y @NCA aerogel catalyst after the electrochemical test is further developed to evaluate the stability. As seen in Figure  S7a,b, no obvious changes including the particle size, pore size, and morphologies have been observed, further verifying the stability of the aerogel-based electrocatalyst. EIS is used to study the charge transport kinetics of the electrocatalysts. The Nyquist plots and the equivalent electrical circuit are shown in Figure 6b. The charge-transfer resistance (Rct) value of the FeO x (OH)y@NCA sample has the smallest Rct, which is due to its mesoporous and homogeneous structure with smooth morphology, therefore showing better electric conductivity than the optimal Ni 7 FeO x (OH) y @NCA. [43,44]. It is mentioned that the optimal Ni 7 FeO x (OH) y @NCA exhibits a proper Rct of 28.0 Ω, which is comparable with those in the previously reported literature [45][46][47], therefore, it has an excellent charge transfer property during the OER process. Furthermore, the TOF value is further tested at an overpotential of 300 mV (Figure 6c). The optimal Ni 7 FeO x (OH) y @NCA shows a large turnover fre-Gels 2023, 9,190 9 of 17 quency (TOF) value of 0.506 s −1 , which is remarkably larger than that of FeO x (OH) y @NCA (0.167 s −1 ), NiO x (OH) y @NCA (0.067 s −1 ) and Ni 7 FeO x (OH) y @CA (0.08 s −1 ), demonstrating its rather outstanding OER activity. It is found that the Ni 7 FeO x (OH) y @NCA has the best mass activity at an overpotential of 300 mV (Figure 6d). Table S2 compares the OER activity of the optimal Ni 7 FeO x (OH) y @NCA to other similar electrocatalysts reported in the literature. Clearly, the as-prepared Ni 7 FeO x (OH) y @NCA shows strongly competitive and even better OER performance compared to the reported electrocatalysts, which can be a promising electrocatalyst for wide applications in energy saving and energy storage. Gels 2023, 9,190 9 of 17 activity is dependent on the moderate content of the Ni/Fe oxyhydroxide loaded on the NCA substrate. The effect of heat-treatment temperature on OER performance is also investigated and it is found that the electrocatalyst obtained by heat-treatment at 900 °C shows the best OER activity among the three samples ( Figure S6). Stability is a crucial factor in assessing the feasibility of OER electrocatalytic performance. The polarization curve of the optimal Ni7FeOx(OH)y@NCA only slightly drifts after extended scanning of 2000 cycles (Figure 6a). The time-dependent potential profile of the Ni7FeOx(OH)y@NCA is also evaluated at 10 mA·cm −2 over 8 h (Figure 6a, inset), which further verifies that the optimal Ni7FeOx(OH)y@NCA is stable during the OER process. Furthermore, the SEM images of the resulting Ni7FeOx(OH)y@NCA aerogel catalyst after the electrochemical test is further developed to evaluate the stability. As seen in Figure S7a,b, no obvious changes including the particle size, pore size, and morphologies have been observed, further verifying the stability of the aerogel-based electrocatalyst. EIS is used to study the charge transport kinetics of the electrocatalysts. The Nyquist plots and the equivalent electrical circuit are shown in Figure 6b. The chargetransfer resistance (Rct) value of the FeOx(OH)y@NCA sample has the smallest Rct, which is due to its mesoporous and homogeneous structure with smooth morphology, therefore showing better electric conductivity than the optimal Ni7FeOx(OH)y@NCA. [43,44]. It is mentioned that the optimal Ni7FeOx(OH)y@NCA exhibits a proper Rct of 28.0 Ω, which is comparable with those in the previously reported literature [45][46][47], therefore, it has an excellent charge transfer property during the OER process. Furthermore, the TOF value is further tested at an overpotential of 300 mV (Figure 6c). The optimal Ni7FeOx(OH)y@NCA shows a large turnover frequency (TOF) value of 0.506 s −1 , which is remarkably larger than that of FeOx(OH)y@NCA (0.167 s −1 ), NiOx(OH)y@NCA (0.067 s −1 ) and Ni7FeOx(OH)y@CA

Theoretical Calculations
To gain insight into the synergistic effect of NiFeO x (OH) y and NCA for enhancing the OER performance, the DFT calculations are further carried out. The original lattice constants of the NiFeO x (OH) y @NCA heterojunction are a = 9.28 Å, b = 9.28 Å, and c = 14.78 Å, respectively, and the lattice parameters are a = 9.93 Å, b = 9.93 Å, and c = 18.78 Å after geometry optimization. The optimized interlayer spacing is about 2.525 Å (Figure 7a,b), which verifies that a strong electronic interaction is generated between the two layers. To deeply study the electron transfer between the NiFeO x (OH) y and the NCA layer, as well as the electron transfer within the active NiFeO x (OH) y layer, the electron density difference (Figure 7c) is also calculated via the optimized NiFeO x (OH) y @NCA. It can be found that electron depletion occurs around the upper NCA layer and electrons accumulation occurs near the the Ni/Fe oxyhydroxide layer, indicating that electrons tend to flow from the upper NCA layer to the bottom NiFeO x (OH) y layer, which is consistent with the XPS results. Since the spin densities of the atoms have a great impact on the OER performance of the electrocatalyst, the spin densities of the optimized structures have shown in Figure 7d. It is observed that the NiFeO x (OH) y @NCA sample echibits strong magnetism ( Figure S8), mainly resulting from the center Fe atoms and the adjacent Ni also contributes a small amount of the magnetism. electrocatalysts reported in the literature. Clearly, the as-prepared Ni7FeOx(OH)y@NCA shows strongly competitive and even better OER performance compared to the reported electrocatalysts, which can be a promising electrocatalyst for wide applications in energy saving and energy storage.

Theoretical Calculations
To gain insight into the synergistic effect of NiFeOx(OH)y and NCA for enhancing the OER performance, the DFT calculations are further carried out. The original lattice constants of the NiFeOx(OH)y@NCA heterojunction are a = 9.28 Å, b = 9.28 Å, and c = 14.78 Å, respectively, and the lattice parameters are a = 9.93 Å, b = 9.93 Å, and c = 18.78 Å after geometry optimization. The optimized interlayer spacing is about 2.525 Å (Figure 7a,b) which verifies that a strong electronic interaction is generated between the two layers. To deeply study the electron transfer between the NiFeOx(OH)y and the NCA layer, as wel as the electron transfer within the active NiFeOx(OH)y layer, the electron density difference (Figure 7c) is also calculated via the optimized NiFeOx(OH)y@NCA. It can be found that electron depletion occurs around the upper NCA layer and electrons accumulation occurs near the the Ni/Fe oxyhydroxide layer, indicating that electrons tend to flow from the upper NCA layer to the bottom NiFeOx(OH)y layer, which is consisten with the XPS results. Since the spin densities of the atoms have a great impact on the OER performance of the electrocatalyst, the spin densities of the optimized structures have shown in Figure 7d. It is observed that the NiFeOx(OH)y@NCA sample echibits strong magnetism ( Figure S8), mainly resulting from the center Fe atoms and the adjacent Ni also contributes a small amount of the magnetism.   Figure 8a shows the total density of states (TDOS) and projected density of states (PDOS) of the resulting NiFeO x (OH) y @NCA, as well as the individual NiFeO x (OH) y and NCA, which are used to study the changes in the electronic structure of the surface. Gels 2023, 9, 190 11 of 17 Compared with individual NiFeO x (OH) y and NCA, the NiFeO x (OH) y @NCA shows an enhanced DOS near the Fermi level due to electronic interactions, which is beneficial for charge transfer in both layers. Norskov [48] et al. have proposed that the d-band center of the metal-based active site has a significant impact on the OER performance of the electrocatalyst by tuning the binding strength of intermediates such as *OOH, *OH, and *O between the catalyst surface. According to the PDOS results shown in Figure 8b,c, the d-band center of the Ni atom of NiFeO x (OH) y @NCA positively shifts to −3.93 eV as compared to individual NiFeO x (OH) y (−4.07 eV), and the d-band center of the Fe atom of NiFeO x (OH) y @NCA positively shifts from −3.78 eV to −1.62 eV as compared to the individual NiFeO x (OH) y . This leads to an increase in the adsorption strength of intermediates on the catalyst surface, therefore accelerating the OER performance. As shown in Figure 8d-f, it is observed that the Ni d and Fe d orbitals contribute mainly to the top of the valance band, while the bottom of the conduction band is mainly caused by the N p orbitals. Figure 8a shows the total density of states (TDOS) and projected density of s (PDOS) of the resulting NiFeOx(OH)y@NCA, as well as the individual NiFeOx(OH) NCA, which are used to study the changes in the electronic structure of the su Compared with individual NiFeOx(OH)y and NCA, the NiFeOx(OH)y@NCA show enhanced DOS near the Fermi level due to electronic interactions, which is benefici charge transfer in both layers. Norskov [48] et al. have proposed that the d-band cen the metal-based active site has a significant impact on the OER performance o electrocatalyst by tuning the binding strength of intermediates such as *OOH, *OH *O between the catalyst surface. According to the PDOS results shown in Figure 8b, d-band center of the Ni atom of NiFeOx(OH)y@NCA positively shifts to −3.93 e compared to individual NiFeOx(OH)y (−4.07 eV), and the d-band center of the Fe ato NiFeOx(OH)y@NCA positively shifts from −3.78 eV to −1.62 eV as compared to individual NiFeOx(OH)y. This leads to an increase in the adsorption strengt intermediates on the catalyst surface, therefore accelerating the OER performanc shown in Figure 8d-f, it is observed that the Ni d and Fe d orbitals contribute mainly t top of the valance band, while the bottom of the conduction band is mainly caused b N p orbitals.  It is worth noting that the Fermi level of NiFeO x (OH) y is lower than NCA, which is energetic for electrons migration from NCA to NiFeO x (OH) y . Therefore, the energy bands of NiFeO x (OH) y shift upward, while the energy bands of NCA shift downward, and finally, the two phases reach the equilibrium Fermi level (Figure 9c). Figure 9d shows the PDOS of the generated NiFeO x (OH) y @NCA sample of O. The spin-up and spin-down electronic asymmetries suggest that the strong spin densities of oxygen atoms have been induced by the adjacent Ni and Fe atoms in the resulting NiFeO x (OH) y @NCA sample, which are also responsible for the much-enhanced OER activity. NiFeOx(OH)y. Therefore, the energy bands of NiFeOx(OH)y shift upward, while the energy bands of NCA shift downward, and finally, the two phases reach the equilibrium Fermi level (Figure 9c). Figure 9d shows the PDOS of the generated NiFeOx(OH)y@NCA sample of O. The spin-up and spin-down electronic asymmetries suggest that the strong spin densities of oxygen atoms have been induced by the adjacent Ni and Fe atoms in the resulting NiFeOx(OH)y@NCA sample, which are also responsible for the much-enhanced OER activity.

Conclusions
A novel type NiFeOx(OH)y@NCA electrocatalyst is prepared by a facile and costeffective strategy in this work. The optimal NiFeOx(OH)y@NCA displays a typical aerogel porous structure composed of interconnected nanoparticles with a large BET specific surface area of 231.16 m 2 ·g −1 . The resulting NiFeOx(OH)y@NCA exhibits a low overpotential of 304 mV at 10 mA·cm −2 , a small Tafel slope of 72 mV·dec −1 , and excellent stability after 2000 CV cycles, which is superior to the commercial RuO2 catalyst and most of the reported OER electrocatalysts. The excellent performance is attributed to the welldesigned three-dimensional structure of the carbon aerogel, the extremely large electrochemical active surface area, the synergistic effect of Ni/Fe oxyhydroxide with the porous NCA, and the change of electronic structures of the adjacent carbon via nitrogen introduction. DFT calculations show that the d-band center of the metal-based active site

Conclusions
A novel type NiFeO x (OH) y @NCA electrocatalyst is prepared by a facile and costeffective strategy in this work. The optimal NiFeO x (OH) y @NCA displays a typical aerogel porous structure composed of interconnected nanoparticles with a large BET specific surface area of 231.16 m 2 ·g −1 . The resulting NiFeO x (OH) y @NCA exhibits a low overpotential of 304 mV at 10 mA·cm −2 , a small Tafel slope of 72 mV·dec −1 , and excellent stability after 2000 CV cycles, which is superior to the commercial RuO 2 catalyst and most of the reported OER electrocatalysts. The excellent performance is attributed to the well-designed three-dimensional structure of the carbon aerogel, the extremely large electrochemical active surface area, the synergistic effect of Ni/Fe oxyhydroxide with the porous NCA, and the change of electronic structures of the adjacent carbon via nitrogen introduction. DFT calculations show that the d-band center of the metal-based active site positively shifts and the binding strength of intermediates and the catalyst surface is greatly enhanced, which is responsible for the greatly enhanced OER activity. Therefore, this work may provide a new strategy for the development of low-cost and highly efficient carbon aerogel-based advanced electrocatalysts by tuning electronic structures. The resulting NiFeO x (OH) y @NCA sample was obtained by in situ preparation of Ni/Fe oxyhydroxides on NCA via the sol-gel method. FeCl 3 ·6H 2 O (27 mg), and urea (875 mg) were dissolved in the 5 mL of DI water, and then different amounts of Ni(NO 3 ) 2 ·6H 2 O were added into the obtained mixed solution with the Ni/Fe molar ratios of 1, 3, 5, 7,10, and 12. The above reaction mixture was further stirred for 10 min to obtain the homogeneous solution A. Resorcinol (900 mg) and formaldehyde (1.6 mL) were dissolved in H 2 O (6.8 mL) and stirred for 10 min to form a homogeneous solution B. Then, solutions A and B were mixed, and the pH value was adjusted to 7.0 with the addition of Na 2 CO 3 (10 mg) to obtain the RF solution containing N-doped Ni/Fe metal salt (Ni/Fe/NRF sol). The solution was further stirred for 0.5 h under room temperature, and the mixture was transferred into a plastic mold until gelation was under 50 • C. The wet gels were aged at 50 • C for three days and three times each day, during which period the wet gels were washed with ethanol to remove the impurities, organic solvents, and water inside the pores. Subsequently, the wet gels were dried by CO 2 supercritical drying to obtain the as-dried aerogels. Finally, the as-dried aerogels were carbonized under an N 2 atmosphere at 900 • C for 2 h to obtain NiFeO x (OH) y @NCA. In addition, the FeO x (OH) y @NCA, NiO x (OH) y @NCA, and NiFeO x (OH) y @NCA samples were also produced by a similar process in the absence of Ni(NO 3 ) 2 ·6H 2 O, FeCl 3 ·6H 2 O, and urea, respectively.

Characterizations
The X-ray diffraction (XRD) patterns of the nanostructures were performed on a Rigaku Ultima X-ray diffractometer with Cu Kα radiation in the 2θ range of 10-80 • , and the operating voltage and current were 40 kV and 40 mA, respectively, with a step size of 0.02 • . X-ray photoelectron spectroscopy (XPS) measurements were performed using a Thermo Scientific K-Alpha. The XPS test was carried out with an Al Kα radiation, a spot size of 400 µm, the analyzer mode of CAE (pass energy 150.0 eV), and the step size of 1.0 eV, respectively. The microstructures and morphologies of the samples were investigated using a Scanning Electron Microscope (SEM, ZEISS Sigma 304). The N 2 adsorption/desorption isotherms, including BET specific surface area, pore volume, and pore size distribution were performed on a V-Sorb 2800P surface area and pore distribution analyzer. Raman spectra were conducted on the Horiba Evolution equipment at an excitation laser wavelength of 532 nm. The morphology and compositions were further obtained using transmission electron microscopy (TEM, FEI TF20), and the EDS mapping images were also tested in the equipment. The magnetic properties of the resulting electrocatalyst were tested on the HH-15 vibrating sample magnetometer with a magnetic field between −10,000 to 10,000 Oe.

Electrochemical Measurements
Electrochemical measurements were performed with an electrochemical CS2350 workstation in a three-electrode setup, using 1 M KOH as the electrolyte solution (pH = 14). The Hg/HgO (1 M NaOH) electrode and platinum wire were used as the reference electrode and counter electrode, respectively. The catalyst ink was prepared by mixing 10 mg of catalyst, 600 µL of ethanol, 400 µL of deionized water, and 40 µL 5 wt% Nafion solution, and ultrasonic for 1 h. An amount of 10 µL of catalyst ink was then loaded onto a glassy carbon electrode (3 mm in diameter), with a loading mass density of 1.37 mg/cm 2 . The linear sweep voltammetry (LSV) curves were recorded at a scan rate of 10 mV s −1 after electrochemical conditioning by 10 cyclic voltammetric (CV) scans reaching a stable state. The double-layer capacitances (C dl ) for estimating the electronic catalytic surface area (ECSA) were tested by CV curves at different scan rates (10,20,30,40, and 50 mV·s −1 ) in an electrochemical window of 1.175-1.215 V versus RHE. The experimental conditions for electrochemical impedance spectroscopy (EIS) was: Initial voltage (V) = 0, High frequency (Hz) = 100,0000, Low frequency (Hz) = 0.01, Amplitude (V) = 0.005, Quiet Time (s) = 2. The reference potential was calibrated to the reversible hydrogen electrode (RHE) based on the Nernst equation [49]: The turnover frequency (TOF) was calculated from the equation [50]: where J is the current density (A·cm −2 ) at a given overpotential (η = 300 mV), A is the surface area of the electrode, F is the Faraday constant (96,485 C·mol −1 ), and m is the number of moles of metal on the electrode. The mass activity (J m , A·mg −1 ) was calculated from the active mass deposited on the electrode surface (m, g) and the measured current I (A), as the following equation [51]:

Theoretical Calculations
Spin-polarized density functional theory (DFT) calculations were performed by the CASTEP module in the Materials Studio 8.0 package. Periodic geometry and cell optimization of NiFO x (OH) y @NCA were first performed, followed by electronic property calculations and analysis. The plane wave basis with an energy cutoff of 400 eV and ultrasoft pseudopotential was performed during all the calculations, and the exchangecorrelation energy was described by the generalized gradient approximation of Perdew, Burke, and Ernzerhof (GGA-PBE) functional. To evaluate the on-site Coulomb interactions in the 3D states of NiFeO x (OH) y @@NCA hybrid, the DFT + U approach with the Hubbard parameter U = 6.45 eV, and U = 5.30 eV for Ni and Fe atom in NiFeO x (OH) y structure. A dispersion-corrected semi-empirical TS scheme was employed to further characterize the interaction between the two layers. To simulate the structure of NiFeO x (OH) y @NCA, the (110) plane of NiOOH was selected, in which one of the eight atoms was replaced by a Fe atom, with a mismatch rate of less than 5%. The Brillouin zone was sampled through a 4×4×1 uniform k-point grid for geometric optimization and electronic structures calculations. The model structures were optimized with a total energy threshold of 10-5 eV/atom, a maximum force of 0.03 eV/Å, a maximum stress of 0.05 GPa, and a maximum displacement of 0.001 Å, respectively.