In situ Hydrothermal Oxidation of Ternary FeCoNi Alloy Electrode for Overall Water Splitting

Exploring noble metal‐free catalyst materials for high efficient electrochemical water splitting to produce hydrogen is strongly desired for renewable energy development. In this article, a novel bifunctional catalytic electrode of insitu‐grown type for alkaline water splitting based on FeCoNi alloy substrate has been successfully prepared via a facile one‐step hydrothermal oxidation route in an alkaline hydrogen peroxide medium. It shows that the matrix alloy with the atom ratio 4:3:3 of Fe:Co:Ni can obtain the best catalytic performance when hydrothermally treated at 180°C for 18 h in the solution containing 1.8 M hydrogen peroxide and 3.6 M sodium hydroxide. The as‐prepared Fe0.4Co0.3Ni0.3‐1.8 electrode exhibits small overpotentials of only 184 and 175 mV at electrolysis current density of 10 mA cm−2 for alkaline OER and HER processes, respectively. The overall water splitting at electrolysis current density of 10 mA cm−2 can be stably delivered at a low cell voltage of 1.62 V. These characteristics including the large specific surface area, the high surface nickel content, the abundant catalyst species, the balanced distribution between bivalent and trivalent metal ions, and the strong binding of in‐situ naturally growed catalytic layer to matrix are responsible for the prominent catalytic performance of the Fe0.4Co0.3Ni0.3‐1.8 electrode, which can act as a possible replacement for expensive noble metal‐based materials.


Introduction
Hydrogen is an important raw material for chemical synthesis and is widely used in metal smelting, welding, and other fields. [1]In addition, its application as a new energy source has received wide attention in recent years. [2]However, currently, the most widely used technologies for industrial hydrogen production, such as partial oxidation of heavy hydrocarbons, steam reforming, and coal gasification using fossil fuels usually require large energy consumption owing to the harsh reaction conditions including high temperature and high pressure.4] Therefore, it is urgent to develop new low cost and environmentally friendly hydrogen producing technologies, among which, electrolytic hydrogen production, as a kind of expedient and CO 2 -free technology, has attracted extensive attention. [5]he challenge of enabling electrolytic water splitting to prepare hydrogen on a large industrial scale is to develop cheap, highly active, and stable non-noble metal catalytic electrode materials, which can be alternatives to noble catalysts represented by Pt for cathodic hydrogen evolution reaction (HER), and IrO 2 /RuO 2 for anodic oxygen evolution reaction (OER), [6][7][8][9][10] so as to not only improve the kinetic rate of water decomposition but also obtain high-efficiency hydrogen production by hydrolysis at reasonable costs.
Various nonprecious metal compounds including oxides, [11,12] hydroxides, [13,14] sulfides, [15,16] selenides, [17,18] nitrides, [19,20] carbides, [21,22] phosphides, [23,24] and borides [25,26] of transition metal elements have been studied as catalysts for HER or/and OER in alkaline solution, which were often loaded on conductive substrates such as carbon or metal materials.For this kind of loading-type catalytic electrode, sharp changes in microstructure and composition at the substrate/catalytic layer interface usually lead to weak interfacial binding.The corresponding negative effects are increased electron transmission resistance and even interface debinding which slow down the electrochemical reaction.When the catalytic electrode is used for water decomposition, the escaping of gas from the electrode/electrolyte interface will accelerate the debinding of deposited catalytic layers.Usually, the hydrogen evolution in the cathode region is more violent than oxygen evolution in the anode region under the same current density because the number of electrons involved in generating one mole of hydrogen is half that involved in generating one mole of oxygen.That is to say, the stability of the loading-type hydrogen evolution electrode faces greater challenges.This should be an important reason why the number of reported studies on HER catalytic materials is much smaller than that on OER catalytic materials.
Therefore, creating a natural interface transition between the carrier and the catalytic layer has become an important breakthrough point for improving the catalytic performance and service life of the catalytic electrode.Preparing a catalytic layer integrated with the substrate by in-situ growth process involving the reaction of matrix elements can overcome the performance bottleneck caused by the interface problem of loadingtype catalytic electrode.For example, X20CoCrWMo10-9/Co 3 O 4 composite fabricated via a straightforward in-situ electro-oxidation of a Exploring noble metal-free catalyst materials for high efficient electrochemical water splitting to produce hydrogen is strongly desired for renewable energy development.In this article, a novel bifunctional catalytic electrode of insitu-grown type for alkaline water splitting based on FeCoNi alloy substrate has been successfully prepared via a facile one-step hydrothermal oxidation route in an alkaline hydrogen peroxide medium.It shows that the matrix alloy with the atom ratio 4:3:3 of Fe:Co:Ni can obtain the best catalytic performance when hydrothermally treated at 180°C for 18 h in the solution containing 1.8 M hydrogen peroxide and 3.6 M sodium hydroxide.The as-prepared Fe 0.4 Co 0.3 Ni 0.3 -1.8 electrode exhibits small overpotentials of only 184 and 175 mV at electrolysis current density of 10 mA cm À2 for alkaline OER and HER processes, respectively.The overall water splitting at electrolysis current density of 10 mA cm À2 can be stably delivered at a low cell voltage of 1.62 V.These characteristics including the large specific surface area, the high surface nickel content, the abundant catalyst species, the balanced distribution between bivalent and trivalent metal ions, and the strong binding of in-situ naturally growed catalytic layer to matrix are responsible for the prominent catalytic performance of the Fe 0.4 Co 0.3 Ni 0.3 -1.8 electrode, which can act as a possible replacement for expensive noble metal-based materials.
Co-based tool steel in NaOH medium showed natural homogeneous transition at the substrate/catalytic layer interface due to the intrinsic formation of the Co-enriched outer zone and exhibited much superior electrocatalytic properties to the Fe/Ni-based single-phase Co 3 O 4 catalysts that present an island aggregation form with microcracks. [27]esides CoFe alloy, insitu oxidized electrodes based on some industrial Fe/NiFe alloy substrates including S235 mild steel, [28] 304/316 stainless steel, [29][30][31][32][33] Ni42 alloy, [34] etc. have also been investigated as efficient OER or/and HER catalysts.However, those used substrates were industrial alloys with a designated brand, and the composition ratio of the alloy elements was fixed, then the influence of the matrix alloy composition on the properties of the catalytic layer was not discussed in these studies.
[37][38] Also, attempts have been made to construct insitu oxidation catalytic electrodes based on FeCoNi ternary alloy. [39,40]Zhao et al. synthesized amorphous oxides by electrochemical anodization of FeCoNi alloy plate (mole ratio of Fe:Co:Ni = 1:1:1) followed by lowtemperature (100°C) annealing, which showed activity toward OER with an overpotential of 170 mV in 0.1 M KOH electrolytic. [39]Habazaki's group found that the optimum anodizing treatment for FeNiCo Kovar alloy (29.72 wt.% Ni and 16.28 wt.% Co) in a fluoridecontaining ethylene glycol electrolyte reduced the overpotential of OER to 245 mV at 10 mA cm À2 in 1.0 M KOH electrolyte. [40]However, the anodization processes in these two reports were both performed in organic solution with environmentally unfriendly fluoride ions; in addition, the results showed that the alloy element composition ratio had great influence on catalytic performance.Thus, to exploit more environmentally friendly synthetic methods and to further reveal the influence of alloy element composition ratio on electrochemical catalytic properties of the electrodes are both of great significances for developing FeNiCo alloy based in-situ grown catalysts.
Herein, in this work, we prepared a series of alloy substrates with different element composition ratios and attempted to synthesize insitu grown oxidation catalyst layers by one-step hydrothermal method in a simple solution containing only two substances sodium hydroxide and hydrogen peroxide.The influence of the amount of oxidative reagents used and of the alloying element ratio on the OER and HER catalytic properties of the catalytic electrode were studied.As the result, the optimized parameters for the hydrothermal oxidation route were determined, and a high-performance FeCoNi alloy based in-situgrown-type electrode with catalytic activities toward both HER and OER in alkaline media was obtained.The investigation puts forward some new understandings for the design of in-situ-grown-type composite catalysts.

Catalysis Properties for Water Electrolysis
The fabrication process of the catalytic electrode is schematically portrayed in Figure 1.The alloys with different molar ratios of Fe, Co, and Ni were prepared by melting methods, cut into plates, and then treated by hydrothermal oxidization in medium containing sodium hydroxide and hydrogen peroxide to obtain the catalytic electrodes.
First, the effect of oxidant concentrations on the catalytic properties of the in-situ oxidized Fe 0.4 Co 0.3 Ni 0.3 sample with the Fe:Co:Ni atomic ratio 4:3:3 was investigated.The molar ratio of sodium hydroxide to hydrogen peroxide was fixed at 2:1, and catalytic electrodes were synthesized by heating the alloy matrix at 180 °C for 18 h in four different concentration conditions: 2.4 M NaOH +1. Figure 2 shows the catalytic properties of the blank and hydrothermally treated samples.The values of overpotential to achieve a current density of 10 mA cm À2 (ƞ 10 ) and of Tafel slope according to OER (Figure 2b,c) and HER (Figure 2e,f) were determined by the iRcompensated LSV curves (Figure 2a,d).It demonstrates that the catalytic activity of the alloy is significantly improved by oxidation treatment, and the concentration of oxidant has an obvious influence on the electrochemical catalytic performance.The ƞ 10 and Tafel slope values measured in both OER and HER processes gradually decrease as the hydrogen peroxide concentration increases from 1.2 to 1.8 M, but when the hydrogen peroxide concentration continuously increases to 2.1 M, the ƞ 10 and Tafel slope do not decrease further and contrarily increase slightly.For OER process, the lowest ƞ 10 and Tafel slope values obtained at 1.8 M hydrogen peroxide are 184 mV and 48 mV dec À1 , respectively.And for HER, the best performance also corresponds to 1.8 M hydrogen peroxide, represented as ƞ 10 of 175 mV and Tafel  S2, Supporting Information), which is attributed to the electrode acting as a nonideal capacitor. [41,42]So the data were fitted by the Randles equivalent circuit based on a single time constant model (Figure S3, Supporting Information) for both OER and HER processes. [43]It is validated that the Fe 0.4 Co 0.3 Ni 0.3 -1.8 sample possesses the lowest charge transfer resistances (R ct ) of 1.30 Ω cm 2 in OER process and 1.04 Ω cm 2 in HER process among all the samples (Figures S1c and S2c 3), the lowest R ct (Figures S8 and S9, Supporting Information), and the largest ECSA (Figure S10, Supporting Information).The result turns out that not the more nickel and cobalt elements are contained in the matrix, the better the performance of the finally prepared electrode.And the presence of iron with the appropriate content as 40 at.% in the alloy matrix is beneficial for achieving an optimized catalytic layer under the given reaction conditions.
Finally, the electrochemical stability of Fe 0.4 Co 0.3 Ni 0.3 -1.8 sample with the most outstanding performance during OER and HER processes was furtherly evaluated as shown in Figure 4.When the as-prepared Fe 0.4 Co 0.3 Ni 0.3 -1.8 electrode is used as the anode and the platinum electrode as the cathode to conduct the water electrolysis experiment at a current density of 10 mAÁcm 2 , the Fe 0.4 Co 0.3 Ni 0.3 -1.8 exhibits excellent stability with a negligible OER g 10 increase of 3.3 mV after 48 h testing (Figure 4a), and the overpotential value for achieving 50 mAÁcm À2 OER current density (g 50 ) evaluated by LSV curve increases only 17 mV (Figure 4b).When Fe 0.4 Co 0.3 Ni 0.3 -1.8 electrode is used as the cathode and a commercial IrO 2 /Ti electrode as the anode to conduct water electrolysis at 10 mA cm 2 for 48 h, HER g 10 value of Fe 0.4 Co 0.3 Ni 0.3 -1.8 increases only 6.3 mV (Figure 4c) and g 50 increases only 13 mV (Figure 4d).Finally, the water electrolysis experiment was conducted using simultaneously Fe 0.4 Co 0.3 Ni 0.3 -1.8 at the anode and cathode sides, the applied voltage to achieve 10 mA cm 2 Comparisons of overpotential and Tafel slope values between our optimized catalyst with some typical reported catalysts in literatures are listed in Tables S1 and S2, Supporting Information.It shows that the performance of the as-prepared Fe 0.4 Co 0.3 Ni 0.3 -1.8 electrode is among the advanced lists of FeCoNi ternary oxide/hydroxide catalysts (Table S1, Supporting Information) and of alloy-based in-situ-growntype catalytic electrodes (Table S2, Supporting Information).When specifically compared with the electrochemically grown catalytic layers from FeCoNi alloys in the fore-mentioned reports, [39,40] the method developed in this study outstands in terms of easy fabrication, more environmentally friendly solution composition, and high-performance bifunctionality toward both electrocatalytic HER and OER in alkaline solution.

Microstructure and Composition Characterization
The morphologies of these as-prepared catalytic electrodes were observed with SEM (scanning electron microscopy).The surfaces of oxidized Fe 0.4 Co 0.3 Ni 0.3 alloys obtained at different H 2 O 2 concentrations show significant changes in the micro-structure (Figure S11, Supporting Information).At lower H 2 O 2 concentrations of 1.2 and 1.5 M, the surface shows particles with a shape similar to the octaheon, and the grain size is in the range from about 200 nm to 1 lm, while H 2 O 2 concentration reaching 1.8 M, the surface micro-structure changes to a mixed pattern containing nanosheets and nanorods.Both the diameter of the nanorods and the thickness of the nanosheets are less than 100 nm, which helps to obtain a larger specific surface area.At 2.1 M H 2 O 2 concentration, this kind of surface nano coarse structure remains, but it seems that the density of the distributed nanostructures per unit area decreases in some degree.For the four electrodes prepared from substrates with different element composition ratios all at 1.8 M H 2 O 2 concentration, their surface morphologies also show significant differences (Figure S12, Supporting Information).It demonstrated that when the iron content in the alloy base reaches or exceeds 60%, that is, for samples Fe 0.6 Co 0.2 Ni 0.2 -1.8 and Fe 0.8 Co 0.1 Ni 0.1 -1.8, the surface does not appear morphology structure of nanosheet and nanorod, and the grains stick into a piece, resulting in remarkably reduced electrode specific surface area.The changes in surface structures reflected by SEM are consistent with the comparison results of ECSA investigations as shown in Figures S4 and S10, Supporting Information.This indicates that the surface micro-structure is an important factor affecting the electrode catalytic properties because of its impact on the ECSA value.
According to Winzer's law, [44,45] surface roughening will result in the enhancement of hydrophilicity of a hydrophilic surface.So another positive effect of the nanostructured morphology is to improve the adsorption properties and wettability of electrolyte on the electrode surface, which enables the generated gas to quickly leave the surface, and thus to maintain rapid electrolysis of water molecules.As shown in Figure 5, for Fe 0.4 Co 0.3 Ni 0.3 alloy, the oxidation treatment reduced the contact angle of water from 77.78°to 47.81°and of 1 M KOH from 71.88°to 33.28°.But it is worth mentioning that another important reason for the improved surface hydrophilicity is hydroxy groups generated on the surface of the catalytic layer, which will be proved by the subsequent X-ray photoelectron spectroscopy (XPS) results.The Energy Environ.Mater.2024, 7, e12590 increasing hydroxy groups lead to stronger hydrogen bonding interaction between the surface with water molecules, thus reflecting a stronger affinity for electrolyte.
In order to understand the growth mechanism of the catalytic layer, sectional analysis of the Fe 0.4 Co 0.3 Ni 0.3 -1.8 sample was carried out.It illustrates that the thickness of the generated oxide film is about 9 lm (Figure 6a).The element line scanning result in Figure 6b demonstrates that the oxide film can be divided into two layers including the bottom dense layer with thickness of about 5 lm as shown by AB section, and the surface loose layer with thickness of about 4 lm in BC section.This kind of micro-structure feature that the film density gradually decreases from inside to outside is advantageous for obtaining excellent catalytic performance.Firstly, the dense in-situ grown inner layer can strongly bind to the substrate, and on the other hand, the outside loose layer is conducive to gain large ECSA, thus, both the interface binding during the long-term operation and the electrochemical activity of the electrode are guaranteed.The signal strength of Fe, Co and Ni detected in the catalytic layer decreases against the alloy matrix because of the presence of the oxygen element in the oxidation film.The contents of Fe, Co and Ni remain nearly unchanged in the inner dense layer, but gradually decreased from the inside to the outside surface in the case of the outer loose layer, proving that the density of oxide film in the loose layer decreases from inside to outside.As can be seen from Figure 6b, while within the region from a depth of about 2 lm extending to the interface of the matrix/oxide film, the order of the element content in the film is roughly the same as that in the matrix: Fe > Co % Ni.However, in the region extending from the topmost layer to the depth of about 2 lm, the red line stands for the nickel content moves to the highest position, suggesting the topmost surface is enriched in nickle.Energy-dispersive X-ray (EDX) spectra were collected from three random regions on the sample surface (Figure 6d and Figure S13, Supporting Information), and the element percentage data (Table S3, Supporting Information) show much higher Ni (Avg.66.88 at.%) content than Fe (Avg.15.22 at.%) and Co (Avg.3.34 at.%), also agreeing with the nickel enrichment in the surface layer because the detection depth limit of EDX is generally less than   Fe 0.4 Co 0.3 Ni 0.3 -1.8 exhibits diffraction peaks of the largest number, indicating that the corresponding catalytic species on the surface are the most abundant, which is beneficial to building a better catalytic performance.While compared among the four groups of electrodes with different matrix composition as shown in Figure 5b, it is visible that Fe 0.4 Co 0.3 Ni 0.3 -1.8 electrode presents a greater number of diffraction peaks than Fe 0.2 Co 0.4 Ni 0.4 -1.8 electrode.When the iron content increases to 60%, that is, for Fe 0.6 Co 0.2 Ni 0.2 -1.8, the number of diffraction peak significantly reduced, suggesting the crystallization degree of the catalytic material in the membrane is greatly reduced.And as the iron content continually increases to 80% (Fe 0.8 Co 0.1 Ni 0.1 -1.8), much more compounds of iron were generated, leading the significantly strength increasing of diffraction associated with iron compounds.On the other hand, from comparison between new diffraction peaks generated by oxidation treatment of the same sample, it can be judged that the difference in signal intensity between different diffraction peaks for Fe 0.4 Co 0.3 Ni 0.3 -1.8 is smaller than other samples, implying the content distribution of catalytic species on the Fe 0.4 Co 0.3 Ni 0.3 -1.8 surface is more uniform.Thus, the XRD results summarily demonstrate that the existence of multivalent states and multiple species has a positive effect on improving the catalytic efficiency.
The electrochemical reaction is carried out at the electrode/electrolyte interface; therefore, it is particularly important to study the most superficial chemical composition of the catalytic layer by XPS for understanding the catalytic mechanism.The as-prepared electrode surfaces were tested by XPS, and the atomic percentages of Fe, Co, and Ni among the total metal amount were counted.As revealed in Figure 8a, for the oxidized Fe 0.4 Co 0.3 Ni 0.3 samples, the surface content of the elements followed the same order: Ni > Fe > Co under any H 2 O 2 concentration condition, which is consistent with the inference about nickel enrichment in outerside surface layer drawn from SEM analysis.The content ratio of Fe:Co:Ni varies little at H 2 O 2 concentration of 1.2, 1.5, and 1.8 M.But when increasing from 1.8 M to 2.1 M H 2 O 2 , the surface Ni content decreases form 80 to 71 at.%, and Fe increases from 12 to 19 at.%, with Co still changing not obviously (from 8% to 10%).So, when combining SEM and XPS results with the influence law of H 2 O 2 concentration on the catalytic performance, it can be inferred that microstructure and enrichment degree of nickel element on the surface are two important factors affecting the catalytic properties of the electrode.At 2.1 M H 2 O 2 concentration, the electrode ECSA is smaller (Figure S4, Supporting Information) and the surface nickel content is lower (Figure 8a) than at 1.8 M H 2 O 2 concentration, so the catalytic performance of Fe 0.4 Co 0.3 Ni 0.3 -2.1 is inferior to that of Fe 0.4 Co 0.3 Ni 0.3 -1.8.When the H 2 O 2 concentration is much lower as 1.2 M and 1.5 M, although the surface Ni contents are close to that at 1.8 M, the surface larger particle size leads to significantly reduced ECSAs (Figure S4, Supporting Information), and finally, as the comprehensive result, their catalytic performance was further reduced.As shown in Figure 8b, comparison among samples prepared at the same condition from different alloy matrix indicates that the element composition ratio of the matrix also has a great influence on the superficial element ratio.It is clearly demonstrated that electrodes Fe 0.8 Co 0.1 Ni 0.1 -1.8 and Fe 0.6 Co 0.2 Ni 0.2 -1.8 show much higher iron contents on the top surface and relatively lower Ni content on the surface.However, for electrodes Fe 0.2 Co 0.4 Ni 0.4 and Fe 0.4 Co 0.3 Ni 0.3 , a significant surface enrichment of nickel occurs.This implies that when the content of iron in the matrix is more than or equal to three times that of cobalt/nickel, it is not conducive to obtain a significant nickel surface enrichment.Meanwhile, the surface cobalt amount still maintains the most moderate change among the three elements with changing elements ratio of the substrate.So, here, for the four electrodes with different matrix components, the difference in catalytic properties is still relative to varieties of nickel element content and micro-structure of the top surface.The electrode with higher Ni surface content and higher ECSA presents more superior catalytic performance.
In order to understand the change of valence states of elements in the composite catalytic layer, the narrow spectrum of Fe, Co, and Ni measured on the surface of oxidized electrode is deconvoluted using the Gaussian-fitting program (Figures S14 and S15, Supporting Information), and the ratio of M 2+ to M 3+ of each element in the top surface was statistically obtained as shown in Figure 8c,d.It can be distinguished from Figure 8c that with the increase of H 2 O 2 concentration, Fe 3+ /Fe 2+ ratio increases, whereas the Ni 3+ /Ni 2+ and Co 3+ /Co 2+ ratios decrease.This is related to the relative weaker stability of Fe 2+ , which is more easily to be oxidized to high valence states than Co 2+ and Ni 2+ .For Fe 2+ , Co 2+ , and Ni 2+ , the number of electrons needed to be lost to reach the half full stable structure of the outermost electron orbit (3d 5 ) Energy Environ.Mater.2024, 7, e12590 is 1, 2, and 3, respectively.From the perspective of ionization energy, it means that the ability to lose electrons is gradually weakened, and the order of reducibility is Fe 2+ > Co 2+ > Ni 2+ .Therefore, compared with Co 2+ and Ni 2+ , oxidation from divalent to trivalent is easier for Fe 2+ .As a result, the proportion of Fe 3+ /Fe 2+ increases with the rising of oxidant H 2 O 2 concentration.Similarly, the element composition ratio of the alloy matrix is also found to have an impact on the valence state of the surface elements, as shown in Figure 8d.It can be seen that when the iron content is equal to or less than 40%, that is, in the case of Fe 0.  Energy Environ.Mater.2024, 7, e12590 depths (Figure 8f) that O mainly exists in the form of OH À at the outermost layer, and O 2À while reaching the depth of 40 nm O 2À .This proves that in the alkaline-oxidizing medium, the oxidized metal elements are easily combined with hydroxyl groups to form hydroxyl compounds, which helps to enhance the surface hydrophilicity and accelerate the escape of the produced hydrogen or oxygen in electrolysis.During the process of film thickening, the inner layer is constantly dehydrated to become oxide.Thus, as the result, the main components of the outmost surface layer within about 40 nm thickness are hydroxides, and the film substances below 40 nm depth are mainly oxides.The signal intensity of all metal elements including iron, cobalt, and nickel gradually increases with the sputtering from outerside surface to the inside as shown in Figure 8g-i, which is consistent with the trend observed by SEM that the density of the top loose oxide layer gradually increases from the outside to the inside.It is worth noting that in the sputtering depth range from 0 to 100 nm, there are basically no Fe and Co elements in metal valence detected.The exception is that nickel on the outer surface is all in oxidation valence state, but as sputtering proceeds inward, the existence of metal valence nickel gradually appears to coexist with Ni 2+ and Ni 3+ .The phenomenon that only Ni 0 and not Co 0 and Fe 0 is detected can also be explained by the reducibility order Fe 2+ > Co 2+ > Ni 2+ .Because Fe 2+ has the relatively strong reducibility; thus, the existence of Ni 0 in the film is due to the reduction of Ni 2+ by Fe 2+ , which is thermodynamically possible. [46,47]

Conclusions
In summary, we have prepared an in-situ-grown-type ternary catalytic electrode by hydrothermal oxidation treatment of FeCoNi alloy matrix in an alkaline oxidation medium composed of sodium hydroxide and hydrogen peroxide for water electrolysis application.By exploring the influence of different oxidant concentration and different element composition ratio of the alloy matrix on the catalytic performance of the electrode, it is found that when alloy with Fe:Co:Ni atomic ratio of 4:3:3 is heat-treated at 180°C for 18 h in solution containing 1.8 M H 2 O 2 and 3.6 M NaOH, the as-prepared bifunctional catalytic electrode exhibits the optimal electrochemical catalytic performance for water splitting with small overpotentials of only 184 mV and 175 mV at electrolysis current density of 10 mA cm 2 for alkaline OER and HER, respectively.It exhibits a satisfied stability for at least 48 h.The overall water splitting at electrolysis current density of 10 mA cm 2 can be delivered at a low cell voltage 1.62 V. Based on the mechanism analyses, it can be concluded that the large ECSA, the high surface nickel content, the abundant catalyst species, the distribution equilibrium between M 2+ and M 3+ metal ions, and the natural strong binding of catalytic layer to matrix are contribution factors for the as-prepared Fe 0.4 Co 0.3 Ni 0.3 -1.8 electrode to achieve excellent catalytic performance.The designed in-situ-grown-type bifunctional water splitting electrocatalyst Fe 0.4 Co 0.3 Ni 0.3 -1.8 can act as a potential replacement for expensive noble metal-based materials.

Figure 1 .
Figure 1.Schematic illustration of the preparation process for the catalytic electrode.
, Supporting Information).To assess the ECSA (electrochemical active surface area) of the oxidized Fe 0.4 Co 0.3 Ni 0.3 catalytic electrode, the double-layer capacitance C dl was measured by the cyclic voltammogram (CV) curves recorded at scan rate from 10 to 120 mV S À1 in the non-Faradaic region within a potential window of 0.05-0.20 V vs. RHE.The Fe 0.4 Co 0.3 Ni 0.3 -1.8 sample consistently exhibits the maximum ECSA as shown in Figure S4, Supporting Information.So all the electrochemical characterization results manifest that for the Fe 0.4 Co 0.3 Ni 0.3 alloy matrix, the optimal reaction conditions are as follows: 180 °C, 18 h, and 3.6 M NaOH +1.8 M H 2 O 2 , which is conducive for obtaining the most significant catalytic activity of the electrode for both OER and HER processes.To further investigate the influence of alloy element composition ratio on the catalytic properties of the electrode, another three sets of alloys including Fe 0.2 Co 0.4 Ni 0.4 , Fe 0.6 Co 0.2 Ni 0.2 , and Fe 0.8 Co 0.1 Ni 0.1 in addition to Fe 0.4 Co 0.3 Ni 0.3 were prepared also by melting and then treated by hydrothermal reaction at the optimal reaction conditions (180 °C, 18 h, 3.6 M NaOH +1.8 M H 2 O 2 ).Their OER/HER catalytic performance is illustrated in Figure 3. Comparing to unoxidized alloy substrates (Figures S5 and S6, Supporting Information), the ƞ 10 and Tafel slope values of all the samples subjected to oxidation treatment (Figure 3) show a downward trend.However, among all alloys, Fe 0.4 Co 0.3 Ni 0.3 presents the largest difference between the ƞ 10 values measured before and after oxidation treatment (Figure S7, Supporting Information), and the as-prepared Fe 0.4 Co 0.3 Ni 0.3 -1.8 exhibits the most outstanding activity toward both OER and HER among the four oxidized electrodes, with possessing the lowest ƞ 10 , the smallest Tafel slopes (Figure

Figure 2 .
Figure 2. Effect of the reagent concentration in hydrothermal process on the catalytic performance of oxidized Fe 0.4 Co 0.3 Ni 0.3 electrode: a) Linear sweep voltammetry (LSV) curves, b) ƞ 10 , and c) Tafel slopes of catalytic electrode for oxygen evolution reaction (OER), and d) LSV curves, e) ƞ 10 , and f) Tafel slopes of catalytic electrode for HER.

Figure 3 .
Figure 3.Effect of the alloy element composition ratio on the catalytic performance of the prepared catalytic electrode: a) Linear sweep voltammetry (LSV) curves, b) ƞ 10 , and c) Tafel slopes of catalytic electrode for OER, and d) LSV curves, e) ƞ 10 and f) Tafel slopes of catalytic electrode for HER.

Figure 4 .
Figure 4. a) Chronpotentiometric curves of Fe 0.4 Co 0.3 Ni 0.3 -1.8 for oxygen evolution reaction (OER) at a constant current of 10 mAÁcm À2 ; b) OER LSV curves of Fe 0.4 Co 0.3 Ni 0.3 -1.8 before and after stability test; c) Chronpotentiometric curves of Fe 0.4 Co 0.3 Ni 0.3 -1.8 for hydrogen evolution reaction (HER) at a constant current current of 10 mAÁcm À2 ; d) HER LSV curves of Fe 0.4 Co 0.3 Ni 0.3 -1.8 before and after stability test; e) Linear sweep voltammetry (LSV) curves of the anode in different electrolytic cells for overall water splitting; f) Stability test of the Fe 0.4 Co 0.3 Ni 0.3 -1.8 | Fe 0.4 Co 0.3 Ni 0.3 -1.8 electrolyzer recorded at a current density of 10 mAÁcm À2 .

Figure 5 .
Figure 5.The contact angle measurements of water and 1 M KOH on the surfaces of a,c) bare Fe 0.4 Co 0.3 Ni 0.3 and b,d) oxidized Fe 0.4 Co 0.3 Ni 0.3 -1.8.

Figure 6 .
Figure 6.The morphologies and composition analysis for Fe 0.4 Co 0.3 Ni 0.3 -1.8 electrode: a) The cross-section micrographs of the oxide film; b) Line scanning Energy-dispersive X-ray (EDX) analysis of the cross section; c) Surface micrograph; d) EDX spectra measured on the surface; e-h) EDX surface elemental mapping results of O, Fe, Co, and Ni.

Figure 7 .
Figure 7. X-ray diffraction (XRD) patterns of catalytic electrodes prepared a) from Fe 0.4 Co 0.3 Ni 0.3 alloy substrate at different concentrations of hydrogen peroxide and b) from alloys with different element composition ratios at a concentration of 1.8 M hydrogen peroxide; c) The standard XRD patterns of oxides/hydroxides of iron, cobalt, and nickel.

2
Co 0.4 Ni 0.4 -1.8 and Fe 0.4 Co 0.3 Ni 0.3 -1.8 electrodes, all the metal elements including Fe, Co, and Ni exist in the coexistence of bivalent and trivalent states.When the iron content in the alloy matrix is equal to or exceeds 60%, namely for Fe 0.6 Co 0.2 Ni 0.2 -1.8 and Fe 0.8 Co 0.1 Ni 0.1 -1.8 electrodes, almost all the iron ions detected exist in the form of trivalent state, and all cobalt ions exist in bivalent state, meanwhile, the ratio of Ni 2+ /Ni 3+ increased.Based on the above observations, a conclusion can be drawn that higher H 2 O 2 concentration or excessive iron content in the matrix can lead to the preferential oxidation of Fe 2+ to Fe 3+ and then elevates the ratios of Ni 2+ /Ni 3+ and Co 2+ /Co 3+ .Obviously, the catalytic performance is superior when the three elements all exist in two valence states, and the ratio of M 2+ /M 3+ (M = Fe + Co + Ni) is at the middle level, such as for Fe 0.4 Co 0.3 Ni 0.3 -1.8 and Fe 0.2 Co 0.4 Ni 0.4 -1.8.That is to say, the mixed existence of multivalent states is beneficial to the improvement of catalytic performance, and this is coherent with the inference drawn from XRD results.For the purpose of getting details about the depth distribution of elements in the outermost surface of the oxide layer, XPS sputtering experiments within 100 nm depth were carried out for sample Fe 0.4 Co 0.3 Ni 0.3 -1.8.The calculated percentages of elements at different depths are shown in Figure 8e.It shows that nickel content decreases gradually from the outside to the inside; meanwhile, Fe content increases gradually and Co content varies very little, further confirming the nickel surface enrichment phenomenon observed by SEM.It can be seen from the O 1s XPS spectra colletced at different sputtering

Figure 8 .
Figure 8. X-ray photoelectron spectroscopy (XPS) analysis of catalytic electrodes a,c) prepared from Fe 0.4 Co 0.3 Ni 0.3 alloy substrate at different concentrations of hydrogen peroxide and b,d) prepared from alloys with different element composition ratios at concentration of 1.8 M hydrogen peroxide; e) The change of element contents with sputtering depth for sample Fe 0.4 Co 0.3 Ni 0.3 -1.8; f-i) The O, Fe, Co, and Ni XPS spectra recorded at different sputtering depths for sample Fe 0.4 Co 0.3 Ni 0.3 -1.8.