Long-Term Stability of Ni–Sn–Fe-Based Coatings Prepared by Electrodeposition for the Oxygen Evolution Reaction

A coating comprising Ni, Sn, and Fe was deposited on a Ni mesh by cathodic polarization from a bath containing Ni2+, Sn2+, and Fe3+ salts in solution, and the oxygen evolution reaction properties of the thus obtained electrode were investigated. The lattice volume of the Ni–Sn–Fe-based coating was estimated to be 77.1 Å3, a much larger value than that (44.0 Å3) of the Ni–Fe-based coating obtained in the absence of Sn2+. The Ni–Sn–Fe-coated electrode manufactured by cathodic polarization at a current density of −120 mA cm−2 afforded a current density of 10 mA cm−2 at an overpotential of 276 mV in 1 M KOH. The said electrode’s Tafel slope was estimated to be 37 mV dec−1. When electrolyzed at constant current density (+50 mA cm−2 to +800 mA cm−2), it exhibited a stable potential for at least 162 h. The Ni–Sn–Fe-coated electrode was also used as the anode in a two-electrode cell (80 °C; 30 wt% KOH) and electrolyzed for 3 d. Evidence indicated a low cell voltage of 1.81 V at a current density of +600 mA cm−2, which is at an industrial level.

Hydrogen is attracting attention as a clean energy carrier, amid concerns over the depletion of fossil fuels.Green hydrogen can be produced by the electrolysis of water using renewable energy as a power source.Hydrogen can be converted to energy by internal combustion engines and fuel cells in the absence of CO 2 emissions.Efforts are being made globally to produce green hydrogen at low cost so as to realize a hydrogen-based society.In order to reduce the cost of hydrogen production, the performance of electrocatalysts needs to be improved.In water electrolysis, an oxygen evolution reaction (hereafter referred to as OER) with four-electron-fourproton transfer takes place at the anode, leading to a large overpotential.Ruthenium-based and iridium-based oxides are regarded as the best OER catalysts. 1However, the said metals are rare, expensive, and less durable.Therefore, transition metal-based catalysts have attracted attention as alternatives to precious metalbased catalysts.3][4][5] Recently, the results of ab initio density functional theory calculations have pointed to the reason that Ni and Fe complexation leads to improvements in OER catalytic activity. 6However, most of these transition metal-based catalysts are synthesized by hydrothermal methods, and synthesizing them requires multiple steps.Furthermore, the thus produced powder samples have to be supported on current collectors together with a polymer binder like Nafion.This process is not suitable for industrial-scale electrolysis.Therefore, a need exists for a technology affording the synthesis of Ni and Fe composite catalysts by electrodeposition.2][13] Wu et al. reported the OER performance of an electrode obtained via the deposition of a Ni-Sn-Fe-based coating on Ni mesh from a bath containing Ni 2+ , Sn 2+ , and Fe 3+ salts. 14In the study by Wu and co-workers, the Ni-Sn-Fe-coated electrode manufactured via cathodic polarization at a current density of −20 mA cm −2 flowed a current density of 10 mA cm −2 due to OER at an overpotential of 253 mV.However, the effect of the value of the current density applied during the electrodeposition process was not investigated.Consequently, the advantages of the electrodeposition method were not exploited.In the present study were investigated the effects on the obtained electrodes' OER properties (including stability) of the composition of the bath containing the Ni, Sn, and Fe precursor ions as well as the experimental parameters of the electrodeposition.Moreover, the effects of electrolyte temperature, concentration, and electrolysis conditions on the stability of the obtained catalysts were also explored.

Experimental
Reagents.-Allchemicals were of reagent grade, and they were sourced from Wako Pure Chemicals.They were used without further purification.All solutions were prepared with ultrapure water (18.2MΩ cm) purified with Advantec RFU464TA.
Catalyst synthesis.-Beforeuse, the working electrode, a 30 mesh Ni mesh, was electrolytically degreased in an alkaline solution and acid-washed.The bath for preparing a Ni-Sn-Fe-based coating (denoted as sample a in Fig. 2) consisted of 0.06 M NiCl 2 , 0.02 M SnCl 2 , 0.02 M FeCl 3 , 0.5 M K 2 P 2 O 7 , and 0.1 M C 2 H 5 NO 2 in water, the pH of which was 8.5.The above-treated Ni mesh (3 × 3 cm 2 ) was immersed in the said bath and polarized at a constant current density of −120 mA cm −2 until an electrical charge of 48 C cm −2 was passed.In the latter part, the current density was varied from −10 to −200 mA cm −2 to optimize the catalyst with respect to the OER activity.The bath for the Ni-Sn-coated electrode (denoted as sample b in Fig. 2) consisted of 0.06 M NiCl 2 , 0.02 M SnCl 2 , 0.5 M K 2 P 2 O 7 , and 0.1 M C 2 H 5 NO 2 in water (pH 8.5).The bath for the Ni-Fe-coated electrode (denoted as sample b in Fig. 2) consisted of 0.06 M NiCl 2 , 0.02 M FeCl 3 , 0.5 M K 2 P 2 O 7 , and 0.1 M C 2 H 5 NO 2 in water (pH 8.5).Conditions for electrodeposition were the same as those for the Ni-Sn-Fe-based coating.
ECS Advances, 2023 2 040504 scan speed were applied in the 25°-80°2θ region.Sample surface observations were made by scanning electron microscopy (SEM) using a JEOL JSM-7000F microscope.Component analysis was carried out using a JED-2300F attached to the JEOL JSM-7000F microscope.X-ray photoelectron spectra (XPS) were obtained using a Thermo Scientific K-Alpha spectrometer with a monochromatic Al Kα source (1486.6 eV).A pass energy of 50 eV and channel widths of 1.0 and 0.1 eV were adopted to collect wide-and narrow-range spectra, respectively.XPS fitting was made using CasaXPS software by setting the adventitious carbon peak to 284.8 eV.All spectra were deconvoluted using a Gaussian-Lorentzian line shape and a Shirley background.
Electrochemical measurements.-Allelectrochemical measurements were performed using a standard three-electrode cell connected to a Bio-Logic SP-300 potentiostat.A Ni-mesh-supported catalyst was used as the working electrode; a Hg/HgO electrode (filled with 1 M KOH) and a Pt coil were used as the reference and counter electrodes, respectively.In order to evaluate the OER performance, linear sweep voltammetry (LSV) experiments were conducted at room temperature in 1.0 M KOH solution (pH = 14).Potentials were recorded with 60% iR compensation.All potentials measured vs the Hg/HgO electrode (E(Hg/HgO)) were converted to reversible hydrogen electrode (RHE) standards (E(RHE)) using the following equation: Hg HgO 0.059 pH 0.098 1 Prior to each measurement, the samples were conditioned by a potential scan (0.05-1.2 V vs RHE) at a sweep rate of 150 mV s −1 (200 cycles).After confirming that the current values were stable, the OER performance was evaluated.The overpotential (ƞ) was equal to E(RHE) −1.23 V. Tafel plots were extracted from the rising part of the LSV. 15System durability was evaluated by the potential cycling and galvanostatic methods.
Performance evaluation using a two-electrode cell.-Electrolytictests were also carried out using a two-electrode cell setup under conditions simulating industrial usage.The appearance of the whole system (a) and the internal configuration of the twoelectrode cell (b) are depicted in Fig. 1.Electrolysis was carried out for 3 d at 80 °C in a 30 wt% KOH solution.The current density was controlled by Kikusui Electronic Corp. PAS10-35.Specifically, the current density was made to increase in 100 mA cm −2 steps, up to a final value of +600 mA cm −2 , while the voltage between the two electrodes was measured.Notably, the cell voltage was the value displayed when a digital multimeter was pressed against the current collector, and the voltage value was equal to the average measured over 24 min at each current density during 3 d of continuous operation.

Results and Discussion
Structural characterization of the catalysts.-InFig. 2 16 We recorded CV curves without the Faraday processes in the 0-0.1 V potential range (vs the Hg/HgO electrode) at sweep rates ranging from 2 to 10 mV s −1 (Fig. S1).There, the difference in current density (Δj = (j anodic − j cathodic )/2) at 0.  4a, which is reported in Fig. 4b, a current resulting from the oxidation of Ni species was detected around 1.45 V (vs RHE). 17herefore, LSV experiments were conducted at a sweep rate of 0.01 mV s −1 to minimize the effects of the faradaic oxidation reactions of Ni species and capacitive currents (see data in Fig. 4c).Indeed, in Fig. 4c, the peaks associated with the oxidation of Ni species are not detected.For all samples, the overpotential required to reach a current density of 10 mA cm −2 for the OER was similar to that obtained at a sweep rate of 1 mV s −1 .In Fig. 4d are reported the Tafel plots of the LSV curves in Fig. 4c.The Tafel slope for Ni-Sn-Fe had a value of 37 mV dec −1 , which corresponds to the maximum reaction rate.The OER properties reported so far for the Ni-Sn and Ni-Fe catalysts are summarized in Table II. 18Results indicate that the overpotentials for the OER activity of the Ni-Sn-Fe catalyst synthesized in this study are lower than those of the Ni-Sn catalyst.The Tafel slope was comparable to that of the Ni-Fe catalyst.
The catalytic activities of different materials should be compared based on the turnover frequency (TOF), as this parameter reflects the specific properties of the materials.The TOF is calculated based on the Faraday efficiency of the OER as 100%: TOF = (j × A) where j is the current density at a given overpotential (A cm −2 ), A is the geometric area of the electrode (0.25 cm 2 ), 4 is the number of electrons transferred in the OER reaction, F is the Faraday constant (96,485 C), and m is the total number of moles of Ni and Fe in the catalyst (all transition metals in the catalyst are assumed to contribute to the reaction).It is assumed that the surface of the Ni mesh substrate is completely covered by the Ni-Sn-Fe, Ni-Sn, and Ni-Fe coatings, and does not participate in the OER process.The TOF of each catalyst was calculated based on the LSV curves reported in Fig. 4c and plotted against the overpotential (see Fig. 5).
The TOF value determined from the current density obtained at ƞ = 300 mV increased in the following order: Ni-Sn (0.03 s −1 ) < Ni-Fe (0.04 s −1 ) < Ni-Sn-Fe (0.24 s −1 ).In Fig. S2 are reported the LSV curves (a) and relevant Tafel plots (b) after 1,000 CV cycles.The sweep rate was 0.01 mV s −1 .The OER characteristics determined for the Ni-Sn-Fe, Ni-Sn, and Ni-Fe catalysts are listed in Table III, together with the results obtained before the CV experiments: no significant changes were observed in the overpotential and Tafel slope values required to reach a current density of 10 mA cm −2 , before, and after CV.These data are indicative of high catalyst durability.However, in the case of the unmodified Ni mesh, the overpotential required to generate a current density of 10 mA cm −2 increased from 317 mV before any CV cycles to 352 mV after 1,000 CV cycles.For this sample, the Tafel slope value also increased from 46 to 59 mV dec −1 (before and after the CV cycles, respectively).When the crystallite size was calculated using the Scherrer equation, the value for this parameter was found to decrease as the applied current density increased (Table IV).The ECSA values of the Ni-Sn-Fe samples manufactured by cathodic polarization were calculated at various current densities.The CV curves without the Faraday processes and the plot of the applied current density vs the scan rate are reported in Figs.S3 and S4a, respectively.Evidence indicates that the ECSA increased as current density increased (Fig. S4b).The mentioned trend can be easily ascribed to the decrease in crystallite size, as also indicated by the surface morphology of Ni-Sn-Fe samples in Fig. 7.In fact, the data therein indicate that the spherical morphology of the crystallites becomes evident, when the applied current density has a more negative value than −80 mA cm −2 .LSV measurements were performed at a scan rate of 0.01 mV s −1 .The values for the overpotential required to reach a current density of 10 mA cm −2 , from the LSV (a) and the corresponding Tafel plot (b) in Fig. S5, and the Tafel slope value vs the current density during electrodeposition are reported in the bar graph in Fig. 8a.The obtained LSV data indicate that the overpotential and the Tafel slope decreased as the applied current density increased and that performing the cathodic polarization at more negative values for the current density than −120 mA cm −2 resulted in a substantially reduced value for the overpotential.Furthermore, the values of the overpotential affording a current density of 10 mA cm −2 were calculated from the LSV (Fig. S6) which were normalized by the corresponding ECSA values.As a result, the overpotential value was found to decrease with cathodic polarization above a current density of −120 mA cm −2 even after the ECSA normalization (Fig. 8a).These values were then utilized to draw a plot of TOF (Fig. S7).The TOF calculated from the current density at ƞ = 300 mV at each current density are shown in Fig. 8b.These data indicate that the TOF values increased substantially when the cathodic polarization was conducted at values for the current density of −120 mA cm −2 or more negative.Evidence thus suggests that the enhanced OER activity of Ni-Sn-Fe is not due simply to the effect of the increase in surface area, but also to the crystallinity of the catalyst and/or the optimal electronic structure due to the composition (Ni, Sn, Fe) ratio.

Effect of the current density on
Stability tests.-InFig. 9a are reported the potential profiles of Ni-Sn-Fe, Ni-Sn, Ni-Fe, and untreated Ni mesh polarized at a constant current.The retention time at each current density was 18 h.The potentials of Ni-Sn and Ni-Fe increased after 40 h and 90 h, respectively, and their final values were similar to that of the untreated Ni mesh.This observation means that the catalyst has dropped out.On the other hand, in the case of Ni-Sn-Fe, the potential shifted to positive with the applied current density, but no catalyst dropout occurred for at least 162 h.This indicates that the coating prepared as described in the present study renders the Ni-Sn-Fe catalyst more stable than the other two and the previously reported the Ni-Sn and Ni-Fe catalysts (Table II).The XRD pattern (Fig. 9b) and SEM image (Fig. 9c) of the Ni-Sn-Fe after the stability test are shown in Fig. 9a.As shown in Fig. 9b, the XRD pattern due to the hexagonal Ni 3 Sn 2 system before electrolysis was also observed after electrolysis, although its intensity became smaller.As can be evinced from the surface SEM image (Fig. 9c), the spherical morphology was retained after electrolysis.When the Ni-Sn-Fe before electrolysis was subjected to XPS measurements, a peak appeared at 852.7 eV in the Ni 2p 3/2 region (Fig. 9d).It was attributed to the zero-valent nickel (Ni 0 ), 19 which is in good agreement with the Ni 3 Sn 2 crystalline structure.After electrolysis, the main peak shifted to higher binding energy (855.1 eV), corresponding to the higher valence state of Ni, assignable to Ni 3+ . 20This strongly suggests that the surface of Ni-Sn-Fe coating was oxidized and acted as the OER catalyst during electrolysis, which can be reflected in the decrease in the diffraction intensity (Fig. 9b).In other words, the bulk of the catalyst remained crystalline Ni 3 Sn 2 .The stability observed for Ni-Sn-Fe can be explained by the ease of bubble release, in addition to its activity as described above.As seen from Fig. 3, Ni-Sn-Fe has a much more porous morphology than Ni-Fe.
Performance of a two-electrode cell comprising a Ni-Sn-Fe anode.-InFig. 10 are reported the performance evaluation results of the two-electrode cells constructed using Ni-Sn-Fe and bare Ni as anodes.Notably, in both cases, the cathode was Ni-Sn.][13] The cell whereby the anode consisted of the Ni-Sn-Fe sample afforded lower voltages at all current densities than the cell whereby the anode consisted of bare Ni mesh.Indeed, the voltage measured for the two-electrode cell comprising Ni-Sn-Fe was 1.81 V at a current density of +600 mA cm −2 .by cathodic polarization at a current density of −120 mA cm −2 was 37 mV dec −1 , while the overpotential was 276 mV required for generating a current density of 10 mA cm −2 .During electrolytic tests conducted at current densities ranging from +50 mA cm −2 to +800 mA cm −2 , the potential was observed to be stable for at least 162 h.This stability level is higher than those previously reported for Ni and Fe composite oxide catalysts synthesized implementing the electrodeposition method.Furthermore, during water electrolysis, a two-electrode cell with Ni-Sn-Fe as the anode and Ni-Sn as the cathode exhibited a cell voltage value of 1.81 V at a current density of +600 mA cm −2 in 30 wt% KOH solution at 80 °C.

Figure 1 .
Figure 1.Schematic representation of the utilized two-electrode system (a) and photo of the external configuration of the said system (b).

c
= 5.18 Å, and V = 75.5 Å 3 ).On the other hand, sample c can be assumed to consist of FeNi 3 (ICSD No. 01-077-7971 a = 3.55 Å, b = 3.55 Å, c = 3.55 Å, and V = 44.7 Å 3 ).The lattice volumes of samples a and b were estimated to be 77.1 and 77.0 Å 3 , respectively.These values are substantially larger than that of 44.0 Å 3 estimated for sample c.The crystallite size was calculated based on the full width at half maximum (fwhm in radian) of the (102) peak of samples a and b and the (111) peak of sample c, utilizing the Scherrer formula; i.e., crystallite size = 0.89λ/fwhm × cosθ, where λ is the X-ray wavelength (0.154051 nm) and θ is the Bragg angle.The crystallite size (4.56 nm) of sample a was slightly larger than that (4.21 nm) of sample b.These values are substantially smaller than the value of 10.22 nm estimated for sample c.Compositional analysis of the catalyst surfaces was carried out by conducting energy-dispersive X-ray spectroscopy experiments (TableI).The compositional ratios of Ni, Sn, and Fe in samples b and c were in close agreement with those of the precursor solution.However, the Fe content (3%) in sample a was much smaller than the concentration percentage (20%) of the precursor ions ([FeCl 3 ] × 100/([NiCl 2 ] + [SnCl 2 ] + [FeCl 3 ])) in the bath for electrodeposition.Notably, hereafter, samples a, b, and c are referred to as Ni-Sn-Fe, Ni-Sn, and Ni-Fe, respectively.In Fig. 3 are reported scanning electron microscopy (SEM) images of the surfaces of Ni-Sn-Fe, Ni-Sn, and Ni-Fe.Evidence indicates that Ni-Sn-Fe and Ni-Sn are composed of spherical particles smaller than 10 μm in diameter.On the other hand, Ni-Fe has a smoother appearance than Ni-Sn-Fe and Ni-Sn.The electrochemically active surface area (ECSA) of each catalyst was calculated based on the relevant estimated electrochemical double layer capacity (C dl ) in N 2 -saturated 1 M KOH acting as the electrolyte.
05 V was plotted as a function of scan rate; based on the equation according to which ECSA = C dl /C s , wherein C s is the specific capacitance of atomically smooth surfaces, the ECSA values for Ni-Sn-Fe, Ni-Sn, and Ni-Fe were estimated to be 1.7, 1.7, and 0.3 m 2 g −1 , respectively.OER performance of the catalysts.-InFig. 4a are reported the LSV curves obtained for Ni-Sn-Fe, Ni-Sn, Ni-Fe, and bare Ni electrodes.Measurements were made in O 2 -saturated 1.0 M KOH solution at a sweep rate of 1 mV s −1 .The iR drop between the working and reference electrodes was corrected.The lowest value, 276 mV, was obtained for Ni-Sn-Fe.Based on the blow-up of the data in Fig.
Ni-Sn-Fe deposition on Ni mesh.-TheNi-Sn-Fe samples were prepared by conducting cathodic polarization experiments at different current densities, from −10 to −200 mA cm −2 , in a situation whereby the delivered

Figure 4 .
Figure 4. Linear sweep voltammograms of Ni mesh electrodes modified with Ni-Sn-Fe, Ni-Sn, and Ni-Fe coatings recorded in 1.0 M KOH solution at scan rates of (a and b) 1 mV s −1 and (c) 0.01 mV s −1 .(d) Tafel plots of the linear sweep voltammograms in Fig. 4c.Results obtained for the bare Ni mesh electrode were included.

Figure 5 .
Figure 5. Plots of the turnover frequency of the oxygen evolution reaction as a function of the overpotential afforded by Ni mesh electrodes modified with Ni-Sn-Fe, Ni-Sn, and Ni-Fe coatings in the case of experiments conducted in 1.0 M KOH solution at a scan rate of 0.01 mV s −1 .

Figure 6 .
Figure 6.(a, b) X-ray diffraction patterns of the Ni-Sn-Fe samples obtained by applying the different noted current densities.(c) Compositions of the said samples plotted against the applied current density.

Figure 7 .
Figure 7. SEM images of the Ni-Sn-Fe samples obtained by applying the noted current densities.

Figure 8 .
Figure 8.(a) Overpotential geometric required to reach a current density of 10 mA cm −2 (obtained from Fig. S5a), overpotential normalized to ECSA required to reach a current density of 10 mA cm −2 (obtained from Fig. S6), and Tafel slope (obtained from Fig. S5b) of the Ni-Sn-Fe samples obtained imposing the different noted applied current densities.(b) TOF of the Ni-Sn-Fe samples obtained imposing the different noted applied current densities.

Figure 9 .
Figure 9. (a) Potential profiles of bare Ni mesh and Ni mesh electrodes modified with Ni-Sn-Fe, Ni-Sn, and Ni-Fe when polarized at the noted current densities (red).(b) XRD patterns of the electrode modified with Ni-Sn-Fe before and after the stability test in Fig. 9a.(c) SEM image of the electrode modified with Ni-Sn-Fe after the stability test.(d) XPS spectra of the same samples as in Fig. 9b.

Figure 10 .
Figure 10.The cell voltage of untreated Ni mesh and the Ni mesh electrode modified with a Ni-Sn-Fe at each current density when used as the anode of a two-electrode cell.

Table III .
OER properties of various electrodes measured before and after 1,000 CV cycles.

Table IV .
Crystal data on the Ni-Sn-Fe samples obtained at different values for the current density.