Solution-Processed Ni-Based Nanocomposite Electrocatalysts: An Approach to Highly E ﬃ cient Electrochemical Water Splitting

,


INTRODUCTION
−3 One of the most promising alternatives involves the use of renewable electricity (i.e., wind, solar, and hydropower) to power electrochemical conversion processes, which convert abundant molecules (e.g., water, carbon dioxide, and nitrogen) into higher-value products (e.g., hydrogen, hydrocarbons, oxygenates, and ammonia).In all these processes, water oxidation stands out as the preferred reaction to provide the protons and electrons needed for the target reduction reactions, although this process is considered a kinetic bottleneck, and consequently, the development of electrocatalytic materials that effectively oxidize water is essential for improving the efficiency of the overall electrochemical conversion process. 4urrently, the most efficient water oxidation catalysts (WOCs) are based on iridium and ruthenium oxides, IrO 2 and RuO 2 , which operate in acidic/neutral conditions and are the preferred choice for the state-of-the-art polymer electrolyte membrane (PEM) electrolyzers. 5,6However, the high cost and scarceness of these materials justify the intensive search for electrocatalytic materials based on earth-abundant elements and low-cost synthetic processes to find sustainable and costeffective alternatives.First-row transition metals oxides (Ni, Co, Fe, and Mn) afford a promising alternative to IrO 2 and RuO 2 as WOCs.Among them, Ni-based materials constitute one of the best choices due to their high electrocatalytic activity and stability under alkaline conditions, related to their high electrical conductivity and corrosion resistance. 7,8Indeed, large-scale commercial liquid alkaline (LA) electrolyzers preferentially use nickel-based anodes. 9,10Furthermore, the solution processing of oxide semiconductors is an exciting fabrication method to improve the efficiency and decrease costs by direct-writing and high-throughput roll-to-roll printing technologies such as inkjet printing and flexography.
Several strategies have been proposed for the synthesis of Ni-based anodes, such as pulsed laser deposition, 11 sputtering, 12 electrodeposition, 13,14 chemical vapor deposition, 15 and sol−gel method. 16Among them, low-temperature processed materials have shown the highest activity toward oxygen evolution reaction (OER).Berlinguette et al. reported a lowtemperature photochemical metal−organic deposition, leading to amorphous mixed metal oxide films based on NiO x for OER electrocatalysis. 17Liu et al. introduced a room-temperature solution multistep procedure to grow an amorphous nickel− iron alloy catalyst on Ni foam for water oxidation that outperforms its crystalline counterpart.The same authors reported a ball-milling method to synthesize a NiFe hydroxide catalyst, inducing lattice tensile strain, which enhanced the adsorption of oxygenated intermediates for water splitting. 18imilarly, Zhou et al. reported the synthesis of amorphous mesoporous Ni/Fe (oxy)hydroxide [(Ni,Fe)OOH] films on Ni foam by room-temperature mechanical stirring processing with an exceptional activity and durability toward OER. 19argent et al. developed a room-temperature synthesis to produce gelled Fe, Co, and Ni oxyhydroxide materials with high-valence transition-metal distribution (i.e., W, Mo, Nb, Ta, and Re), reporting a ∼17-fold mass activity enhancement compared to that for the OER catalysts widely employed in industrial water-splitting electrolyzers. 20More recently, Duan et al. reported non-stoichiometric NiO x nanocrystals with unusual activity and stability, stemming from the high density of defects combined with oxygen atoms to form NiOOH species. 21Shudo et al. also reported a 3D porous Ni/NiO x bifunctional oxygen electrocatalyst derived from freeze-dried Ni(OH) 2 with exceptional activity.The stacking of 2D sheets into 3D mass seemed to play a vital role behind this excellent bifunctionality of freeze-dried Ni/NiO x . 22Although these electrocatalysts offer excellent OER catalytic performance, most of the synthetic approaches rely on multistep procedures with severe limitations to be compatible with industrial fabrication methods for large-scale production.
Here, we report a low-temperature, straightforward, and fast synthetic method with an extraordinary ability to in situ generate homogeneous Ni-based electrocatalytic thin films from a precursor solution by spin-coating and compatible with different industrial printing techniques such as inkjet printing and slot-die coating. 23Remarkably, this in situ synthesis approach can be easily integrated into scale-up processes like roll-to-roll printing for large-area and cost-effective mass production.The low baking temperatures employed are particularly beneficial since these allow deposition onto flexible plastic substrates, which cannot be processed above 150 °C without compromising the excellent film adherence over different materials, resulting in homogeneous catalytic electrodes.The OER activity of the as-synthesized electrocatalysts was proportional to the density of structural/electronic defects, which increases at lower baking temperatures, together with Fe incorporation, leading to a competitive undoped NiO x catalyst with 358 mV of overpotential at 10 mA cm −2 and more than 60 h of continuous operation without significant losses.O, as a precursor of NiO x , MeOEtOH as a solvent, and MEA as a chelating and stabilizer agent.Before deposition, 1 cm 2 of nickel foil (99.6% purity, from US Solid, Ohio) was ultrasonicated in a 3 M HCl solution for 10 min, rinsed with 3:1 isopropanol/acetone, and dried with compressed air.Then, the Ni foil was exposed to a UV/ozone cleaning for 10 min (Ossila Ozone Cleaner).The typical NiO x precursor formulation consists of a solution of 0.9 M Ni(OAc) 2 and 0.9 M MEA in MeOEtOH.The precursor solution was aged in a dry bath block for 15 min at 60 °C.During this process, subsequent hydrolysis and polycondensation reactions of Ni(OAc) 2 in MeOEtOH led to the formation of a stable sol of Ni-oxo-acetate and/or Ni-hydroxo-acetate species of different sizes.As a result of the formation of a sol, the viscosity of the precursor solution increased.Here, the presence of MEA as a chelating ligand is critical to control the kinetics of the polycondensation and avoid the transition from sol to gel before the deposition step.Afterward, this solution was spincoated on the Ni foil (or FTO substrate) at 2000 rpm for 30 s and soft-baked at 50 °C for 30 min on a hot plate to remove MEA and MeOEtOH.Finally, the sample was baked in air at different temperatures from 50 to 500 °C for 30 min.The thickness of the NiO x film on the Ni foil and FTO substrate after spin-coating is identical for a given baking temperature, indicating that a similar amount of the electrocatalyst is coated on the electrode.The overall process for the synthesis of the electrocatalyst is shown in Supporting Information, Figure S1.

EXPERIMENTAL METHODS
2.3.Structural, Optical, and Electrical Characterization.Field emission scanning electron microscopy (SEM) performed with an S-4800 instrument from HITACHI (Tokyo, Japan) operating at 20 kV was used for the morphological characterization of the samples.Their crystalline structure was assessed by X-ray diffraction (XRD) collected on a Bruker D8 Advanced X-ray diffractometer with copper K α radiation (λ = 1.5418Å) operating at a grazing incidence of 1°, at a scan rate of 5°•min −1 for 2θ angles ranging from 20 to 80°.The chemical composition of the resulting electrocatalytic electrodes was investigated by confocal multi-spectral imaging (CMSI) Raman spectroscopy and X-ray photoelectron spectroscopy (XPS).The Raman measurements were performed with a WITec Apyron confocal microscope using a 532 nm laser with a 40 mW power, a grating of 1800 lines/mm, a spectrometer of 300 mm, and an optical objective Zeiss EC Epiplan-Neofluar Dic 100×/0.9.X-ray photoelectron spectroscopy (XPS) was performed in an ultrahigh vacuum system (ESCALAB210; base pressure 1.0 × 10 −10 mbar) from Thermo VG Scientific.Samples were mounted on a conducting sample holder and introduced into the vacuum chamber.The measurements were taken over an area of 1 mm 2 .High-resolution transmission electron microscopy (HR-TEM) images were obtained using the model FEI TALOS F200X from ThermoFisher Scientific that combines highresolution S/TEM and TEM imaging with industry-leading energydispersive X-ray spectroscopy (EDX) signal detection and 3D chemical characterization with compositional mapping.Photoluminescence measurements were performed in a Horiba Jobin Yvon Fluorolog system.The electrical conductivity (σ) was calculated with the Van der Pauw equation using the Keithley 2400 SourceMeter equipment to measure the electrical resistances R 1 and R 2 along the vertical and horizontal edges, respectively: where d is the thickness of the film.Four contacts were deposited with silver paint on the film surface.2.4.Electrochemical Characterization of the Electrocatalysts.Linear sweep voltammetry (LSV) and cyclic voltammetry (CV) measurements were performed with the Autolab potentiostat/ galvanostat PGSTAT302 at 10 and 50 mV/s −1 scan rate, in a threeelectrode configuration cell, using a Pt wire as the counter electrode and a Ag/AgCl (3 M KCl) electrode as the reference.A 1 M KOH solution at pH 13.6 (without Fe purification) was used as the electrolyte.All potentials were referred to the reversible hydrogen electrode (RHE) through the Nernst equation: The electrode/electrolyte contact area was 1 cm 2 .The turnover frequency (TOF) was calculated as TOF j nq geo = , where j geo is the geometrical current density at an overpotential of 400 mV (selected for comparison), n is the number of electrons transferred during the reaction (for water oxidation, n = 4), and q is the integrated area under the cathodic redox wave divided by the scan rate.The O 2 evolution at the electrocatalyst surface was determined by gas chromatography measurements using a sealed cell coupled to an Agilent Micro-GC gas chromatograph during a chronoamperometric measurement at 1. Electrochemical impedance spectroscopy (EIS) measurements were performed between 0.1 Hz and 1 MHz with 20 mV of amplitude perturbation, with a step potential of 50 mV in the anodic direction.The EIS raw data were analyzed with the ZView software (Scribner Associates), fitting the raw data to an equivalent circuit model for extracting both capacitances and resistances.With the extracted capacitance values from EIS, Mott−Schottky analysis was carried out using the following expression: C SC represents space charge capacitance, e is the elementary charge, ε 0 is the vacuum permittivity, ε r is the relative permittivity of NiO x (taken as 9.1), 24 N A is the acceptor density, A is the geometrical area, k is the Boltzmann constant, and T is the absolute temperature (taken as 298 K).From this analysis, the values of the flat-band potential (V FB ) and the acceptor density (N A ) were extracted.

RESULTS AND DISCUSSION
NiO x nanocomposite films spin-coated on planar Ni substrates were tested as water oxidation electrocatalysts in 1 M KOH solution (pH 13.6) after baking at different temperatures between 50 and 500 °C.The cyclic voltammograms in Figure 1a clearly show the significant impact of the baking temperature on the electrocatalytic performance, leading to a systematic increase in the catalytic current upon decreasing the temperature.The baking step after the coating is necessary to enhance the adhesion of the thin film onto the substrate since unbaked films were not mechanically stable (not shown).The well-known redox wave related to the reversible reaction Ni(OH) 2 + OH − ⇋ NiOOH + H 2 O + e − , characteristic of the Ni-based electrocatalysts, can be observed between 1.2 and 1.5 V RHE . 25Higher peak intensities were correlated to lower temperatures.At 50 °C, the recorded overpotential from linear sweep voltammetry (LSV; Supporting Information, Figure S2) at 10 mA•cm −2 reached 358 mV, as shown in Figure 1b.The turnover frequency (TOF) at a 400 mV overpotential, which reflects the amount of evolved molecular oxygen per mole of the catalyst per second, was also strongly dependent on the baking temperature, increasing upon decreasing the temperature, with the highest value (0.95 s −1 ) at 50 °C.Tafel analysis (from cyclic voltammograms of Figure 1a) shows lower Tafel slopes as the baking temperature decreases (Figure 1c).These results are consistent with enhanced water oxidation kinetics. 26,27At baking temperatures of 200−500 °C, Tafel slopes around 100 mV•dec −1 (close to 120 mV•dec −1 ) suggest that the rate-determining step is the first electron transfer step in the reaction.Conversely, at 50−100 °C, the Tafel slopes were around 78 mV•dec −1 , suggesting a change of the water oxidation mechanism, with the rate-determining step involving a chemical step after the first electron transfer. 28Chronopotentiostatic tests at 10 mA•cm −2 confirmed the long-term stability of the champion NiO x films (Figure 1d), without any significant variation of the overpotential after 60 h of testing.The excellent film adherence onto the Ni substrate is believed to significantly account for this promising long-term stability, highlighting the competitive performance of an easily upscalable and stable earth-abundant water oxidation electrocatalyst solution-processed at nearly room temperature.Indeed, no degradation or surface amorphization was identified after operation.Furthermore, the Faradaic efficiency for oxygen evolution on the optimal electrocatalyst under operation reached 100% after 1 h of testing, as shown in Supporting Information, Figure S3.The overpotential and stability of our low-cost and up-scalable electrocatalysts are competitive with those of other reported pure NiO x -based electrocatalysts, as shown in Supporting Information, Table S1.
A detailed structural and spectroscopic characterization was carried out to understand the origin of the electrocatalytic performance of the synthesized materials.SEM investigation revealed minor morphological changes in the NiO x nanocomposite films for different baking temperatures (Supporting Information, Figure S4).A more detailed morphological and crystalline characterization was obtained by HRTEM (Figure 2).Negligible differences were observed between 50 and 100 °C baked films, and only results at 100 °C are shown here.NiO x films appear as polycrystalline layers at any baking temperature.Crystal growth is promoted with temperature as expected, and HRTEM images show a noticeable increase in the mean crystallite size from <2 nm at 100 °C up to 14 nm at 500 °C (Figure 2, left panel and Supporting Information, Figure S5), in good correspondence with XRD results (Figure 3a,d), which clearly show the progressive development of crystalline domains within the nanocomposite films with baking temperature.
Indeed, the X-ray diffractograms of the NiO x films baked at 300−500 °C showed three significant peaks corresponding to the (111), (200), and (220) crystal planes of the NiO cubic structure (JCPDS 78-0423), 29 but no crystalline peaks were observed below 300 °C (Figure 3a), suggesting the short-range structural order of these films.The different crystallinities of the samples are also evident from the electron diffraction patterns or calculated fast Fourier transforms (FFTs), where the {111} and {200} planes can be identified (Figure 2, right panel).The EDS chemical analyses allow the mapping of the Ni and O distribution within the films (Figure 2, middle panel).Additionally, their estimated thickness by HAADF-STEM measurements (Supporting Information, Figure S5) decreases monotonically with the baking temperature from 75 nm at 100 °C down to 25 nm at 500 °C.This effect relates to the more efficient thermal decomposition of organic compounds (mainly acetates and methoxyethanol) at higher temperatures, in agreement with the TGA results (Supporting Information, Figure S6).Therefore, a compositional transition from an organic−inorganic hybrid nanocomposite is observed where NiO x NCs are embedded inside an acetate matrix (at low temperatures) to an all-inorganic NiO x film with a microstructure formed by round-shaped interconnected particles (at high temperature).Comparing the solid content of samples baked at low temperatures (50, 100, and 200 °C) and high temperatures (300, 400, and 500 °C), a constant wt.ratio of 2.7 is obtained, which is in good agreement with the theoretical 2.9 wt.ratio of Ni(OAc) 2 /NiO (Supporting Information, Table S2).This confirms that the solid content of thin films baked at low and high temperatures mainly corresponds to Ni(OAc) 2 and NiO, respectively.It must be noted that the large grain boundary area (related to the decrease in the crystallite size) and the higher concentration of short-range order domains within the films have been reported to enhance the electrocatalytic activity of metal oxide electrocatalysts, 30−33 in excellent agreement with the performance metrics reported in Figure 1.
Complementary structural information was obtained with Raman spectroscopy of the as-prepared electrocatalysts (Figure 3b).At the lowest baking temperatures (50−100 °C), different vibrational modes were observed at 560, 945, and 1434 cm −1 .The presence of nickel acetate complexes is confirmed by peaks at 945 cm −1 (C−C str) and 1434 cm −1 (C−O str). 34The peak located at 560 cm −1 is attributed to Ni−O stretching.We observe that this peak gradually weakens in intensity with the decrease in the baking temperature (and hence with the crystal size), which is in good agreement with previous studies. 35These results support the formation of an organic− inorganic hybrid nanocomposite where NiO x NCs are embedded inside an acetate matrix.Conversely, at higher baking temperatures (200−500 °C), the observed vibrational modes were located at 460, 498, and 1098 cm −1 .All the observed peaks are associated with the Ni−O stretching vibrational modes. 36,37On the other hand, the band located at 450−500 cm −1 increased with the baking temperature, reflecting the increase in the crystallinity of the material and confirming that a higher catalytic activity correlates to a higher fraction of short-range order domains.Additional evidence of the highly defective nature of the electrocatalysts baked at lower temperatures was obtained from photoluminescence (PL) measurements on the as-prepared samples deposited on glass substrates.−40 Note that oxygen vacancies in metal oxides can also be monitored by other techniques like electron paramagnetic resonance, 41,42 cathodoluminescence, 43 transient absorption spectroscopy, 44 spectroelectrochemistry, 45 and impedance spectroscopy. 46,47Supporting Information, Figure S7 shows the PL spectra of the different NiO x films and the glass substrate, where the shoulder at higher energies corresponds to the glass substrate (∼375 nm) and the radiative band-to-band emission expected at 335 nm (3.7 eV 48,49 ) was not identified.Moreover, the observed emission peak intensity clearly decreased with the baking temperature (Figure 3c), confirming that the density of surface oxygen vacancies decreases with the baking temperature.Collectively, XRD, HRTEM, Raman, and PL results suggest that the catalytic activity of our NiO x films scales with the density of structural defects (oxygen vacancies), in good agreement with previous studies. 50,51Figure 3d shows the evolution of the film thickness, crystallite size (measured from XRD and HRTEM), and electrical conductivity with the baking temperature.The electrical conductivity increases from 2 × 10 −5 at 175 °C to 0.1 S•cm −1 at 500 °C as a result of the transition from a hybrid NiO x nanocomposite to an all-inorganic NiO x film.Both the growth of the crystallite size of NiO x and the elimination of the organic matrix (that leads to a significant decrease in the interparticle spacing within the NiO x thin film) favor charge carrier transport within the film. 52urface properties of the as-synthesized NiO x films were also evaluated by X-ray photoelectron spectroscopy (XPS).Figure 4a,b shows the XPS spectra at the energy region of the O 1s and Ni 2p core levels measured in the samples baked from 50 to 500 °C.Below 200 °C, both core levels show a single component.Whereas the O 1s has a singlet peak centered at 532.0 eV, Ni 2p exhibits a spin-orbit doublet whose j = 3/2 broad component lies at 856.5 eV with a spin-orbit splitting of 17.7 eV.The Ni 2p main doublet is accompanied by a typical shake-up satellite located at ∼6 eV from the main doublet to higher binding energies.The presence of these core levels is characteristic of Ni(II) species, which we attribute to Ni(OH) 2 . 53,54t temperatures ≥200 °C, both the O 1s and Ni 2p spectra are red-shifted.The presence of new O 1s and Ni 2p core-level components located at lower binding energies is observed.As From XPS measurements, the surface oxygen vacancy ratio in the NiO x samples can also be estimated (Supporting Information, Figure S8).From this analysis, it is clear that samples prepared at temperatures below 200 °C show an excess in O of around 20−40% attributable to adsorbed adventitious water.In contrast, samples prepared at higher temperatures (where the presence of adsorbed water is less probable) show a surface O vacancy ratio of ∼20%, suggesting a surface NiO 0.8 local configuration.These results align well with the hypothesis that bulk defect states may also participate in catalysis, as recently reported for amorphous electrocatalysts. 57Additionally, the more hydrophilic surface that resulted from low-temperature processing can be beneficial for electrocatalysis.More details on XPS spectra are provided in Supporting Information, Section S8.
Further mechanistic insights were obtained by impedance spectroscopy (IS).The density of catalytic sites has been used as a descriptor for catalytic activity, 45 and there is a direct link between this quantity and the chemical capacitance, which can be determined by IS. 58 NiO x films were deposited on FTO substrates, and the catalytic currents followed the same trends compared to those using the Ni substrate, although absolute values were significantly lower, as expected for a less conductive substrate (see Supporting Information, Figure S9a).The series resistance of FTO/NiO x is around 20 Ω cm 2 (Supporting Information, Figure S10), while that of Ni/ NiO x films is 9 Ω cm 2 .The experimental impedance data were fitted to a simple Randles circuit (Supporting Information, Figure S11) since only a single arc was observed in the Nyquist plots, and both resistances and capacitances were extracted.The resistances (Figure 5a), which are controlled by the charge transfer kinetics, are directly correlated to the j−V curves shown in Supporting Information, Figure S9a, through the expression ( ) . Consequently, higher slopes of the catalytic currents are related to lower resistances.At relevant potentials for water oxidation (1.4−2 V vs RHE), the series resistance (R s ) controls the total resistance, and charge transfer kinetics does not limit the overall reaction.At moderate potentials (1−1.4V vs RHE), the capacitance follows a Mott− Schottky dependence.Consequently, both flat-band potential (V FB ) and acceptor density (N A ) were estimated at each baking temperature, confirming the p-type behavior of the NiO x / Ni(OH) 2 catalytic films (Supporting Information, Figure S12), as previously reported. 59It has been reported that p-type behavior plays a beneficial role on electro-oxidation processes since a positively charged surface will attract negatively charged intermediates prone to water oxidation. 60,61Although V FB does not show a clear trend with baking temperatures, N A increases with lower baking temperatures, consistent with a higher density of structural/electronic defects, as recently reported for sputtered NiO x films. 58At more anodic potentials (1.4−2 V vs RHE), the capacitance increases with applied potential with a well-defined peak over the redox wave and a gradual decrease over the catalytic region, as shown in Supporting Information, Figure S9b for the NiO x nanocomposite film baked at 50 °C.This behavior has been recently observed on NiO x electrocatalysts 58 and holds at each baking temperature (Figure 5b).
The capacitive peak over the region of the redox wave is associated with the bulk reversible transformation of Ni(OH) 2 to NiOOH before the onset of catalysis.The decrease in capacitance at the catalytic region is ascribed to the decreased hole population due to hole injection to the solution.Furthermore, the exponential increase in the capacitance with film thickness is typical of a chemical capacitance, 62 from which an approximate density of states (DOS) can be calculated since C = e • DOS (Figure 5c). 63This procedure has been successfully employed on different materials like TiO 2 , 62 silicon, 64 conducting polymers, 65 etc.The obtained DOS, in good agreement with previous reports for highly defective Ni(OH) 2 /NiOOH electrocatalysts, 58 reflects an increase in the density of catalytic sites for films baked at lower temperatures, with a higher density of structural/ electronic defects (oxygen vacancies) as confirmed by XRD, TEM, Raman, conductivity, and PL.
Furthermore, it has been recently reported that amorphous Ni-based catalysts can expose bulk catalytic active sites, leading to a higher OER activity, while their crystalline counterpart only exposes surface sites. 57Our XPS results are consistent with this behavior, and to further understand this behavior, the electrochemical surface area (ECSA) of the NiO x films baked at the different temperatures was assessed by cyclic voltammograms at the non-faradaic region (1.1−1.2V vs RHE).The results, shown in Supporting Information, Figures S13 and S14 and Table S3, reflect an increase in the surface area with the baking temperature, clearly indicating that the higher catalytic currents obtained at low temperatures are not related to higher surface area but to enhanced catalytic activity.
The acceptor density (N A ) vs the baking temperature is shown in Figure 5d.Two different regimes can be observed below and above 200 °C, also observed in XPS measurements.Below 200 °C, the acceptor density of the NiO x films was 2 orders of magnitude higher compared to those baked at higher temperatures due to the higher concentration of oxygen vacancies, responsible for the higher catalytic activity toward water oxidation.Oxygen vacancies have been found to dramatically enhance the catalytic activity of several metal oxides for different added-value chemical reactions such as oxygen evolution reaction, 66 oxygen reduction reaction, 67 N 2 reduction, 68 and methanol oxidation, 69 among others.Their role in the final performance of (photo)electrocatalytic materials and devices is critical for the design of novel electrocatalysts.Specifically for OER, the concentration and location of oxygen vacancies within the crystalline lattice have been found to correlate to the final performance of the (photo)electrocatalysts 70,71 by increasing the density of catalytic sites, 72 improving the electronic conductivity, 73 and promoting the adsorption of OH − species onto lowcoordinated metals. 74Here, the concentration of oxygen vacancies was found to increase when decreasing the baking temperature, leading to a less stoichiometric cubic NiO phase and a concomitant increase in the density of short-range order domains and active catalytic sites, as observed by the higher intensity of the Ni(II)/Ni(III) redox wave and the increase in the acceptor density of 2 orders of magnitude.These effects can also be probably related to a favorable modification of the electronic environment of Ni centers, compatible with lattice tensile stress, which has been related to a more favorable adsorption of oxygenated intermediates for water splitting. 18ll these factors led to a dramatic enhancement of the OER activity of the NiO x catalysts baked at temperatures below 200 °C, providing a nearly room-temperature solution-processed method to deliver highly competitive electrocatalysts for water oxidation.
Finally, it is well known that unintentional Fe incorporation from the unpurified KOH electrolyte solution plays a key role in the electrocatalytic activity of Ni-based catalysts, which cannot be neglected. 75,76A higher Fe concentration was indirectly detected by Raman spectroscopy in the films baked at lower temperatures.Fe incorporation, as an impurity in the host Ni lattice, improves the OER activity by providing active sites modifying the local Ni environment. 76,77The Raman spectra of the electrocatalytic films baked at 50 and 300 °C after OER can be observed in Supporting Information, Figure S15a.−80 After the normalization of the Raman spectra of both materials (Supporting Information, Figure S15b), the band at 498 cm −1 associated to Ni−O vibrations is more clearly defined for the electrocatalyst baked at a higher temperature.Additional vibrational modes at 470 and 690 cm −1 were identified after baking at 50 °C.The first peak has been associated with the E g mode in SnO 2 , as detailed above, while the second peak accounts for Fe−O vibrations in OER electrocatalysts. 81This higher Fe incorporation, together with the higher concentration of oxygen vacancies, synergistically interacts to provide the observed enhanced catalytic activity toward water oxidation for the electrocatalysts baked at lower temperatures (<200 °C). 76cheme 1 summarizes the general correspondence between synthesis, structure, and performance for the different baking conditions tested.The short-range crystallinity of the organic− inorganic nanocomposite films processed at lower temperatures (50−200 °C), compatible with lattice stress, low electrical conductivity, and high density of defects/catalytically active surface species, and higher Fe incorporation are responsible for the very competitive electrocatalytic performance of the material toward water oxidation.

CONCLUSIONS
A straightforward, solution-processed method has been developed for the in situ synthesis of an organic−inorganic nanocomposite thin film of NiO x nanocrystals embedded in an acetate-based organic matrix as a competitive low-cost and upscalable electrocatalyst for water oxidation.The reaction takes place after spin-coating during a soft bake at low temperatures (50−200 °C), allowing control of the film thickness, crystallinity, and electrical conductivity of the nanocomposite through the baking conditions.The NiO x -based nanocomposite thin film coated on the Ni foil showed a promising electrocatalytic performance when baked at low temperature (50 °C), reflected on a competitive OER overpotential (358 mV at 10 mA•cm −2 ) in alkaline conditions (pH = 13.6) for more than 60 h without significant degradation.The small size of the NiO x NCs (2−5 nm) accounts for the high density of crystalline/electronic defects, compatible with lattice stress, low conductivity, and high density of catalytically active surface species, and higher Fe incorporation prompting a more efficient, robust, and economic OER nickel-based electrocatalyst.This synthetic strategy, combined with relevant guidelines for catalyst design and optimization, may pave the way for the future design of highly efficient, durable, and economically competitive water oxidation electrocatalysts to be used in a wide range of electrosynthetic applications.

■ ASSOCIATED CONTENT
* sı Supporting Information 6 V vs RHE in 1 M KOH solution.The faradaic efficiency (FE) was estimated through the relation FEO 2 ( exp ) is the amount of evolved O 2 in mol, monitored every 5 min, and O 2 (theo) is the theoretical O 2 evolution calculated with Faraday's Law: n(mol) j O2 is the current density recorded in the chronoamperometry measurement, t is the time in seconds, n is the number of electrons transferred in the reaction, and F is the Faraday constant, 96,485.33C mol −1 .

Figure 1 .
Figure 1.Electrochemical characterization of NiO x films on a planar Ni substrate, baked at different temperatures in KOH 1 M (pH 13.6).(a) Cyclic voltammograms (iR corrected).(b) Overpotential (squares) at 10 mA•cm −2 and calculated TOF values (diamonds) at 400 mV overpotential vs baking temperature.(c) Tafel analysis showing the Tafel slopes for the different baking temperatures.(d) Chronopotentiostatic test at 10 mA•cm −2 on the champion catalyst and reference Ni substrate (baked at 50 °C).

Figure 2 .
Figure 2. (S)TEM characterization of the NiO x films on Si substrates, synthesized at different temperatures (100−500 °C): HRTEM images showing film thickness and particle size (left), the corresponding EDS maps (Ni, O, and Si displayed in green, red, and blue, respectively) of the samples (middle), and FFTs showing (111) and (200) NiO x atomic planes (right).

Figure 3 .
Figure 3. (a) XRD, (b) Raman, and (c) PL measurements on the NiO x samples baked at different temperatures (50−500 °C).(d) Evolution of the film thickness, crystalline size (from XRD and TEM), and electrical conductivity with the baking temperature.
the baking temperature increases, the low binding-energy O 1s peak becomes more intense at the expense of the broad peak.The deconvolution of the O 1s spectra (Figure 4a) shows three singlets from 200 °C.The one located at 532.1 eV is attributed to OH − species, whereas the other two peaks located at 531.4 and 529.7 eV are ascribed to the presence of O −2 species.The integrated intensity of the O −2 component located at 531.4 eV and that of OH − located at 532.1 eV show a 1:1 ratio for samples baked above 200 °C.This is indicative of the presence of NiOOH. 55The third O −2 singlet, located at 529.7 eV, is assigned to NiO. 54−56 We can conclude that Ni(OH) 2 is the main form of Ni(II) in the samples baked at 50−100 °C.From 200 °C, we observe the formation of non-stochiometric NiO with the formation of Ni(III) species in the form of NiOOH together with Ni(II) in the form of Ni(OH) 2 and NiO.Therefore, the deconvolution area of OH − in the O 1s spectra corresponds to both Ni(OH) 2 and NiOOH.As the baking temperature increases, NiO becomes the most dominant compound.A higher density of Ni(II) sites in the pristine material is consistent with the formation of NiO, which converts into Ni(OH) 2 upon immersion in the alkaline electrolyte solution and into the catalytically active NiOOH phase upon the application of anodic bias.

Figure 4 .
Figure 4. (a, b) O 1s and Ni 2p spectra were acquired by XPS in samples synthesized at different baking temperatures.The result of the deconvolution process for each specific spectrum is shown below each particular spectrum.The different O 1s components identified have been identified by dashed vertical lines.

Figure 5 .
Figure 5. (a) Resistance (R) and (b) capacitance (C) vs applied potential, extracted from EIS measurements, for the analyzed samples.(c) Calculated DOS for the different films.(d) Acceptor density (N A ) vs baking temperature showing two clear regimes of carrier concentration.All results have been normalized to the electrochemical surface area (ECSA). 2 2