Nanoparticles of Mixed-Valence Oxides MnXCO3-XO4 (0 ≤ X ≤ 1) Obtained with Agar-Agar from Red Algae (Rhodophyta) for Oxygen Evolution Reaction

The development of efficient electrocatalysts for the oxygen evolution reaction (OER) is of paramount importance in sustainable water-splitting technology for hydrogen production. In this context, this work reports mixed-valence oxide samples of the MnXCo3-XO4 type (0 ≤ X ≤ 1) synthesized for the first time by the proteic sol-gel method using Agar-Agar as a polymerizing agent. The powders were calcined at 1173 K, characterized by FESEM, XRD, RAMAN, UV–Vis, FT-IR, VSM, and XPS analyses, and were investigated as electrocatalysts for the oxygen evolution reaction (OER). Through XRD analysis, it was observed that the pure cubic phase was obtained for all samples. The presence of Co3+, Co2+, Mn2+, Mn3+, and Mn4+ was confirmed by X-ray spectroscopy (XPS). Regarding the magnetic measurements, a paramagnetic behavior at 300 K was observed for all samples. As far as OER is concerned, it was investigated in an alkaline medium, where the best overpotential of 299 mV vs. RHE was observed for the sample (MnCo2O4), which is a lower value than those of noble metal electrocatalysts in the literature, together with a Tafel slope of 52 mV dec−1, and excellent electrochemical stability for 15 h. Therefore, the green synthesis method presented in this work showed great potential for obtaining electrocatalysts used in the oxygen evolution reaction for water splitting.


Introduction
In recent decades, with the exponential growth of the population, the intensification of using fossil fuels has generated great impacts on the environment [1]. Given this, the energy transition from fossil fuels to clean energy sources has become necessary, and major renewable energy technologies have been developed [2], such as geothermal [3], wind [4], solar [5,6], and biomass [7]. However, to benefit resources even more, it is necessary to use efficient devices for energy storage and conversion [8,9].
One of the energy conversion processes that has received a lot of attention is water splitting via electrolysis for the production of hydrogen [10], which consists of two semireactions: the hydrogen evolution reaction (HER) [11] and the oxygen evolution reaction (OER) [12][13][14]. During the water-splitting process, the kinetic reaction that takes place at the anode (OER) is slow and requires high energy consumption (overpotential) due to the four electrons transferred in the reaction, causing its efficiency to decrease [15]. Therefore,

Structural and Morphological Characterization
X-ray powder diffraction patterns (XRD) were obtained by a Shimadzu XRD-7000 diffractometer using Kα(Cu) = 1.5481 Å radiation. The 2θ range was investigated from 10° to 80° with a step size of 0.02° and acquisition time of 1 s. The crystallite size, lattice parameters, and atomic positions were determined by Rietveld refinement using the software Materials Analysis Using Diffraction (TOPAS). FT-IR spectra were performed by a Shimadzu IRPrestige21 spectrophotometer between 500 and 4000 cm −1 , using KBr pellets. The ultraviolet-visible spectra (UV-Vis) were obtained in the UV-2600i spectrophotometer from Shimadzu. SEM images were obtained by a field-emission

Structural and Morphological Characterization
X-ray powder diffraction patterns (XRD) were obtained by a Shimadzu XRD-7000 diffractometer using Kα(Cu) = 1.5481 Å radiation. The 2θ range was investigated from 10 • to 80 • with a step size of 0.02 • and acquisition time of 1 s. The crystallite size, lattice parameters, and atomic positions were determined by Rietveld refinement using the software Materials Analysis Using Diffraction (TOPAS). FT-IR spectra were performed by a Shimadzu IRPrestige21 spectrophotometer between 500 and 4000 cm −1 , using KBr pellets. The ultraviolet-visible spectra (UV-Vis) were obtained in the UV-2600i spectrophotometer from Shimadzu. SEM images were obtained by a field-emission scanning electron microscope (FESEM, Carl Zeiss, Supra 35-VP Model) equipped with a Bruker EDS detector (XFlash 410-M). Surface chemical states were studied by X-ray photoelectron spectroscopy (XPS) using a SPECS Phoibos 150 spectrometer with a high-intensity monochromatic Al-Ka X-ray source (1486.6 eV). Samples were dispersed in acetone and deposited on silicon by drop-coating. Adventitious carbon C 1s with binding energy at 284.8 eV was used as reference energy. CasaXPS software was used for spectra deconvolution, thus obtaining the height, area, and position of the analyzed peaks. All the symmetric peaks were fitted Nanomaterials 2022, 12, 3170 4 of 23 using Gaussian and Lorentzian functions. Magnetic measurements were obtained using a vibrating sample magnetometer (VSM) from Lakeshore, model 7400, at room temperature, with a maximum magnetic field applied up to +15.0 KOe.

Electrochemical Characterization
All electrochemical studies were performed in an alkaline aqueous solution (KOH, 1 M pH = 13.6) at room temperature by a PGSTAT204 with FRA32M module (Metrohm Autolab) using a three-electrode setup with a platinum plate and Hg/HgO as counter and reference electrodes, respectively. The samples Mn X Co 3-X O 4 (0 ≤ X ≤ 1) were used for the fabrication of the working electrodes. Catalytic inks were prepared by mixing 5 mg of each catalyst with 50 µL of Nafion solution (5 wt%) and dispersing the mixture in 500 µL of isopropyl alcohol. Then, inks were drop-casted onto Ni foams (1 × 1 cm) on clean substrates and dried at room temperature for 5 h to prepare the working electrodes. Linear sweep voltammetry (LSV) was performed at 5 mV s −1 . Electrochemical impedance spectroscopy (EIS) was carried out using dc potentials (1.4 V vs. RHE) in the frequency range of 0.1 Hz-10 kHz and voltage amplitude of 10 mV. All measured potentials (with iR correction) were converted to the reversible hydrogen electrode (RHE) using the Nernst equation (E RHE = E Hg/HgO + 0.059×pH + 0.098). Overpotential (ï) values were calculated by the equation ï= E RHE -1.23 V. The stability tests were conducted by chronopotentiometry analysis using multi-steps of 10-20 mA cm −2 . Figure 2 shows the complete characterization process. photoelectron spectroscopy (XPS) using a SPECS Phoibos 150 spectrometer with a highintensity monochromatic Al-Ka X-ray source (1486.6 eV). Samples were dispersed in acetone and deposited on silicon by drop-coating. Adventitious carbon C 1s with binding energy at 284.8 eV was used as reference energy. CasaXPS software was used for spectra deconvolution, thus obtaining the height, area, and position of the analyzed peaks. All the symmetric peaks were fitted using Gaussian and Lorentzian functions. Magnetic measurements were obtained using a vibrating sample magnetometer (VSM) from Lakeshore, model 7400, at room temperature, with a maximum magnetic field applied up to +15.0 KOe.

Electrochemical Characterization
All electrochemical studies were performed in an alkaline aqueous solution (KOH, 1 M pH = 13.6) at room temperature by a PGSTAT204 with FRA32M module (Metrohm Autolab) using a three-electrode setup with a platinum plate and Hg/HgO as counter and reference electrodes, respectively. The samples MnXCo3-XO4 (0 ≤ X ≤ 1) were used for the fabrication of the working electrodes. Catalytic inks were prepared by mixing 5 mg of each catalyst with 50 μL of Nafion solution (5 wt%) and dispersing the mixture in 500 μL of isopropyl alcohol. Then, inks were drop-casted onto Ni foams (1 × 1 cm) on clean substrates and dried at room temperature for 5 h to prepare the working electrodes. Linear sweep voltammetry (LSV) was performed at 5 mV s −1 . Electrochemical impedance spectroscopy (EIS) was carried out using dc potentials (1.4 V vs. RHE) in the frequency range of 0.1 Hz-10 kHz and voltage amplitude of 10 mV. All measured potentials (with iR correction) were converted to the reversible hydrogen electrode (RHE) using the Nernst equation (ERHE = EHg/HgO + 0.059×pH + 0.098). Overpotential (ղ) values were calculated by the equation ղ = ERHE-1.23 V. The stability tests were conducted by chronopotentiometry analysis using multi-steps of 10-20 mA cm −2 . Figure 2 shows the complete characterization process. Figure 2. Physical, chemical, and electrochemical characterization of MnXCo3-XO4 (0 ≤ X ≤ 1). Figure 2. Physical, chemical, and electrochemical characterization of Mn X Co 3-X O 4 (0 ≤ X ≤ 1).

X-ray Diffraction (XRD)
The refined X-ray diffraction patterns of the Mn X Co 3-X O 4 (0 ≤ X ≤ 1) are shown in Figure 3a. As noted, all peaks are characteristic of the cubic phases of Co 3 O 4 (structure of Spinel#MgAl 2 O 4 type, with lattice parameter a = b = 8.072(3) Å, ICSD n • 36256, space group Fd-3mS (227)) [56] and MnCo 2 O 4 (structure of Spinel#MgAl 2 O 4 type, with lattice parameter a = b = 8.28(2) Å, ICSD n • 18544, space group Fd-3mZ (227)) [57]. No secondary phases corresponding to impurities were detected. The ICSD n • 36,256 was used to fit the Mn X Co 3-X O 4 samples (X < 1), while the 18,544 ICSD file was applied to refine the sample with composition X = 1. The observed patterns are similar to those reported previously for pure and doped cobaltites [16,20,58]. All crystallographic parameters, including crystallite size and lattice parameter, as well as the agreement indices (R wp , R exp e χ 2 ) for samples of Mn X Co 3-X O 4 (0 ≤ X ≤ 1) are gathered in Table 1.
phases corresponding to impurities were detected. The ICSD n° 36,256 was used to fit the MnXCo3-XO4 samples (X < 1), while the 18,544 ICSD file was applied to refine the sample with composition X = 1. The observed patterns are similar to those reported previously for pure and doped cobaltites [16,20,58]. All crystallographic parameters, including crystallite size and lattice parameter, as well as the agreement indices (Rwp, Rexp e χ 2 ) for samples of MnXCo3-XO4 (0 ≤ X ≤ 1) are gathered in Table 1.    Figure 3b shows the magnification of the most intense diffraction peak (311), located between 36 • and 37. nm for x = 0.8, 66 nm for x = 1, with the only exception of the sample corresponding to X = 0.4 for which the 2θ angle is slightly higher than the one displayed by the sample with X = 0.2). The largest variations of FWHM occur in the 0 ≤ X ≤ 0.6 range. The Mn +2 has a radius of 0.80 Å and it is larger than the radii of Co +3 (0.63 Å) and Co +2 (0.65 Å); thus, it causes distortion and strain in the Co 3 O 4 lattice, resulting in a decrease in crystallite size. In addition, other manganese oxidation states may be present in the samples such as Mn +3 and Mn +4 . Mn +3 (3d 4 ) is responsible for the Jahn-Teller phenomenon, which also develops a distortion in the lattice and an intrinsic strain that leads to a decrease in the crystallite size [20,[59][60][61]. The trends of lattice parameter and crystallite size are in agreement with previous reports [16,20]. The maximum values of the agreement factors R wp and R exp from the Rietveld analyses are 8.14% and 6.92%, respectively. The low values of fitting quality (χ 2 ≤ 1.30) indicate excellent agreement between the data and the refined models.

Field-Emission Scanning Electron Microscopy (FESEM)
The FESEM images of the Mn X Co 3-X O 4 nanoparticles (0 ≤ X ≤ 1) are shown in Figure 4. A non-uniform morphology was observed, specifically polyhedral-shaped particles, and a few smaller spherical-like particles, mostly agglomerated [38]. Another observation is that as the amount of Mn increases, the morphology of particles tends to be octahedral like. The average particle size distribution was 208 nm for X = 0.0, 162 nm for X = 0.2, 145 nm for X = 0.4, 142 nm for X = 0.6, 140 nm for X = 0.8, and 133 nm for X = 1.0. From the size distribution histograms, it is evident the shift of main sizes to smaller values as the Mn content increased.

Transmission Electron Microscopy (TEM)
Additional morphological characterization was carried out by the TEM technique. Typical images of the nanoparticles are presented in Figure 5. They show particles with non-uniform morphologies of different sizes. These pictures agree well with the images acquired by FESEM ( Figure 4). Figure 5b,e,h,k,n,q shows high-resolution TEM images (5 nm scale) of particles larger than 10 nm in size, with fringes related to atomic planes with spacings of 0.24, 0.296, 0.282, 0.252, 0.303, and 0.486 nm that may be due to the planes (311), (220), (220), (311), (220) and (111) for samples X = 0.0, X = 0.2, X = 0.4, X = 0.6, X = 0.8 and X = 1.0, respectively. Furthermore, it appears that the particles are coated with a carbon layer with a thickness smaller than 5 nm. Figure 5c,f,i,l,o,r shows the small-area electron diffraction (SAED) patterns of the samples. They exhibit diffraction rings originating from crystal planes (111), (220), (311), (400), (422), (333), and (440). The planes are listed beginning from the smallest ring.

Fourier-Transform Infrared (FT-IR) Spectroscopy
The FT-IR technique shows the vibrational fingerprint of the sample, with absorption peaks that correspond to the frequencies of vibrations of the bonds among the atoms that make up the material [62]. Figure 6 shows the spectra of the Mn X Co 3-X O 4 samples (0 ≤ X ≤ 1) in the range from 400 to 4000 cm −1 , where two bands with the highest intensities are located at 552-570 and 643-663 cm −1 , which are related to the stretching vibrations of the metal-oxygen bond, which confirms the formation of the pure Co 3 O 4 phase [63]. The v 1 band at 552-570 cm −1 is characteristic of the vibration of Co 3+ at the octahedral site, and the ν 2 band at 643-663 cm −1 is related to the vibration of Co 2+ at the tetrahedral site, confirming the formation of the spinel-like oxide [64], in agreement with the XRD study. The low-intensity band that appears at 1100 cm −1 is due to C-O stretching vibrations. The band at 1383 cm −1 is attributed to the symmetric deformations of C-N and CH 2 groups, originating from the residues of nitrate ions and agar-agar [65][66][67]. The band at 1635 cm −1 was attributed to the angular deformation of the adsorbed water molecules [68]. The broad absorption band in the region of about 3440 cm −1 is due to OH stretching of the water molecules adsorbed from the moisture during the storage process [69]. Furthermore, it is observed that the bands at 554-643 cm −1 for sample X = 0.8 and the bands at 552-643 cm −1 for sample X = 1.0 are similar, which may be related to the oxidation states of manganese (Mn +2 , Mn +3 , and Mn +4 ) or to the vibrational intensity between the manganese and oxygen bond. In general, the frequency of the peaks of the absorption bands at (552-570) and (643-663) cm −1 (Table 2) decreases with the replacement of cobalt with manganese ions, i.e., they shift to the right as the amount of manganese increases, and this is related to the increase in the metal-oxygen distance, as indicated by the increase in the lattice parameter of the unit cells (Table 1), since Mn ions are larger than Co ions [21].

Transmission Electron Microscopy (TEM)
Additional morphological characterization was carried out by the TEM technique. Typical images of the nanoparticles are presented in Figure 5. They show particles with non-uniform morphologies of different sizes. These pictures agree well with the images acquired by FESEM (Figure 4). Figure 5b,e,h,k,n,q shows high-resolution TEM images (5 spacings of 0.24, 0.296, 0.282, 0.252, 0.303, and 0.486 nm that may be due to the planes (311), (220), (220), (311), (220) and (111) for samples X = 0.0, X = 0.2, X = 0.4, X = 0.6, X = 0.8 and X = 1.0, respectively. Furthermore, it appears that the particles are coated with a carbon layer with a thickness smaller than 5 nm. Figure 5c,f,i,l,o,r shows the small-area electron diffraction (SAED) patterns of the samples. They exhibit diffraction rings originating from crystal planes (111), (220), (311), (400), (422), (333), and (440). The planes are listed beginning from the smallest ring.  for sample X = 1.0 are similar, which may be related to the oxidation states of manganese (Mn +2 , Mn +3 , and Mn +4 ) or to the vibrational intensity between the manganese and oxygen bond. In general, the frequency of the peaks of the absorption bands at (552-570) and (643-663) cm −1 ( Table 2) decreases with the replacement of cobalt with manganese ions, i.e., they shift to the right as the amount of manganese increases, and this is related to the increase in the metal-oxygen distance, as indicated by the increase in the lattice parameter of the unit cells (Table 1), since Mn ions are larger than Co ions [21].

Ultraviolet-Visible Spectroscopy (UV-Vis)
The electronic properties of Mn X Co 3-X O 4 samples, as illustrated in Figure 7, were investigated by UV-Vis spectroscopy in the wavelength range from 300 to 1400 nm. The absorptions at 528 and 792 nm for the sample X = 0 correspond to the ligand-metal O(-II) → Co(III) and O(-II) → Co(II) electron transfer, respectively [76][77][78]. The variation of absorption in the range from 1033 to 1110 nm shows that when the amount of manganese increases, the absorption peak wavelength increases, and this is related to the higher O 2 /O 1 ratios according to the XPS results, and it will affect the overpotential in the oxygen evolution reaction for the MnCo 2 O 4 (X = 1.0) sample, which would mean a better catalytic activity for the oxidation reactions.   (Table 3), confirming the formation of the pure phase of mixed-valence oxides of spinel-like structure [79][80][81][82][83]. The most intense band at 688-660 cm −1 is assigned to the octahedral site MO6 related to the A1g mode of the O7h spectroscopic symmetry, which corresponds to the stretching vibrational modes of these oxides M-O, where M = {Co, Mn}, thus substantiating the formation of MnCo2O4. The Raman bands with medium intensity in the intervals 468-488 cm −1 and 508-518 cm −1 are assigned to Eg and F2g, respectively; meanwhile, the Raman bands with lower intensities in the interval 603-617 cm −1 are caused by the F2g mode. Moreover, the Raman bands with very low intensity at 182-193 cm −1 are attributed to the F2g mode related to the tetrahedral sites of CoO4 [82][83][84]. In general, when comparing the positions of the peaks, it is observed that as the amount of manganese increases, the peaks shift to the left, analogously to what was noticed in the FT-IR spectra. This change may be due to the greater ionic radius of Mn 2+ , in comparison to that of Co 2+ /Co 3+ [85][86][87], which when entering the structure of Co3O4, generates a large distortion in the crystalline structure and increases the distance between the metal and the oxygen, and consequently a weakening of bonds occurs. Another reason would be due to vibrations in the structure, where the Co 2+ and Co 3+ cations are located in tetrahedral and octahedral sites in the cubic crystal structure [88].   (Table 3), confirming the formation of the pure phase of mixed-valence oxides of spinel-like structure [79][80][81][82][83]. The most intense band at 688-660 cm −1 is assigned to the octahedral site MO 6 [82][83][84]. In general, when comparing the positions of the peaks, it is observed that as the amount of manganese increases, the peaks shift to the left, analogously to what was noticed in the FT-IR spectra. This change may be due to the greater ionic radius of Mn 2+ , in comparison to that of Co 2+ /Co 3+ [85][86][87], which when entering the structure of Co 3 O 4 , generates a large distortion in the crystalline structure and increases the distance between the metal and the oxygen, and consequently a weakening of bonds occurs. Another reason would be due to vibrations in the structure, where the Co 2+ and Co 3+ cations are located in tetrahedral and octahedral sites in the cubic crystal structure [88].

X-ray Photoelectron Spectroscopy (XPS)
The surface oxidation states of the samples were analyzed by XPS. Figure 9 shows the high-resolution Co 2p, Mn 2p, and O 1s spectra obtained from the analysis. All data were corrected for the carbon peak position. In the case of the Co 2p spectra (Figure 9a), four peaks were deconvoluted, at lower binding energies, corresponding to Co 3+ and Co 2+ , as well as two satellite peaks at higher binding energies. The binding energies obtained for Co 3+ were found to fall in the 779.64-780.19 eV range, while for Co 2+ , in the 781.19-781.85 eV range, in agreement with previous work [90]. We also found a Co 2+ /Co 3+ ratio varying from 0.31 to 0.40 among the samples, for which the sample X = 0.8 was found to have the lowest value of Co 2+ /Co 3+ = 0.31. Higher oxidation states can induce more bonded oxygen species, which may have a positive impact on oxidation reactions.

X-ray Photoelectron Spectroscopy (XPS)
The surface oxidation states of the samples were analyzed by XPS. Figure 9 shows the high-resolution Co 2p, Mn 2p, and O 1s spectra obtained from the analysis. All data were corrected for the carbon peak position. In the case of the Co 2p spectra (Figure 9a), four peaks were deconvoluted, at lower binding energies, corresponding to Co 3+ and Co 2+ , as well as two satellite peaks at higher binding energies. The binding energies obtained for Co 3+ were found to fall in the 779.64-780.19 eV range, while for Co 2+ , in the 781.19-781.85 eV range, in agreement with previous work [90]. We also found a Co 2+ /Co 3+ ratio varying from 0.31 to 0.40 among the samples, for which the sample X = 0.8 was found to have the lowest value of Co 2+ /Co 3+ = 0.31. Higher oxidation states can induce more bonded oxygen species, which may have a positive impact on oxidation reactions. Nanomaterials 2022, 12, x FOR PEER REVIEW 13 of 25 Conversely, in the case of the Mn 2p spectra (Figure 9b), the data were deconvoluted into three peaks, which were ascribed to Mn 4+ (ranging from 644.154 eV to 645.146 eV), Mn 3+ (ranging from 642.790 eV to 643.125 eV), and Mn 2+ (641.223 eV to 641.459 eV), in agreement with previous reports [90,91]. The lowest oxidation state species, Mn 2+ , was found to represent the largest fraction of the total species present at the surface, with a relative value varying from 0.42 to 0.51 among the samples. Conversely, the presence of Mn 4+ and Mn 3+ oxidation states is related to the relatively high calcination temperature used in this work, i.e., 900 °C, as found in previous literature [92]. In this respect, we also noted a higher Mn 3+ /Mn 4+ ratio for the tested samples, which correlates well with the previously discussed Jahn-Teller phenomenon, with a concurrent distortion of the crystal lattice.

Intensity (a.u.)
Binding energy (eV) Conversely, in the case of the Mn 2p spectra (Figure 9b), the data were deconvoluted into three peaks, which were ascribed to Mn 4+ (ranging from 644.154 eV to 645.146 eV), Mn 3+ (ranging from 642.790 eV to 643.125 eV), and Mn 2+ (641.223 eV to 641.459 eV), in agreement with previous reports [90,91]. The lowest oxidation state species, Mn 2+ , was found to represent the largest fraction of the total species present at the surface, with a relative value varying from 0.42 to 0.51 among the samples. Conversely, the presence of Mn 4+ and Mn 3+ oxidation states is related to the relatively high calcination temperature used in this work, i.e., 900 • C, as found in previous literature [92]. In this respect, we also noted a higher Mn 3+ /Mn 4+ ratio for the tested samples, which correlates well with the previously discussed Jahn-Teller phenomenon, with a concurrent distortion of the crystal lattice.
Finally, the O 1s high-resolution spectra (Figure 9c) [91,93]. From the analysis of the O 1s high-resolution spectra (Table 4), we determined higher O 2 /O 1 ratios with increasing Mn content, with a maximum value obtained for the X = 1.0 sample. This suggests that the compounds with the highest Mn content possess increased catalytic activity, as a likely result of increased oxygen-ion vacancies in these samples.

Vibrating Sample Magnetometer (VSM)
Magnetization measurements were done to study the magnetic behavior of samples at room temperature and to determine the cation magnetic moment in an approximate manner. For all samples, the isothermal magnetization at T = 300 K showed a linear behavior with the magnetic field, and their magnetization at a given field increased with the Mn concentration, as shown in Figure 10. the characteristic binding energies determined for these peaks, they are likely related to surface lattice oxygen (Olat, O 2− ), adsorbed oxygen species (Oads, O 2− , O2 2− , and O − ), and adsorbed water species (OH2O), in agreement with earlier reports on similar compounds [91,93]. From the analysis of the O 1s high-resolution spectra (Table 4), we determined higher O2/O1 ratios with increasing Mn content, with a maximum value obtained for the X = 1.0 sample. This suggests that the compounds with the highest Mn content possess increased catalytic activity, as a likely result of increased oxygen-ion vacancies in these samples.

Vibrating Sample Magnetometer (VSM)
Magnetization measurements were done to study the magnetic behavior of samples at room temperature and to determine the cation magnetic moment in an approximate manner. For all samples, the isothermal magnetization at T = 300 K showed a linear behavior with the magnetic field, and their magnetization at a given field increased with the Mn concentration, as shown in Figure 10.  , and T = 300 K. It is known that L(a) tends to a/3 when a is less than about 0.5 [94]. In the present case, if µ~4.51 × 10 −20 Erg/Oe (theoretical magnetic moment for Mn 3+ and Co 3+ ) and H = 15 × 10 3 Oe, then, one can get a = 0.01633, which is smaller than 0. 5 .899 µ B . Therefore, the magnetic moments for Co and Mn seem to be mainly in the low spin configuration; however, we cannot rule out the presence of some moments in the high spin configuration.

Oxygen Evolution Reaction (OER)
The samples were also evaluated as electrocatalysts for the oxygen evolution reaction (OER). According to the results of the anodic polarization (Figure 11a), the electrodes presented values of 515 (Ni foam), 342 (X = 0.0), 342 (X = 0.2), 339 (X = 0.4), 337 (X = 0.6), 323 (X = 0.8), 299 (X = 1.0) and 235 (RuO 2 /Ni foam benchmark, extracted from reference [95]) mV vs. RHE, respectively, to record a current density J = 10 mA cm 2 . Among the investigated materials, the MnCo 2 O 4 (X = 1.0) samples displayed the best catalytic activity for OER, i.e., the lowest overpotential because the incorporation of Mn into the structure enhanced the defect concentrations, thus increasing the amount of catalytically active sites, which facilitated the mass transfer process, favoring OER [96]. Moreover, the crystalline size decreased with the increase of manganese content, which indicates that the Co 3 O 4 sample (X = 0.0) has larger average crystal sizes than the other samples, especially MnCo 2 O 4 (X = 1.0); thus, Co 3 O 4 was the sample that had the highest overpotential. This indicates that the presence of Mn, has a suppressive effect on Co 3 O 4 [97]. The obtained values are in agreement with others reported in the literature for Mn X Co 3-X O 4 nanostructures, as shown in Table 5. Table 5. Comparison of OER performance of nanostructured Mn X Co 3-X O 4 (0 ≤ X ≤ 1) catalysts reported in the literature. Data refer to an overpotential to generate j = 10 mA cm 2 (η 10 ). * CFP (carbon fiber paper); GC (glassy carbon).  * CFP (carbon fiber paper); GC (glassy carbon).
The electrocatalytic kinetics for OER was investigated by the Tafel plots extracted from the LSV (linear sweep voltammetry) curves (Figure 11a The electrocatalytic kinetics for OER was investigated by the Tafel plots extracted from the LSV (linear sweep voltammetry) curves (Figure 11a), using the Tafel equation (η = a + b log j), where b is the Tafel slope, η is the overpotential, j is the current density, and a is a constant. The values of the Tafel slope ( Figure 11b) were 63 (X = 0.0), 73 (X = 0.2), 72 (X = 0.4), 69 (X = 0.6), 68 (X = 0.8), and 52 mV dec −1 (X = 1.0). Therefore, the results did not follow exactly a sequence like the ï 10 values (Figure 11a), but it can be observed that the electrode based on the MnCo 2 O 4 sample (X = 1.0) exhibited the best reaction kinetics for OER, as it showed the lowest Tafel slope, which demonstrates a higher efficiency for oxygen evolution. The Tafel slope of 63 mV dec −1 for the Co 3 O 4 sample (X = 0.0) corresponds to slightly slower kinetics, indicating limitation in charge and mass transfer processes compared to the x = 1.0 sample. The Co 3 O 4 sample was also the one with the highest overpotential, with no distortion in the structure, which reduces defects and consequently the oxygen vacancies [106,107]. These results are consistent with the XPS values as well as the electrochemical impedance spectroscopy. The samples with X = 0.2, 0.4, 0.6, and 0.8 show values next to 70 mV dec −1 , which means much slower kinetics for OER.
All this evidence can be explained by the distortion of the lattice with the increase of the amount of Mn in the Co 3 O 4 structure, which changes the electronic charge distribution and increases the disorder in the crystalline system [106][107][108][109]. Furthermore, with increasing Mn percentages, the availability of oxygen and flexibility in the lattice is greater, which in turn is related to the M-O bond length [110]. In any case, the samples (X = 0.2), (X = 0.4), (X = 0.6) and (X = 0.8) show values of Tafel slope below 80 mV dec −1 , and these results indicate the adsorption of intermediate species as the rate-determining step (rds), based on the Krasil'shchikov reaction model for OER in alkaline medium [111,112].
The double-layer capacitance (C DL ) can be obtained from the relationship between the anode current density (i a ) and the scan rate (υ), according to (i a = υ x C DL ) [100]. Figure 11c shows the double-layer capacitance values obtained for the samples: 1.78 (X = 0.0), 2.75 (X = 0.2), 2.57 (X = 0.4), 2.41 (X = 0.6), 3.16 (X = 0.8), and 2.03 mF cm −2 (X = 1.0). These results suggest that the largest number of active sites is organized on the electrode surfaces. Although among the samples of this series, the one with X = 0.8 had the second lowest performance for OER, it displays the highest C DL value, which may be linked to the high amount of oxygen vacancies that improves the absorption of reactive species (like OH ) [113]. For the samples (X = 0.2), (X = 0.4), and (X = 0.6), the values of 2.75, 2.57, and 2.41 mF cm −2 , respectively, are consistent with the XPS data (Figure 9b), where the species in the lowest oxidation state, Mn 2+ , represented the largest fraction of the total species present on the surface, with a relative fraction ranging from 0.42-0.51 among the samples. However, even with a low C DL value for the sample (X = 1.0), the presence of Mn ions in the structure is essential for superior electrocatalytic properties. This was proven by the best overpotential extracted from the LSV curves ( Figure 11a) and the XPS data (Figure 9c), as the substitution of Mn in spinel oxide cobalt occurs selectively in the (Co 3+ ) lattices, and the energy required for Mn 2+ to substitute Co 3+ is lower than that of Co 2+ , [114]. Moreover, Mn +2 , Mn +3 , and Mn +4 have ionic ratios of 0.80, 0.66, and 0.60 Å, respectively, and the ratios of Mn +2 and Mn +3 are larger than that of Co 3+ (0.63 Å). Therefore, Mn doping results in the expansion of the Co 3 O 4 lattice, generating defects, which influences the mass diffusion and charge transfer properties, contributing to oxygen-ion vacancies, which are consistent with the O 2 /O 1 ratio that was highest for the MnCo 2 O 4 sample (X = 1.0), with a value of 0.940 [108,109,115,116].
Durability is another important indicator of catalytic performance. The stability of the electrocatalysts was evaluated by chronopotentiometry. Tests were performed at a current density of 10 mA cm −2 for 15 h. According to the curves shown in Figure 11d, it can be seen that the samples (X = 0) and (X = 0.6) exhibited a potential decrease until about 2 h, but then they remained stable, whereas for the samples (X = 0.8) and (X = 1.0), the potential was practically stable for the entire time period tested. In general, all samples showed satisfactory stability over 15 h of testing [117].

Electrochemical Impedance Spectroscopy (EIS)
The electrocatalytic activity also was investigated by electrochemical impedance spectroscopy (EIS). The EIS spectra of all samples were collected at 1.4 V vs. RHE. As seen in the Bode plots (Figure 12b), the OER is composed of complex processes involving electrosorption of intermediate species during the reaction progress. This suggests an equivalent circuit model (ECM) able to describe these processes [118,119] that is composed of R S (uncompensated solution resistance), R P (polarization resistance, which denotes the overall rate of the OER), Q DL (double-layer pseudo-capacitance), R -ad (resistance associated with adsorption of intermediate species), and Q -ad (pseudo-capacitance of these species throughout the reaction). A constant phase element (Q) was used to model an imperfect capacitor, and its impedance was obtained by: throughout the reaction). A constant phase element (Q) was used to model an imperfect capacitor, and its impedance was obtained by: Then, the values were used to calculate true capacitance (CDL or C-ad) by: In Figure 12a, the impedance of the electrodes is composed of two incomplete semicircles. The first is attributed to the polarization process (charge transfer), and the second indicates limitations on mass transfer processes, related to the intermediate species adsorption process [120]. For the electrodes, the obtained RP values were consistent with the OER performance, i.e., the X = 1.0 sample showed the lowest value (6.02 Ω), followed by X = 0.8 (9.70 Ω). The other samples revealed RP values very close, but the result was expected as their overpotential values were close. The CDL values varied slightly (Table 6) due to the oxidation peak shown in Figure 11a. The RadCad loop associated with relaxation, which was attributed to the adsorbed intermediate species, revealed the difficulty of these electrodes to work in the diffusive processes observed at low frequencies. The high R-ad (>1100 Ω) values displayed by the electrodes in those low frequency (>1 Hz) confirm that the adsorption of intermediates should be a rate-limiting step as predicted by Tafel analysis (Figure 11b) [118,121]. The values obtained from the fitting of the spectra are listed in Table 6. Table 6. EIS-Results of fitting of the impedance spectra reported in Figure 12.  Then, the values were used to calculate true capacitance (C DL or C -ad ) by:

RS (Ω) RP (Ω) CDL (mF) R-ad (Ω) C-ad (mF)
In Figure 12a, the impedance of the electrodes is composed of two incomplete semicircles. The first is attributed to the polarization process (charge transfer), and the second indicates limitations on mass transfer processes, related to the intermediate species adsorption process [120]. For the electrodes, the obtained R P values were consistent with the OER performance, i.e., the X = 1.0 sample showed the lowest value (6.02 Ω), followed by X = 0.8 (9.70 Ω). The other samples revealed R P values very close, but the result was expected as their overpotential values were close. The C DL values varied slightly (Table 6) due to the oxidation peak shown in Figure 11a. The R ad C ad loop associated with relaxation, which was attributed to the adsorbed intermediate species, revealed the difficulty of these electrodes to work in the diffusive processes observed at low frequencies. The high R -ad (>1100 Ω) values displayed by the electrodes in those low frequency (>1 Hz) confirm that the adsorption of intermediates should be a rate-limiting step as predicted by Tafel analysis (Figure 11b) [118,121]. The values obtained from the fitting of the spectra are listed in Table 6. Table 6. EIS-Results of fitting of the impedance spectra reported in Figure 12.

Conclusions
The Mn X Co 3-X O 4 (0 ≤ X ≤ 1) samples were synthesized by the proteic sol-gel method (green synthesis) using Agar-Agar as a polymerizing agent in order to investigate their structural, optical, magnetic, and electrochemical properties. X-ray diffraction indicated for all samples the obtainment of the pure cubic phase without any secondary phase, which was also confirmed from Raman, TEM, FT-IR, and UV-Vis studies. Regarding the magnetic measurements, it was observed for all samples a magnetization in a certain field increasing with the Mn concentration, which is typical of a paramagnetic behavior. From the XPS analysis, the species in the Mn 2+ oxidation state represented the largest fraction of the total species present on the surface, and as the amount of Mn increased, the O 2 /O 1 ratio also increased, reaching a value of 0.940 for the sample MnCo 2 O 4 (X = 1.0). For OER, the same sample exhibited the best catalytic activity when compared with the others, with an overpotential of 299 mV, which is lower than those of noble metal electrocatalysts reported in the literature. In addition, the samples showed superior long-term stability for efficient water oxidation activities at J = 10 mA/cm 2 per 15 h. Thus, it can be concluded that proteic sol-gel synthesis is an excellent method to produce nanosized mixed-valence oxides Mn X Co 3-X O 4 for the fabrication of electrodes for water electrolysis. Data Availability Statement: The study did not report any data.