Study of oxygen evolution reaction on thermally prepared x PtO y-( 100x ) IrO 2 electrodes

The mixed coupled xPtOy-(100-x)IrO2 electrodes (x = 0, 10, 20, 30, 40, 50 60, 70, 80, 90 and 100) were thermally prepared at 450 °C on titanium supports. The prepared electrodes were firstly physically characterized by scanning electron microscopy (SEM), X-ray diffraction (XRD) and X-ray photoelectron spectroscopy (XPS). Afterwards, electrochemical characterizations were performed by voltammetric (cyclic and linear) methods in different electrolyte media (KOH and HClO4). It is shown that the prepared electrodes are composed by both PtOy (platinum and platinum oxide) and IrO2 (iridium dioxide). For xPtOy-(100-x)IrO2 electrodes having higher content of IrO2, more surface cracks and pores are formed, defining a higher surface area with more active sites. Higher surface area due to presence of both PtOy and IrO2, is for xPtOy-(100x)IrO2 electrodes in 1 M KOH solution confirmed by cyclic voltammetry at potentials of the oxide layer region. For all prepared electrodes, voltammetric charges were found higher than for PtOy, while the highest voltammetric charge is observed for the mixed 40PtOy-60IrO2 (x = 40) electrode. The Tafel slopes for oxygen evolution reaction (OER) in either basic (0.1 M KOH) or acid (0.1 M HClO4) media were determined from measured linear voltammograms corrected for the ohmic drop. The values of Tafel slopes for OER at PtOy, 90PtOy-10IrO2 and IrO2 in basic medium are 122, 55 and 40 mV dec-1, respectively. For other mixed electrodes, Tafel slopes of 40 mV dec-1 were obtained. Although proceeding by different OER mechanism, similar values of Tafel slopes were obtained in acid medium, i.e., Tafel slopes of 120 60 and 39 mV dec-1 were obtained for PtOy, 90PtOy-10IrO2 and IrO2, and 40 mV dec-1 for other mixed electrodes. The analysis of Tafel slope values showed that OER is more rapid on coupled mixed electrodes than on pure PtOy. For mixed xPtOy-(100-x)IrO2 electrodes, OER is more rapid when the molar J. Electrochem. Sci. Eng. 10(4) (2020) 347-360 OXYGEN EVOLUTION REACTION ON Pt-Ir OXIDE ELECTRODES 348 percent of PtOy meets the following condition: 0 ˂ x ≤ 80. This study also showed that the mixed coupled electrodes are more electrocatalytically active for OER than either PtOy or IrO2 in these two media.


Introduction
Currently, climate changes and energy shortage are some of the greatest challenges for humanity [1][2][3]. Increasing carbon dioxide (CO2) levels in the atmosphere cause severe issues such as global warming, energy security, and sustainability. These issues stimulated researchers to find new clean technologies that are high in efficiency and low in cost [4]. To fulfill this goal without destroying the environment, alternative methods of energy generation, conversion, and storage need to be employed [5][6][7][8]. As a sustainable and environmentally friendly alternative, water electrolysis combined with renewable electricity sources, such as wind and solar power, is expected to provide a practical solution. However, the major drawback of technologies such as solar or wind energy is that they cannot function continuously. Thus, renewable energy from water splitting by the electrochemical route is considered as one of the important techniques to provide a solution for climate change and energy shortage. The overall water splitting reaction includes the hydrogen evolution reaction (HER) and oxygen evolution reaction (OER). To realize the practical and sustainable application of electrocatalytic hydrogen/oxygen evolution reactions (HER/OER), an efficient, durable, and economical electrocatalyst is demanded triggering the reaction with minimal overpotential and fast kinetics [9][10][11].
According to the literature, the high-performance electrocatalysts are mainly based on noble metals such as for example, platinum [12,13]. From the cost perspective, however, it is highly desirable to develop electrocatalysts based on metal oxides.
It was already found that when IrO2 is mixed with Pt oxide, Pt-Ir oxide shows enhanced electrocatalytic activity and rapid oxygen evolution reaction [28,29]. To further research about electrocatalytic properties of mixed PtOy-IrO2, it appears necessary to optimize the ratio of PtOy and IrO2 on this electrode. In the present work, the proportion of each element (Pt and Ir) of the mixed electrode has been varied and the influence of these variations on the kinetics of OER is studied. Mixed xPtOy-(100-x) IrO2 electrodes (x = 0, 10, 20, 30, 40, 50, 60, 70, 80, 90 and 100) where x represents the molar percent of PtOy precursor, have been prepared thermally at 450 °C and the kinetics of OER is followed on these electrodes in 0.1 M KOH and 0.1 M HClO4 media, respectively.

Experiments
Metallic supports (or substrates) of titanium plates of dimensions 16160.5 mm were used. These supports were pretreated by sandblasting and rinsing to remove all the impurities and to increase the roughness which improves adhesion of the deposit. The surfaces were degreased and cleaned with isopropanol, followed by rinsing with distilled water. Precursors of H2PtCl66H2O (Fluka, 99 %), and H2IrCl6H2O (Fluka, 99 %) were applied by brushes to the surfaces of the pretreated electrodes. These precursors were diluted in isopropanol (Fluka, 99.5 %). Two precursor solutions of H2PtCl6H2O and H2IrCl6H2O were prepared separately at 23.30 g/L and 23.25 g/L, respectively. Then, mixtures of these two solutions were prepared, respecting the proportions by moles number of PtOy and IrO2 for the coupled mixed electrodes. The concentrations of precursors in the mixture which served for the preparation of the mixed xPtOy-(100-x)IrO2 electrodes are grouped in Table 1. The solvent of the deposit was evaporated by baking in the oven at 80 °C for 10 min. Then the deposit underwent thermolysis in the furnace for 15 min at 450 °C. Several layers of deposit were thus produced before being calcined in the furnace at 450 °C for one hour. The mixed coupled xPtOy-(100x)IrO2 electrodes (x = 0, 10,20,30,40,50,60,70,80,90 or 100) were prepared by this method, where x is referred to the molar percent of PtOy precursor. For the mixed electrodes, a mixture of different percent molars of the corresponding precursors was used. The deposit loading was about 5 g m -2 on each titanium substrate. High-resolution images of the prepared electrode surfaces were realized by scanning electron microscopy (SEM, ZEISS, SUPRA 40VP). The crystalline structure was examined by X-ray diffraction (XRD) using a Siemens equipment with a Cu cathode. The XPS analysis was carried out with a Kratos Axis-Ultra Spectrometer using a monochromatic Al Kα X-Ray source, operated at 15 kV, and a pass energy of 20 eV. The electrochemical measurements (cyclic and linear sweep voltammetry) were carried out using a conventional three-electrode electrochemical cell and Autolab PGStat 20 (Ecochemie). The prepared xPtOy-(100-x)IrO2 electrodes, having geometric surface area of 0.785 cm 2 , served as the working electrodes (WE). The reference electrode (RE) was a saturated calomel electrode (SCE), mounted in a Luggin capillary and placed close to WE by a distance of 1 mm. The counter-electrode (CE) was the spiral platinum wire, having the area at least hundred times larger than WE. All potential values in this manuscript are given against the reversible hydrogen electrode (RHE) using the following relation: All electrochemical experiments were made at the ambient temperature of 25 °C.

Physical characterization of prepared electrodes
The prepared electrodes were analyzed by scanning electron microscopy. The surface micrographs of IrO2, 40PtOy-60IrO2, 70PtOy-30IrO2, and PtOy electrodes are presented in Figure 1. The micrograph of pure IrO2 in Figure 1a shows superimposed layers of iridium dioxide, cracks and some pores, what is in accordance with the literature [24]. The cracks observed on the surface of pure IrO2 occurred during the cooling of the deposit due to the mechanical stresses produced by the difference in the coefficients of thermal expansion of the substrate and the deposit. In Figure 1b, SEM image of the electrode containing 40PtOy-60IrO2 shows that the surface of this deposit appears more continuous and the cracks are narrower in comparison with pure IrO2. Also, some small Pt crystals dispersed on the surface of the deposit can be observed in Figure 1b. SEM image of the deposit containing 70PtOy-30IrO2 is presented in Figure 1c, showing a porous and very rough surface with almost non-existent cracks. At this stage, the layer of the deposit resembles that of PtOy [25]. On PtOy electrode surface shown in Figure 1d, there are no cracks but the grains are strongly bonded to each other, thus showing the porous and rough nature of the deposited PtOy. These observations indicate that when the molar percentage of PtOy in the deposit increases, the surface structure of the deposit tends to be more like that of PtOy. XPS spectra of PtOy, 50PtOy-50IrO2 and IrO2 electrodes are presented in Figure 2. On the XPS spectrum of PtOy, one can notice the presence of three intense peaks at 70, 312 and 335 eV, which all characterize platinum presence on titanium. Carbon is also present with a peak observed at 286 eV at all presented electrodes. Presence of carbon could be due to the poisoning of the measuring device, isopropanol used in the precursor preparations, and/or contamination in the oven used to prepare the electrodes. Two very close peaks are observed at 218 and 224 eV. These two peaks characterize the presence of platinum oxide. In the XPS spectrum of the mixed electrode 50PtOy-50IrO2 in addition to platinum peaks observed at 70, 312 and 335 eV, there are some new peaks at 61, 296 and 494 eV. These three new peaks also exist in XPS spectrum of pure iridium oxide, approving the presence of iridium oxide in the mixed electrode. The XRD spectrum of PtOy presented in Figure 3 provides more details on the bulk structure of the platinum coating. The XRD pattern of PtOy shows two peaks at 2 = 40 and 68°, which are assigned to the metallic Pt (JCPDS N o 4-802). The peaks located at 35 and 63° correspond to PtO2 (JCPDS N o 38-1355). The peaks located at 2 = 40 and 53 o could be assigned to Ti [29]. Those XRD results indicate that the coating of PtOy is a mixture of metallic Pt and PtO2. In other words, the deposit is a mixture of Pt with zero-valence and its oxidized form, what also confirms the XPS results of PtOy. Generally, both XPS and XRD results reveal that platinum present at the titanium surface exists in different oxidized states. Figure 3. XRD spectrum of PtOy Electrochemical characterization of prepared electrodes with ferri/ferrocyanide redox couple Figure 4a shows the cyclic voltammograms of the mixed 60PtOy-40IrO2 electrode in 1 M KOH, containing different concentrations of potassium ferri-ferrocyanide couple (1 -100 mM). These voltammograms showed oxidation peaks in the forward potential scans and reduction peaks in the reverse potential scans. There is an increase in the current density of both peaks with increase of the concentration of potassium ferri-ferrocyanide couple. The current densities of the oxidation and reduction peaks were plotted as a function of the concentration of potassium ferri-ferrocyanide for 10PtOy-90IrO2, 50PtOy-50IrO2, and 90PtOy-10IrO2 electrodes, and presented in Figures 4b and 4c. The obtained curves are straight lines that pass through the origin of the coordinate system. Note that similar results were obtained with other mixed electrodes studied here (curves not shown). The equations of the lines make it possible to determine the anode (Da) and cathode (Dc) diffusion coefficients, using the following equation [30,31]: jp = 2.69 10 5 n 3/2 AD 1/2 C 1/2 In eq. (2), jP is peak current density, n is number of exchanged electrons, A is geometric surface area, C is concentration of reaction species, D is their diffusion coefficient, and v is scan rate. The values of diffusion coefficients of ferri/ferrocyanide couple listed in Table 2 were determined by eq. (2) with n = 1, and are of the same order of magnitude as those from the literature [33]. The certain difference observed between the values determined in this work and literature data could be linked to the use of the geometric surface and not the real surface of the electrode in the calculation. It is clear from Table 2 that D values are somewhat lower for mixed xPtOy-(100-x)IrO2 electrodes than those obtained on pure IrO2 and PtOy.

2 / o
The ratios of the absolute value of the anode and cathode peak current densities are also shown in Table 2. A value close to 1 was obtained for each electrode. These results confirm an almost reversible behavior of the potassium ferri-ferrocyanide couple on all here studied electrodes.  RHE. This can be ascribed to the reduction of oxides formed in the forward potential scan [33] and also during the electrode calcination. A rectangular shape from 0.91 to 1.47 V/RHE is characterized by an almost constant current density and found similar to that observed previously [29]. In comparison to IrO2 electrode, PtOy electrode has much lower voltammetric charge. At the other side, IrO2 electrode possesses very high voltammetric charge and its voltammetric response is generally characteristic for heat-prepared DSAs (dimensionally stable anodes) [29,35,36]. Much improved voltammetric charge of the pure IrO2 electrode can be associated with the pseudocapacitive reactions at IrO2 and also, due to the presence of cracks and pores that increased the number of electrode active sites at the surface [35]. Figure 5b shows cyclic voltammograms of mixed xPtOy-(100-x)IrO2 electrodes in 0.1 M KOH medium from 0.77 to 1.47 V/RHE. The voltammograms of mixed electrodes have shapes similar to PtOy, indicating that PtOy is present on all the prepared mixed electrodes. However, the voltammetric charges of these electrodes are higher than that of PtOy. This increase in the voltammetric charge is due to the presence of IrO2 on the surface of mixed electrodes. At mole percentages of PtOy greater than 50 %, the voltammetric charge is lower than that of IrO2 but still greater than that of PtOy. For mole percentages of PtOy less than 50 %, the voltammetric charge is higher than that of PtOy and IrO2. The highest voltammetric charge is observed for PtOy with mol percentage of 40 %. The high voltammetric charge of mixed electrodes shows the increase in active sites due to the presence of both PtOy and IrO2.  The studies were also carried out at the lower scan rate because, at a low scan rate, the system is in a quasi-stationary state. In Figure 6, the linear voltammograms obtained with different xPtOy-(100-x)IrO2 electrodes in the basic medium of 0.1 M KOH are presented. These voltammograms show that the potential at j = 1 mA cm -2 related to oxygen evolution varies from one electrode to another. The OER potential was also determined at j = 10 mA cm -2 for each electrode. The potential values of all prepared xPtOy-(100-x)IrO2 electrodes at these two current density values are presented in Table 3.  Table 3 show that OER potentials for xPtOy-(100-x)IrO2 electrodes determined at j = 1 mA cm -2 and j = 10 mA cm -2 vary in a similar way. The highest potential values observed at any current density for PtOy electrode suggest that it is less electrocatalytically active for OER than other electrodes in the basic medium. Also, it seems that 90PtOy-10IrO2 electrode is more electrocatalytically active than PtOy, but less than IrO2 and other mixed electrodes. On the other hand, the electrodes with PtOy percentage lower or equal to 80 % have a potential related to OER close but less than pure IrO2. It is clear that the coupling of PtOy with IrO2 has improved electrocatalytic activity of the electrode. The electrodes which have PtOy percentage lower or equal to 80 % are the most electrocatalytically active for the oxygen evolution in the basic medium.

Figure 6. Linear voltammograms (5 mV s -1 ) of prepared electrodes in 0.1 M KOH medium
The Tafel slope of OER on an electrode can be abnormally high, especially within a high potential domain. According to Kapalka et al. [37], the increase of the Tafel slope is due to the ohmic drop.
The ohmic drop between the anode and the cathode is due to the geometry of the cell, solution conductivity, presence of bubbles in solution, and accumulation of gas bubbles produced on the surface of the electrode (blockage of the surface) [38,39]. Here, the ohmic drop corrections of the measured polarization curves were carried out according to the method already described in the literature [28,29]. Briefly speaking, the overpotential (η / V) observed in the experiment can be defined by where a / V is Tafel constant, b / V dec -1 is Tafel slope, j / A cm -2 is current density and R / Ω cm -2 is the total area-specific, uncompensated resistance of the system, assumed to be constant. The derivative of eq. (3) concerning current density, results in eq. (4) from which b and R can be easily obtained by plotting dη / dj as a function of 1 / j.
The estimation of R allows correcting the experimental overpotential by subtracting the ohmic drop jR according to the following equation: where ηcorr stands for the corrected overpotential.
During the calculations, the derivatives dη / dj were replaced by their finite elements Δη ⁄ Δj estimated from each pair of consecutive experimental points.
After correcting the ohmic drop for each electrode in the electrolyte used, the Tafel slope lines were determined from the corrected E vs. log j dependences that are for some xPtOy-(100-x)IrO2 electrodes presented in Figure 7a. Note that Tafel slopes vary from electrode to electrode, suggesting that the kinetics of OER varies from one electrode to another Thus, it can be seen from Figure 7a that the Tafel slope of pure IrO2 electrode is 40 mV dec -1 . According to Figure 7b, Tafel slope remains almost constant at 40 mV dec -1 for all mixed electrodes which have PtOy percentage lower or equal to 80 %. This result indicates that OER at IrO2 and all these mixed electrodes proceeds practically with the same kinetics. At the other side, Tafel slopes of 55 and 122 mV dec -1 were respectively obtained for 90PtOy-10IrO2 and PtOy electrodes. These results suggest that the electrodes which have the voltammetric response close to PtOy have also high Tafel slope similar to PtOy. This is the case for electrodes whose surfaces are rich in PtOy. Also, when the surface is dominated by IrO2, the Tafel slopes tend towards that of pure IrO2.
Generally, the values of the obtained Tafel slopes suggest that the kinetics of OER on mixed electrodes is faster on the electrodes having the percentage of PtOy lower or equal to 80 % than on the electrodes with the percentage of PtOy higher than 80 %.
Tafel slopes presented in Figure 7b suggest that mixed xPtOy-(100-x)IrO2 electrodes have probably the same reaction pathway for OER, but different rate determining steps. For PtOy anode, the first step described by eq. (6) can be considered as the rate determining step, because the observed Tafel slope is close to 120 mV dec -1 . On the pure IrO2 and the anodes with PtOy percentage less or equal to 80 %, the observed Tafel slope is 40 mV dec -1 , indicating that the step described by eq. (8) could be considered as the rate determining step. For the electrode containing 90 % of PtOy, the second step described by eq. (7) will be considered as the rate determining step, because its Tafel slope was found close to 60 mV dec -1 .

Electrochemical characterization of prepared electrodes in perchloric acid medium
The oxygen evolution has also been studied in an acid medium. The linear voltammograms of mixed xPtOy-(100-x)IrO2 electrodes in 0.1 M HClO4, recorded in a quasi-stationary manner are presented in Figure 8.  Figure 8 shows that the onset potential related to the oxygen evolution (the potential of the sharp current increase) varies from one electrode to another but is the highest for PtOy electrode.
The potential values at j = 1 mA cm -2 and j = 10 mA cm -2 related to OER were determined for all xPtOy-(100-x)IrO2 electrodes and the obtained values are represented in Table 4.
Data in Table 4 show that potential values of OER vary from one electrode to another similarly and regardless of the current density. PtOy electrode has the highest potential related to oxygen evolution, suggesting that PtOy is the least electrocatalytically active electrode for OER in an acid medium. Except for the electrode which contains 90 % PtOy, the mixed xPtOy-(100-x)IrO2 electrodes have lower onset potentials related to OER than either pure PtOy or pure IrO2. This pointed out that the mixed Pt-Ir oxide electrode in an acid medium is generally more electrocatalytically active for O2 evolution than its pure components. These results are fully consistent with those observed for OER in the basic medium. The Tafel slopes for OER in 0.1 M HClO4 were determined for all studied xPtOy-(100-x)IrO2 electrodes. As shown in Figure 9a, plotting of the corrected E vs. log j gave straight lines in the Tafel region. The particular values of Tafel slopes for all studied xPtOy-(100-x)IrO2 electrodes are recorded in Figure 9b. Tafel slope of 120 mV dec -1 was obtained for pure PtOy, while 60 mV dec -1 was obtained for the mixed electrode with 90 % of PtOy. The other mixed electrodes and pure IrO2 gave values of 40 mV dec -1 . All these values obtained in the acid medium (0.1 M HClO4) are almost identical to those obtained in the basic medium (0.1 M KOH). Also, the Tafel slopes show that the oxygen evolution reaction in the acidic and basic medium is faster on the electrodes which have a percentage in PtOy lower or equal to 80 % than on the electrodes which have a percentage in PtOy higher than 80 %.
The obtained Tafel slopes, together with the literature results [27,29,40] point towards the following mechanism of OER on the prepared electrodes in an acid medium: S + H2O → S-OH + H + + e - S-OH → SO + H + + e - 2SO → 2 S + O2 (12) where S stands for electrode active sites, while OH and O represent adsorbed intermediates. The mechanism described by eqs. (10)(11)(12) predicts the Tafel slope of 120 mV dec -1 if the first step described by eq. (10) is the rate determining step (rds), 40 mV dec -1 if the second step described by eq. (11) is rds, and 30 mV dec -1 if the third step described by eq. (12) is rds.
For OER on PtOy in HClO4 medium, the rate determining step is the first step described by eq. (10), because the Tafel slope is 120 mV dec -1 . For the pure IrO2 electrode and the mixed xPtOy-(100-x)IrO2 electrodes except that with 90 % PtOy, the rate determining step is the second step described by eq. (11), because their Tafel slopes are very close to 40 mV dec -1 . With the platinumiridium dioxide coupling with a proportion of 90 % PtOy, however, the Tafel slope is 60 mV dec -1 . This specific value is between that obtained for pure PtOy (120 mV dec -1 ) and IrO2 (39 mV dec -1 ), but closer to IrO2 than PtOy. This result is in agreement with the voltammogram for 90PtOy-10IrO2 electrode that is closer to IrO2 than PtOy.
Generally, Tafel slopes of mixed xPtOy-(100-x)IrO2 electrodes are close to that of IrO2 and it is just IrO2 which determines the oxygen evolution rate. The rate determining step is similar to that obtained with IrO2 electrode, but OER is slower on 90PtOy-10IrO2 anode than IrO2 anode. The possible reason could be the mixed kinetics due to the adsorption of unstable species such as S-OH* (equations 13a and 13b), or the synergistic effect of PtOy and IrO2 [22]. The mechanism of OER involving eq. (13) instead eq. (10) accounts for possible surface chemical rearrangements. Therefore, the elemental step described by eq. (13b) probably controls the overall rate of OER [41,42].
The electrodes with molar percentage of PtOy lower or equal to 80 % have the lowest Tafel slope values. This shows that the kinetics of OER is faster for these electrodes than for the electrodes which have PtOy percentage greater than 80 %. This result is in agreement with the voltammetric curves where the onset potential related to oxygen evolution reaction on these electrodes is lower than that obtained with the electrodes with PtOy content greater than 80 % mol. It is clear from these results that the electrodes which have a voltammetric behavior close to that of iridium oxide have faster OER kinetics than the electrodes which have the PtOy content greater than 80 % mol. When the mixed oxide surface is dominated by certain metal oxide, the slope of Tafel tends towards that oxide. These observations agree with the scanning electron micrographs because on the electrodes having content less or equal to 80 % there are prominent pores and cracks on these electrodes. This has increased the number of active sites, hence the rapid kinetics of oxygen production.
Presence of Pt in different oxidation states (Pt and PtO2) and presence of pure IrO2 is indicated for mixed coupled xPtOy-(100-x)IrO2 deposits, while their surface morphology was changed from the highly cracked and porous surface characteristic for pure IrO2 (x = 0) to more continuous and much less cracked PtOy (x = 100).
All prepared mixed coupled xPtOy-(100-x)IrO2 electrodes showed the metallic character by supporting perfectly the reversible ferri/ferrocyanide reaction taken as the redox probe. Voltammetric charges measured in the oxide region of potentials for all mixed coupled xPtOy-(100-x)IrO2 electrodes in 0.1 M KOH were found higher than for pure PtOy. For mixed electrodes having the mol percentage of PtOy equal or less than 50 % (x  50), the voltammetric charge is found even higher than for pure IrO2. The highest voltammetric charge was observed for the mixed of 40PtOy-60IrO2 electrode, i.e. the electrode having PtOy mol percentage of 40 % (x = 40). This could be ascribed to the increase of electrochemical surface area of the electrode due to the presence of both PtOy and IrO2.
In either basic (0.1 M KOH) or acid (0.1 M HClO4) media, all mixed xPtOy-(100-x)IrO2 electrodes were found more electrocatalytically active for oxygen evolution reaction (OER) than pure PtOy electrode. The mixed electrode having mol percentage of PtOy equal or lower than 80 % (x  80) showed similar, or for (x  40) even better electrocatalytic activity for OER than pure IrO2. Tafel slopes of mixed xPtOy-(100-x)IrO2 electrodes obtained from linear voltammograms corrected for ohmic drops, suggest faster OER on coupled mixed xPtOy-(100-x)IrO2 electrodes having (x  80) than pure PtOy. This suggests that just presence of IrO2 determines high oxygen evolution rate at coupled mixed xPtOy-(100-x)IrO2 electrodes.