TECHNICAL FIELD

[0001] The present disclosure relates to electrodes and more particularly to electrodes for use in electrolysis processes.

BACKGROUND

[0002] Hydrogen is emerging as an important part of the clean energy paradigm. It can be used as a high efficiency, low polluting fuel, as well as an energy carrier for moving, storing and delivering energy from renewable energy sources. Moreover, hydrogen can be produced from water, which is an abundant and relatively inexpensive feedstock.

[0003] A number of different processes can be used for producing hydrogen from water. These processes include water electrolysis, photo-electrolysis, photo-biological production and high temperature decomposition. Of these processes, water electrolysis is the most practical way to produce hydrogen using renewable energy sources, such as power from wind turbines and/or photovoltaic solar panels.

[0004] There are several types of water electrolysis processes that are used for hydrogen generation, namely alkaline electrolysis, proton exchange membrane (PEM) electrolysis, and solid oxide electrolytic cell (SOEC) electrolysis. Alkaline electrolysis and PEM electrolysis are more commonly utilized, with alkaline electrolysis being the most advanced and most suitable for large-scale implementation.

[0005] In water electrolysis, a DC electrical power source is connected to two electrodes, which are placed in water. Theoretically, a potential difference between the electrodes of 1 .23 volts will split water into hydrogen and oxygen. Hydrogen is produced at the cathode, while oxygen is produced at the anode. More specifically, there is a hydrogen evolution reaction ("HER") at the cathode and an oxygen evolution reaction ("OER") at the anode. The HER is a reduction reaction in which electrons (e") from the cathode are given to hydrogen cations to form hydrogen gas, while the OER is an oxidation reaction in which electrons are given to the anode and oxygen is generated.

[0006] While 1 .23 volts will theoretically split water, a higher voltage is required in practice. The amount that exceeds 1 .23 volts is called overpotential or overvoltage and represents lost energy or inefficiency. In the water electrolysis process, the largest overvoltage is the reaction overvoltage for the oxidation of water to oxygen at the anode. As such, significant effort has been exerted to reduce the overvoltage at the anode.

[0007] In alkaline electrolysis, the water contains an alkaline electrolyte, such as potassium hydroxide (KOH) or sodium hydroxide (NaOH), and the electrodes are separated by a diaphragm. In PEM electrolysis, the water is deionized, and the electrodes are separated by a solid polymer electrolyte that permits protons from the anode to pass to the cathode, while insulating the electrodes electrically.

[0008] Recent advancements in water electrolysis have resulted in the process being operated at a higher current density. In addition, renewable energy sources are often used to provide electrical power for performing water electrolysis. The use of higher current densities, coupled with the intermittent nature of renewable energy, puts enhanced stress on the electrodes, causing conventional electrodes to deteriorate more rapidly. As such there is a need for more robust electrodes that provide good performance, while being less susceptible to degradation under conditions of high current density and intermittent power, i.e. numerous power shutdowns. In particular, there is a need for an anode with reduced oxygen overvoltage that is not susceptible to degradation due to high current density and shutdowns.

[0009] EP 3296431 A1 describes an oxidized nickel foam electrode having a coating layer comprising crystalline nanoparticles which include nickel, nickel oxide and ion oxide. WO 2019/172160 A1 describes an alkaline water electrolysis anode comprising a conductive substrate having an intermediate layer made of nickel oxide and a catalyst layer. EP 3064614 A1 describes an anode for alkaline water electrolysis having a nickel substrate and a lithium-containing nickel oxide catalytic layer formed on the substrate. EP 3375906 A1 describes an anode for electrochemical reactions which includes a nickel substrate and a nickel oxide layer on the conductive substrate and an oxidation catalyst layer disposed on the oxide layer.

SUMMARY

[0010] In accordance with the present disclosure, an electrode is provided for use in an alkaline electrolysis process. The electrode includes a metal substrate and a catalytic layer disposed on the metal substrate. The catalytic layer includes nickel and nickel oxide and has a porosity less than about 1 m ² /g measured by BET. An active composition is disposed both on and within the catalytic layer. The active composition includes one or more metal compounds selected from the group consisting of a cobalt compound, an iridium compound, a rhodium compound, an iron compound, a platinum compound, a lithium compound and a manganese compound.

[0011] Also provided in accordance with the present disclosure is an alkaline water electrolysis unit that includes the electrode, which functions as an anode. The unit also includes a cathode and an electrolyte solution that is substantially free of chlorine.

[0012] A method of forming an electrode is also disclosed herein. In accordance with the method, a metal substrate is provided and a catalytic layer is formed on the metal substrate. The catalytic layer includes nickel and nickel oxide formed by thermal spraying, cold spraying or other surface finishing process such as laser cladding and electroplating, and has a porosity less than about 1 m ² /g (BET). An active composition is applied to the catalytic layer and is thermally decomposed. The active composition includes one or more metal compounds selected from the group consisting of a cobalt compound, an iridium compound, a rhodium compound, an iron compound, a platinum compound, a lithium compound and a manganese compound.

BRIEF DESCRIPTION OF THE DRAWINGS [0013] The features, aspects, and advantages of the present invention will become better understood with regard to the following description, appended claims, and accompanying drawings where:

[0014] Fig. 1 shows a schematic representation of a water electrolysis unit;

[0015] Fig. 2 shows a schematic representation of an electrode of the water electrolysis unit;

[0016] Fig. 3 shows a graph of average capacitance for a plurality of test electrodes having different catalytic layers;

[0017] Fig. 4 shows a graph of released charge for the test electrodes;

[0018] Fig. 5 shows a graph of capacitance versus released charge for the test electrodes.

[0019] Fig. 6 shows an SEM image of a cross-section of an electrode according to the present invention;

[0020] Fig. 7 shows an SEM/EDAX image of the nickel and indium distribution of the same region of the electrode of Fig. 6; and

[0021] Fig. 8 shows an SEM/EDAX image of the oxygen distribution of the same region of the electrode Fig. 6, overlaid to the SEM image of Fig. 6.

DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS

[0022] It should be noted that in the detailed descriptions that follow, identical components have the same reference numerals, regardless of whether they are shown in different embodiments of the present disclosure. It should also be noted that for purposes of clarity and conciseness, the drawings may not necessarily be to scale and certain features of the disclosure may be shown in somewhat schematic form.

[0023] The present disclosure is directed to an electrode 10 constructed for use in an electrolysis process. The electrode 10 is especially suited for use in an alkaline electrolysis process and, more particularly, as an anode in such a process. A unit 12 that may be used for performing alkaline water electrolysis is shown in Fig. 1. The unit 12 may include the electrode 10, functioning as an anode, and a cathode 14. The electrode (anode) 10 and the cathode 14 are separated by a diaphragm 16 and are disposed in a vessel 18 containing an electrolyte solution. The electrolyte solution used in the electrolysis process preferably does not include chlorine or other halides. The electrolyte solution may comprise water and potassium hydroxide (KOH) or water and sodium hydroxide (NaOH). The diaphragm 16 is composed of a microporous material with an average pore size of less than 1 pm, which allows ions to move between the electrode (anode) 10 and the cathode 14, but is impermeable to gas, thereby preventing the intermixing of generated hydrogen and oxygen gases. The diaphragm 16 may be composed of a porous polymer, such as polytetrafluoroethylene (PTFE), or a composite material comprising zirconium dioxide (ZrC ) and polysulfone, which at the time of writing is commercially sold under the trademark Zirfon®.

[0024] The reactions that occur at the electrode (anode) 10 and the cathode 14 are as follows: cathode: 2H2O + 2e~ - > H2 + 2OH’ anode: 2OH’ - > 1402 + H2O + 2e ^_

[0025] The cathode 14 is composed of conductive metal, which may be nickel, low carbon steel, or stainless steel. The cathode 14 may have a structure as described below for the electrode 10 and may be coated with nickel or a nickel-iron alloy.

[0026] Referring now to Fig. 2, the electrode 10 includes a substrate 22 upon which a catalytic layer 24 is disposed. The catalytic layer 24 has pores formed therein, but is only moderately porous. An active composition 26 is disposed in and/or on the catalytic layer 24. The active composition 26 may form a layer and/or its constituent components may be interspersed into the pores of the catalytic layer 24. In some embodiments, the active composition 26 does not form a layer and its constituent parts are distributed throughout the porosity of the catalytic layer 24. In other embodiments, the active composition 26 forms a layer on the catalytic layer 24 and its constituent parts are also distributed throughout the porosity of the catalytic layer 24. The catalytic layer 24 may be disposed over all of the outer surfaces of the substrate 22 and, similarly, the active composition 26 may be disposed in and/or on all of the catalytic layer 24. In Fig. 2, the electrode 10 is shown as having a rectangular configuration, with opposing major surfaces being covered with the catalytic layer 24 and the active composition 26. While not shown, end surfaces of the electrode 10 may also be covered with the catalytic layer 24 and the active composition 26. The catalytic layer 24 and the active composition 26 are applied to the substrate 22 so as to be substantially uniform in thickness and composition over the exterior of the substrate 22. Preferably, deiviations in thickness are mostly within a +/- 5 pm range.

[0027] Structurally, the substrate 22 may be punched plate, woven wire, wire mesh, metal sponge, expanded metal, perforated or unperforated metal sheet, flat or corrugated lattice work, spaced apart rods or strips, or other structural forms. The wire mesh has been found to work particularly well in the execution of the invention. The thickness of the substrate 22 is preferably from about 0.15 mm to about 3.0 mm and more preferably from about 0.6 mm to about 2.0 mm. Preferably, the substrate 22 has openings to reduce current density. The substrate 22 is composed of a conductive metal, such as iron, iron alloy, nickel, nickel alloy or stainless steel. Iron alloys that may be used include iron-nickel alloys, iron-chromium alloys and iron-nickel-chromium alloys. Nickel alloys that may be used include nickel-copper alloys and nickel chromium alloys. Preferably, the substrate 22 is comprised of nickel, a nickel alloy or an iron alloy. A nickel plain weave wire mesh has been found especially suitable for use as the substrate 22.

[0028] The catalytic layer 24 is formed from a metal using a suitable thermal spraying process, such as flame spraying, wire-arc spraying, plasma spraying, high velocity oxyfuel (HVOF) spraying, high velocity air-fuel (HVAF) spraying, detonation gun and combustion wire spraying. Wire-arc spraying and plasma spraying have been observed to work particularly well in the execution of the invention. The catalytic layer 24 may alternatively be formed from a metal using cold spraying or another type of surface finishing process, such as laser cladding and electroplating.

[0029] The metal from which the catalytic layer 24 is formed may be nickel or a nickel alloy and/or nickel oxide (NiO). However, nickel is preferred. In this regard, it has been found that using a nickel-aluminum alloy that is de-alloyed after deposition with a caustic solution produces a porous nickel layer that tends to be too brittle and deteriorates too quickly. Preferably, nickel is deposited on the substrate 22 using nickel powder in an atmospheric plasma spraying (APS) process, a cold gas spraying process or a wire-arc spraying process. More preferably, nickel is deposited on the substrate 22 using nickel powder in an APS process. The nickel powder that is used to form the catalytic layer 24 has a mean particle size of from about 10 pm to about 150 pm, more preferably from about 45 pm to about 90 pm.

[0030] Before the metal (e.g., nickel) powder is sprayed onto the substrate 22, the substrate 22 is treated to remove contaminants and condition its surface for receipt of the metal powder. The treatment may include removing oil and grease through a degreasing process using a solvent. A grit blasting process is preferably performed on the substrate 22 after any degreasing process has been performed. In the grit blasting process, abrasive media, such as chilled iron or aluminum oxide, is pressurized with compressed air and directed against the surfaces of the substrate 22. The grit blasting further removes any contaminants and roughens the surfaces to form pits and crevices that promote mechanical bonding of the metal powder that is sprayed on the surfaces. Debris from the grit blasting may be removed by vacuum cleaning, brushing and/or air blowing.

[0031] An APS process may be performed with a plasma torch in which a strong electric arc is generated between a positively charged pole (anode) and a negatively charged pole (cathode). The arc ionizes flowing process gas (such as a mixture of argon, hydrogen, nitrogen and helium) into a plasma state to form a plasma jet, which may have a temperature in range of from about 10,000°C to about 30,000°C. Typically, the anode is circular and helps form a nozzle through which the plasma jet is ejected. The anode may be composed of copper, while the cathode may be composed of thoriated tungsten. Metal (e.g. nickel) powder is fluidized in a carrier gas (which may include hydrogen) and fed into the plasma jet, melting the powder particles and propelling them at high speeds (such as between 300 to 550 m/s) onto the substrate 22. During their travel to the substrate 22, some of the molten metal particles are oxidized to form oxides, such as nickel oxide (NiO). As such, when the catalytic layer 24 is formed using APS, the catalytic layer 24 includes both nickel and nickel oxide.

[0032] The molten metal particles sprayed from the plasma torch impact the substrate 22, where they flatten into "splats", which shrink and solidify to form stacked lamellae with micro-features, such as pores. As the splats shrink and solidify, they mechanically bond to the substrate 22 through mechanical hooking. As such, the adhesion of the sprayed metal particles is primarily due to mechanical bonding.

[0033] In a cold spraying process, metal (e.g. nickel) powder is fluidized in a high- pressure stream of carrier gas, which is preheated to a temperature of up to 1 ,000°C and may be comprised of nitrogen, helium or air. The stream of carrier gas and metal particles is passed through a converging-diverging DeLaval nozzle, where the stream is cooled and accelerated to a supersonic velocity of over 1 ,000 m/s. The supersonic stream of metal particles and carrier gas is directed onto the substrate 22, where the metal particles undergo plastic deformation upon impact. The accelerated metal particles impact the substrate 22 with enough kinetic energy to induce mechanical and/or metallurgical bonding between the metal particles and the substrate 22. Unlike in a thermal spraying process, such as APS, the metal particles are not melted during the cold spraying process.

[0034] In some embodiments, after the catalytic layer 24 has been formed, such as by APS or cold spraying, the substrate 22 coated with the catalytic layer 24 is heated in an oven at a temperature of from about 300°C to about 500°C for a period of from about 10 to about 20 minutes. In some embodiments, the substrate 22 coated with the catalytic layer 24 may be heated in an atmosphere comprising hydrogen.

[0035] The catalytic layer 24 is formed so as to have a thickness on one side of the substrate 22 of between about 10 pm to about 300 pm. Surprisingly, it has been found that with a thickness of 50 pm or less on each side of the substrate 22, the electrode 10 exhibits surprisingly good properties, in that it may reduce the reversal currents generated during shutdowns. As such, the catalytic layer 24 may have a thickness of 50 pm or less and more specifically in a range of from about 10 pm to about 50 pm, even more specifically in a range of from about 30 pm to about 50 pm.

[0036] As set forth above, the catalytic layer 24 is only moderately porous. Porosity may be measured by different methods, such as the well-known Brunauer-Emmett- Teller (BET) method or mercury (Hg) porosimetry. The BET method determines the overall specific external and internal surface area of a porous solid (per unit mass) from a measurement of the physical adsorption of gas molecules (e.g. nitrogen) by the porous solid. The determined BET surface area in m ² /g provides a measure of porosity. Using the BET method, the catalytic layer 24 has a porosity of less than 1 m ² /g.

[0037] Porosity can be correlated with electrochemical properties. As is known, the double layer capacitance of an electrode is directly related to the total surface area of the electrode. With respect to the BET, the measurement of the double layer capacitance allows to more specifically gauge the “active porosity” of the electrode, i.e. the surface area that is actually accessible to the electrolyte for the electrochemical reaction. Therefore, in alternative or in addition to the the determined BET surface area in m ² /g, the porosity of catalytic layer 24 may be characterized in terms of its double layer capacitance, before or after being coated with the active composition 26, normalisied by the loading of the catalytic layer 24. The loading of the catalytic layer 24 is measured as the amount of metal (in grams) per cm ² . The normalised double layer capacitance of the catalytic layer 24 is between from about 1 .0 to about 10.0 mF/g, and preferably between from about 3.0 to about 7.0 mF/g. The double layer capacitance before normalization is measured in mF/cm ² according to the procedure set forth below. Typical loading of the catalytic layer 24 is 50 and 1200 g/m ² , more preferably between 100 and 600 g/m ² Lower loadings of the catalytic layer 24 are more cost efficient as they obviously require less material in the catalytic layer but also less of the the usually costly active composition 26, which otherwise may get excessively diluted. In some applications, for example, without limitation, applications involving leaching of the active composition 26, typical loadings may be higher, between 100-3000 g/m ² referred to the metal.

[0038] The double layer capacitance of catalytic layers 24 having different compositions was studied in relevant conditions, i.e. KOH 25% and T = 80°C, using test electrodes 11 . Each test electrode 11 included a substrate 22 of nickel expanded mesh. In a first test electrode 11 a, the substrate 22 was plasma coated with nickel; in a second test electrode 11 b, the substrate 22 was plasma sprayed with a nickel-aluminum alloy; in a third test electrode 11 c, the substrate 22 was wire-arc sprayed using two nickel wires; in a fourth test electrode 11 d, the substrate was wire-arc sprayed with two wires, each composed of a nickel-aluminum alloy. The test electrodes 11 were initially conditioned by performing 10 minutes polarization at 2 kA/m ² Successively, voltammetric scans were performed in the potential region from 0.25 V to 0 V vs Hg/HgO at different scan rates (it is understood that, in general, the skilled person will select appropriate potential regions compatible with the active composition 26, in order to avoid interference due to possible faradic processes). The electrochemical double layer capacitance was estimated using the standard equation dE/dt = Cdi ■ dQ/dt. The test electrodes 11 showed a specific capacitance (normalized by the catalytic layer 24 load) in a range of from about 0.5 to about 30 mF/g, as shown in Fig. 3, which plots the average capacitance versus electrode type (11 a-11d). Notably, the test electrodes 11a, 11 c, 11 d were desirably in a range of from about 1.0 to about 10.0 mF/g. The electrodes 11 b and 11d containing Al were preventively leached in a KOH (30%) solution at 80°C for 2 hours. [0039] The test electrodes 11 were characterized by the amount of charge they could release when subjected to a reverse current in relevant conditions, i.e. KOH 25% and T = 80°C. The test electrodes 11 were each initially conditioned by performing 2 hours of anodic polarization at 10 kA/m ² Successively, a chronopotentiometric step was performed, by applying a cathodic current of 120 A/m ² until the electrode potential was recorded as -0.8V vs SHE (standard hydrogen electrode). The total amount of charge released from the test electrode 11 was then calculated by integrating the current over time. The test electrodes 11 were able to release a controlled charge (normalized by the coating load) in a range of from about 1 ,000 to about 70,000 mC/g, as shown in Fig. 4, which plots the normalized average charge versus electrode type (11 a-11 d). Notably, the test electrodes 11 a, 11 c and 11 d were desirably in a range of from about 2,000 to about 10,000 mC/g.

[0040] In certain embodiments, more than 90 percent by weight of nickel in the catalytic layer is in its metal form while up to 10 percent by weight of nickel is in its oxide form, predomently NiO, i.e. in the latter case, the weight percentage refers to NiO. Preferably, the metallic nickel content in the catalytic layer is in the range from 90 to 99.5 wt%, preferably in the range from 95 to 99 wt.%. The nickel oxide content in the catalytic layer can be in a range from at least 0.5 wt% to 10 wt% and is preferably in the range from 1 to 5 wt%.

[0041] It is believed that the optimal performances of the electrode configuration is due to the controlled charge release from the porous nickel layer. Also, a certain amount of porosity of the nickel/nickel oxide layer is desired in order to enhance the number of accessible catalytic sites. It can be expected that the relation between porosity and charge release will vary depending on nature of the layer. The control of the porosity/capacitance is used to control electrode robustness (charge). As shown in Fig. 5, which plots average capacitance (y-axis) versus average charge (x-axis) for electrodes 11 a-11 d, a linear relationship exists between capacitance and released charge for the electrodes 11 having a catalytic layer 24. It was thus demonstrated that the examined electrode configurations are characterized by the same, appropriate relationship between porosity/capacitance and charge release.

[0042] The active composition 26 includes one or more metals selected from the group consisting of iridium (Ir), cobalt (Co), rhodium (Rh), Iron (Fe), Platinum (Pt), Lithium (Li), manganese (Mn). The indium in the active composition 26 may be an iridium compound, such as indium oxide (IrC ). The cobalt in the active composition 26 may be a cobalt compound, such as a cobalt oxide, namely cobalt (II) oxide (CoO) or cobalt (II, III) oxide (CO3O4), or nickel cobaltite (NiCo2O4). Preferably, cobalt that is in the active composition 26 is nickel cobaltite. Even more preferably, nickel cobaltite that is in the active composition 26 is in the form of spinels. The rhodium in the active composition 26 may be a rhodium compound, such as rhodium (III) oxide (Rh20s); the manganese in the active composition 26 may be a manganese compound, such as manganese oxide (MnO ₂ ); the iron in the active composition 26 may be an iron compound, such as iron oxide (Fe2O ₃ ) or FesC ; the platinum in the active composition 26 may be a platinum compound, such as platinum oxide (PtO ₂ ); the lithium in the active composition 26 may be a mixed compound, such as lithium nickel oxide (LiNiC>2).

[0043] In some embodiments, the active composition 26 may consist essentially of a cobalt compound, a nickel compound, an iridium compound, a lithium compound (e.g. lithium nickel oxide), or an iron compound (e.g. iron oxide). Alternately, the active composition 26 may consist essentially of a cobalt compound and an indium compound, or the active composition 26 may comprise a cobalt compound or an iridium compound or both a cobalt compound and an iridium compound. Preferably, the active composition 26 comprises a cobalt compound or an iridium compound, more preferably both a cobalt compound and an indium compound, still more preferably the active composition 26 comprises nickel cobaltite and indium oxide. Thus, the active composition 26 may comprise: 0-100 mole percent of an indium compound, 0-100 mole percent of a cobalt compound and 0-40 mole percent of one or more transition metal compounds selected from the group consisting of a rhodium compound, a manganese compound, an iron compound, a platinum compound, a lithium compound. More preferably, the active composition 26 may comprise: about 10-50 mole percent of an indium compound, about 40-90 mole percent of a cobalt compound and 0-20 mole percent of one or more transition metal compounds selected from the group consisting of a rhodium compound, a manganese compound, an iron compound, a platinum compound, a lithium compound. More preferably, the active composition 26 may comprise: about 10-30 mole percent of an indium compound, more than 70 mole percent of a cobalt compound and 0-10 mole percent of one or more transition metal compounds selected from the group consisting of a rhodium compound, a manganese compound, an iron compound, a platinum compound, a lithium compound. Still more preferably, the active composition 26 may comprise: 10-30 mole percent of an indium compound, 70-90 mole percent of a cobalt compound and 0-10 mole percent of one or more transition metal compounds selected from the group consisting of a rhodium compound, a manganese compound, an iron compound, a platinum compound, a lithium compound.

[0044] The catalytic layer 24 and the active layer 26, taken together as a combination, comprise from about 80 mole percent to about 99.9 mole percent nickel, from 0 to about 2 mole percent nickel oxide, from about 0 mole percent to about 0.8 mole percent of an iridium compound, from about 0 mole percent to about 2 mole percent of a cobalt compound, and from about 0 to about 2 mole percent of one or more transition metal compounds selected from the group consisting of a rhodium compound, a manganese compound, an iron compound, a platinum compound, a lithium compound.

[0045] The active composition 26 may be formed from one single precursor composition that is applied to the catalytic layer 24 in one or more application processes, or the active composition 26 may be formed from a plurality of different precursor compositions that are applied to the catalytic layer 24 in a plurality of application processes. As set forth above, the active composition 26 may form a layer on top of the catalytic layer 24 as schematically depicted in Fig. 2 and/or the active composition 26 may be absorbed into the catalytic layer 24 and become distributed throughout the porosity of the catalytic layer 24. In this manner, the layer of the active composition 26 may have a thickness from about 0 to about 20 pm, more preferably from about 0 to about 10 pm.

[0046] Generally, the precursor composition(s) include precursor compounds (such as organic or inorganic metal salts) that are thermally decomposable to form the metal compounds in the active composition 26. The metal salts may be chlorides and may be dissolved in solvents, including acids (such as hydrochloric acid and nitric acid) and alcohols, such as isopropyl alcohol, n-propyl alcohol, n-butyl alcohol and ethyl alcohol. Some examples of the precursor compound(s) are given below.

[0047] For iridium oxide, the precursor compound(s) may include the chloride, sulfate or nitrate salt of iridium dissolved in a solvent, such as an acid or an alcohol. More specifically, the precursor compound(s) may include a solution of iridium (III) chloride (IrCh) in its trihydrate form (lrCl3(H2O)3), together with hydrochloric acid and isopropyl alcohol and/or ethyl alcohol.

[0048] For cobalt (II) oxide, the precursor compound(s) may include cobalt chloride (C0CI2) and water. For cobalt (II, III) oxide, the precursor compound(s) may include cobalt (II) acetate tetrahydrate (C4HeCoO4_4H2O), ethyl alcohol and oxalic acid (C2H2O4).

[0049] For nickel cobaltite, the precursor compound(s) may include nickel (II) acetate tetrahydrate (C4HeNiO4 _4H2O) and cobalt (II) acetate tetrahydrate (C4HeCoO4_4H2O) mixed with urea (CO(NH2)2, deionized water, ethyl alcohol, glycerol and tetraethylene glycol (TEG), wherein the nickel (II) acetate and the cobalt (II) acetate are co-precipitated. Alternately, the precursor compound(s) may include nickel (II) nitrate hexahydrate (Ni(NO3)2*6H2O) and cobalt(ll) nitrate hexahydrate (Co(NO3)2*6H2O), which may be dissolved in NH4OH. [0050] The precursor composition(s) may be applied to the catalytic layer 24 using a brush, or by electrospraying, roller coating or dip coating.

[0051] After the precursor composition(s) have been applied to the catalytic layer 24, the substrate 22 coated with the catalytic layer 24 and the precursor composition(s) is heated to thermally decompose the precursor compound(s) to form the active composition 26. The heating may take place in an oven at a temperature of from about 300°C to about 500°C for a period of from about 10 to about 20 minutes.

EXAMPLE 1

[0052] A 100X100 mm piece of nickel wire woven mesh, with a 0.17 mm diameter wire, was plasma sprayed with 99.9% purity nickel powder with a particle size of -45 / +10 pm (Fe <0.5, CKO.4, C<0.02, S<0.01 in ambient air on both sides in an amount of 4.8 ± 0.5 g/dm ² and with a target thickness of 50 pm (on each side). Afterwards, the sprayed wire mesh was heated in an oven at 350°C for 15 minutes in air. The plasma-sprayed woven mesh was allowed to cool and then was coated with a precursor composition, by means of a brush, in a series of coating, heating and cooling steps. The precursor composition comprised 19.77 g of nickel(ll) nitrate hexahydrate and 39.56 g of cobalt(ll)/nitrate hexahydrate, brought to a volume of 100mL with DI water. Initially, the plasma-sprayed wire mesh was coated with the precursor composition and then heated in air in an oven at 350°C for 15 minutes. After cooling, the wire mesh was again coated with the precursor composition and again heated in the oven at 350°C for 15 minutes. This process of coating, heating and cooling was repeated to achieve a load of 7.52 gMETAL/m ² .

[0053] The foregoing process produced an electrode E1 having, on each side, a catalytic layer and an active composition. The catalytic layer was mostly nickel with a small amount of NiO. The active composition comprised 0.0427 mol/ m ² of nickel cobaltite. When viewed with a scanning electron microscope (SEM), the catalytic layer was a nickel matrix formed by nickel particles and a thin crust of NiO, with the nickel cobaltite being distributed into the porosity of the catalytic layer.

[0054] The combination of the catalytic layer and the active composition had a thickness (on one side) of 51 ± 6 pm and had a porosity of 0.3 m ² /g (BET). The combination comprised 97.6 mole percent nickel, 0.9 mole percent nickel oxide and 1.5 mole percent nickel cobaltite.

EXAMPLE 2

[0055] A 100X100 mm piece of nickel expanded mesh was plasma sprayed with 99.9% purity nickel powder with a particle size of -45 / +10 pm (Fe <0.5, CKO.4, C<0.02, S<0.01 in ambient air on both sides in an amount of 4.5 ± 0.5 g/dm ² and with a target thickness of 45 pm (on each side). Afterwards, the sprayed wire mesh was heated in an oven at 350°C for 15 minutes in air. The plasma-sprayed expanded mesh was allowed to cool and then was coated with first and second precursor compositions, as described below.

[0056] The first precursor compostion was the same as the precursor composition applied in Example 1 , i.e. the solution of nickel (II) nitrate hexahydrate and cobalt (II) nitrate hexahydrate. The first precursor composition was also applied to the mesh in the same manner as in Example 1 to achieve a load of 7.52 gMETAi_/m ² .

[0057] The second precursor composition comprised 4.27g of hexa- ammineiridium(lll) hydroxide [lr(NHs)6](OH)3 brought to a volume of 10OmL with deionized water. After the application of the first precursor composition, the second precursor composition was applied by brush in a series of coating, heating and cooling steps. Initially, the plasma-sprayed and coated mesh was coated with the second precursor compositions and then heated in an oven at 350°C for 15 minutes in air. After cooling, the mesh was again coated with the second precursor composition and again heated in the oven at 350°C for 15 minutes. This process of coating, heating and cooling was repeated to achieve a load of 1 .88 gMETAi_/m ² .

[0058] The foregoing process produced an electrode E2 having, on each side, a catalytic layer and an active composition. The catalytic layer was mostly nickel with a small amount of NiO. The active composition was 9.78 mmol/m ² of indium oxide and 42.7 mmol/m ² of nickel cobaltite, which corresponds to 18.6 mole percent indium oxide and 81 .4 mole percent nickel cobaltite. When viewed with a scanning electron microscope (SEM), the catalytic layer was a nickel matrix formed by nickel particles and a thin crust of NiO, with the iridium oxide and the nickel cobaltite being distributed into the porosity of the catalytic layer. The combination of the catalytic layer and the active composition had a thickness (on one side) of just less than 50 pm and had a porosity of 0.4 m ² /g (BET). The combination comprised comprised 98.8 mole percent nickel, 0.12 mole percent iridium oxide and 0.53 mole percent cobaltite.

EXAMPLE 3

[0059] A 100X100 mm piece of nickel expanded mesh was plasma sprayed with 99.9% purity nickel powder with a particle size of -45 / +10 pm (Fe <0.5, CKO.4, C<0.02, S<0.01 in ambient air on both sides in an amount of 4.5 ± 0.5 g/dm ² and with a target thickness of 45 pm (on each side). Afterwards, the sprayed expanded mesh was heated in an oven at 350°C for 15 minutes in air. The plasma-sprayed expanded mesh was then allowed to cool and then was coated with a precursor composition, by means of a brush in a series of coating, heating and cooling steps. The precursor composition was the same as the second precursor composition of Example 2, i.e., hexa-ammineiridium(lll) hydroxide [lr(NH3)e](OH)3 brought to a volume of 100mL with deionized water. Initially, the plasma-sprayed expanded mesh was coated with the precursor composition and then heated in an oven at 350°C for 15 minutes in air. After cooling, the mesh was again coated with the precursor composition and again heated in the oven at 350°C for 15 minutes. This process of coating, heating and cooling was repeated to achieve a load of 1.88 gMETAL/m ² .

[0060] The foregoing process produced an electrode E3 having, on each side, a catalytic layer and an active composition. The catalytic layer was mostly nickel with a small amount of NiO. The active composition was 9.78 mmol/m ² of iridium oxide. When viewed with a scanning electron microscope (SEM), the catalytic layer was a nickel matrix formed by nickel particles and a thin crust of NiO, with the indium oxide being distributed into the porosity of the catalytic layer (see Figures 6 - 8). The combination of the catalytic layer and the active composition had a thickness (on each side) of just less than 50 pm and had a porosity of 0.4 m ² /g (BET). The combination comprised comprised 98.7 mole percent nickel, 0.9 mole percent nickel oxide and 0.4 mole percent iridium oxide.

COMPARATIVE EXAMPLE 1

[0061] A 100X100 mm piece of nickel expanded mesh was plasma sprayed with 99.9% purity nickel powder with a particle size of -45 / +10 pm (Fe <0.5, O<0.4, C<0.02, S<0.01 in ambient air on both sides in an amount of 13.0 ± 0.9 g/dm ² and with a target thickness of 300 pm and 100 pm on the other side. Afterwards, the sprayed mesh was heated in an oven at 350°C for 15 minutes. The sprayed expanded mesh was then allowed to cool, thereby forming an electrode CE1 with only a catalytic layer comprising nickel and nickel oxide. The catalytic layer had a thickness (on each side) of just less than 200 pm and had a porosity of 0.4 m ² /g (BET).

COMPARATIVE EXAMPLE 2

[0062] A 100X100 mm piece of nickel expanded mesh was plasma sprayed with 99.9% purity nickel powder with a particle size of -45 / +10 pm (Fe <0.5, O<0.4, C<0.02, S<0.01 in ambient air on both sides in an amount of 4.5 ± 0.5 g/dm ² and with a target thickness of 45 pm. Afterwards, the sprayed expanded mesh was heated in an oven at 350°C for 15 minutes. The sprayed wire mesh was then allowed to cool, thereby forming an electrode CE2 with only a catalytic layer comprising nickel and nickel oxide. The catalytic layer had a thickness (on one side) of just less than 50 pm and had a porosity of 0.4 m ² /g (BET).

[COMPARATIVE EXAMPLE 3]

[0063] Except as noted below, Example 2 was re-performed on another 100X100 mm piece of nickel expanded mesh. The mesh was not plasma sprayed, or otherwise coated with a layer of nickel/nickel oxide or other metal. The active composition was formed directly on the expanded mesh. The same first and second precursor compositions were applied to the expanded mesh in the same manner as in Example 2, except the there was no plasma sprayed layer.

[0064] The process produced an electrode CE3 having, on each side, an active composition. The active composition was 9.78 mmol/m ² of iridium oxide and.42.7 mmol/m ² of nickel cobaltite, which corresponds to 18.6 mole percent indium oxide and 81 .4 mole percent nickel cobaltite.

TESTING

[0065] The electrodes E1 , E2, E3, , CE1 , CE2 and CE3 were each used as anodes in electrolysis cells and the oxygen overvoltages at the anodes were measured in KOH 25% at 80°C with a three electrodes set-up, using a SCE reference electrode and a VMP3 Biologic potentiostat equipped with a 20A Booster. The iR drop was measured by means of Electrochemical Impedance Spectroscopy (EIS). The resulting electrode potential determined at 10 kA/m ² was subtracted by the thermodynamic oxygen evolution potential at pH14. Results of these measurements are shown in Table 1 below. In addition, electrolysis cells with the E2, E3, CE1 , CE2 and CE3 anodes, respectively, were connected to receive electric power having a current of 10 kA/m ² and were then subjected to a series of 50 shutdowns of electric power, with on/off cycles of the duration of 6 hours. The voltages at 10 kA/m ² of the cells were measured after each shutdown. The results of these measurements are also shown in Table 1 below. TABLE 1

[0066] The results of the tests show that while the catalytic layer of electrode CE2 is substantially thinner than the catalytic layer of electrode CE1 («50 m versus «200 pm), the oxygen overvoltage of electrode CE2 was not that much greater than electrode CE1 (380 mV versus 350 mV). The electrodes E1 -E3 had substantially better oxygen overvoltages than both the electrodes CE1 and CE2. The electrode E2 had significantly better oxygen overvoltage than CE1 and CE2 and had better shutdown tolerance than the electrodes E3, and CE3. The electrode CE3 had good oxygen overvoltage, but not very good shutdown tolerance. The results show that an electrode with only the active composition 26, without the catalytic layer 24, has a good oxygen overvoltage, but poor shutdown tolerance. Conversely, electrodes with only catalytic layer 24, without the active composition 26, has good shutdown tolerance, but poor overvoltage. The synergic combination of catalytic layer 24 and active composition 26 yields an electrode with enhanced properties, i.e. both good shutdown tolerance and good overvoltage., particularly an active layer comprising both a cobalt compound and an iridium compound. Such an electrode (e.g. electrode E2) also has an improved oxygen overvoltage.

[0067] Figures 6 to 8 show SEM/EDAX images of a cross-section of an electrode according to Example 3 which comprises a nickel substrate 30, a catalytic layer 31 made from nickel and a small amount of nickel oxide and an iridium-based active composition thermally applied to the catalytic coating. The SEM image of Fig. 6 shows the morphology of the catalytic layer which was obtained by depositing nickel on a substrate made of a coarse nickel mesh via thermal spray. As can be taken therefrom, the catalytic layer 31 which exhibits some porosity is clearly distinguished by a well-defined interface 32 from the dense nickel substrate 30 which exhibits essentially no porosity. The porosity of the catalytic layer 31 essentially results from cracks 33 being formed within the catalytic layer 31. Predominantly, these cracks 33 are essentially orientated parallel to the interface 32 between substrate 30 and catalytic layer. The black area 34 corresponds to a resin required for sample preparation. The active composition (not visible in Fig. 6) can penetrate into the catalytic layer via cracks 33 as can be taken from the SEM/EDAX image of Fig. 7 which shows the same cross-sectional region of the electrode as Fig. 6. Gray areas correspond to nickel of the catalytic layer while white areas/dots correspond to iridium of the active composition. Although the indium EDAX map ay is quite noisy, distinct patches with increased indium content, indicated by black circles in Fig. 7, are clearly visible indicating that the active composition has penetrated into the catalytic layer and accumulates in regions corresponding to porous regions in Fig. 6. Fig. 8 is an SEM/EDAX image showing the oxygen distribution in the same region of the electrode of Fig. 6. The oxygen EDAX image (white) is overlaid to the SEM image (gray) of Fig. 6. Disregarding noise, it can be taken from Fig. 8 that the substrate 30 consists of pure nickel while the catalytic layer 31 comprises patches containing oxygen (black circles) which correspond to the nickel oxide component of the catalytic layer.

[0068] It is to be understood that the description of the foregoing exemplary embodiment(s) is (are) intended to be only illustrative, rather than exhaustive. Those of ordinary skill will be able to make certain additions, deletions, and/or modifications to the embodiment(s) of the disclosed subject matter without departing from the spirit of the disclosure or its scope.