A method for pretreating a nickel foam for alkaline electrolysis

Abstract

The invention relates to a method for pretreating a nickel foam (NF) for use as an electrode in alkaline electrolysis by a) performing a pretreatment by exposing the NF to a vaporized solution of iron(III) (Fe3+), such as ferric nitrate, or b) performing a pretreatment by exposing the NF to a liquid solution of iron(III) (Fe3+) for a duration of 59 second or less. The pretreatment (a, b) results in growth of an electro-catalytic iron compound on, or in, the nickel foam (NF) surface so as to reduce an overvoltage for water electrolysis after for example 10 seconds with spraying of ferric nitrate. Thus, the invention yields a fast method for pretreatment of nickel foam for the electrodes in an alkaline electrolysis system for water electrolysis with significant reductions in the required overvoltage, e.g. in some embodiments an improvement of 167 mV is observed.

Description

Technical field of the invention

[0001] The present invention relates to a method for pretreating a nickel foam for alkaline electrolysis increasing the energy efficiency for hydrogen formation by chemical modification of the anode. In particular, the present invention relates to a corresponding system for performing alkaline water electrolysis (AWE), and a corresponding nickel foam suitable for use in such an alkaline electrolysis system.Background of the invention

[0002] Hydrogen is a very interesting and important part of the energy system - especially in the upcoming Power-to-X (PtX) strategies. By converting green electricity from wind turbines, photovoltaic cells and water-based power plants into hydrogen, the energy can either be used directly in the transport sector, stored for later use, or converted together with carbon dioxide (CO2) into so-called green methane, green methanol, green diesel, green aviation fuel, etc. Hereto comes that it is relatively cheaper to transport energy as hydrogen through piping systems as compared to transporting energy as electricity through copper wires through the electric grid.

[0003] Water electrolysis can be broadly classified into three different technologies: Solid Oxide Electrolysis (SOE), Proton Exchange Membrane (PEM) Electrolysis and Alkaline Water Electrolysis (AWE). Furthermore, on a more academic level, water can also be split into H2 and O2 by thermolysis, photolysis and microbial electrolysis although these technologies are much less matured.

[0004] Hydrogen production by alkaline water electrolysis (AWE) is a well-established technology up to the megawatt level. H2 is produced at the cathode surface forming hydroxyl ions or hydroxide (OH-), which migrate through a porous diaphragm membrane to the anode under the influence of the electrical field between the anode and cathode. Subsequently, two hydroxyl ions (OH-) are discharged forming oxygen (02) and water (H2O). Oxygen atoms combines at the surface of the anode and escapes as oxygen (02).

[0005] The underlying reactions are: Anode reaction:2OH- → H2O + 1 / 2O2 + 2e-(1)Cathode reaction:2H2O + 2e- → H2 + 2OH-(2)Overall cell reaction:H2O → H2 + 1 / 2O2(3)

[0006] The chemical reaction (1) generating oxygen is often called the oxygen evolution reaction (OER), and the chemical reaction (2) generating hydrogen is similarly called the hydrogen evolution reaction (HER). The thermodynamic voltage for this combined water splitting is 1.23 V (at 25 °C / 1 atm), but in reality significantly higher voltage is needed for effectively driving the water splitting under practical and industrial scale of operation i.e. a so-called overvoltage or overpotential is needed. Hereto comes also the ohmic resistance of the electrolyte, the formed bobbles, and the used membrane.

[0007] The anode / cathode for alkaline water electrolysis is typically a nickel-based material. Nickel is stable in the concentrated KOH solution (typically 20-35 wt.%). Lifespans beyond 25 years and decades of stability of Ni-based electrodes in alkaline water electrolysis have been reported. Conventionally, porous nickel, such as e.g. a nickel foam (NF) or different Ni alloys, has been used as standard electrode material to increase the surface area.

[0008] There is continuously ongoing research and development activities focusing on improved electrodes for alkaline electrolysis by lowering the overpotential for hydrogen and oxygen formation in the above chemical reactions, respectively (2) and (1).

[0009] Overall, two main strategies may be used for applying an electrochemical catalyst at the electrodes for water splitting; (1) the first and most common approach is to electrodeposit a powder-based catalyst on one or both electrodes, but typically a conductive binder is needed, which may block electrochemically active sites and have limited stability over time, especially with high currents. (2) Secondly, the electrodes may be improved by applying noble metals, typically platinum group metals (PGM), with highly active electrocatalytic sites on the substrate of the electrodes, e.g. nickel-foam, copper foils, stainless-steel, etc. However, the complications with controlling the produced electrocatalytic sites, and the costs of using noble metals are normally not feasible for industrial scale electrolysis. For a recent review of electrocatalysts for water splitting in electrolysis, the skilled reader is for example referred to "Nanostructured materials on 3D nickel foam as electrocatalysts for water splitting" by Chaudhari et al. in Nanoscale, 2017, 9, 12231.

[0010] In Nano Research, November 2017, Yin & Jiang et al., reported that a simple immersion of nickel foam (NF) into ferric nitrate could enhance the anode performance remarkably by creating an amorphous layer of Ni-Fe Hydroxide (α-NiFeOxHy) on a nickel-based substrate, i.e. not by adding a powder-based catalyst. By immersing NF into an aqueous ferric nitrate solution, the overpotential was shown to have a potential decrease of between 101-175 mV (measured at 10mA cm-2 in 1M KOH at ambient temperatures), enhancing the activity of the anode, and thereby, lowering the overpotential for the oxygen evolution reaction (OER) during hydrogen production. The results also showed this worked with different concentrations of ferric nitrate solutions. Furthermore, it was also reported that there was a significant improvement in the performance of the anode after immersion for longer periods of time, and it was concluded that a longer immersion time results in further increase in the OER activity. However, if the immersion time increases using the method suggested from this paper in Nano Research, this method is not very useful in industrial applications, where large volumes of nickel foam has to be prepared for the electrodes in a fast and efficient manner.

[0011] Hence, an improved method for pretreating a nickel foam for alkaline electrolysis would be advantageous, and in particular a more efficient and / or reliable method would be advantageous.Summary of the invention

[0012] Thus, an object of the present invention relates to providing an improved method for pretreating a nickel foam for alkaline electrolysis.

[0013] In particular, it is an object of the present invention to provide a faster and more efficient method that solves the above mentioned problems of the prior art with preparing and upscaling a pretreatment of nickel foam to industrial implementation.

[0014] Thus, one aspect of the invention relates to a method for pretreating a nickel foam (NF) for alkaline electrolysis, the method comprising: providing a NF, the NF being suitable for functioning as a first and / or a second electrode in an alkaline electrolysis system for water electrolysis, and a) performing a pretreatment of said NF by exposing the NF to a vaporized solution of iron(III) (Fe 3+< ), alternatively or additionally a solution of iron(II) (Fe 2+< ), or b) performing a pretreatment of said NF by exposing the NF to a liquid solution of iron(III) (Fe 3+< ), alternatively or additionally a solution of iron(II) (Fe 2+< ), for a duration of 59 second or less, wherein said pretreatment (a, b) results in growth of an electro-catalytic iron compound on, or in, the nickel foam (NF) surface so as to reduce an overvoltage for water electrolysis.

[0015] The invention is particularly, but not exclusively, advantageous for providing a faster method for pretreatment of nickel foam for the electrodes in an alkaline electrolysis system for water electrolysis with significant reductions in the required overvoltage, e.g. in some embodiments an improvement of 167 mV is observed.

[0016] In particular, it is an advantage of the present invention that it provides a relatively simple method that solves the above mentioned problems of the prior art with industrial scale pretreatment of nickel foam for the electrodes in a fast and efficient way, i.e. fast pretreatment on a scale of seconds can be sufficient, which will be apparent from the obtained experimental results explained further below.

[0017] Firstly, it is highly surprising starting from the results presented in Nano Research, November 2017, Yin & Jiang et al., that a short pretreatment of nickel foam according to the present invention is sufficient to electrochemically activate the nickel surface because numerous places in the 2017 article, it is explained that the pretreatment in Fe(NO 3 ) 3 should be relatively long, e.g. "A longer immersion time results in further increased OER activity" cf. abstract. Thus, there is an explicit teaching away in Nano Research, November 2017, Yin & Jiang et al., for the skilled person to consider performing a pretreatment of nickel foam at a short time scale of just a few seconds, or for example by providing the iron (III) in vaporised manner, i.e. by a spraying method.

[0018] Secondly, the surprising element of the present invention is further emphasised by the advantage of reduced corrosion damage on the nickel foam by the short pretreatment or vaporised way of providing the iron(III), e.g. with a ferric nitrate modification. Thus, from the teaching of Nano Research, November 2017, Yin & Jiang et al., it is evident that the controlled but negative corrosive effect on the nickel foam of the Fe(NO 3 ) 3 pretreatment is to be balanced with the advantageous effect of reduced overvoltage, but there is an inherent conflict between these two effects / processes. Nevertheless this is resolved by the present invention, where little or no corrosion will take place while still providing a significant reduction of the overvoltage.

[0019] It is contemplated that the short exposure of the nickel foam by either a spray, or a short immersion, with a Fe(III) solution may similarly be implemented with a short exposure to a Fe(II) solution, such as FeSO 4 , FeCl 2 , etc., as shown in Nano Research, November 2017, Yin & Jiang et al., though this has not been tested by the present inventors. However, more than sufficient experimental evidence is provided below to satisfactorily demonstrate the usability of Fe(III) solutions to quickly and efficiently perform pretreatment of a nickel foam for much improved alkaline electrolysis. Thus, there is reason to believe that a Fe(II) solution may be also be applied quickly, i.e. by a) spray or b) immersion for 59 seconds or less, to yield improved results for alkaline electrolysis. It is contemplated that a combined solution with both Fe(II) and Fe(III) may be applied with advantageous results.

[0020] Without being bound to any specific theory, it is therefore contemplated that the suggested mechanism in Nano Research, November 2017, Yin & Jiang et al. also to some extent may explain the results of the present invention, i.e. that short exposure to Fe(NO 3 ) 3 results in the formation of a similar nickel-iron-hydroxide compound, preferably on the form FeNiO x H Y , more preferably on a substantially amorphous form on, or in, the nickel foam (NF).

[0021] As mentioned in the embodiments below alternative spray chemistries may also be adapted to further improve the performance, hereunder spraying a combined solution of Fe(III) and Ni(II), e.g. with Fe(NO 3 ) 3 and Ni(NO 3 ) 2 , of an optimal Fe / Ni ratio, which could potentially increase the anode performance further. Alternatively, a solution with Fe(III) may be applied, and at another point in time, before or after, a solution with Ni(II) be applied. Clearly other chemicals or compounds may be applied to obtain an optimised or even optimal Fe / Ni ratio during fast spraying or fast dipping according to the present invention.

[0022] Another aspect of the present invention relates to an alkaline electrolysis system with a first and / or a second electrode(s) comprising nickel foam (NF) being pretreated according to the first aspect, the alkaline electrolysis system comprising: a first electrode and a second electrode positioned in a electrolysis chamber, and wherein the first and / or the second electrode comprises a pretreated nickel foam (NF) surface.

[0023] Yet another aspect of the present invention is to provide a nickel foam (NF) being pretreated according to the first aspect, said nickel foam comprising an electrocatalytic iron compound on, or in, the nickel foam (NF) so as to reduce an overvoltage for water electrolysis, preferably an electro-catalytic compound being a nickel-iron-hydroxide, more preferably on the form FeNiO x H Y , and most preferably on a substantially amorphous form on, or in, the nickel foam (NF).Brief description of the figures

[0024] Figure 1 shows an overview of the three different scales of an industrial alkaline electrolysis system, Figure 2 shows a more detailed view of an electrolysis chamber according to the present invention, Figure 3 shows A) a laboratory implementation of a spraying system for implementing the present invention, and a B) suitable nozzle system for implementing the present invention, Figure 4 shows A) schematic drawing of the immersion cell, and B) shows the disassembled immersion cell with the current collector having nickel foam anode pretreated according to the present invention, Figure 5 shows electrochemical characteristics (A, B) and two SEM magnifications (C, D) of untreated nickel foam for comparison with nickel foam pretreated according to the present invention, Figure 6 shows a Table of various experimental samples to demonstrate the performance of the anodes made by the present invention, Figure 7 shows the iron content on the surface obtained by EDX of the samples shown in the Table of Figure 6, Figure 8 shows some SEM images of various representative samples, Figures 9 and 10 shows various polarization curves of some representative anodes made by the present invention, Figure 11 shows A) polarization curves of some advantageous representative samples according to the present invention together with B) corresponding voltages at 200 mA / cm2 from the samples shown in part A, Figure 12 show schematically a roll-to-roll (R2R) production line for implementing a pretreatment of nickel foam according to an embodiment of the present invention, and Figure 13 shows a flow-chart of a method according to an aspect of the present invention.

[0025] The present invention will now be described in more detail in the following.Detailed description of the invention Definitions

[0026] Prior to discussing the present invention in further details, the following terms and conventions will first be defined: Alkaline elektrolyzer: As the skilled person will understand an industrial scale alkaline electrolysis system typically comprises a stacked system of anodes, membranes and cathodes sandwiched between bipolar plates held together between two endplates.

[0027] In context of the present application, it may be understood that the term 'electrodes' is to be interpreted broadly as the skilled person in electrochemistry will readily understand. Thus, electrodes according to the present invention, i.e. the first and / or the second electrode, may form part of one, or more, electrolytic cell(s), for example a single electrolytic cell, stacked electrolytic cells, low pressure electrolytic cells, high pressure electrolytic cell, a traditional electrolytic cell, a zero gap electrolytic cell, an electrolytic cell with gas diffusion layers in between, etc.

[0028] Lye: With the term lye is meant KOH or NaOH dissolved in water as the skilled person will understand. Typical lye concentrations could be around 30wt%, but other concentrations such as 20wt%, 25wt%, 35wt% or 40wt% could be applied for some embodiments.

[0029] In context of the present application, it may be understood that the term vaporized is to be understand in a broad manner i.e. a liquid being transformed generally into a gas state, e.g. by an appropriate nozzle or nebulizer designed for that purpose. Alternatively or additionally, the term may understood as being atomized or being aerified as other ways of forming a gaseous state or being in a aeriform state. In some embodiment, there may be formed a visible fog of particles / droplets suspended in the air as typically seen from a standard nozzle used for vaporising a liquid.

[0030] The skilled person will readily understand that the present invention has been demonstrated at 70-90 deg. Celsius typically of industrial alkaline water electrolysis, but generally other temperatures, both higher, such as 100, 110, or 120 deg. Celsius, and lower, such as 40, 50 or 60 deg. Celsius, may be implemented within the context of the present invention.

[0031] Alkaline Water Electrolysis (AWE) with electrodes being nickel foam (NF) pretreated according to the present invention can be performed in various pressure regimes, as the skilled person in electrolysis will readily understand, i.e. often a division into three pressure regimes is made: Low Pressure Electrolysis (LPE): In low-pressure alkaline water electrolysis, the operating pressure is usually between 1 and 10 bar. LPE systems are often used for small-scale or laboratory applications.

[0032] Moderate Pressure Electrolysis (MPE): In moderate-pressure alkaline water electrolysis, the operating pressure is typically between 10 and 30 bar. MPE systems are commonly used in industrial and pilot-scale applications.

[0033] High Pressure Electrolysis (HPE): High-pressure alkaline water electrolysis involves operating pressures that can exceed 30 bar, 40 bar, or 50 bar, or often reaching up to 100 bar or higher. HPE systems are employed for large-scale hydrogen production and are used in industries like power generation and transportation where high-pressure hydrogen is required.

[0034] It should be noted that embodiments and features described in the context of one of the aspects of the present invention also apply to the other aspects of the invention.

[0035] All patent and non-patent references cited in the present application, are hereby incorporated by reference in their entirety.

[0036] Figure 13 shows a flow-chart of a method according to an aspect of the present invention; i.e. a method for pretreating a nickel foam NF, 5 cf. Figures 1 and 2, for alkaline electrolysis, the method comprising: S1 providing a NF, the NF being suitable for functioning as a first 1a and / or a second 1b electrode in an alkaline electrolysis system for water electrolysis, and S2 a performing a pretreatment of said NF by exposing the NF to a vaporized solution of iron(III) (Fe3+), such as with a Fe(NO 3 ) 3 spray as seen in Figure 3A or 3B, Or S2 b performing a pretreatment of said NF by exposing the NF to a liquid solution of iron(III) (Fe3+) for a duration of 59 second or less, S3 wherein said pretreatment S2 a) or S2 b) results in growth of an electrocatalytic iron compound on, or in, the nickel foam surface (NF) so as to reduce anovervoltage for water electrolysis.

[0037] In one embodiment, there may be a) performed a pretreatment of said NF by exposing the NF to a vaporized solution of iron(III) (Fe 3+< ) for a duration of 59 second or less, preferably for a duration of 55 second or less, more preferably for a duration of 50 seconds or less, most preferably for a duration of 40 seconds or less, or even 30 seconds or less.

[0038] In yet another embodiment, there may be a) performed a pretreatment of said NF by exposing the NF to a vaporized solution of iron(III) (Fe 3+< ) for a duration of 25 seconds or less, preferably for a duration of 20 seconds or less, more preferably for a duration of 15 seconds or less, most preferably for a duration of 10 seconds or less, or even more preferably for a duration of 5 seconds or less.

[0039] Similarly, in one embodiment wherein there may be b) performed a pretreatment of said NF by exposing the NF to a liquid solution of iron(III) (Fe 3+< ) for preferably a duration of 55 second or less, more preferably for a duration of 50 seconds or less, most preferably for a duration of 40 seconds or less, or even 30 seconds or less.

[0040] In yet another embodiment, there may be b) performed a pretreatment of said NF by exposing the NF to a liquid solution of iron(III) (Fe 3+< ) for preferably for a duration of 25 seconds or less, preferably for a duration of 20 seconds or less, more preferably for a duration of 15 seconds or less, most preferably for a duration of 10 seconds or less, or even more preferably for a duration of 5 seconds or less.

[0041] Thus, for both options a) and b) of the present invention, the exposure to the Fe(III) source, optionally the Fe(II) source, may be of a relatively short period, typically just a few seconds like 2 or 10 seconds can in some situations be sufficient, as it will be demonstrated below.

[0042] More particularly for some embodiments both options a) and b) of the present invention the period of exposure to the Fe(III) source, optionally the Fe(II) source, may be described as intervals like 1-59 seconds (s); 1-50 s, 1-40s, 1-30 s, preferably 1-20s, or more preferably 1-15 s. Alternatively, it may be described as 1.5-59 seconds (s); 1.5-50 s, 1.5-40s, 1.5-30 s, preferably 1.5-20s, or more preferably 1.5-15 s. More alternatively, it may be described as 2-59 seconds (s); 1.9-50 s, 1.9-40s, 1.9-30 s, preferably 1.9-20s, or more preferably 1.9-15 s. Even more alternatively, it may be described as 5-59 seconds (s); 5-50 s, 5-40s, 5-30 s, preferably 5-20s, or more preferably 5-15 s. Clearly, the chosen spray interval will also be a function of the applied spray flow or fluid pressure, i.e. higher flows corresponds to shorter exposure times. Advantageously the present invention works beneficially at 2 or 10 seconds, as it will be demonstrated below in the Table of Figure 6.

[0043] In another particular embodiment, it is contemplated that for both options a) and b) of the present invention, the exposure to the Fe(III) source, optionally the Fe(II) source, may be a minimum period of exposure, preferably at least 1 second (s), more preferably at least 1.5 s, most preferably at least 1.9 or around 2 s, sufficient to result in growth of an electro-catalytic iron compound on, or in, the nickel foam surface (NF) so as to reduce an overvoltage for water electrolysis according to first aspect of the present invention.

[0044] In an advantageous embodiment, the overvoltage may be reduced by at least 50 mV, preferably at least 100 mV, more preferably at least 150 mV, and in the best example 167 mV was obtained. This was at 200 mA / cm2, 80-90 deg. C, and at atm. pressure, but there is no reason to believe that this will be significantly different for typical conditions for industrial alkaline water electrolysis at higher temperature and pressure. First estimates of the resulting energy saved are around 8% if applied in an industrial AWE context, which is a relatively high reduction for such an easy-to-implement modification of the electrodes made of NF.

[0045] In another advantageous embodiment, the electro-catalytic iron compound may be a nickel-iron-hydroxide, preferably on the form FeNiO x H Y , , more preferably on a substantially amorphous form on, or in, the nickel foam (NF).

[0046] In another embodiment, the a) vaporized solution of iron(III) (Fe3+), or b) liquid solution of iron(III) (Fe3+) may be chosen from the list of: a solution with ferric chloride (FeCl 3 ), a solution with ferric sulfate (Fe 2 (SO 4 ) 3 ), and a solution with ferric nitrate (Fe(NO 3 ) 3 ) or other suitable solutions with a source of Fe(III) readily available to the skilled person in electrochemistry. It is contemplated that a solution of Fe(II) may also be applied if oxidised to Fe(III) immediately before and / or during the application on or in the nickel foam surface. Advantageously, the solution may a water solution with ferric nitrate, Fe(NO 3 ) 3 (aq), and the concentration may be 0.2-0.8M, preferably 0.3-0.7M, most preferably 0.4-0.6M, such as around 0.5M.

[0047] In some advantageous embodiments, the nickel foam (NF) may be cleaned prior to the said pretreatment according to the first aspect of the invention, preferably cleaned with an acid, more preferably cleaned with a hydrogen chloride solution (HCl(aq)), preferably by a vaporised solution or by immersion in a liquid. In fact, there will be presented experimental results below for the beneficial effect of acid cleaning with hydrogen chloride, cf. Figure 9 left part. Other acids may be used for cleaning prior to spraying.

[0048] In other advantageous embodiments, the nickel foam after being pretreated (a, b) may be left undisturbed for a predefined reaction time, said predefined reaction being preferably at least 1 minute, at least 2 minutes, at least 3 minutes, at least 4 minutes, or at least 5 minutes. Thus, there is no further contact with the Fe(III) source and the already deposited Fe is allowed to react with the nickel foam over time to form nickel-iron surface compounds with advantageous electrocatalytic properties. Additionally with the above minimum time, or alternatively, this reaction time may be limited to preferably a maximum of 20 minutes, max. 10 minutes, max. 5 minutes, or max. 2 minutes, particular if the pretreatment time can thereby be optimised without losing the advantageous electrocatalytic properties.

[0049] In some advantageous embodiments, the nickel foam (NF) may be air dried, or oven dried, after said pretreatment, preferably at an oven temperature below 150 deg. Celsius, more preferably below 110 deg. Celsius, such as around 100 deg. Celsius, preferably being air or oven dried after said predefined reaction time. Lower oven temperatures are also contemplated, preferably below 90 deg. C, more preferably below 80 deg. C, or most preferably below 70 deg. C, such as around 60 deg. C or around 50 deg. C.

[0050] In some advantageous embodiments, the nickel foam (NF) may be pretreated by option a) and / or option b) a plurality of times, preferably at least two times, more preferably at least three times, optionally interrupted by a reaction time as mentioned above, and / or being air dried or oven dried as also mentioned above after said one, or more, application(s) of a pretreatment option a) and / or b). Thus, the mentioning of 'a pretreatment' according to the present invention do not in any way preclude that additionally one, or more, pretreatment(s) may be contemplated with the context of the present invention. In fact, there will be presented beneficial experimental results below for multiple pretreatments of NF with a Fe(III) source, e.g. a Fe(NO 3 ) 3 solution.

[0051] In some embodiments, the solution of iron(III) (Fe 3+< ) in said above mentioned option a) or b) may further comprise a source of Ni, preferably Ni(II) or Ni(III), e.g. with added Ni-salts in a same solution like NiCl 2 , NiSO 4 , Ni(NO 3 ) 2 etc., so as to modify the Fe / Ni ratio in the solution and thereby of the resulting electrocatalytic iron compound grown in, or on, the NF surface and thereby reduce the overvoltage for water electrolysis in an advantageous manner. Optionally, the solution of the solution of iron(III) (Fe 3+< ) in said above mentioned option a) or b) may be modified by various acids, or buffers, for adjusting the pH value of the solution, e.g. by HCl, HNO 3 , H 2 SO 4 solutions etc.

[0052] In other advantageous embodiments the pretreatment, and optionally said cleaning, preferably acid cleaning, mentioned above, and / or optionally the drying mentioned above, may be performed in a production line with a substantially continuous pretreatment of nickel foam (NF), e.g. in a production line where NF is pretreated in a continuous manner over time by having for example an open end of nickel foam passing by a spraying post with Fe(NO 3 ) 3 on a production line. Thus, unlike a batch pretreatment like suggested in Nano Research, November 2017, Yin & Jiang et al. with immersion into a liquid with Fe(NO 3 ) 3 , i.e. iron (III) nitrate for half an hour, or even longer, the present invention is much more readily implemented in an industrial up-scaling of a pretreatment method for nickel foam. Beneficially, the substantially continuous pretreatment of nickel foam (NF) may be performed in a roll-to-roll (R2R) production line comprising an initial unwinding of non-pretreated flexible nickel foam, and a final rewinding of the pretreated flexible nickel foam for even further advantageous implementation of the present invention.

[0053] The invention will now be described in further details in the following non-limiting examples.

[0054] Figure 1 shows an overview of the three different scales, A, B and C, of an industrial alkaline electrolysis system.

[0055] Part A to the left shows a perspective drawing of an alkaline electrolysis system 10 with a length being typically around 10 m. The electrolyzer or electrolysis system includes a plurality of cells arranged in an elongate stack; each cell having a bipolar plate, anode, membrane, cathode and bipolar plate again. The electric power is supplied to the outermost cells in the stack and at the centre. The electric potential will be distributed along the stack, ideally supplying each cell with the same current density and a potential between 2-2.5 volt depending on the applied current density. Lye is supplied at inlet channel 12 to the cells in common internal channels running through the stack. Likewise, there are common internal channels 11 conducting gases and excess lye / water from the cells, i.e. a channel conducting hydrogen from all the cathodes in the stack, and a channel conducting oxygen from the anodes.

[0056] Part B in the middle shows a schematic layout of an electrolysis chamber 2 according to the present invention with a negative cathode 1a to the left, where hydrogen gas is produced, and a positive anode 1b to the right, where the oxygen gas is produced. The anode and the cathode are separated by a porous membrane or diaphragm, which allows hydroxide OH -< to pass from the cathode part to the anode part as indicated by the small arrow across the diaphragm. As explained above, the electrolysis system 10 will have a plurality of electrolysis chambers, e.g. 50, 100 cells or even up to 500 cells, in a stacked configuration but only one cell is shown here for clarity.

[0057] Part C to the right shows a SEM image of the nickel foam (NF) 5 forming part of the first and the second electrodes 1a, 1b, like in a conventional electrolysis system.

[0058] Figure 2 shows a more detailed view of an electrolysis chamber 2 according to the present invention. Thus, the invention relates to a method for pretreating the nickel foam (NF) 5) for alkaline electrolysis, the method comprising / providing a NF, the NF being suitable for functioning as a first (1a) and / or a second (1b) electrode in an alkaline electrolysis system for water electrolysis, and a) performing a pretreatment of said NF by exposing the NF to a vaporized solution of iron(III) (Fe 3+< ), or b) performing a pretreatment of said NF by exposing the NF to a liquid solution of iron(III) (Fe 3+< ) for a duration of 59 second or less.

[0059] This pretreatment, cf. option a and / or b above, results in growth of an electrocatalytic iron compound on, or in, the nickel foam surface (NF) 5 so as to reduce an overvoltage for water electrolysis.

[0060] Figure 3A shows a laboratory implementation of a spraying system for implementing the present invention. Thus, the spraying system has an (A) immobilizing structure holding the (B) handheld sprayer containing ferric nitrate. The nozzle is pointed through a hole in the (C) lid, which is lined with Veiton. The nickel foam (NF) to be plate / exposed is placed on a (D) beaker with PEEK rods across, which is encapsulated in a (E) 15L polypropylene box.

[0061] The process is started with pumping the sprayer to maximum pressure (1.5 bar) and spraying a short burst of acid into a pitcher while holding the sprayer vertically, allowing liquid to fill the spray tube inside the reservoir, but other pressures may be applied as well, e.g. a pressure around 1 bar, or 1.1, 1.2, 1.3, 1.4, 1.6, 1.7, 1.8, 1.9 or 2.0 bar.

[0062] Fig. 3B shows a suitable nozzle system for implementing the present invention, which has also been tested. In this particular nozzle system, the flow and / or pressure can be optimized to provide a vapour of iron(III), e.g. from ferric nitrate, possibly the vapour was visible as a fog with droplets. Figure 3B shows an improvement of the present invention where the nozzle is moved over the NF to simulate the movement of the NF in a roll-to-roll configuration while spraying ferric nitrate.

[0063] Figure 4 shows A) schematic drawing of the immersion cell, and B) shows the disassembled immersion cell with the current collector having nickel foam pretreated according to the present invention. Thus, the CAD rendering of A) shows an immersion cell inserted into a beaker. Notice the holes in the top of the lid, which allow access to the electrodes and current collectors. Part B) shows the disassembled immersion cell, with the Teflon frames encapsulating the current collector, electrode, membrane, electrode, current collector assembly. Nickel screws hold the assembly together, tightened to 0.45Nm by nickel nuts on the other side.

[0064] Testing was done by connecting the current collectors to a battery tester (Neware, BTS4000 5V / 12A) and applying 800mA to the 2x2 cm square of nickel foam NF. This equals the 200mA / cm 2< , which is typically used in industrial full scale electrolysers, e.g. by the present applicant HydrogenPro. Cables were attached to each of the current collectors, as well as the NF to be able to calculate the ohmic loss between them and be able to exclude that from the final calculation, since the ohmic losses found in this system will not be like the industrial counterpart. Polarization curves were used in this setup to gather data on the performance of the electrodes at different current densities. The cells were tested for between 6-12 days. Higher currents were also applied to simulate accelerated aging.

[0065] Figure 5 shows electrochemical characteristics A and B and two SEM magnifications C and D of nontreated nickel foam called 'NNF' for comparison with nickel foam pretreated according to the present invention. Thus, the different electrochemical characteristics of untreated NF at A) with differing amperages over 14 days, B) at 200mA / cm 2< for each day. Electron microscope pictures taken using SEM microscope at (C) 50x magnification and (D) 1000x magnification. Data for A were made using two polarization curved made consecutively at the differing amperages, while B shows the stabilized regions between each polarization round. Lines between data points are meant only as a guide to the eye.

[0066] Figure 6 shows a Table of various experimental samples to demonstrate the performance of the anodes made by the present invention. From this base line of untreated NF, a series of iterations using different concentrations of ferric nitrate were made, as well as a variety of different pretreatments, spraying / dipping techniques, reaction period and drying method. The Table of Figure 6 shows the full overview of the different combinations, which were tested and showed advantageous results. From the treatment, it was noticed that there was a general rust coloring of the NF samples, which was very common to see on all the electrodes that received the 0.5M spray. An electrode from the same plate was tested in an immersion cell and came out looking darker in color than the pretesting sample, losing some of the characteristic rust color.

[0067] In the Table, acid cleaned' denotes an immersion in 2M HCl for 15 minutes under sonication, single pass in one pass back and forth using another handheld system for acid cleaning, saturation is five passes back and forth using the same system. Twice saturated means two rounds of saturation with the handheld system with an oven drying between each round. 2s spray is a 2 second spray using the system shown in Figure 3A, and 10s is 10 seconds using the same spraying system of Figure 3A. The dips are simple immersions in 1.5L ferric nitrate for the stated amounts of time where the NF plates were suspended in the middle of the container vertically. The reaction period denotes the amount of time the nickel foam had to react with the ferric nitrate from first contact to rinsing with demineralized water. Oven dried means rinsed with water and left in a 100 deg. C oven for 2+ hours, while air dried is the same rinse, whereafter the NF sample is left in a fume hood overnight.

[0068] Figure 7 shows iron content on the surface obtained by EDX of the samples shown in the Table of Figure 6 i.e. with same sample number on the horizontal scale. Figure 7 shows the iron content in each sample in atomic %. The data were gathered using EDX. Five different spots on each sample were investigated to get an average of the sample. Error bars display the standard deviation for the five different places on each NF sample that were investigated. Each sample is of an electrode after spraying, but before testing in an immersion cell.

[0069] Figure 7 shows that there was a significant increase in iron content for all the sprayed electrodes which utilized 0.5M ferric nitrate. There is a large degree of error for most of the samples, which comes from the variations of surface structure, which can be found on the treated NF. No major statistical variation could be found, except for CLO053 which had a large concentration of iron in all sites investigated, as well as CLO028, which had little to none. CLO025 also had a larger than average amount of iron content, which can be attributed to the double coating it received, which was speculated to give it an even larger amount of iron than the rest. This was not the case, as it is barely statistically significant when compared to most other sprays. In general, the large degree of variation in the iron content makes it hard to conclude anything about the correlation between iron content and electrode performance for OER, but Figure 7 shows that there is indeed a significant amount of iron deposited when utilizing the methods of the present invention.

[0070] Figure 8 shows some SEM images of various representative samples intended to given an overview of the structures of some of the experimental electrodes made with SEM pictures. Part A shows the structure of CLO016 (single pass, not acid cleaned), B shows CLO017 (saturated not, acid cleaned) and C and D shows CLO033 (saturated). E and F shows CLO025 (saturated, double coated), G shows CLO028 (30 min dip 0.05M) and H shows CLO060 (5 min dip). All samples used 0.5M ferric nitrate except CLO028, and all samples except CLO016 and CLO017 were acid cleaned.

[0071] Figure 8 shows the different structures made when applying the different experimental combinations of pretreatments of NF according to present invention. These structures show that overall, there was indeed a difference in the surface structure of the treated NF, though the results vary. Figure 8A shows that there are some amorphous structures that have been built in a layer on top of the NF, though it was only in Figure 8B, where the sample had been saturated instead of just receiving a single pass that there was some delamination, which show the layer much more. Figure 8C and 8D show the impact of acid cleaning versus not acid cleaning, as the acid seems to have 'eaten away' at the nickel to a higher degree. This was also shown experimentally, as after switching to acid cleaning, the liquid runoff from the sprayed NF plate was much greener, implying that there was more nickel that had been dissolved in the acid. There were still similar structures, and the layer formed on top of the NF could be seen clearly in a few places. It is worth noting that the delamination seen on Figure 8D is not characteristic for the sample, as the layer applied to all the experimental electrodes was very well adhered to the NF. No delamination on a large scale was observed, even when handling the electrodes for testing purposes. Figure 8E shows the effect of applying two layers of ferric nitrate at 0.5M using the saturation technique. This led to a much higher level of delamination, which is speculated to be caused by the second layer not adhering to the first, hence the very thin film that can be seen better in Figure 8F. Figure 8G shows the effect of dipping NF in 30 minutes in ferric nitrate at a low concentration (0.05M) This led to some dissolving of the NF at the surface and an amorphous structure, but it did not seem like a large noticeable layer was formed like on the ones that received the higher concentration. Finally, Figure 8H shows the impact of 5 minutes of immersion in 0.5M ferric nitrate, where the general structure of the foam was apparently heavily damaged, leaving the foam structurally unsound. It was difficult to handle and would collapse / break apart at the minutest of physical manipulation. It was thus deemed unworkable for future experiments because of these properties, and therefore spraying is consider a more viable option for this concentration.

[0072] Figures 9 and 10 shows various polarization curves of some representative samples.

[0073] Figure 9, left: Polarization curve showing the performance of unmodified NF (acid cleaned) also called "NNF", CLO016 (single pass not acid cleaned), CLO017 (Saturation not acid cleaned) and CLO033 (saturation acid cleaned). All experimental samples had 5-minute reaction time and used 0.5M solution of ferric nitrate. All data taken from two polarizations which were made on the 6 th< day. Lines between data points are meant only as a guide to the eye.

[0074] Figure 9, right: Polarization curve comparing the different dip concentrations, showing the performance of CLO028 (dipped in 0.05M solution for 30 minutes and CLO060 (dipped in 0.5M solution for 5 minutes). Both electrodes received acid cleaning before immersion. Also here, all data are taken from two polarizations which were made on the 6 th< day. Lines between data points are meant only as a guide to the eye. In Figure 9, right, the different concentration of dips was tested to get a baseline of the performance of nickel electrodes when treated in the same way as Yin & Jiang et al. (2017). This was done at two different concentrations, 0.05M and 0.5M ferric nitrate, which had every other parameter constant except the reaction time, as 30 minutes in 0.5M ferric nitrate will destroy the structure of the nickel foam by dissolution of nickel into the solution. These experiments showed that indeed, simply immersing nickel foam in ferric nitrate for a period without added heat or electricity did in fact improve performance. In fact, CLO028 also showed an improvement of 30mV, which can be substantial on an industrial scale.

[0075] Figure 10 shows various polarization curves of some representative samples.

[0076] Figure 10, left, is comparing different spray variations. CLO025 (saturation spray twice coated), CLO033 (saturation spray), CLO035 (saturation spray 0.5M 10-minute reaction period), CLO040 (2s spray) and CLO041 (10s spray). All samples were acid cleaned before spraying, sprayed with 0.5M solution and oven dried after spraying and all except CLO035 had a 5-minute reaction period. Again, all data taken from two polarizations which were made on the 6 th< day. Lines between data points are meant only as a guide to the eye. Thus, the Figure shows a variety of different sprays, with changes in reaction time, number of coats, changes in experimental spray setup from handheld to mechanical and changes in the amount of acid sprayed. The experiments with changing the reaction were done to see if a longer reaction time could give the acid time enough to create a rougher surface with a thicker amorphous layer and possibly create more sites for the OER to occur. The data did not support this, as the curves for CLO033 and CLO035 are almost identical. They both reached a 130mV decrease in overpotential, which is among the better results found. Sample CLO025, where the electrode was sprayed, rinsed and dried, before receiving a second coating was done to attempt to create a thicker layer of the amorphous layer described in Yin & Jiang et al. (2017). This also proved futile, as the electrode had a worse performance at only approximately 50mV decrease from NF, which is 80mV less than CLO033, which received otherwise identical treatment.

[0077] Figure 10, right, is comparing oven-dried and air-dried samples from the spray method according to the present invention; e.g. sample CLO052 (2s spray, air dried) and CLO053 (10s spray, air dried) versus CLO041 (10s spray, oven dried) and CLO040 (2s spray, oven dried). All samples were acid cleaned before spraying, sprayed with 0.5M solution and had a 5-minute reaction period. All data taken from two polarizations which were made on the 6thday. Lines between data points are meant only as a guide to the eye.

[0078] Figure 11 shows A) polarization curves of some advantageous representative samples according to the present invention together with B) a plot with corresponding voltages at 200 mA / cm2 from the sample shown in part A.

[0079] Thus, this figure is comparing the best performing electrodes to non-treated NF (NNF). To the left, graph A) shows the polarization curves for the best of each comparison made, with NNF (untreated NF) as a reference. CLO033 (saturation spray, 5-minute reaction time, oven dried), CLO035 (saturation spray, 10-minute reaction time, oven dried), CLO053 (10 second spray, 5-minute reaction time, air dried) and CLO060 (5-minute dip, 5-minute reaction time, oven dried) were compared to NNF, which was only acid cleaned before testing. All experimental electrodes used 0.5M ferric nitrate and were acid cleaned. All data taken from two polarizations, which were made on the 6 th< day. Lines between data points are meant only as a guide to the eye.

[0080] To the right, B) shows the voltage at 200mA / cm 2< for each of the samples shown for ease of comparison. Thus, Figure 11 shows the final comparison between the better performance electrodes from the previous figures. There was no large difference between using 5 or 10 minutes of reaction time, as CLO035 again had almost the same, if not slightly worse performance at 200mA / cm 2< . This is advantageous for possible industrial applications, as it means that even 5 minutes is enough to have the desired effect, and this number could potentially be lowered with further studies, especially as seen by 2 seconds and 10 seconds of spraying. Figure 11B shows the performance of each electrode at 200mA / cm 2< , which further supports the conclusion that any of the sprays where 0.5M ferric nitrate was used leads to a significant decrease in overpotential, though with some variation between samples.

[0081] Figure 12 show schematically a roll-to-roll (R2R) production line for implementing a pretreatment of nickel foam according to an embodiment of the present invention. Thus, the invention is performed in a production line with a substantially continuous pretreatment of nickel foam NF, where untreated NF 5' is shown in a winding configuration to the left, and then unwinded into the R2R production line , e.g. in a production line where NF is pretreated in a continuous manner over time by passing a spraying post or station 100 with Fe(NO 3 ) 3 on a production line, where the solid arrow represent the overall direction of the NF on the line. To the right, the treated NF 5" is finally winded onto a suitable roll for further use in system for AWE like shown in Figures 1 and 2. Also shown in Figure 12 is an initial cleaning station 110 on the production line, and a drying station 120 with nozzles for directing the air, optionally heated air or a heated oven, onto the NF after the pretreatment at the station 100. If multiple steps with initial cleaning, pretreatment with Fe(NO 3 ) 3 and / or drying are required, these additionally steps can either be implemented by passing the NF by the same station several time or inserting more cleaning, pretreatment and / or drying stations, or a combination of such variants (neither shown in Figure 12). The present invention is thereby readily implemented in an industrial up-scaling of a pretreatment method for nickel foam.

[0082] In short the present invention relates to a method for pretreating a nickel foam (NF) for use as an electrode in alkaline electrolysis by a) performing a pretreatment by exposing the NF to a vaporized solution of iron(III) (Fe 3+< ), such as ferric nitrate, or b) performing a pretreatment by exposing the NF to a liquid solution of iron(III) (Fe 3+< ) for a duration of 59 second or less. The pretreatment (a, b) results in growth of an electro-catalytic iron compound on, or in, the nickel foam (NF) surface so as to reduce an overvoltage for water electrolysis after for example 10 seconds with spraying of ferric nitrate. Thus, the invention yields a fast method for pretreatment of nickel foam for the electrodes in an alkaline electrolysis system for water electrolysis with significant reductions in the required overvoltage, e.g. in some embodiments an improvement of 167 mV is observed, as seen in Fig. 11.References

[0083] "Nanostructured materials on 3D nickel foam as electrocatalysts for water splitting" by Chaudhari et al. in Nanoscale, 2017, 9, 12231. "Remarkably enhanced water splitting activity of nickel foam due to simple immersion in a ferric nitrate solution" by Yin & Jiang et al. in Nano Research, November 2017.

Claims

1. A method for pretreating a nickel foam (NF, 5) for alkaline electrolysis, the method comprising: - providing a NF, the NF being suitable for functioning as a first (1a) and / or a second (1b) electrode in an alkaline electrolysis system (10) for water electrolysis, and - a) performing a pretreatment of said NF by exposing the NF to a vaporized solution of iron(III) (Fe3+), or - b) performing a pretreatment of said NF by exposing the NF to a liquid solution of iron(III) (Fe3+) for a duration of 59 second or less, wherein said pretreatment (a,b) results in growth of an electro-catalytic iron compound on, or in, the nickel foam surface (NF) so as to reduce an overvoltage for water electrolysis.

2. The method for pretreating according to claim 1, wherein there is a) performed a pretreatment of said NF by exposing the NF to a vaporized solution of iron(III) (Fe3+) for a duration of 59 second or less, preferably for a duration of 55 second or less, more preferably for a duration of 50 seconds or less, most preferably for a duration of 40 seconds or less, or even more preferably 30 seconds or less.

3. The method for pretreating according to claim 1, wherein there is a) performed a pretreatment of said NF by exposing the NF to a vaporized solution of iron(III) (Fe3+) for a duration of 25 seconds or less, preferably for a duration of 20 seconds or less, more preferably for a duration of 15 seconds or less, most preferably for a duration of 10 seconds or less, or even more preferably for a duration of 5 seconds or less.

4. The method for pretreating according to claim 1, wherein there is b) performed a pretreatment of said NF by exposing the NF to a liquid solution of iron(III) (Fe3+) for preferably a duration of 55 second or less, more preferably for a duration of 50 seconds or less, most preferably for a duration of 40 seconds or less, or even more preferably 30 seconds or less.

5. The method for pretreating according to claim 1, wherein there is b) performed a pretreatment of said NF by exposing the NF to a liquid solution of iron(III) (Fe3+) for preferably for a duration of 25 seconds or less, preferably for a duration of 20 seconds or less, more preferably for a duration of 15 seconds or less, most preferably for a duration of 10 seconds or less, or even more preferably for a duration of 5 seconds or less.

6. The method for pretreating according to claim 1, wherein the overvoltage is reduced by at least 50 mV, preferably at least 100 mV, more preferably at least 150 mV.

7. The method for pretreating according to claim 1, wherein said electrocatalytic iron compound is a nickel-iron-hydroxide, preferably on the form FeNiOxHY, more preferably on a substantially amorphous form on, or in, the nickel foam (NF).

8. The method for pretreating according to claim 1, wherein the a) vaporized solution of iron(III) (Fe3+), or b) liquid solution of iron(III) (Fe3+) is chosen from the list of: a solution with ferric chloride (FeCl3), a solution with ferric sulfate (Fe2(SO4)3), and a solution with ferric nitrate (Fe(NO3)3).

9. The method for pretreating according to claim 8, wherein the solution is a water solution with ferric nitrate, Fe(NO3)3 (aq), and the concentration is 0.2-0.8M, preferably 0.3-0.7M, most preferably 0.4-0.6M, such as around 0.5M.

10. The method for pretreating according to claim 1, wherein the nickel foam (NF, 5) is cleaned prior to the said pretreatment, preferably cleaned with an acid, more preferably cleaned with a hydrogen chloride solution (HCl(aq)), preferably by a vaporised solution or by immersion in a liquid.

11. The method for pretreating according to claim 1, wherein the nickel foam after being pretreated (a, b) is left undisturbed for a predefined reaction time, preferably said predefined reaction being at least 1 minute, at least 2 minutes, at least 3 minutes, at least 4 minutes, or at least 5 minutes.

12. The method for pretreating according to claim 1, wherein the nickel foam (NF, 5) is air dried, or oven dried, after said pretreatment, preferably at an oven temperature below 150 deg. Celsius, more preferably below 110 deg. Celsius, such as around 100 deg. Celsius, preferably being air or oven dried after said predefined reaction time according to claim 11.

13. The method for pretreating according to any of the preceding claims, wherein the nickel foam (NF, 5) is pretreated by option a) and / or option b) a plurality of times, preferably at least two times, more preferably at least three times, optionally interrupted by a reaction time according to claim 11, and / or being air dried or oven dried according to claim 12 after said one, or more, application(s) of a pretreatment option a) and / or b).

14. The method for pretreating according to any of the preceding claims, wherein the said pretreatment, and optionally said cleaning according to claim 10, and / or optionally said drying according to claim 12, is performed in production line with a substantially continuous pretreatment of nickel foam (NF, 5').

15. The method for pretreating according to claim 14, wherein said substantially continuous pretreatment of nickel foam (NF, 5) is performed in a roll-to-roll (R2R) production line comprising an initial unwinding of non-pretreated flexible nickel foam (5'), and a final rewinding of the pretreated flexible nickel foam (5").

16. The method for pretreating according to claim 1, wherein said solution of iron(III) (Fe3+) in said option a) or b) further comprises a source of Ni, preferably Ni(II) or Ni(III), so as to modify the Fe / Ni ratio of the resulting electro-catalytic iron compound in, or on, the NF surface and thereby reduce the overvoltage for water electrolysis.

17. An alkaline electrolysis system (10) with a first and / or a second electrode(s) comprising nickel foam (NF, 5) being pretreated according to any of claims 1-16, the alkaline electrolysis system comprising: - a first electrode (1a) and a second electrode (1b) positioned in a electrolysis chamber, and wherein the first and / or the second electrode (1a, 1b) comprises a pretreated nickel foam (NF, 5) surface.

18. A nickel foam (NF, 5) being pretreated according to any of claims 1-16, said nickel foam comprising an electro-catalytic iron compound on, or in, the nickel foam (NF) so as to reduce an overvoltage for water electrolysis, preferably an electro-catalytic compound being a nickel-iron-hydroxide, preferably on the form FeNiOxHY, more preferably on a substantially amorphous form on, or in, the nickel foam (NF).

Images