Electrolyzing device, gas diffusion electrode for used in said device and method for manufacturing the gas diffusion electrode

Abstract

Described is an electrolyzing device comprising a first electrode configured to generate a first gas and a second electrode configured to generate a second gas. The first and second electrodes are arranged to face each other to define an electrolyte space for receiving an electrolyte between the first electrode and the second electrode. At least one of the first and second electrode is a gas diffusion electrode comprising a gas diffusion backbone layer facing away from the electrolyte space and comprising conductive porous material; a catalyst layer facing the electrolyte space and comprising an electrochemically active catalyst, wherein the catalyst layer further comprises hydrophobic particles with an average size of 10 nm - 5 µm, and a microporous layer comprising conductive particles with an average size of 15 nm - 2 µm and hydrophobic particles with an average size of 10 nm - 5 µm and arranged between the gas diffusion backbone layer and the catalyst layer. Also described are a gas diffusion electrode for use in said device, as well as a method of manufacturing said gas diffusion electrode.

Description

[0001] ELECTROLYZING DEVICE AND METHOD FOR MANUFACTURING THE

[0002] DEVICE

[0003] TECHNICAL FIELD OF THE INVENTION

[0004] The invention relates to an electrolyzing device comprising a first electrode configured to generate a first gas and be in fluid contact with a first gas collector for the first gas; a second electrode configured to generate a second gas and be in fluid contact with a second gas collector for the second gas; wherein the first and second electrodes are arranged to face each other to define an electrolyte space for receiving an electrolyte between the first electrode and the second electrode. The invention further relates to a gas diffusion electrode for use as the first or the second electrode or the first and the second electrode in the device. The invention further relates to a proton exchange membrane electrolyser, an anion exchange membrane electrolyser, a hybrid electrolyzer or a metal air battery, comprising the gas diffusion electrode. The invention finally relates to a method of manufacturing the gas diffusion electrode.

[0005] BACKGROUND OF THE INVENTION

[0006] The invention is in the field of electrolyzing devices comprising a gas diffusion electrode. The advantages of the invention become particularly apparent in membrane- free electrolyzing devices, as will be explained further below.

[0007] Bubble formation and separator stability in electrolyzing devices such as alkaline water electrolyzers significantly impacts their performance, safety, and efficiency. Membrane- free and membrane-less electrolyzers have been suggested as a candidate to eliminate performance losses and improve electrolyzer efficiencies. Several cell and electrode designs for membrane-free concepts have been proposed, in particular for producing hydrogen and oxygen gas, but industrial implementation of these designs has been limited by issues relating to gas crossover and scalability. Gas crossover may occur due to the absence of a physical separator in membrane-free electrolyzers, leading to mixing of a first gas such as hydrogen and a second gas such as oxygen, which may result in efficiency losses as well as potential safety hazards. In addition, scaling up of these systems while maintaining efficient and stable operation remains a challenge. To date, practical solutions to solve the issue of gas crossover, without compromising on scalability, leakage, and performance are not available.

[0008] Prior art membrane-free and membrane-less electrolyzers are known in different configurations. The buoyancy flow-by electrolyzer, as disclosed in W02024 / 100678 consists of two vertical parallel plate electrodes. A flowing electrolyte carries the produced hydrogen and oxygen gas bubbles into separate downstream effluent channels. In the electrolyte, hydrogen and oxygen gas bubbles are concurrent and gas crossover therefore is a major bottleneck for the scale-up of these systems.

[0009] Capillary-fed electrolyzer systems, as proposed by A. Hodges et al., “A high- performance capillary-fed electrolysis cell promises more cost-competitive renewable hydrogen,” Nat. Commim., vol. 13, no. 1, pp. 1-11, 2022, doi: 10.1038 / s41467-022- 28953-x, operate on the principle of capillary action, where a liquid is drawn through a porous separator arranged between electrodes. The wicking material provides electrolyte to the two adjacent electrodes, where hydrogen and oxygen production for instance may occur. The height of these systems is limited to less than 16 cm due to the restricted supply of electrolyte, which is caused by flow limitations induced by the capillary action. Additionally, lifetime degradation in this design is expected to result in an increased heat generation and an associated efficiency drop, which these systems will not be able to manage effectively.

[0010] Divergent flow-through electrolyzer systems have also been proposed, for instance by M. I. Gillespie and R. J. Kriek, “Scalable hydrogen production from a mono-circular filter press Divergent Electrode-Flow-Through alkaline electrolysis stack,” J. Power Sources, vol. 397, no. April, pp. 204-213, 2018, doi: 10.1016 / j.jpowsour.2018.07.026. These systems employ a divergent flow design in which a liquid compartment separates the two electrodes, and the fluid is forced through the electrodes. Management of the fluid transport in the system eliminates the need for a membrane, while hydrogen and oxygen are produced in two separate product streams. Divergent flow-through electrolyzer systems may effectively separate hydrogen and oxygen, however long-term stable deployment is hindered by fluid malmanagement resulting from a heavy sensitivity to pressure distribution in the cell and local turbulence formation. To combat hydrogen crossover and safety hazards, larger interelectrode gaps are required, typically > 2.5 cm, which in turn hinders efficient operation.

[0011] Although hydrogen and oxygen may be separated in the above systems with some efficiency, they suffer from bubble formation, and the gases produced must be separated from their electrolyte stream before a pure product stream is obtained.

[0012] Gas diffusion (or gas-breathing) electrolyzers have been proposed to solve this issue. Gas-breathing electrolyzer cells are designed to facilitate the release of hydrogen and oxygen as gases from the electrolyzer that are directly stripped out of the electrolyte solution. In these systems, the permeance of produced oxygen and hydrogen and the electrochemical efficiency of the electrode must be balanced. Initial systems have focused on the usage of Gore ePTFE membranes as a gas-breathing backbone of the electrode, as proposed by P. Tiwari et al., “A new class of bubble-free water electrolyzer that is intrinsically highly efficient,” Int. J. Hydrogen Energy, vol. 44, no. 42, pp. 23568-23579, 2019, doi: 10.1016 / j.ijhydene.2019.07.100. Current densities up to 0.2 A cm-2were achieved at 80°C and 1.66V (89% efficiency). However, ePTFE membranes are intrinsically non-conductive and can therefore not readily be used as gas diffusion media in a scalable electrochemical stack design. More recently, K. Deng et al., “Ampere-level membrane-less water electrolysis enabled by rose-petal-effect- mimetic interface,” Joule, vol. 7, no. 8, pp. 1852-1866, 2023, doi:

[0013] 10.1016 / j.joule.2023.06.010 developed a novel type of catalyst that is deposited on a carbon gas diffusion electrode for acidic media. They demonstrate effective gas wicking in such media up to a current density of 4.2 A cm'2, while keeping acceptable hydrogen gas crossover and a performance of 1 A cm'2at 61.5% efficiency.

[0014] Ideally, gas-breathing electrolyzers should also be able to operate in alkaline media to limit the dependency on the use of rare-earth metal catalysts and to ensure low gas crossover of hydrogen (hydrogen has a lower solubility and diffusivity coefficient in alkaline electrolytes compared to acidic electrolytes). However, to date there are no available gas-breathing electrodes that can facilitate effective alkaline electrolysis.

[0015] Thus, there is a need to develop novel gas diffusion electrodes and cell concepts to improve the separation of a first gas such as hydrogen and a second gas such as oxygen in membrane-free electrolysis with limited bubble formation, particularly in alkaline electrolysis.

[0016] Since the filing of the priority application, further developments have been made to the devices and methods described herein. Simulation results and experimental work - which are now described herein - have established a relationship between electrode micro structure and bubble-free operation that is not known from the prior art.

[0017] SUMMARY OF THE INVENTION

[0018] It is an object of the invention therefore to provide a novel gas diffusion electrode with gas-breathing properties that can enable electrolysis, particularly alkaline water electrolysis, at scale without excessive gas crossover and bubble formation. A further object relates to providing a proton exchange membrane electrolyser, an anion exchange membrane electrolyser, a hybrid electrolyzer, a metal air battery, or any other electrosynthesis device that requires a gas-liquid interface comprising the gas diffusion electrode. Yet another aim relates to providing an electrolyzing device comprising a first electrode configured to generate a first gas and be in fluid contact with a first gas collector for the first gas; a second electrode configured to generate a second gas and be in fluid contact with a second gas collector for the second gas; wherein the first and second electrodes are arranged to face each other to define an electrolyte space for receiving an electrolyte between the first gas and the second electrode, and wherein at least one of the first and second electrode comprises the novel gas diffusion electrode. The concerning electrolyte space optionally consists of a highly porous spacer material. The invention also aims at providing a method of manufacturing the gas diffusion electrode.

[0019] According to a first aspect, the invention provides for this purpose an electrolyzing device comprising a first electrode configured to generate a first gas and be in fluid contact with a first gas collector for the first gas; a second electrode configured to generate a second gas and be in fluid contact with a second gas collector for the second gas; wherein the first and second electrodes are arranged to face each other to define an electrolyte space for receiving an electrolyte between the first electrode and the second electrode, wherein at least one of the first and second gas diffusion electrode is a gas diffusion electrode comprising: a gas diffusion backbone layer facing away from the electrolyte space and comprising a conductive porous material; a catalyst layer facing the electrolyte space and comprising an electrochemically active catalyst, wherein the catalyst layer further comprises hydrophobic particles with an average particle size measured according to ISO 22412 between 10 nm - 5pm and a microporous layer arranged between the gas diffusion backbone layer and the catalyst layer and comprising conductive particles with an average particle size measured according to ISO 22412 between 15 nm - 2 pm and hydrophobic particles with an average particle size measured according to ISO 22412 between 10 nm - 5 pm.

[0020] In a second aspect of the invention, there is provided a gas diffusion electrode for use as the first or the second electrode or the first and the second electrode in a device according to the first aspect.

[0021] In a third aspect of the invention, there is provided a proton exchange membrane electrolyser, an anion exchange membrane electrolyser, a hybrid electrolyzer, a metal air battery, or any other electrosynthesis device requiring a gas-liquid interface and comprising the invented gas diffusion electrode.

[0022] A fourth aspect of the invention finally provides a method of manufacturing a gas diffusion electrode as claimed according to the second aspect.

[0023] The gas diffusion electrode may be used advantageously in a water splitting process comprising bubble-free alkaline water electrolysis. The inventors found out that an electrolysis device that comprises the novel gas diffusion electrode allows to break apart water into hydrogen and oxygen gases without producing a substantial amount of bubbles, which is a common performance limiter in prior art methods. Bubble formation may also be a source for gas crossover, such as hydrogen gas crossover, leading to a potential formation of explosive mixtures.

[0024] Prior art electrolysis techniques, such as alkaline or proton exchange membrane electrolysis, typically rely on a physical barrier (in the form of a membrane or separator) to keep the hydrogen and oxygen gases produced during the process separate. The present invention however does away with the need for such barriers. It is not excluded however that the invention according to the first aspect uses a barrier, particularly in the form of a highly porous spacer.

[0025] A key feature of the novel gas diffusion electrode is its ability to "wick away" or remove first and second gases, such as hydrogen and oxygen gases, as they are produced, substantially preventing bubble formation in the liquid electrolyte, while at the same time preventing liquid electrolyte to leak out of the device. This may be achieved through an inventively designed microstructure and composition of the catalyst layer and water-repellent backbone layer within the gas diffusion electrode, as claimed. The gas diffusion electrode as claimed allows the produced gases to move away from the site where the electrolyte is being split, enabling the electrolysis process to occur more efficiently and at a faster rate. The gas diffusion electrode as claimed in embodiments thereof may further withstand harsh conditions, such as in (strong) alkaline solutions and high temperatures, typical of alkaline electrolysis.

[0026] By operating efficiently at high current densities (typically >100 mA per square cm), the invented device and gas diffusion electrode may produce a first gas, such as hydrogen, and a second gas, such as oxygen, at scales and speeds that are relevant for industrial applications, all while avoiding the complications and inefficiencies associated with gas bubbles in the electrolyte solution, such as gas crossover. The invented device and gas diffusion electrode for instance allows producing a first and a second gas at a relatively high purity. For instance, the first and / or second gas may have a purity of > 98 vol.%, meaning that < 2 vol.% of other ingredients besides the gas is present in the gas. For instance, the first gas may comprise < 2 vol.% of the second gas, or the second gas may comprise < 2 vol.% of the first gas.

[0027] The invented gas diffusion electrode according to the second aspect was shown to be resistant to the corrosive alkaline conditions, ensuring longevity and stability during operation. The invented gas diffusion electrode according to the second aspect may also be used advantageously in other electrochemical devices, such as in metal-air batteries, anion-exchange membrane (AEM) and proton-exchange membrane (PEM) electrolysis devices, such as hybrid water electrolysis devices with oxidation species such as urea, ethanol, methanol, nitric oxide, nitrogen, or hydrazine as well as in carbon dioxide electrolysis or any other electrosynthesis device necessitating a gas-liquid interface.

[0028] According to a first aspect of the further developments, a distance between the surface where the product (hydrogen or oxygen) is produced and the aerophilic, hydrophobic surface may be kept smaller than 5 pm and preferably smaller than 2 pm. When the diffusive path length exceeds multiple pm, local supersaturation values may exceed the critical supersaturation limit, triggering bubble nucleation.

[0029] In an embodiment, in addition to the distance between the catalyst particle and hydrophobic network, the hydrophobic surface may be interconnected to form a hydrophobic network towards the electrode backbone / gas phase. If this condition is not met, bubbles may still be able to form.

[0030] In a further embodiment, the hydrophobic network may be properly distributed to ensure the condition for the maximum distance is met across the entire electrode surface.

[0031] Some of the controllable experimental properties and procedures to achieve the features of the above embodiments include: i) catalyst ink composition, e.g., hydrophobic particle type and size, catalyst particle type and size, hydrophobic particle to catalyst particle ratio, catalyst ink loading, additives, in particular rheology agents such as methyl cellulose and surfactants such as trition x-100; ii) spray conditions, e.g., ultrasonic spray coating, hand spray coating, plate temperature, etc.; and iii) sintering temperature profile.

[0032] According to a second aspect of the further developments, spray coating PTFE and catalyst may create maximum distances between the electrochemically active site and the hydrophobic backbone of less than a pm, maintaining supersaturation below the nucleation threshold even at high current densities. This theoretically allows to not only operate at -100 mA / cm2 bubble-free but also up to 2000 mA / cm2 (and beyond). This maximum distance is supported by SEM and EDX and simulation results. If these conditions are not met, difficulties may be experienced in operating bubble-free at higher current densities. By means of microstructural engineering and using the knowledge on the abovementioned structure-function relationship bubble formation may be effectively prevented.

[0033] DETAILED DESCRIPTION OF THE INVENTION

[0034] The invention as disclosed above will now be elucidated further.

[0035] The electrolyzing device as claimed in claim 1 comprises a first gas diffusion electrode configured to generate a first gas and be in fluid contact with a first gas collector for the first gas; a second electrode configured to generate a second gas and be in fluid contact with a second gas collector for the second gas; wherein the first and second electrodes are arranged to face each other to define an electrolyte space for receiving an electrolyte between the first electrode and the second electrode, wherein at least one of the first and second electrode is a gas diffusion electrode comprising: a gas diffusion backbone layer facing away from the electrolyte space and comprising conductive porous material a catalyst layer facing the electrolyte space and comprising an electrochemically active catalyst, wherein the catalyst layer further comprises hydrophobic particles with an average particle size measured according to ISO 22412 between 10 nm - 5pm, and a microporous layer, arranged between the gas diffusion backbone layer and the catalyst layer, and comprising conductive particles with an average particle size measured according to ISO 22412 between 15 nm - 2 pm and hydrophobic particles with an average particle size measured according to ISO 22412 between 10 nm - 5 pm.

[0036] The conductive porous material is electrically conductive, and preferably also thermally conductive. Suitable conductive porous material may be selected from felt, cloth, paper or foam, non-fibrous material or combinations thereof, without however being limited to these.

[0037] The electrochemically active catalyst is, potentially, supported on a catalyst support.

[0038] The electrolyzing device may be used in alkaline water electrolysis. The gas diffusion electrode may be configured to be particularly advantageous for this purpose. The first gas diffusion electrode is configured to generate a first gas and be in fluid contact with a first gas collector. By this is meant that the first gas generated by the first gas diffusion electrode may be carried to the first gas collector through a fluid connection, being a connection for a liquid (such as electrolyte) or a gas (such as the first gas). Likewise, the second gas diffusion electrode is configured to generate a second gas and be in fluid contact with a second gas collector. By this is meant that the second gas generated by the second gas diffusion electrode may be carried to the second gas collector through a fluid connection, being a connection for a liquid (such as electrolyte) or a gas (such as the second gas).

[0039] The first and / or second electrode may be positioned adjacent to a first and / or second gas collector respectively, or they may be placed at a distance therefrom.

[0040] The electrolyzing device as claimed may be used with advantage for the splitting of water or an aqueous solution as electrolyte into hydrogen and oxygen gas. It is particularly suitable for use in alkaline, optionally membrane-free, electrolysis. Apart from having performance benefits, alkaline membrane-free electrolyzers have several intrinsic advantages over other electrolyzer types used for green hydrogen production. Compared to proton exchange membrane (PEM) electrolyzers, they may exhibit less dependence on critical materials, such as Iridium, a rare-earth metal with limited global supply, and Nafion™, a perfluoroalkyl substance (PFAS) that is typically used as membrane and ionomer. Compared to ordinary alkaline water electrolyzers, new operational modes can be implemented to improve the load cycling capability (i.e., responsiveness to fluctuations in the power supply and integration with renewable energy sources). Compared to anion exchange membrane (AEM) electrolyzers, the projected system lifetime is significantly increased, due to the elimination of the unstable membrane. Compared to solid oxide electrolyzers and all low-temperature electrolyzer types disclosed above, a membrane-free electrolyzer has a lower capital expenditure, due to enhanced performance and a simplified balance-of-plant.

[0041] The gas diffusion backbone layer facing away from the electrolyte space comprises conductive porous material, such as felt, cloth, paper or foam, a non-fibrous material or combinations thereof. The conductivity of the porous material may be selected according to the purpose. In this respect it is worthwhile to mention that the conductivity of the solid gas diffusion electrode is typically much larger than the conductivity of the liquid electrolyte. Indeed, a typical conductivity of the gas diffusion electrode would be around l-2xl07S / m at 20°C, while a typical conductivity of the liquid electrolyte would be around 50-100 S / m at 20°C. These values are not to be interpreted as limiting the invention in any way but are only given for informative reasons, i.e. to illustrate what is meant by ‘much larger’. The conductivity of the gas diffusion backbone layer (and of the catalyst and microporous layers) may be selected according to the above. A suitable conductivity of a carbon gas diffusion backbone layer (GDL) could be in the range of 200 to 500 S / m for instance, while the conductivity of a pure metal diffusion backbone layer could be as high as l.OxlO7S / m or more, such as 1.43xl07S / m for a pure nickel gas diffusion backbone layer. This illustrates the broad range of suitable conductivities for the gas diffusion backbone layer, and also for the catalyst layer, the microporous layer, and eventually the complete assembled gas diffusion electrode. Given the typical conductivity range of an electrolyte, any conductivity of the gas diffusion backbone layer, as well as the catalyst and microporous layer, above 50 S / m is suitable, with the above-disclosed conductivity of pure nickel as a theoretical maximum. A more preferred conductivity of the gas diffusion backbone layer is preferably such that the conductivity is 100 S / m or more, more preferred 500 S / m or more, and most preferred 1000 S / m or more.

[0042] According to an embodiment of the invention, the conductive material of the gas diffusion backbone layer is selected from nickel, titanium or carbon, or combinations thereof. Other materials such as conductive polymers, metals like steel, and combination thereof may also be suitable. For instance, a suitable gas diffusion backbone layer may comprise a carbonous backbone and microporous layer, such as Freudenberg H15C14, a fluorinated polyolefin, such as Dupont Teflon PTFE Disp 30, 220 nm average particle size, as hydrophobic agent, and platinum on carbon as catalyst (Tanaka TEC10V40E). Carbon however may be susceptible to oxidation and be relatively vulnerable to corrosion. For this reason, a carbonous backbone layer is preferably used in environments with relatively mild alkalinity and low oxidative conditions only. A carbonous backbone appears to be particularly limited for use as a hydrogen evolution electrode in a PEM electrolyzer, a hydrogen evolution electrode in a AEM electrolyzer, and as a hydrogen evolution electrode in a membrane-free alkaline electrolyzer. Another electrode configuration may comprise a nickel backbone, for instance Bekaert Currento® 2NI18-0,25 Sintered Metal Fiber Matrix for the gas diffusion backbone layer and Ni nanoparticles with an average individual particle size of 90 nm for the microporous layer, PTFE as hydrophobic agent and either nickel or nickel-iron double layered hydroxide or palladium nanoparticles as catalyst particles. This electrode configuration was shown to be less susceptible to oxidation and may therefore be suitable for operation in heavy alkaline and oxidative conditions. Thus, the use of a gas diffusion electrode with a nickel backbone layer may be particularly suitable as a substantially bubble-free hydrogen evolution or oxygen evolution electrode in a membrane-free alkaline electrolyzer, an oxygen evolution and reduction electrode in metal-air batteries, and as a hydrogen evolution electrode in an AEM electrolyzer. Yet another electrode configuration comprises a gas diffusion backbone layer made of titanium particles and fibres and may be used in PEM electrolysis in particular.

[0043] According to the invention, the gas diffusion backbone layer comprises conductive porous material, such as felt, cloth, paper, any non-fibrous material or foam or combinations thereof. Such porous material may contain fibres. When fibrous, the individual fibre diameter may conveniently be selected within a broad range. Preferred individual fibre diameters may range from 1-50 pm (microns), more preferably from 2- 40 pm, even more preferably from 5-30 pm, and most preferably from 10-23 pm.

[0044] The conductive porous material may be intrinsically conductive or may be provided to be conductive by a conductive coating. The wording ‘conductive’ may comprise ‘electrically conductive’ or ‘thermally conductive’ or both.

[0045] The uncompressed porosity of the porous material may be defined as the fraction of void space in the uncompressed gas diffusion backbone layer. The uncompressed porosity of the gas diffusion backbone layer may conveniently be selected to be higher than 20 vol.%, preferably higher than 40 vol.%, more preferably higher than 60 vol.%, and even more preferably higher than 80 vol.%. A preferred uncompressed porosity is lower than 95 vol.%, more preferably lower than 90 vol.%, and most preferably lower than 85 vol.%.

[0046] The gas diffusion backbone layer in an embodiment according to the invention is hydrophobic in that it comprises hydrophobic particles, either consisting of separate individual particles or of particles that are sintered, or of combinations of individual and sintered particles. The particles may be provided in a coating material. In another embodiment of the invention, the hydrophobicity is not provided by hydrophobic particles but by a hydrophobic coating and / or by a hydrophobic agent. Hydrophobicity of the gas diffusion backbone layer however is not essential to the invention.

[0047] The hydrophobicity of the catalyst layer, the microporous layer, and, optionally, the gas diffusion backbone layer, may be quantified by means of a contact angle, suitably measured according to the sessile drop method (OCA30, developed and manufactured by Data-physics). Suitable contact angles are defined to be at least 90°, preferably at least 105°, more preferably at least 110°.

[0048] The degree of hydrophobization may be selected within wide limits. Particularly preferred degrees of hydrophobization are selected in the range of 0.00001-50 wt.%, more preferably from 2-40 wt.%, even more preferably from 3-35 wt% and most preferably from 5-30 wt.%. The degree of hydrophobization is herein defined as the weight ratio of hydrophobic material relative to the total weight of the material in a given layer.

[0049] The thickness of the gas diffusion backbone layer may conveniently be selected to preferably be at least 50 microns. In view of the added costs of having a thicker electrode and the effect on a total stack thickness, the thickness of the gas diffusion backbone layer may preferably be limited to 2000 microns. However, a higher thickness is possible. More preferably, the thickness may be selected from 100 to 1000 microns. Most preferably, the thickness may be selected from 100 to 250 microns.

[0050] According to the invention, the gas diffusion electrode(s) of the electrolytic device comprise a catalyst layer facing the electrolyte space and comprising an electrochemically active catalyst and hydrophobic particles, and a microporous layer comprising conductive particles and / or agglomerates thereof. The catalyst and microporous layers are typically two separate layers. However, it is not excluded that the catalyst layer and the microporous layer are combined into a single microporous / catalyst layer, according to an embodiment of the invention. The combined microporous / catalyst layer then comprises the conductive and hydrophobic particles, as claimed.

[0051] In addition, the electrolyzing device according to the first aspect may comprise further layers than the backbone, catalyst and microporous layers. It is not excluded that additional layers are interspersed with two of the backbone, catalyst and microporous layers. For instance, a hydrophilic bubble suppression layer may be added to the gas diffusion electrode layers.

[0052] A preferred embodiment of the invention according to the first aspect however provides a device wherein the gas diffusion backbone layer, the catalyst layer and the microporous layer are positioned in direct contact with each other. This does not exclude the presence of additional layers positioned outside such contiguous assembly of the gas diffusion backbone layer, the catalyst layer and the microporous layer. For instance, a hydrophilic bubble suppression layer may be arranged on top of the catalyst layer to face the electrolyte space, further limiting gas leakage in the electrolyte. Additional layers may also be provided at the side facing the gas diffusion backbone layer.

[0053] The device as claimed according to the first aspect further comprises a catalyst layer facing the electrolyte space. This catalyst layer comprises an electrochemically active catalyst, potentially provided on a conductive catalyst support, and further comprises hydrophobic particles, preferably made of a fluorinated polyolefin such as poly(tetrafluoroethylene) (PTFE), with an average particle size measured according to ISO 22412 between 10 nm - 5 pm, preferably between 10 nm - 4 pm, more preferably between 10 nm - 3 pm, more preferably between 10 nm - 2 pm, more preferably between 10 nm - 1 pm, more preferably between 20 nm - 1 pm, more preferably between 30 nm - 1 pm, more preferably between 40 nm - 1 pm, more preferably between 50 nm - 1 pm more preferably between 100 nm - 1 pm, even more preferably between 100 nm - 750 nm, more preferably between 100 nm - 500 nm, and most preferably between 150 - 400 nm, such as about 230 nm. The wording ‘between’ should be read at all instances to also encompass the extremes of the indicated range. In other words, ‘between 10 nm - 5 pm’ should be interpreted to include 10 nm and 5 pm.

[0054] The catalyst used in the catalyst layer may be chosen in accordance with the specific application involved. Those skilled in the art will know which catalyst to select for the purpose. According to embodiments of the invention, a device is provided wherein the catalyst is provided in the form of catalyst particles, and the catalyst particles are selected from the group consisting of platinum-coated carbon particles, nickel-coated carbon particles, iridium oxide nanoparticles, ruthenium nanoparticles, nickel nanoparticles and nickel-iron layered double hydroxides, or combinations thereof.

[0055] An advantageous embodiment provides a device wherein the electrochemically active catalyst in the catalyst layer comprises catalyst particles with an average hydrodynamic particle size as measured according to ISO 22412 of 1 - 500 nm, more preferably of 5 - 250 nm, even more preferably of 10 - 100 nm, and most preferably of 15 - 30 nm. The average hydrodynamic particle size also relates to the size of supported catalyst particles. The average hydrodynamic particle size relates to the size of individual particles. However, the supported catalyst particles may agglomerate to agglomerates of approximately 100 nm size or more.

[0056] The catalyst layer as claimed exhibits mixed wettability, and may be applied by, for example but not limited to, (ultrasonic) spray coating, blade coating, stencil coating, decal transfer method, electroless deposition or electrochemical deposition. The purity, uniformity, and size of the hydrophobic particles and catalyst particles may be important in further improving the invented gas diffusion electrode. To prevent bubble formation, gas that is produced on the catalyst particles preferably has to be transported to a hydrophobic network formed by the hydrophobic particles before critical supers aturation level is reached and bubbles are formed. When the size of the catalyst particles and the hydrophobic particles is relatively small, as claimed, the diffusive transport pathway of the dissolved gases is shortened, thus favouring the gas removal rate and preventing bubble formation in the liquid. The inventors have also found that potential impurities that may be present, for instance in a dispersion carrying the catalyst and / or hydrophobic particles, or as dust particles, may obstruct bubble-free performance at industrially relevant conditions (of typically >100 mA / cm2). Furthermore, a potential hydrophobic particle network in one or more of the gas diffusion electrode layers is preferably sufficiently open to favour gas transport out of the electrolytic device or cell.

[0057] Given the above, a preferred embodiment of the invention provides a device wherein at least one of the catalyst layer and the microporous layer comprises impurities other than the hydrophobic particles, and that the impurities have a size < 200 pm, more preferably < 175 pm, even more preferably < 150 pm, even more preferably < 125 pm, and most preferably < 100 pm.

[0058] The fraction of (supported) catalyst particles in the catalyst layer may be selected from 0.2 up to to 0.95, relative to the total volume of the solid phase of the catalyst layer, which comprises hydrophobic particles and possibly also conductive fillers and catalyst support material. In a further embodiment, the fraction of (supported) catalyst particles in the catalyst layer may be selected from above 0.05 to 0.8. The fraction of hydrophobic particles in the dried catalyst layer may be selected from 5 to 80 vol%, preferably from 29 to 79 vol.%, more preferably from 29 to 70 vol%, and most preferably from 29 to 50 vol.%, such as about 35 vol.%. In a further embodiment, the fraction of hydrophobic particles in the dried catalyst layer may be selected from 6 to 80 vol.%, preferably from 10 to 40 vol.%, more preferably from 10 to 20 vol.% and most preferably from 10 to 15 vol.%, such as about 15 vol.%. The remaining fraction may then be occupied by the electrochemically active catalyst, optional catalyst support material, and possibly conductive fillers. During the formation of the final layer, the catalyst ink may be supplemented with pore-forming and templating agents, which are subsequently removed prior to the completion of the final layer.

[0059] The thickness of the catalyst layer may conveniently be defined in terms of areal loading or areal density, which describes the amount of dried catalyst layer material deposited on the microporous or gas diffusion backbone layer per square cm. Dividing the areal density by density yields an average layer thickness. The thickness of the catalyst layer (expressed in areal density) may suitably be selected from 0.5 - 20 mg / cm2, preferably from 0.8 - 10 mg / cm2, more preferably from 1.0 - 3 mg / cm2, such as about 1.1 mg / cm2. Although the hydrophobic particles of the microporous layer, the catalyst layer and optionally the gas diffusion backbone layer may be selected to differ from each other (either in particle size distribution or in material or in both) a practical embodiment relates to a device wherein the hydrophobic particles of the microporous layer, the catalyst layer and optionally the gas diffusion backbone layer are made from the same material. More preferably, their particle size distribution is also substantially the same.

[0060] Suitable hydrophobic particles for use in one, two or all of the backbone, microporous and catalyst layer may be selected from polymers such that a resulting water contact angle as measured according to the sessile drop method (OCA30, developed and manufactured by Data-physics) fulfils a minimum contact angle requirement, and may preferably be selected from the group of polyolefins, such as polypropylene (PP) and polyethylene (PE), fluorinated polyolefins such as polytetrafluoroethylene (PTFE) and fluorinated ethylene propylene (FEP), polystyrene (PS), polyvinylchloride (PVC), polyethylene terephthalate (PET), polyurethane (PU), polypropylene oxide (PPO), polysiloxanes such as polydimethylsiloxane (PDMS), and polysulfones, such as polysulfone (PSU), polyethersulfone (PES / PESU) and polyphenylene sulfone (PPSU), and sulfonated polymers, such as sulfonated poly(ether ether ketone) (SPEEK), PTFE being preferred.

[0061] Other suitable hydrophobic particles may for instance comprise (crystalline) SiO2 microspheres. In preferred embodiments however, (crystalline) SiO2 microspheres are excluded from at least one of the backbone, catalyst and microporous layers.

[0062] At least one of the gas diffusion electrodes of the electrolytic device according to the invention also comprises a microporous layer. The microporous layer comprises conductive particles with an average particle size according to ISO 22412 between 15 nm - 2 pm and hydrophobic particles with an average particle size according to ISO 22412 between 10 nm - 5 pm. The microporous layer is arranged between the gas diffusion backbone layer and the catalyst layer, but may also be combined with the catalyst layer, if desired. In the latter embodiment, the particle sizes and thickness disclosed hereinabove apply to the combined catalyst / microporous layer. The claimed particle size of the conductive particles in the microporous layer applies to individual particles. Indeed, it is not excluded that the conductive particles (and maybe the hydrophobic particles in the appropriate layers also) agglomerate into agglomerates of some size. Preferably however, agglomeration of particles is not favoured and may be substantially prevented or hindered, for instance by stirring, ultrasonic probing treatment, ultrasonic bath, planetary mixer or ball milling. The same holds for the catalyst particles or catalyst support material of the catalyst layer. For instance, in case of Pt / C catalyst particles, the carbon particles may agglomerate, not the Pt.

[0063] The hydrophobic microporous layer is typically applied on top of the gas diffusion backbone layer by methods such as spray coating, drop casting, blade coating, slot-die coating, stencil coating, any other coating technique, or a sequential combination of the aforementioned coating techniques, specifically blade coating followed by spray coating. Alternatively, the microporous layer may be applied by dynamic hydrogen bubble templating or electrochemical co-deposition of nickel and a hydrophobic agent. Microscopic cracks may occur during the application process and curing process of the microporous layer. The applied hydrophobic microporous layer is preferably sufficiently ‘crack- free’, meaning that only cracks with a size of smaller than 3 pm are allowed to prevent liquid leakage. This may be achieved by avoiding or removing impurities in the microporous layer before providing it to the other gas diffusion electrode layers.

[0064] The fraction of hydrophobic particles in the dried microporous layer may be selected from 10-75 wt.%, more preferably from 15-45 wt.%, and most preferably from 15-25 wt.%, such as about 20 wt.% for instance, the wt.% being relative to the total weight of the dried microporous layer.

[0065] The microporous layer comprises conductive particles with an average particle size according to ISO 22412 between 15 nm - 2 pm and hydrophobic particles with an average particle size according to ISO 22412 between 10 nm - 5 pm. In a preferred embodiment of the device as claimed, the conductive particles of the microporous layer have an average particle size according to ISO 22412 between 10 nm - 4 pm, more preferably between 10 nm - 3 pm, more preferably between 10 nm - 2 pm, more preferably between 10 nm - 1 pm, more preferably between 20 nm - 1 pm, more preferably between 30 nm - 1 pm, more preferably between 40 nm - 1 pm, more preferably between 50 nm - 1 pm more preferably between 100 nm - 1 pm, even more preferably between 100 nm - 750 nm, more preferably between 100 nm - 500 nm, and most preferably between 150 - 400 nm, such as about 230 nm.

[0066] The fraction of conductive particles in the dried microporous layer may be selected from 50-95 wt.%, more preferably from 60-90 wt.%, and most preferably from 75-85 wt.%, relative to the total weight of the dried microporous layer.

[0067] According to another embodiment, a device is provided wherein the conductive particles of the microporous layer are selected from nickel, titanium or carbon particles, or combinations thereof, preferably nickel particles.

[0068] The thickness of the microporous layer may be selected from 10 pm to 200 pm, preferably from 20 to 80 pm, more preferably from 30 to 70 pm, such as about 50 pm for instance. The lower limit of the thickness of the microporous layer preferably corresponds to a minimal thickness sufficient for covering a potential roughness of the gas diffusion backbone layer. The combined thickness of the microporous layer and the gas diffusion backbone layer are preferably chosen based on a number of criteria that may be conflicting. For instance, the gas diffusivity of the gas diffusion electrode should be high enough without however incurring a too high electrical resistivity. Also, the combines thickness may be instrumental in maintaining mechanical integrity of the gas diffusion electrode. Those skilled in the art will be able to select the combined thickness of the microporous layer and the gas diffusion backbone layer without undue burden.

[0069] The device as claimed according to the first aspect of the invention may be used with a membrane separator between the electrodes. However, its advantages become in particular apparent in an embodiment wherein the electrolyte space for receiving the electrolyte between the first gas diffusion electrode and the second gas diffusion electrode does not contain a membrane separator, or, alternatively, does contain a spacer that is highly porous. In a membrane-free or membrane-less electrolyzer, the electrolyte fluid is managed differently than in a conventional alkaline, AEM or PEM electrolyzer. In traditional alkaline, PEM and AEM electrolyzers, the electrolyte is fed from the back of (one or both of) the electrodes, resulting in a flooded electrode and the generation of gaseous hydrogen and oxygen as bubbles. In membrane-free electrolyzers, the electrolyte is fed in between the two electrodes, creating a natural barrier for hydrogen and oxygen mixing. The inventors in the present disclosure propose the use of an electrode in such a system where the electrode composition is finely tuned to maximize hydrogen or oxygen gas extraction from the liquid electrolyte in the gas diffusion backing layer, as claimed.

[0070] The claimed invention according to the first and second aspects has several desirable properties. At least one, and preferably both, of the gas diffusion backbone and microporous layers for instance are substantially impenetrable by the liquid electrolyte to prevent flooding of the backing layer, and complete flooding in particular. The catalyst and backbone layer are sufficiently electrically conductive, and the first and second gas, such as hydrogen and oxygen respectively, are transported out as a gas before the concentrations in the liquid electrolyte reach their critical supers aturation levels, resulting in the formation of bubbles, preferably at a separation efficiency larger than 99%.

[0071] In addition to the above, the materials used in the catalyst, microporous and backbone layer are stable in industrial operating conditions, which may be summarized as 1-60 bar, 20-120°C, 1-6M KOH for instance. The gas diffusion electrode as claimed may survive erosion rates of ±1.5 pm / year and retain functionality.

[0072] The device as claimed according to the first aspect may comprise two gas diffusion electrodes that are substantially the same. In another embodiment, the gas diffusion electrodes have the same layers apart from their thickness. In yet another embodiment, the gas diffusion electrodes use different hydrophobic particles. They may also differ in other aspects, if desired.

[0073] The device according to the first aspect of the invention may be applied as a hydrogen or oxygen-evolving electrode in a gas-breathing configuration. The opposing half-cell can either be in a gas-breathing or flow-through configuration. Such an embodiment provides a device wherein one of the first or second electrodes comprises: a gas diffusion backbone layer facing away from the electrolyte space and comprising conductive porous material; a catalyst layer facing the electrolyte space and comprising an electrochemically active catalyst, wherein the catalyst layer further comprises hydrophobic particles with an average particle size measured according to ISO 22412 between 10 nm - 5pm, and a microporous layer comprising conductive particles with an average particle size measured according to ISO 22412 between 15 nm - 2 pm and hydrophobic particles with an average particle size measured according to ISO 22412 between 10 nm - 5 pm and arranged between the gas diffusion backbone layer and the catalyst layer, and the other electrode comprises a flow-through electrode.

[0074] Providing such a hybrid configuration of a gas-breathing hydrogen electrode and a flow-through oxygen electrode may aid in mitigating liquid breakthrough in the hydrogen compartment. Further, it may reduce a number of design constraints (related to scaling, material choice and the influence of hydrodynamic pressure) from the electrolyzing device and may therefore lead to operation with improved stability and durability. Additionally, it may be beneficial in the use of the system to produce more valuable oxidative products on the anode.

[0075] The invention also provides an electrolytical multi-cell stack, comprising a plurality of devices as claimed in accordance with the first aspect, whereby the plurality of the devices are electrically connected.

[0076] According to a second aspect of the invention, a gas diffusion electrode for use as the first or the second gas diffusion electrode or the first and the second electrode in a device as claimed according to the first aspect of the invention is provided, the gas diffusion electrode comprising: a gas diffusion backbone layer facing away from the electrolyte space and comprising conductive porous material; a catalyst layer facing the electrolyte space and comprising an electrochemically active catalyst, wherein the catalyst layer further comprises hydrophobic particles with an average particle size measured according to ISO 22412 between 10 nm - 5pm, and a microporous layer comprising conductive particles with an average particle size measured according to ISO 22412 between 15 nm - 2 pm and hydrophobic particles with an average particle size measured according to ISO 22412 between 10 nm - 5 pm, and arranged between the gas diffusion backbone layer and the catalyst layer. It is expressly stated that all embodiments disclosed above in the context of the device according to the first aspect of the invention also apply to the gas diffusion electrode, as claimed in accordance with the second aspect of the invention, when applicable.

[0077] The gas diffusion electrode as claimed according to the second aspect of the invention may also be applied in other gas-liquid electrochemical systems than the claimed electrolytic device. According to the third aspect of the invention, a proton exchange membrane (PEM) electrolyser, an anion exchange membrane (AEM) electrolyser, a hybrid electrolyzer or a metal air battery is provided, comprising a gas diffusion electrode as claimed in accordance with the second aspect.

[0078] The function of the gas diffusion electrode may slightly change with respect to its final application. In a PEM or AEM electrolyzer, the electrode may facilitate asymmetric feeding of water in the cell and promote production of high-purity hydrogen, while retaining water close to and in the membrane, potentially preventing dry-out and local heating of the cell. In a metal air battery, the electrode configuration may be used as a reversible oxygen-electrode. In discharge mode, the hydrophobic functionality of the electrode may prevent flooding of the electrode and leakage of electrolyte out of the cell. In charge mode, the electrode may effectively extract produced oxygen from the liquid electrolyte, without losing oxygen as gas bubbles in the liquid, reducing mass transport limitations and gas hold-up in the system. In a hybrid electrolyzer, the electrode may facilitate the separation of gaseous hydrogen from the production of a valuable anodic by-product. To improve electrochemical activity of the electrode, either one or a mixture of catalyst particles may be required in this application.

[0079] According to a fourth aspect of the invention, there is provided a method of manufacturing a gas diffusion electrode as claimed according to the second aspect, the method comprising providing a gas diffusion backbone layer comprising conductive porous material; applying a microporous layer comprising conductive particles with an average particle size measured according to ISO 22412 between 15 nm - 2 pm and hydrophobic particles with an average particle size measured according to ISO 22412 between 10 nm - 5 pm onto the gas diffusion backbone layer; applying a catalyst layer comprising an electrochemically active catalyst, wherein the catalyst layer further comprises hydrophobic particles with an average particle size measured according to ISO 22412 between 10 nm - 5 pm onto the microporous layer.

[0080] The gas diffusion electrode may be produced from any non-reactive, conductive backbone layer, any non-reactive conductive particle as a filler, any non-reactive, non- conductive hydrophobic particle, and an electrocatalyst for the desired reaction. Any non-reactive non-conductive particle may act as a filler or a forming agent.

[0081] The manufacturing method allows producing a gas diffusion electrode with a catalyst layer with mixed wettability (hydrophobic non-conductive particles and hydrophilic catalyst particles, as claimed in claim 1), wherein the size and composition of the hydrophobic and catalyst particles have been selected as claimed.

[0082] In an embodiment of the method as claimed, the gas diffusion backbone layer is hydrophobized by contacting the conductive porous material with an aqueous dispersion of a hydrophobic polymer at a concentration of 0.00001-20 wt.%.

[0083] Yet another embodiment relates to a method wherein the gas diffusion backbone layer is hydrophobized by contacting the conductive porous material with an aqueous dispersion of a hydrophobic polymer at a concentration of 0.1-20 wt.%.

[0084] Yet another embodiment provides a method wherein the microporous layer is hydrophobized by contacting the conductive particles with an aqueous dispersion. Preferably the conductive particles are nickel-based. This embodiment offers a method of manufacturing nickel-based, crack- free hydrophobic microporous layers.

[0085] In yet another embodiment, a composition that comprises both conductive particles, such as nickel particles, and hydrophobic particles, such as PTFE particles, is prepared, and said composition is then sprayed or blade-coated. Apart from dip-coating the microporous layer, it may also be spray-coated. In a preferred embodiment, the hydrophobic particles in the catalyst layer, the microporous layer and, optionally, the gas diffusion backbone layer, may be sintered. Sintering of the hydrophobic particles may prolong the durability of any one of the catalyst, microporous and gas diffusion backbone layer, and hinder or even prevent erosion of the gas diffusion electrode.

[0086] BRIEF DESCRIPTION OF THE FIGURES

[0087] Examples of the invention will now be elucidated with reference to the following figures, without however being limited thereto. In the figures:

[0088] Figure 1 schematically shows an individual electrolyzer cell equipped with gas diffusion electrodes in accordance with an embodiment of the invention;

[0089] Figure 2 schematically shows an electrolyzing device comprising the electrolyzer stack according to an embodiment of the invention for use in hydrogen production;

[0090] Figures 3A-C schematically show cross-sectional views of a gas diffusion backbone layer, a catalyst layer and a microporous layer according to an embodiment of the invention;

[0091] Figures 4A-C schematically show different configurations of a membrane-less electrolysis cell in accordance with an embodiment of the invention, wherein (a) shows a gas-breathing configuration, (b) a flow-through configuration, and (c) a hybrid configuration;

[0092] Figures 5A-C schematically show different configurations of an electrochemical device according to embodiments of the invention with (a) a proton exchange membrane electrolyzer where the hydrogen-side is operated dry, (b) an anion exchange membrane electrolyzer where the hydrogen-side is operated dry, and (c) a metal-air battery where the air- side is operated dry;

[0093] Figure 6 schematically illustrates an experimental window-cell setup used for the preliminary screening of the gas diffusion electrodes;

[0094] Figure 7 shows a graph of the measured hydrogen gas flow and calculated separation efficiency, defined as the measured hydrogen gas flow divided by the total hydrogen production rate by Faraday’s law; and Figure 8 shows a graph of a temperature-time profile used in preparing the hydrophobized gas-diffusion backbone layer, according to an embodiment of the invention.

[0095] Figure 9 shows a graph of a temperature-time profile used in preparing the hydrophobized nickel gas diffusion backbone layer, according to an embodiment of the invention.

[0096] Figure 10 shows a schematic representation of an electrolyzer cell setup for a flow cell experiment, according to an embodiment of the invention.

[0097] Figure 11 shows a bar chart representing catalyst and gas diffusion media support layer area specific resistance with 5 and 80 vol.% PTFE content, according to an embodiment of the invention.

[0098] Figure 12 shows a bar chart representing separation efficiency of manufactured gas extraction electrodes as a function of PTFE content in the catalyst layer, according to an embodiment of the invention.

[0099] Figure 13 shows a graph representing IR-corrected current density-voltage curves for cathodes with varying PTFE content, according to an embodiment of the invention. Figure 14 shows a comparison of saturation profiles in a traditional catalyst layer and a mixed wettability catalyst layer, according to an embodiment of the invention.

[0100] Figure 15 shows a graph representing hydrogen removal as a function of hydrophobic particle radius and a graph representing maximum saturation as a function of current density, according to an embodiment of the invention.

[0101] DESCRIPTION OF EXEMPLARY EMBODIMENTS

[0102] Characterization Methods

[0103] Gas diffusion backbone thickness measurement

[0104] In all examples, the thickness of the gas diffusion layer backbone was provided by the supplier. In addition, the thickness of the gas diffusion backbone layer was measured by obtaining a cross-sectional image of the substrate with a scanning electron microscope (SEM, JSM IT-100 manufactured by JEOL). SEM images were taken with a secondary electron detector at an acceleration voltage of 10 kV, with a working distance of 10 mm and a probe current of 35. A perpendicular cross-section was cut out using an ion milling apparatus. The distance between the outermost fibers on both sides of the conductive porous substrate was measured to obtain the thickness of the backbone layer in um.

[0105] Microporous layer thickness measurement.

[0106] The microporous layer thickness was provided by the supplier. It was verified in a similar fashion as disclosed hereinabove for the gas diffusion backbone layer. First, an ion milling apparatus was used to cut out a cross-section of a sample and an image was obtained with the scanning electron microscope disclosed hereinabove. The thickness of the gas diffusion backbone layer was subtracted from the combined thickness of the gas diffusion backbone layer and microporous layer to obtain the microporous layer thickness in gm. Possible penetration of the microporous layer in the gas diffusion layer was neglected.

[0107] Catalyst layer thickness (loading) measurement

[0108] The catalyst layer loading was measured using an analytical balance (AC210S manufactured by Satorius). The combined weight of the gas diffusion backbone layer and the microporous layer before coating was subtracted from the weight of the final gas diffusion electrode to obtain a catalyst layer loading in weight units. The total loading was divided by the geometrical surface area of the substrate to obtain the areal catalyst layer loading in mg / cm2.

[0109] Water permeation pressure measurement

[0110] To quantify the water retention ability of the produced gas diffusion electrodes, a capillary flow porometer (Porolux 500, manufactured by IB -FT) was used to quantify the water permeation pressure of the material. Water was dropped onto the catalyst layer of the gas diffusion electrode and the air pressure on this side of the electrode was increased up to the point that air started flowing through the conductive porous substrate. The breakthrough pressure was noted as the water permeation pressure (kPa).

[0111] Contact angle measurement

[0112] Contant angle measurements were carried out using the sessile drop method (OCA30, developed and manufactured by Data-physics). A 5 pL droplet consisting of ultra-pure water (18.2 MQ cm) was deposited on the sample surface with a microsyringe. An image of the static droplet was captured after 30 seconds and the contact angle was calculated with SCA20 software, provided by Dataphysics.

[0113] Electrochemical cell measurement

[0114] The separation efficiency of the gas diffusion electrode was measured utilizing an analytical single-cell set-up with a window for visualization, as shown in figure 6 for instance. This experimental setup allows for quick screening and visual inspection of the porous electrode, however, due to the large gap width between both electrodes and the inability to flow fresh electrolyte, this setup is not suitable for efficient operation.

[0115] A set-up that describes real operation, both in a single-cell setup and in an electrochemical stack configuration is disclosed hereinbelow. A 1 cm2sample of a gas diffusion electrode as produced was mounted in an analytical ‘window’ cell as the cathode, where a 5 cm2untreated, Bekaert Currento® 2NI18-0,25 Sintered Metal Fiber Matrix was used as the anode. The anode area was purposefully kept larger than the tested cathode area to limit the effect of gas bubble coverage on the anode on performance. The electrochemical cell was compressed to 2 Nm using a Toolcraft torque wrench.

[0116] A potentiostat (VMP-300, manufactured by BioLogic) equipped with a booster module was employed to conduct a chronoamperometry (controlled voltage) measurement. The electrolyte separating both half-cells consisted of a 2M potassium hydroxide solution. All tests were performed at room temperature and atmospheric pressure. The voltage was set to 5.0V, and the corresponding current was recorded. The relatively high voltage was induced by contact resistances and a large inter-electrode distance in the analytical test cell. It is expected that lower operating voltages (1.48-2.00 V) can be achieved when moving the electrodes closer together, reducing the contact resistance between the electrode and the current collector and reducing the electrical resistance of the electrode itself. Strategies to reduce the contact resistance include using conductive pastes, soldering, or sintering the electrode on the current collector. Strategies to reduce the internal electrical resistance of the electrode include reducing the hydrophobic particle volume percentage towards the lower spectrum (30 vol%). A gas collection chamber (Storage 80, manufactured by H-Tec) was mounted in line with the cathode gas outlet. The time and electrical charge required to accumulate a 30 mL gas volume were documented to determine the experimental flow rate (<p^fsured). The maximum theoretical flow rate for each operating current density (j) was obtained

[0117] The separation efficiency (??) was subsequently determined by:

[0118] The electrolyte separating both half-cells consisted of a 2M potassium hydroxide solution in demineralized water. All tests were performed at room temperature and atmospheric pressure.

[0119] To obtain insights into the electrochemical efficiency and high power-density performance, the following protocol was used:

[0120] A 1 cm2sample of the gas-breathing electrode was mounted in a flow cell as the cathode, where a 1 cm2gas-breathing or flow-through electrode was used as the cathode. The two electrodes are separated by a 250 to 2000 pm flow field, preferably 500 pm made of polytetrafluoroethylene sheet and optionally a spacer, made of a high porosity, porous filler such as nylon or polypropylene. The cell was compressed to 2 Nm using a Toolcraft torque wrench. A 6M KOH aqueous electrolyte reservoir was pumped through the cell using a peristaltic pump (Easyload II, Model 77202-60, Manufactured by Masterflex) and 14 L / S™ tubing at a flow rate between 0.28 and 8.36 ml min1, and preferably 4.18 ml min1. The cell temperature was kept at 80°C. The EU harmonised polarisation curve test method for low-temperature water electrolysis was followed to obtain the polarization curve from the system using a potentiostat (VMP- 300, manufactured by BioLogic) with booster module

[0016] . Prior to each polarization curve measurement, an electrochemical impedance spectroscopy sweep was performed from 100 khz to O.lhz with a voltage amplitude of 10 mV. Produced gases were collected at the anode and cathode and samples were taken for gas chromatography (CompactGC 40, manufactured by Global Analyser Solutions) to determine their purity.

[0121] In a stacked configuration, multiple electrochemical cells are connected in series. The expected electrochemically active size of the stack cells is 0.5 m2(70x70 cm). The required flow rate per individual cell is between 0.28 and 8.4 1 / min and preferably about 4.2 1 / min.

[0122] Hydrophobic agent average particle size measurement was carried out according to ISO 22412 .

[0123] The average hydrodynamic particle size of individual catalyst particles was measured according to ISO 22412.

[0124] Electrolyzer Cell Setup Electrolysis tests were realized using a 5 cm2active area single cell hardware comprising two polymethyl methacrylate (PMMA) housings with titanium (anode) and steel (cathode) current collectors. The electrolyzer cell incorporated a PMMA electrolyte chamber equipped with a Mini-HydroFlex Hydrogen Reference Electrode (Gaskatel) and rubber (EPDM) o-rings for sealing. A 1.8 M potassium hydroxide electrolyte was circulated from a 75 mL electrolyte reservoir through the electrolyte chamber at 20°C, maintaining an interelectrode distance of 7 mm between the anode and cathode. A schematic representation of the experimental setup can be found in Figure 10. Optionally, a spacer could be used in the electrolyte flow field and the gap size could be reduced to enhance performance, such as to a range of 200-1000 pm. In a stacked configuration, multiple electrochemical cells are connected in series. The expected electrochemically active size of the stack cells is 0.5 m2 (70x70 cm). The required flow rate per individual cell is between 0.28 and 8.4 1 / min and preferably about 4.2 1 / min.

[0125] The carbon-based Pt / C electrode served as the cathode, while a Ni foam (80-120 PPI, FuelCellStore) functioned as the anode. All cells were compressed using a torque of 3 Nm. Produced hydrogen and oxygen gases were extracted from the cell housing at both electrode sides through dedicated gas outlets. Electrical Conductivity Measurements

[0126] Electrical conductivity measurements were conducted on carbon-based Pt / C samples using a BioLogic SP-300 potentiostat coupled with EC-Lab software. The electrodes (containing 5 vol% and 80 vol% PTFE) were compressed between two current collectors using PTFE gaskets to achieve a controlled compression of approximately 19% of the initial electrode thickness. The electrode was pressed against another Freudenberg H23C2 (see “Carbon-based gas diffusion media” under Manufacturing methods)_gas diffusion medium, with the catalyst layer facing inwards.

[0127] Conductivity measurements were performed via electrochemical impedance spectroscopy (EIS) with the following parameters: frequency range of 100 kHz to 1000 Hz, amplitude of 10 mV, potential range of 0 to 20 mV, and measurement frequency of 6 points per decade in logarithmic spacing. After each measurement, the cell was disassembled, and the electrodes were readjusted and centered prior to subsequent measurements. Three replicate measurements were performed for each electrode type. The high frequency resistance of the Freudenberg H23C2 substrate was measured separately using identical conditions to isolate the contribution of the sprayed catalyst layer. Fitting of impedance data was conducted with the Python package DearEIS using a L-R model system.

[0128] Electrochemical Test Procedure

[0129] The electrochemical testing protocol comprised the following steps:

[0130] Cell Conditioning:

[0131] The electrolyte was circulated through the cell for 2 hours at the operating flow rate of 10 mL min1using a peristaltic pump (Easyload II, Model 77202-60, Manufactured by Masterflex) to ensure stable wetting of the catalyst layer and achieve thermal equilibrium.

[0132] Electrochemical Impedance Spectroscopy:

[0133] Potentiostatic electrochemical impedance spectroscopy (PEIS) was conducted at the open-circuit voltage with a 10 mV amplitude perturbation over a frequency range of 100 kHz to 100 mHz. Measurements were acquired at 6 points per decade in logarithmic spacing. The high-frequency resistance between the working and reference electrodes was extracted from the PEIS data to enable iR-corrected overpotential calculations.

[0134] Fitting of impedance data was conducted with the Python package DearEIS using a L- R-Tlmbo model system.

[0135] Break-in Procedure:

[0136] A conditioning step was performed at 100 mA cm2for 10 minutes to activate the electrode surface and establish steady-state conditions.

[0137] Polarization Curve Measurement:

[0138] The EU harmonised polarisation curve test method for low-temperature water electrolysis was followed to obtain the polarization curve from the system using a potentiostat (VMP-300, manufactured by BioLogic) with booster module. Currentvoltage characteristics were recorded by stepping the current density through the following sequence: 10, 15, 20, 30, 40, 50, 60, 70, 80, 90, 100, 120, 140, 160, 180, 200, 220, 240, 260, 280, 300, 320, 340, 360, 380, and 400 mA cm2. Each current density was held for two minutes before recording the voltage response throughout one minute. The voltage response in the last minute was averaged to obtain the polarization curve point and associated standard deviation.

[0139] Gas Separation Efficiency Measurements

[0140] Gas separation efficiency was evaluated at a constant current density of 200 mA cm2over three consecutive runs of 4 minutes each. The cell's hydrogen outlet was connected to a Storage 80 gas collection tank (H-TEC Education, FuelCellStore). The bottom portion of the tank, consisting of a graduated cylinder, was filled with deionized water to the zero marking. Produced hydrogen entering the cylinder displaced an equal volume of water, enabling volumetric quantification of gas production.

[0141] The gas extraction effectiveness of the carbon-based Pt / C gas diffusion electrodes was calculated using the following relationship: where Vmeasured is the measured hydrogen volume (mL), n is the number of electrons participating in the reaction (2), F is Faraday's constant (96,485 C moE1), Q is the total charge passed (C), and Vmis the molar volume of hydrogen at 20°C and 1 atm (24,054 mL mol - methods

[0142] Gas diffusion electrodes according to embodiments of the invention were prepared by manufacturing the layers thereof, as claimed, and assembling the layers into a gas diffusion electrode. Hereinbelow, the manufacturing methods used for obtaining the different layers are disclosed.

[0143] Carbon-based gas diffusion media

[0144] Non-woven Freudenberg H23C2 gas diffusion layers with microporous layer were obtained from Quintech and used without any modifications. of gas diffusion backbone layer with

[0145] Hydrophobization of a carbonous backbone layer was carried out by providing PTFE, Fuel cell company, 60% wt, 220nm, ultra-pure water, carbon paper, Freudenberg H15 or Toray TGP-H060 and a 100 ml Beaker as chemicals and tools.

[0146] A dispersion was made containing 10% wt. PTFE particles by diluting the concentrated (60 wt.%) PTFE dispersion accordingly with ultra-pure water and mixing the new dispersion slightly. The 100 ml beaker was then filled with the newly made diluted PTFE dispersion. Cut to size carbon paper samples were then provided in the dispersion for 10 minutes. After 10 minutes, the samples were taken out of the beaker and hung to fully dry.

[0147] The PTFE particles were then sintered on the as treated carbon paper sample by heating them to a temperature of 38O°C and follow a temperature-time curve as shown in Figure 8. Alternatively, a hotplate may be used.

[0148] The carbon-based electrode is now hydrophobic and a water contact angle of >140° was measured with a roll-off angle of <3°.

[0149] Manufacturing of the microporous layer paste The following chemicals and tools were used in preparing a microporous layer paste. i. Vulcan XC-72 Carbon nanoparticles, 0.02 g ii. Ultra-pure water, 8 g iii. PTFE dispersion 60 wt.%, Fuel cell company 0.25 g iv. Triton X100 dispersant, 0.1 g v. Glass vial 16 ml with sealing cap vi. Roller bench set at 20 to 30 rpm. vii. 14 to 15 gr. Glass beads or equivalent volume of ceramic (Yttrium stabilized ZrOx or similar) beads, 3 to 4 mm, clean. viii. A micro balance ix. A glass pipet fitted with appropriate suction cup x. Plastic dosing pipet

[0150] An amount of 8 g ultra-pure water was placed in the glass vial. Next, an appropriate amount of Triton XI 00 was added and the vial sealed and placed on the roller bench for ~ 40 min until the Triton X100 was dissolved. The indicated quantity of Vulcan XC-72 was then added. Subsequently, the milling beads were added and the vial closed and placed on the roller bench for 24 hrs. Following, the glass pipette was placed in the vial and the dispersion was pipetted from the vial into a second clean vial.

[0151] The amount of PTFE dispersion was added into the vial containing the milled Vulcan dispersion, the vial was sealed and placed on the roller bench for 30 minutes. No milling balls were added at this time, as this would cause the PTFE particles to aggregate over time.

[0152] The below procedure yielded ~ 8.4 g ink with a solid content of ~ 2 wt.% and a C to PTFE weight ratio of 1 to 7.5. By changing the added quantities other values may be obtained. The microporous layer paste or ink is now ready for application on the gas diffusion backbone layer by a suitable spray method.

[0153] Catalyst layer ink formulation

[0154] The chemicals and tools needed for this step were as follows: i. Pt / C Tanaka ~ 40%Pt. 0.025 g ii. Ultra-pure water, 8 g iii. PTFE dispersion 60% wt PTFE 0.058 g iv. Triton X100 dispersant, 0.1 g v. Glass vial 16ml with sealing cap vi. Roller bench set at 20 to 30 rpm. vii. 14 to 15 gr. Glass beads or equivalent volume ceramic (Yttrium stabilized ZrOx or similar) beads, 3 to 4 mm, clean. The vial should be filled about 60% with beads. viii. A micro balance ix. Dosing pipets

[0155] An amount of 8 g ultra-pure water was poured in the glass vial. Then an appropriate small amount of Triton X100 was added. The vial was sealed and placed on the roller bench for ~ 40 min until the Triton X100 was dissolved. The indicated quantity of Tanaka Pt / C was then added, whereafter the milling beads were added. The vial was closed and placed on the roller bench for 24 hrs. The next day, the glass pipette was placed in the vial and the dispersion pipetted from the vial into a new clean empty vial.

[0156] The indicated amount of PTFE dispersion was then added into the vial containing the milled ink, the vial was sealed and placed on the roller bench for 30 minutes. No milling balls were added at this stage as this could cause the PTFE particles to aggregate over time.

[0157] The above procedure yielded ~ 8.1 g ink with a solid content of ~ 0.74% and a C to PTFE ratio of 1 to 2.25. By changing the added quantities other values may be obtained. The Pt / C ink is now ready for application onto the microporous layer using a suitable spray process. layer ink and catalyst ink

[0158] A microporous layer (MPL) and catalyst layer (Pt / C ink) were applied onto the hydrophobic carbon-based backbone layer by the following procedure. Used tools were as follows: i. Air brush spray, pressure 1 atm ii. Hotplate set at 90°C. iii. Masking tape iv. Steel carrier plate v. Microbalance vi. Hydrophobic carbon back bone paper vii. Dust mask. viii. Safety glasses ix. Gloves x. Clamb holder

[0159] The hotplate was placed in a safe fume hood and the temperature set to 90°C. The fume hood was set openbetween20 to 30 cm so that the window speed allowed the spray cone to hit the substrate at the targeted spray distance of about 5 cm. The substrate was placed and adhered onto a metal support plate, and place on the hotplate. The air brush was connected to the air hose and the air pressure set to about 1 atm. The airbrush was filled with ~ 3 ml of ink. The spray distance to the substrate was about 5-8 cm. For optimal uniformity, a spray overlap of about 30% was allowed, while using a left-right top-down spray to achieve the desired layer thickness and ink loading. The weight gain on the substrate was periodically measured on a balance.

[0160] After the above application, the substrate was baked with the freshly sprayed MPL on a hotplate or air oven at 240°C for 60 min to remove the surfactant from the layer without sintering the PTFE. The hydrophobic behaviour of the prepared layer was then checked by a contact angle measurement, as disclosed hereinabove. Optionally, the PTFE in the manufactured layer was sintered in steps by heating the hot plate or air oven first to 33O°C with 10°C min i, followed by heating from 330° to 380° with 2.5°C min1and then a temperature hold at 38O°C for 30 minutes. Afterwards, the layer was allowed to cool down to room temperature.

[0161] After the above procedure, a MPL sprayed on the carbon backbone substrate is now ready. A catalyst layer was produced in an identical manner.

[0162] In a different embodiment, a catalyst layer was applied on top of the carbon gas diffusion medium by means of spray coating a catalyst ink. First, 15.0 g of ultrapure water was placed in a glass vial together with 0.025 g of Triton X-100. The vial was sealed and placed on a roller bench set to 20 rpm for ~40 minutes until all Triton X-100 was dissolved. Subsequently, 0.160 g of Pt / C nanoparticles (Tanaka TEC10V50E nanoparticles with 50wt% Pt loading and VULCANRo XC72 carbon support) was added and the ink was dispersed using a quarter-inch ultrasonic probe (Qsonica) with a 30% amplitude and 3 seconds off and 2 seconds on for a duration of 30 minutes. Following ultrasonication, the appropriate amount of PTFE to obtain the desired PTFE volume fraction was added to the vial and the sealed vial was placed on the roller bench for 30 minutes. Catalyst layers were manufactured with a PTFE content varying between 5, 10, 15, 20, 30, 50 and 80 vol% in the final layer. For example, for a catalyst layer with 20 vol% PTFE in the final layer, 0.42 g of 5 wt.% PTFE was added to the ink.

[0163] The obtained ink was ultrasonically spray coated (Sonotek NovoCoat) on a non-woven Freudenberg H23C2 substrate (gas diffusion layers with microporous layer were obtained from Quintech and used without any modifications) until a loading of ~0.5 mg Pt cm-2was reached using a square stainless steel spraying mask with an exposed spraying area of 6.8 cm2, corresponding to a (sintered) catalyst layer thickness of approximately 4.9 pm. The ultrasonic nozzle was controlled with a speed of 80 mm / second, a liquid flow of 0.5 mF min1, a nozzle power of 60% and a gas flow of 10.01 min1using alternating horizontal and vertical spray direction. After spraying, the substrate was sintered by heating it to a temperature of 360°C in a muffle oven, following the temperature-time curve as shown in Figure 9.

[0164] The application of nickel-based (nickel nanoparticles 10-40 nm, Nanografi) or nickeliron based catalyst layers (nickel-iron alloy 30-70 nm 65:35, Nanografi), either on top of the carbon-based or nickel-based gas diffusion media proceeded in similar fashion. However, before spraying, a fraction of 1.23 wt% methylcellolose (Sigma Aldrich) was added to the ink and placed on the roller bench for another 30 minutes.

[0165] Nickel-based electrode

[0166] A microporous layer (MPL) and catalyst layer (Pt / C ink) were applied onto a nickel- based backbone layer by a similar procedure. Used chemicals and tools were as follows: i. PTFE, Fuel cell company, 60%wt, 220nm ii. Ultra-pure water iii. Ni fiber gas diffusion layer (e.g., Bekipor CURRENTO® 2NI06-0.20

[0167] (215x290mm)) iv. 100 ml Beaker

[0168] The same steps as disclosed hereinabove for a carbon backbone layer were used. The application of the microporous layer paste and the catalyst layer ink onto the nickel gas diffusion backbone layer were carried out following a similar protocol to the carbonbased backbone layers described above. Optionally, alternative surfactants and rheology agents can be used to obtain a better nickel nanoparticle ink dispersion, such as sodium dodecyl sulfate (Sigma- Aldrich) or Hypermer KD-2 (Croda Industrial Specialities) as surfactant and xantham gum (labshop) or methyl cellulose (Sigma- Aldrich) as rheology agent. of nickel gas diffusion backbone layer with

[0169] Sintered Metal Fiber Matrix Currento® 2NI 18-0,25 porous transport layers were obtained from Bekaert. Hydrophobization of the nickel backbone was carried out by dip coating the electrodes for 10 minutes in a 20 wt.% PTFE dispersion, which was obtained from a 60 wt.% PTFE dispersion (FuelCellStore) by mixing with ultrapure water. After dip coating, the samples were taken out of the beaker and hung to fully dry.

[0170] The PTFE particles were then sintered by heating them to a temperature of 360°C in a muffle oven and follow a temperature-time curve as shown in Figure 9: First, the temperature is increased to 200°C with a 10°C min1ramp, then from 200°C to 250°C with a 2.5°C min1ramp, a hold for 10 min at 250°, 10°C min1from 250°C to 33O°C, 2.5°C min1from 33O°C to 360°C, a final hold for 30 minutes at 360°C and a cool down to room temperature.

[0171] The nickel-based electrode is now hydrophobic and a water contact angle of >140° was measured with a roll-off angle of <3°.

[0172] A microporous layer was applied on top of the nickel gas diffusion layer by means of spray coating a nickel ink. First, 8 g of ultrapure water was placed in a glass vial together with 0.05 g of Triton X-100 dispersant (Sigma Aldrich). The vial was sealed and placed on a roller bench set to 20 rpm for ~40 minutes until all Triton X-100 was dissolved. Subsequently, 0.5 gram of nickel nanoparticles (10-40 nm, Nanografi) was added and the ink was dispersed using a quarter-inch ultrasonic probe (Qsonica) with a 30% amplitude and 3 seconds off and 2 seconds on for a duration of 30 minutes.

[0173] Following ultrasonication, the appropriate amount of PTFE to obtain the desired PTFE volume fraction (e.g., 0.21 g of 60 wt.% PTFE) was added to the vial and the sealed vial was placed on the roller bench for 30 minutes. Finally, 0.01 g of 1.23 wt% methylcellolose (Sigma Aldrich) was added to the ink and placed on the roller bench for another 30 minutes.

[0174] The obtained ink was hand spray coated on the hydrophobic nickel substrate using a DAGR Gravity Airbrush (.35 mm, Devilbliss) using a hotplate at 90°C and a square stainless steel spraying mask with an exposed spraying area of 6.8 cm2until a loading of ~8.0 mg cm-2was reached, corresponding to a (sintered) microporous layer thickness of approximately 30 pm. After spraying, the substrate was again sintered by heating it to a temperature of 360 °C in a muffle oven and follow a temperature-time curve as shown in Figure 9.

[0175] Catalyst layer ink preparation and application

[0176] A catalyst layer was applied on top of the nickel gas diffusion medium by means of spray coating a catalyst ink. First, 15.0 g of ultrapure water was placed in a glass vial together with 0.025 g of Triton X-100. The vial was sealed and placed on a roller bench set to 20 rpm for ~40 minutes until all Triton X-100 was dissolved. Subsequently, 0.160 g of Pt / C nanoparticles (Tanaka TEC10V50E nanoparticles with 50wt% Pt loading and VULCANRo XC72 carbon support) was added and the ink was dispersed using a quarter-inch ultrasonic probe (Qsonica) with a 30% amplitude and 3 seconds off and 2 seconds on for a duration of 30 minutes. Following ultrasonication, the appropriate amount of PTFE to obtain the desired PTFE volume fraction was added to the vial and the sealed vial was placed on the roller bench for 30 minutes. Catalyst layers were manufactured with a PTFE content varying between 5, 10, 15, 20, 30, 50 and 80 vol% in the final layer. For example, for a catalyst layer with 20 vol% PTFE in the final layer, 0.42 g of 5 wt.% PTFE was added to the ink. The obtained ink was ultrasonically spray coated (Sonotek NovoCoat) on a non-woven Freudenberg H23C2 substrate (gas diffusion layers with microporous layer were obtained from Quintech and used without any modifications) until a loading of ~0.5 mg Pt cm-2was reached using a square stainless steel spraying mask with an exposed spraying area of 6.8 cm2, corresponding to a (sintered) catalyst layer thickness of approximately 4.9 pm. The ultrasonic nozzle was controlled with a speed of 80 mm / second, a liquid flow of 0.5 mL min1, a nozzle power of 60% and a gas flow of 10.01 min1using alternating horizontal and vertical spray direction. After spraying, the substrate was sintered by heating it to a temperature of 360°C in a muffle oven, following the temperature-time curve as shown in Figure 9.

[0177] The application of nickel-based (nickel nanoparticles 10-40 nm, Nanografi) or nickeliron based catalyst layers (nickel-iron alloy 30-70 nm 65:35, Nanografi), either on top of the carbon-based or nickel-based gas diffusion media proceeded in similar fashion. However, before spraying, a fraction of 1.23 wt% methylcellolose (Sigma Aldrich) was added to the ink and placed on the roller bench for another 30 minutes.

[0178] Experimental results

[0179] Electrical conductivity measurements

[0180] The electrical conductivity of Freudenberg H23C2-based cathodes was investigated by measuring the area specific resistance of two electrodes with 5 vol% (i.e., the lower limit) and 80 vol% (i.e., the upper limit) loading of PTFE. The results are shown in Figure 11.

[0181] Two key observations can be made from Figure 11. First of all, the area specific resistance of the catalyst layer with 80 vol% PTFE shows a 2.9 times increased resistance over the catalyst layer with 5 vol% PTFE content. We attribute this to the fact that PTFE is a non-conductive polymer, therefore, higher volume fractions of PTFE lead to a reduced conductivity of the total layer. Nevertheless, although the overall resistance is higher, both electrodes are still sufficiently conductive to act as electrode. The maintained conductivity at the upper limit of 80 vol% PTFE content can be attributed to aggregation of PTFE particles during depositing and sintering of the catalyst layer. Separation efficiency

[0182] To assess the gas extractability of the manufactured electrodes, we evaluated the ratio of hydrogen captured as gas from the system to the total amount of hydrogen produced for the Pt / C carbon-based cathodes. A separation efficiency lower than 100% is indicative of the formation of hydrogen bubbles, which are transported out of the cell with the liquid electrolyte. The results for the considered PTFE range, between 5 and 80 vol%, are shown in Figure 12.

[0183] Figure 12 highlights that the electrode with 5 vol% of PTFE has a lower separation efficiency than all tested samples with higher PTFE content (10-80 vol%). To extract hydrogen from the cell without forming bubbles, produced hydrogen needs to travel to a hydrophobic domain that is interconnected with the gas phase at the backend of the electrode. The lower separation efficiency of 50% is therefore likely indicative of the incomplete formation of such a network. With the current manufacturing procedure, 5 vol% of PTFE can therefore be seen as below the lower limit to produce effective gas extracting electrodes. All electrodes with a higher PTFE content show a separation efficiency around 100%, indicating near optimal separation efficiency. These results agree with visual inspection of bubble formation in the interelectrode gap, which showed vigorous bubble formation for the 5 vol% PTFE content electrode, but no bubble formation for the other electrodes. Although the experiments were performed at 200 mA cm'2, similar phenomena were observed for the entire current density range used in the polarization curve measurements (0 - 400 mA cm'2). The small variation of 98 to 101% observed for the 10-80 vol% PTFE content electrodes can be attributed to measurement errors. The current measurement setup involves visual inspection of the water meniscus in the gas collection tank. Obviously, a 101% separation efficiency is unphysical. Although the dramatic improvement compared with an imperfect gas extraction electrode (5 vol. %) is observed, an improved setup with digital flow meters is pending to improve the accuracy of the measurement further.

[0184] Polarization curve measurements

[0185] IR-corrected polarization curves were obtained (see FIG. 13) to investigate and compare the electrochemical performance of the Pt / C -based cathodes with varying PTFE content. IR-corrections were performed to omit any variations in cell assembly on the overall results. It was observed that all high-frequency resistance values were between 1.36 and 1.42 Qcm2. This relatively high ohmic resistance may be explained by three limiting factors of the analytical cell:

[0186] Operation at room temperature, where operation at elevated temperatures (i.e., 80°C) would reduce overall ohmic resistances. operation at a molarity of 1.8M KOH in the electrolyte, where a higher molarity (i.e. 6M KOH) would reduce overall ohmic resistances.

[0187] Most importantly, operation with a interelectrode gap width of 7000 pm useful for diagnostics, where operation with a decreased interelectrode gap (i.e., 200-500 pm) would reduce overall ohmic resistances.

[0188] In particular, the current 7 mm interelectrode gap between anode and cathode is significant in the current reactor setup in order to fit a reference electrode. In an industrialized process, the reference electrode will be removed and the interelectrode gap will be reduced.

[0189] Two key trends may be observed from Figure 13. Firstly, the electrode with a 5 vol. % PTFE content shows the worst performance and the largest voltage fluctuation during its measurement. This is in agreement with the separation efficiency measurements, as formation of bubbles may result not only in coverage of electrochemically activate surface area and increase the local supersaturation of products close to the electrode surface, but may also inhibit the transport of liquid electrolyte into the electrode structure. Together, this may lead to a reduced overall performance, particularly at higher current densities, where the formation of bubbles may be more vigorous for this electrode. Secondly, we observe a stable and, compared to the 5 vol. % PTFE electrode, an improved performance (>50%) for all other electrodes at higher current densities. No discernible difference was observed for the performance of the electrodes with 10, 20 and 30 vol. % PTFE content. It is worth noting that for tests at these PTFE loadings, no bubble formation was observed in the interelectrode gap. Given the close performance of the three electrodes, different arguments can be made for making a selection. The lower presence of hydrophobic sites in the electrode with lower PTFE content may intrinsically result in less wettability issues. However, a higher PTFE content may result in a more robust or durable process. Given the risk of in-situ PTFE degradation, a higher PTFE content electrode, such as 30%, may be expected to exhibit gas extracting performance over a longer lifetime. Compared to the electrodes between 10 and 30 vol.%, a slight reduction of the performance is observed for the electrode with 80 vol. % PTFE, which may potentially be attributed to a thicker catalyst layer, an impeded wettability of the catalyst particles and a higher required current per available Pt particle. Although the observed trends clearly discern between improperly designed electrodes (5 vol % PTFE) and electrodes that do exhibit gas extracting properties (> 5 vol.% PTFE content), further process intensification is required to improve the absolute performance of the catalyst layer. In particular, industrialized systems will operate at higher operating temperatures, potentially higher potassium hydroxide concentrations and higher catalyst loadings, which would further decrease the observed overpotential for the gas extracting electrodes.

[0190] Theoretical framework for structure-performance relationships

[0191] A simplified, two-dimensional computational model based on a first-order implicit Euler model finite difference method was constructed in order to gain insight into the physical phenomena responsible for effective gas stripping in the manufactured electrodes. In this model, the concentration of generated hydrogen supersaturated within the liquid electrolyte is modeled to study the effects of the cathode structure and operating conditions on the extraction of hydrogen using gas-breathing electrodes in a membrane-free alkaline electrolyzer.

[0192] In this model, the steady- state concentration profile of hydrogen within the electrode and adjacent electrolyte channel was computed by considering diffusion and convection. A uniform grid spacing for the y-direction is considered, whereas a variable grid spacing in the x-direction is considered to improve the resolution within the catalyst layer. The catalyst layer was modelled as a collection of alternating mix of hydrophobic particles and porous catalyst particles. A constant hydrogen pressure was assumed in the hydrophobic backbone, leading to the use of Henry’s law as a boundary condition at the surface of the hydrophobic particles. Further, it was assumed that there is negligible flux of hydrogen through the anode assuming the electrolyte channel is wide enough for the modelled flow rate so that the hydrogen concentration boundary does not reach the anode. Finally, no flow of hydrogen through the casing next to the electrolyte channel was assumed and it was assumed that the diffusive and convective transport of hydrogen at the electrolyte inlet equals the total convective transport of hydrogen towards the inlet.

[0193] In this simplified two-dimensional model, the inclusion of hydrophobic particles within the catalyst layer was studied, as well as the effect of the hydrophobic particle radius, the impact of the hydrophobic volume fraction within the catalyst layer and the impact of current density on the supersaturated hydrogen concentration and thus attainable separation efficiency within the electrolyzer. A base-case scenario was established, of which the input values are provided in table 1.

[0194] Table 1 - Physical and model input values used for the base -case modelling scenario

[0195]

[0196] Apart from the base-case scenario, simulations were run without hydrophobic particles, with hydrophobic particles with varying size, with a varying hydrophobic particle volume fraction in the catalyst layer and at different current densities. Results of the numerical model are shown in Figure 14 and 15.

[0197] The dramatic improvement achieved through mixed wettability design in the catalyst layer is illustrated in Figure 14, where the traditional catalyst layer shows severe saturation (up to 3000x) and only -80% maximum hydrogen removal efficiency, while the mixed wettability catalyst layer maintains low saturation (~2x) and achieves nearly 100% removal efficiency. It should be noted that bubble formation was not explicitly modelled in this numerical model. It is hypothesized that the poor gas extraction demonstrated for the traditional catalyst layer results in high supersaturations, which directly translate into bubble formation. The model demonstrates how mixed wettability design may impede bubble formation and promote gas extraction from the electrode.

[0198] Figure 15A shows an exponential decrease in hydrogen removal efficiency with increasing hydrophobic particle size. This demonstrates that controlling the hydrophobic particle size may be important for effective removal of hydrogen from the cathode, provided that the catalyst and hydrophobic particles are effectively and homogeneously distributed. Apart from the hydrophobic particle size, also the hydrophobic particle volume fraction and operating current density play a (minor) role in the supersaturation and effective removal of hydrogen from the electrode, where both an increasing operating current density and a decreasing hydrophobic particle volume fraction were found to increase the maximum attainable supersaturation within the porous electrode, and thus may impede bubble-free operation.

[0199] Description of the figures

[0200] Referring to figure 1, an individual electrolyzer cell 100 equipped with gas diffusion electrodes (200, 300) in accordance with an embodiment of the invention is schematically shown.

[0201] The electrolyzer cell 100 according to the depicted embodiment comprises a first electrode 200 and a second electrode 300. The first and second electrodes (200, 300) are both gas diffusion electrodes, but other embodiments of the invention may have one gas diffusion electrode only. The first and second gas diffusion electrodes (200, 300) are arranged to face each other to define an electrolyte space 400 for receiving an electrolyte. It is noted that the electrolyzer cell 100 in the described embodiment is membrane-less in that it lacks a membrane separator provided within the electrolyte space 400.

[0202] According to figure 1, at least one of the first and second electrode (200, 300) comprises a gas diffusion backbone layer (210, 310) facing away from the electrolyte space 400, a catalyst layer (220, 320) facing the electrolyte space 400 and a microporous layer (230, 330) arranged between the gas diffusion backbone layer (210, 310) and the catalyst layer (220, 320).

[0203] In the embodiment shown in figure 1, both the first and second electrode (200, 300) comprise the layers described above. It should be noted that the electrolyzer cell 100 of the invention is not limited to such an embodiment. In a different embodiment, the first electrode 200 or second electrode 300 may comprise a different type of electrode, for example a metal electrode.

[0204] In a preferred embodiment, the thickness of the gas diffusion backbone layer (210, 310) may be between 50 and 2000 pm, preferably between 100 and 1000 pm, more preferably between 100 and 250 pm. The thicknesses of the gas diffusion layers (210, 310) may be the same or they may differ from each other. The preferred loading of the catalyst layer (220, 320) may be between 0.01 and 150 mg cm-2, preferably between 0.1 and 10 mg cm-2, more preferably between 0.5 and 5 mg cm'2. The loading refers to the loading of both (supported) catalyst particles, hydrophobic agent, and any additive present in the layer. The disclosed loadings roughly correspond to an approximate thickness of 100 nm - 1.2 mm, more preferably 1 um and 10 um and more preferably between 400 nm and 1.6 um of the catalyst layer.

[0205] The preferred thickness of the microporous layer (230, 330) may be between 5 pm and 2000 pm, preferably between 10 and 100 pm and more preferably between 20 and 60 pm.

[0206] Referring to figure 2, a schematic setup of an electrolyzing device 1 comprising the electrolyzer cell 100 according to an embodiment of the invention for use in hydrogen production is shown. It is noted that an electrolyte will flow through different components of the device 1 and a new reference number will be used for each stage of the process to refer to this electrolyte, to improve comprehensibility.

[0207] The electrolyzing device 1 may be used in electrolyzing water into hydrogen and oxygen. In a first step of such process, a fresh electrolyte 401 enters into a collection tank 420 for the electrolyte. After exiting the collection tank 420, the electrolyte 402 flows through a heater 430. The heater 430 is set to increase the temperature of the electrolyte 402 to a preferred temperature. Alternatively, the electrolyte 402 may be heated by supplying heat to the collection tank 420. Other options for heating the electrolyte 402 are also possible. After exiting the heater 430, the heated electrolyte 403 flows through a pump 440 to pump the electrolyte into the electrolyte space 400. After exiting the pump 440, the electrolyte 404 flows into a first end of the electrolyte space 400 of the electrolyzer cell 100. There may be three different possible outputs from the electrolyzer cell 100. By means of electrolysis, a first gas 240 such as hydrogen may be formed at the cathode side of the electrolyzer cell 100, which may then be collected at a first gas collector 250. The first gas diffusion electrode 200 of the electrolyzer cell 100 may be configured to be in contact with and adjacent to the first gas collector 250. At the anode side of the electrolyzer cell 100, a second gas 340 such as oxygen may be formed, which may be collected at a second gas collector 350. The second gas diffusion electrode 300 of the electrolyzer cell 100 may be configured to be in contact with and adjacent to the second gas collector 350. The remaining electrolyte 405 may exit the electrolyzer cell 100 at a second end of the electrolyte space 400. In the embodiment described above, the configuration of the gas diffusion electrodes (200, 300) is such that both are in a gas-breathing configuration. In a different embodiment, one of the gas diffusion electrodes (200, 300) may be a flow-through electrode, such that the electrolyte 405 will partially exit the electrolyzer cell 100 at the same side as the gas formed at the flow-through electrode side. After exiting the electrolyzer cell 100 at the second end of the electrolyte space 400, the electrolyte 405 may flow into a separation tank 410. In the separation tank 410, the electrolyte 405 may be separated from any residual first and second gas 411 to obtain purified electrolyte 406. The electrolyte 406 may then flow to the collection tank 420, where it may be mixed with the fresh electrolyte 401, after which the above-described may be repeated.

[0208] Figures 3A-C schematically show cross-sectional views of the gas diffusion backbone layer (210, 310) (figure 3A), the catalyst layer (220, 320) (figure 3B), and the microporous layer (230, 330) (figure 3C) according to an embodiment of the invention. The catalyst layer (220, 320) is instrumental in the conversion of electrolyte in hydrogen or oxygen and the retention of electrolyte in the electrolyte space 400, the microporous layer (230, 330) is instrumental in transporting the gasses out of the catalyst layer (220, 320) and depositing a substantially thickness-uniform catalyst layer (220, 320), while the gas diffusion backbone layer (210, 310) is required for removal of gasses out of the electrode (200, 300) and required for mechanical support of the electrode (200, 300).

[0209] Figure 3A shows a cross-sectional view of the gas diffusion backbone layer (210, 310), clarifying its microscopic structure according to an embodiment of the invention. For reference, a width of 100 pm is indicated in figure 3A. As shown in figure 3A, the gas diffusion backbone layer (210, 310) may comprise a conductive porous material, such as for instance but not limited to, felt, cloth, paper, foam, or combinations thereof. The shown embodiment of the gas diffusion backbone layer (210, 310) comprises conductive fibers (212, 312), such as carbon (C), nickel (Ni) or titanium (Ti) fibers. In a preferred embodiment, the individual fiber diameter may be between 1 and 30 pm, preferably between 10 and 23 pm. The gas diffusion backbone layer (210, 310) may further comprise hydrophobic particles (211, 311). Consequently, the gas diffusion backbone layer (210, 310) may be hydrophobic. The hydrophobic particles (211, 311) may be made of, for example, polytetrafluoroethylene (PTFE) or fluorinated ethylene propylene (FEP). The gas diffusion backbone layer (210, 310) may be porous in that it may contain void space (213, 313) in between the hydrophobic particles (211, 311) and the fibers (212, 312). The fraction of void space in the porous layer may define the uncompressed porosity of the gas diffusion backbone layer (210, 310). In a preferred embodiment, the uncompressed porosity is between 0.5 and 0.95, preferably between 0.6 and 0.9.

[0210] Referring to figure 3B, it shows a cross-sectional view of the catalyst layer (220, 320). The cross-sectional view shows the microscopic structure of the catalyst layer (220, 320) according to an embodiment of the invention. For reference, a width of 5 pm is indicated in figure 3B. The catalyst layer (220, 320) may comprise hydrophobic particles (221, 321). Consequently, the catalyst layer (220, 320) may be macroscopically hydrophobic or may contain hydrophobic parts. It may therefore be of mixed wettability on a mesoscale. The hydrophobic particles (221, 321) may be made of, for example, polytetrafluoroethylene (PTFE) or fluorinated ethylene propylene (FEP). The catalyst layer (220, 320) may further comprise catalyst particles (222, 322). The catalyst particles (222, 322) may be made of, for example, platinum on carbon (Pt / C), nickel (Ni) or nickel iron oxide (NiFeOx), and, further, any include any type of filler.

[0211] Referring to figure 3C, it shows a cross-sectional view of the microporous layer (230, 330). The cross-sectional view shows the microscopic structure of the microporous layer (230, 330) according to an embodiment of the invention. For reference, a width of 5 pm is indicated in figure 3C. The microporous layer (230, 330) may comprise hydrophobic particles (231, 331). Consequently, the microporous layer (230, 330) may be hydrophobic. The hydrophobic particles (231, 331) may be made of, for example, polytetrafluoroethylene (PTFE) or fluorinated ethylene propylene (FEP). The microporous layer (230, 330) may further comprise conductive particles (232, 332). The conductive particles (232, 332) may be made of, for example, carbon (C), nickel (Ni) or titanium (Ti). Additionally, the microporous layer (230, 330) may comprise potential cracks (233, 333). These microscopic cracks (233, 333) may occur during the application process and curing process of the microporous layer (230, 330). The size of the cracks (233, 333) may be limited to a radius of 3 pm to prevent potential liquid leakage. The radius or size of any type of filler may also be limited to 3 pm.

[0212] Figures 4A-C schematically show different configurations of a membrane-less electrolysis cell in accordance with an embodiment of the invention, wherein (a) shows a gas-breathing configuration, (b) a flow-through configuration, and (c) a hybrid configuration.

[0213] All three configurations may comprise the layers in their respective first and second gas diffusion electrode (200, 300) as defined with reference to figure 1. For completeness, the layers are here repeated as being the gas diffusion backbone layer (210, 310), the catalyst layer (220, 320) and the microporous layer (230, 330). The reference numbers are left out of figure 4 in view of redundancy.

[0214] Referring to figure 4A, it shows a gas-breathing configuration of the electrolyzer cell 100. The gas-breathing configuration is characterized by the fact that the electrolyte 404 flows into the first end of the electrolyte space 400 of the electrolyzer cell 100 and that the electrolyte 405 exits the electrolyzer cell 100 at the second end of the electrolyte space 400. The first gas 240 is collected at the first gas collector 250 at the side of the first gas diffusion electrode 200, while the second gas 340 is collected at the second gas collector 350 at the side of the second gas diffusion electrode 300.

[0215] Referring to figure 4B, it shows a flow-through configuration of the electrolyzer cell 100. The flow-through configuration is characterized by the fact that the electrolyte 404 flows into the first end of the electrolyte space 400 of the electrolyzer cell 100, but the electrolyte 405 does not exit the electrolyzer cell 100 at a second end of the electrolyte space 400. Instead, the electrolyte 405 exits the electrolyzer cell 100 by flowing through the first and second (gas diffusion) electrodes (200, 300), in similar fashion to the first and second gas (240, 340).

[0216] Referring to figure 4C, it shows a hybrid configuration of the electrolyzer cell 100. The hybrid configuration is characterized by the fact that the electrolyte 404 flows into the first end of the electrolyte space 400 of the electrolyzer cell 100, to then partially exit the electrolyzer cell 100 at the second end of the electrolyte space 400 in accordance with the gas-breathing configuration, and partially exit the electrolyzer cell 100 by flowing through one of the first and second (gas diffusion) electrodes (200, 300) in accordance with the flow-through configuration. The hybrid configuration may be advantageous in that it may mitigate liquid breakthrough, and it may reduce the set of design constraints related to scaling and the influence of hydrodynamic pressure from the electrolyzer, which may lead to operation with improved stability.

[0217] It is noted that reference is made to the wording ‘hybrid’ to mean different configurations and encompass an electrolysis device that produces hydrogen on the cathode and any valuable product other than oxygen on the anode, as well as an electrolysis device with one gas-breathing electrode and one flow-through electrode.

[0218] The electrolyzer cell 100 of the electrolyzing device 1 of the invention may be in the gas-breathing configuration. In an alternative embodiment, the electrolyzer cell 100 of the electrolyzing device 1 of the invention may be in the hybrid configuration.

[0219] Figures 5A-C schematically show different configurations of an electrochemical device according to embodiments of the invention with (a) a proton exchange membrane electrolyzer where the hydrogen-side is operated dry, (b) an anion exchange membrane electrolyzer where the hydrogen-side is operated dry, and (c) a metal-air battery where the air- side is operated dry.

[0220] All three configurations may comprise the layers in their respective first and second gas diffusion electrode (200, 300) as defined with reference to figure 1. For completeness, the layers are here repeated as being the gas diffusion backbone layer (210, 310), the catalyst layer (220, 320) and the microporous layer (230, 330). The reference numbers are left out of figure 5 in view of redundancy.

[0221] Referring to figure 5A, the electrolyzer cell 100 comprises a PEM electrolyzing cell 100 equipped with a membrane 510. In such a PEM electrolyzer, the electrode may facilitate asymmetric feeding of water in the cell and ensure production of high-purity hydrogen 240, while retaining water close to and in the membrane, potentially preventing dry-out and local heating of the cell. Electrolyte 340 and oxygen 405 leave the electrolyzer cell 100 at the anode, as shown. The same holds for an AEM electrolyzer, equipped with a membrane 520, as shown in figure 5B. Referring to figure 5C, the electrolyzer cell 100 may also comprise a metal air battery equipped with a metal cathode 530. In such a metal-air battery, the electrode configuration can be used as a reversible oxygen-electrode. In discharge mode, the hydrophobic functionality of the electrode prevents flooding of the electrode and leakage of electrolyte 404 out of the cell. In charge mode, the electrode can effectively extract oxygen 340 from the liquid electrolyte (404, 405), without losing oxygen as gas bubbles, reducing mass transport limitations and gas hold-up in the system. To improve electrochemical activity of the electrode, either one or a mixture of catalyst particles may be required in this application.

[0222] Figure 6 schematically illustrates an experimental window-cell setup 600 used for the preliminary screening of the gas diffusion electrodes. The single cell set-up comprises a cathode end-plate 613, a cathode current collector 612, a windowed electrolyte reservoir 630, an anode current collector 622, an anode end-plate 623, a hydrogen-side gas outlet 614, an oxygen-side gas outlet 624, a gas-breathing cathode 611, a gas-breathing or flow-through anode 621 an anode electrical connection 625. In the shown set-up, an electrolyte circulation is not used (static electrolyte). Due to the presence of the window, this set-up was designed to be able to observe potential bubble formation at the electrolyte-electrode interfaces.

[0223] To illustrate the improved separation efficiency of a device according to the invention, a comparison was made with state of the art electrodes by performing comparative experiments, also referred to herein as baseline experiments. For these baseline or comparative experiments, both commercially available nickel electrodes without any modification (Baseline- 1-Ni and Baseline-2-Ni), as well as a platinum-based electrode without the addition of PTFE particles (Baseline-3-Pt) were used. Key elements of the composition of the prepared electrodes are presented in Table 1.

[0224]

[0225] Table 1: composition of electrodes

[0226] Figure 7 shows a graph 700 of the measured hydrogen gas flow and calculated separation efficiency, defined as the measured hydrogen gas flow divided by the total hydrogen production rate by Faraday’s law, as elucidated in the experimental section hereinabove.

[0227] On the horizontal axis, the current density in units of amps per square cm is provided. A vertical axis on the left provides the hydrogen separation efficiency in percentages, and the circular data points in the graph 700 correspond to this axis. A vertical axis on the right provides the measured flow rate in units of ml per min. The triangular data points in the graph 700 correspond to this axis. It may be inferred from figure 7 that a carbonbased electrode is able to separate nearly 100% of the hydrogen produced at the cathode. This finding is supported by the absence of nearly any bubble on the cathode during operation. Higher current densities could not be reached in the window-cell setup, because of the high ohmic resistance induced by the large distance between both half cells and contact resistances.

[0228] Figure 8 shows a graph 800 of a temperature-time profile used in preparing the hydrophobized gas-diffusion backbone layer, according to an embodiment of the invention. For a detailed explanation of the figure, we refer to “Dip coating of gas diffusion backbone layer with hydrophobic agent” in the Manufacturing methods section, disclosed hereinabove. Figure 9 show a graph of a temperature-time profile used in preparing the hydrophobized nickel gas diffusion backbone layer, according to an embodiment of the invention. For a detailed explanation of the figure, we refer “Dip coating of nickel gas diffusion backbone layer with hydrophobic agent” in the Manufacturing methods section, disclosed hereinabove.

[0229] Figure 10 shows a schematic representation of an electrolyzer cell setup for a flow cell experiment, according to an embodiment of the invention. Figure 10A shows a schematic representation of a flow cell experiment where liquid is pumped through the electrolyte flow channel to form an interelectrode gap. It highlights the PMMA cell housing, current collectors, electrolyte flow channel and both electrodes. Figure 10B shows a zoomed-in version where a reference electrode is positioned between a Ni- based anode and Pt / C-based cathode and where the flow of hydrogen and oxygen are indicated with a 100% separation efficiency.

[0230] Figure 10A shows a current collector 1011, an anode 1012, an electrolyte flow channel 1013, a cathode 1014 and a cell housing 1015. Figure 10B shows oxygen 1021, a current collector 1022, nickel (Ni) 1023, a reference electrode 1024, platinum on carbon (Pt / C) 1025 and hydrogen 1026.

[0231] For a detailed explanation of the figure, we refer to “Electrolyzer Cell Setup” in the Characterization Methods section, disclosed hereinabove

[0232] Figure 11 shows a bar chart representing catalyst and gas diffusion media support layer area specific resistance with 5 and 80 vol.% PTFE content, according to an embodiment of the invention. For each of 5 vol.% PTFE and 80 vol.% PTFE, a separate bar is shown. Each bar is divided into two parts, where the lighter grey represents the Freudenberg H23C2 and the darker grey represents the catalyst layer. The number inside each of the parts represents the area specific resistance attributable to specifically the Freudenberg H23C2 or the catalyst layer. The cumulative result of the two numbers for each bar corresponds to the value the bar reaches on the vertical axis representing the area specific resistance. It can be seen that the lighter grey area is equal for both PTFE contents, while the darker grey area is significantly higher for the higher PTFE content, leading to a significantly higher overall area specific resistance. For a detailed explanation of the figure, we refer to “Electrical conductivity measurements” in the Experimental results section, disclosed hereinabove.

[0233] Figure 12 shows a bar chart representing separation efficiency of manufactured gas extraction electrodes as a function of PTFE content in the catalyst layer, according to an embodiment of the invention. Rather than a continuous relationship between separation efficiency and PTFE content, seven individual bars represent the separation efficiency corresponding to seven discrete values of PTFE content. For a detailed explanation of the figure, we refer to “Separation efficiency” in the Experimental results section, disclosed hereinabove.

[0234] Figure 13 shows a graph representing IR-corrected current density-voltage curves for cathodes with varying PTFE content, according to an embodiment of the invention. The PTFE content varied between 5, 10, 20, 30 and 80 vol.%. The displayed polarization curves concern the performance of Pt / C carbon-based electrodes used as a cathode (hydrogen evolution reaction electrode). Error bars indicate the time-based variation in measured voltage throughout the measurement. For a detailed explanation of the figure, we refer to “Polarization curve measurements” in the Experimental results section, disclosed hereinabove.

[0235] Figure 14 shows a comparison of saturation profiles in a traditional catalyst layer and a mixed wettability catalyst layer. The results follow from a simulation based on the theoretical framework elucidating on the structure-performance relationships.

[0236] In both Figures 14A and 14B, the vertical axis represents the channel length, while the horizontal axis represents the channel width, both in pm. Furthermore, the 10 pm catalyst layer (CE) is indicated with a dashed line.

[0237] Referring to Figure 14A, the traditional catalyst layer shows severe hydrogen accumulation within the electrode and electrolyte of up to a factor 3000. In a physical scenario, this would result in bubbles. A hydrogen removal efficiency of around 80% is reached. Referring to Figure 14B, the mixed wettability catalyst layer maintains a low saturation with a maximum of around a factor 2, while achieving a nearly 100% hydrogen removal efficiency.

[0238] For a detailed explanation of the figure, we refer to “Theoretical framework for structure-performance relationships” in the Experimental results section, disclosed hereinabove.

[0239] Figure 15 shows a graph representing hydrogen removal as a function of hydrophobic particle radius and a graph representing maximum saturation as a function of current density, according to an embodiment of the invention.

[0240] Referring to Figure 15 A, it shows hydrogen removal as a function of hydrophobic particle radius, highlighting an inverse relationship between hydrophobic particle size and hydrogen removal efficiency. The graph demonstrates the importance of using smaller particles (e.g. less than 2 p.m) for maintaining high removal efficiency.

[0241] Referring to Figure 15B, it shows maximum supersaturation computed for the base-case (see “Theoretical framework for structure-performance relationships”) for different PTFE volume fractions. The three lines represent PTFE volume fractions of 30 vol.%, 50 vol.% and 70 vol.%, respectively, over a current density range of 0 to 15 A cm-2.

[0242] For a detailed explanation of the figure, we refer to “Theoretical framework for structure-performance relationships” in the Experimental results section, disclosed hereinabove.

[0243] Additional embodiments are described below, the features of which may be optionally combined with, or substituted for, features of any of the embodiments described above or in the appended claims:

[0244] According to an embodiment, there is provided a gas diffusion electrode configured for bubble-free operation in alkaline water electrolysis at current densities including exceeding 300 mA / cm2, the electrode comprising: a) a gas diffusion backbone layer comprising conductive porous material; b) a microporous layer facing an electrolyte contact surface; and c) a controlled catalyst micro structure characterized by: i) a maximum diffusive path length between any catalyst site and a connected hydrophobic region not exceeding 3 pm; ii) a hydrophobic region comprising PTFE particles at a concentration of 6-50 vol% of the catalyst-containing layer; iii) catalyst particles or supported catalyst particles with average diameter of 15-30 nm dispersed such that the maximum distance between catalyst particles and closest hydrophobic regime does not exceed > 5 pm.

[0245] According to an embodiment, there is provided a gas diffusion electrode for electrochemical gas evolution without bubble formation, the electrode comprising a catalyst layer with a micro structure engineered to maintain local gas supersaturation below the critical supersaturation limit at current densities above 300 mA / cm2, wherein: a) diffusive path length from any catalyst site to a connected hydrophobic region does not exceed 3 pm; b) hydrophobic regions form an interconnected network throughout the catalyst layer and microporous layer.

[0246] According to an embodiment, there is provided a nickel-based gas diffusion electrode for oxygen evolution in alkaline electrolysis without bubble formation, comprising: a) a nickel fiber or foam gas diffusion backbone layer; b) an electrodeposited nickel microporous layer c) a catalyst layer comprising nickel-iron layered double hydroxide with PTFE content of 6-50 vol% and d) wherein the electrode enables bubble-free oxygen evolution at current densities exceeding 300 mA / cm2.

[0247] According to an embodiment, there is provided an electrolyzing system comprising: a) a first electrode configured as a gas-breathing electrode for bubble-free gas extraction; b) a second electrode configured as either a flow-through electrode or a bubble-evolving electrode, optionally with a hydrophilic porous separator or layer with pore sizes smaller than 200 pm; c) an electrolyte space defined between the first and second electrodes between 100 pm and 4 mm; and d) wherein the system maintains gas separation efficiency exceeding 99% at current densities above 200 mA / cm2.

[0248] According to an embodiment, there is provided a method of manufacturing a gas diffusion electrode for high-current-density bubble-free operation, comprising: a) ultrasonically dispersing catalyst particles and hydrophobic particles at a predetermined ratio to create a catalyst ink with precisely controlled particle distribution; b) spraycoating the dispersed catalyst ink onto a microporous layer under controlled evaporation conditions that create a maximum diffusive path length between catalyst sites and hydrophobic regions not exceeding 3 pm; c) applying a sintering protocol that maintains or improves the predetermined spatial relationship between catalyst and hydrophobic regions; and d) wherein the resulting electrode enables bubble-free operation at current densities exceeding 300 mA / cm2.

[0249] According to an embodiment, there is provided a gas diffusion electrode for a rechargeable metal-air battery comprising: a) a gas diffusion backbone layer with hydrophobic content of 5-25 wt%; b) a microporous layer with controlled crack- free morphology; c) a bifunctional catalyst layer enabling both oxygen reduction and evolution with either one or a mixture of catalysts; and d) wherein the electrode prevents electrolyte leakage during discharge while enabling bubble-free oxygen evolution during charge at current densities exceeding 10 mA / cm2.

Claims

CLAIMS1. An electrolyzing device (1) comprising a first electrode (200) configured to generate a first gas and be in fluid contact with a first gas collector (250) for the first gas; a second electrode (300) configured to generate a second gas and be in fluid contact with a second gas collector (350) for the second gas; wherein the first and second electrodes (200, 300) are arranged to face each other to define an electrolyte space (400) for receiving an electrolyte (404) between the first electrode (200) and the second electrode (300), wherein at least one of the first and second electrode (200, 300) is a gas diffusion electrode comprising: a gas diffusion backbone layer (210, 310) facing away from the electrolyte space (400) and comprising conductive porous material; a catalyst layer (220, 320) facing the electrolyte space (400) and comprising an electrochemically active catalyst, wherein the catalyst layer (220, 320) further comprises hydrophobic particles with an average particle size measured according to ISO 22412 of 10 nm - 5pm and a microporous layer (230, 330) arranged between the gas diffusion backbone layer (210, 310) and the catalyst layer (220, 320) and comprising conductive particles with an average particle size measured according to ISO 22412 of 15 nm - 2 pm and hydrophobic particles with an average particle size measured according to ISO 22412 of 10 nm - 5 pm.

2. A device (1) as claimed in claim 1, wherein the gas diffusion backbone layer is hydrophobic in that it comprises hydrophobic particles.

3. A device (1) as claimed in claim 1 or 2, wherein the gas diffusion layer, the catalyst layer and the microporous layer are positioned in direct contact with each other.

4. A device (1) as claimed in any one of the preceding claims, wherein the electrochemically active catalyst in the catalyst layer comprises catalyst particles with an average hydrodynamic particle size as measured according to ISO 22412 of 1 - 500 nm.

5. A device (1) as claimed in claim 4, wherein the catalyst particles are selected from the group consisting of platinum-coated carbon particles, platinum nanoparticles, iridium nanoparticles, iridium oxide nanoparticles, ruthenium nanoparticles, ruthenium oxide nanoparticles, nickel nanoparticles and nickel-iron layered double hydroxides, or combinations thereof.

6. A device (1) as claimed in any one of the preceding claims, wherein the hydrophobic particles of at least one of the catalyst layer, the microporous layer and, optionally, the gas diffusion layer, have an average particle size measured according to ISO 22412 of between 10 nm - 4 pm, more preferably between 10 nm - 3 pm, more preferably between 10 nm - 2 pm, more preferably between 10 nm - 1 pm, more preferably between 20 nm - 1 pm, more preferably between 30 nm - 1 pm, more preferably between 40 nm - 1 pm, more preferably between 50 nm - 1 pm more preferably between 100 nm - 1 pm, even more preferably between 100 nm - 750 nm, more preferably between 100 nm - 500 nm, and most preferably between 150 - 400 nm.

7. A device (1) as claimed in any one of the preceding claims, wherein at least one of the catalyst layer and the microporous layer comprises impurities other than the hydrophobic particles, and that the impurities, either spherical or non-spherical, have a characteristic size < 200 pm, more preferably < 175 pm, even more preferably < 150 pm, even more preferably < 125 pm, and most preferably < 100 pm.

8. A device (1) as claimed in any one of the preceding claims, wherein the hydrophobic particles are selected from polymers having a contact angle as measured according to the sessile drop method OCA30 by Data-physics of at least 90°, preferably at least 105°, more preferably at least 110°, and preferably selected from the group of polyolefins, such as polypropylene (PP) and polyethylene (PE), fluorinated polyolefins such as polytetrafluoroethylene (PTFE), polystyrene (PS), polyvinylchloride (PVC), polyethylene terephthalate (PET), polyurethane (PU), polypropylene oxide (PPO), polysiloxanes such as polydimethylsiloxane (PDMS), and polysulfones, such as polysulfone (PSU), polyethersulfone (PES / PESU) and polyphenylene sulfone (PPSU), and sulfonated polymers, such as sulfonated poly(ether ether ketone) (SPEEK), PTFE being preferred.

9. A device (1) as claimed in any one of the preceding claims, wherein the hydrophobic particles of the microporous layer, the catalyst layer and optionally the gas diffusion backbone layer are made from the same polymer.

10. A device (1) as claimed in any one of the preceding claims, wherein the conductive particles of the microporous layer have an average particle size according to ISO 22412 of 250 nm - 2 pm, more preferably of 75 nm - 1 pm, and most preferably of 100 nm - 500 nm.

11. A device (1) as claimed in any one of the preceding claims, wherein the conductive particles of the microporous layer are selected from nickel, titanium or carbon particles, or combinations thereof, preferably nickel or carbon particles.

12. A device (1) as claimed in any one of the preceding claims, wherein the conductive material of the gas diffusion backbone layer is selected from nickel, titanium or carbon, or combinations thereof.

13. A device (1) as claimed in any one of the preceding claims, wherein the electrolyte space for receiving the electrolyte between the first electrode and the second electrode does not contain a membrane separator or a highly porous spacer.

14. A device (1) as claimed in any one of the preceding claims, wherein the first electrode (200) is a gas diffusion electrode comprising: a gas diffusion backbone layer (210, 310) facing away from the electrolyte space (400) and comprising conductive porous material; a catalyst layer (220, 320) facing the electrolyte space (400) and comprising an electrochemically active catalyst, wherein the catalyst layer further comprises hydrophobic particles with an average particle size measured according to ISO 22412 of 10 nm - 5pm, and a microporous layer (230, 330) comprising conductive particles with an average particle size measured according to ISO 22412 of 15 nm - 2 pm and hydrophobic particles with an average particle size measured according to ISO 22412 of 10 nm - 5pm and arranged between the gas diffusion backbone layer (210, 310) and the catalyst layer (220, 320), and the second electrode (300) comprises a flow-through electrode.

15. An electrolyzing device (1) comprising a first electrode (200) configured to generate a first gas and be in fluid contact with a first gas collector (250) for the first gas; a second electrode (300) configured to generate a second gas and be in fluid contact with a second gas collector (350) for the second gas; wherein the first and second electrodes (200, 300) are arranged to face each other to define an electrolyte space (400) for receiving an electrolyte (404) between the first electrode (200) and the second electrode (300), wherein at least one of the first and second electrode (200, 300) is a gas diffusion electrode comprising: a gas diffusion backbone layer (210, 310) facing away from the electrolyte space (400) and comprising conductive porous material; a catalyst layer (220, 320) facing the electrolyte space (400) and comprising an electrochemically active catalyst, wherein the catalyst layer (220, 320) further comprises hydrophobic particles with an average particle size measured according to ISO 22412 of 10 nm - 5pm and a microporous layer (230, 330) comprising conductive particles with an average particle size measured according to ISO 22412 of 15 nm - 2 pm and hydrophobic particles with an average particle size measured according to ISO 22412 of 10 nm - 5 pm, wherein the catalyst layer and the microporous layer combine into a single microporous / catalyst layer.

16. An electrolytical multi-cell stack, comprising a plurality of devices (1) as claimed in any one of claims 1 - 15, whereby the plurality of the devices are electrically connected.

17. A gas diffusion electrode for use as the first or the second electrode or the first and the second electrode (200, 300) in a device (1) as claimed in any one of claims 1- 14, the gas diffusion electrode comprising:a gas diffusion backbone layer (210, 310) facing away from an electrolyte space (400) and comprising conductive porous material, such as felt, cloth, paper or foam or combinations thereof; a catalyst layer (220, 320) facing the electrolyte space (400) and comprising an electrochemically active catalyst, wherein the catalyst layer (220, 320) further comprises hydrophobic particles with an average particle size measured according to ISO 22412 of 10 nm - 5pm, and a microporous layer (230, 330) comprising conductive particles with an average particle size measured according to ISO 22412 of 15 nm - 2 pm and hydrophobic particles with an average particle size measured according to ISO 22412 of 10 nm - 5 pm, and arranged between the gas diffusion backbone layer (210, 310) and the catalyst layer (220, 320).

18. A proton exchange membrane electrolyser, an anion exchange membrane electrolyser or a metal air battery, comprising a gas diffusion electrode as claimed in claim 17.

19. A method of manufacturing a gas diffusion electrode as claimed in claim 17, the method comprising providing a gas diffusion backbone layer (210, 310) comprising conductive porous material; applying a microporous layer (230, 330) comprising conductive particles with an average particle size measured according to ISO 22412 of 15 nm - 2 pm and hydrophobic particles with an average particle size measured according to ISO 22412 of 10 nm - 5 pm onto the gas diffusion backbone layer; applying a catalyst layer (220, 320) comprising an electrochemically active catalyst, wherein the catalyst layer further comprises hydrophobic particles with an average particle size measured according to ISO 22412 of 10 nm - 5 pm onto the microporous layer.

20. A method as claimed in claim 19, wherein the gas diffusion backbone layer is hydrophobized by contacting the conductive porous material with an aqueous dispersion of a hydrophobic polymer at a concentration of 0.0001-20 wt.%.

21. A method as claimed in claim 19 or 20, wherein the gas diffusion backbone layer is hydrophobized by contacting the conductive porous material with an aqueous dispersion of a hydrophobic polymer at a concentration of 1-20 wt.%.

22. A method as claimed in any one of claims 19 - 21, wherein the microporous layer is hydrophobized by contacting the conductive particles with an aqueous dispersion comprising a hydrophobic agent.

Images