A FLOW ARRANGEMENT FOR AN ELECTROLYSER, AN ELECTROLYSER, ELECTROLYSIS INSTALLATION, OPERATING METHOD AND METHOD OF MANUFACTURE

Field of Invention

The invention relates to a flow arrangement for an electrolyser, an electrolyser, an electrolysis installation, an operating method for such an electrolyser or installation, and a method of manufacture. In particular but not exclusively, the disclosure relates to such systems and methods for performing continuous electrolysis at supercritical conditions, particularly electrolysis of water and aqueous electrolyte fluids.

Background

One use of electrolysis is for the generation of hydrogen for energy storage. Electricity can be used to separate hydrogen and oxygen from water. Stored hydrogen and oxygen can be recombined in a fuel cell to generate electricity. In the meantime, the hydrogen (and oxygen) can be stored and transported. With improving efficiencies of electrolysis and fuel cell technologies, hydrogen energy storage is being proposed as a solution for many energy storage problems, particularly for the storage of energy from renewable energy sources.

The efficiency of electrolysis is dependent on the losses inherent to the design of the electrolytic cell. Such losses can be considered to impart overpotentials in the electrolytic cell, which represent the cell requiring more energy than theoretically thermodynamically required in order to continue the electrolytic reaction.

Overpotentials can arise owing to a number of different factors. For example, the pre-eminent method of electrolysis of water relies on the use of a polymer electrolyte membrane (PEM electrolysis) which separates the cathode and anode to prevent mixing of hydrogen and oxygen reaction products, while permitting ion transport. However, the presence of a PEM membrane introduces an overpotential into the system. While electrolytic cell designs have been considered which do not use a PEM membrane, such designs face issues in separating the reaction products and maintaining them separate (e.g. oxygen and hydrogen), while permitting efficient ion exchange for electrolysis (i.e. minimising overpotentials).

Further overpotentials are introduced owing to the formation of reaction product bubbles (e.g. hydrogen and oxygen bubbles) that reside on and occlude the electrode surfaces, and owing to the presence of an electrolyte (e.g. in an electrolyte fluid surrounding the electrodes). It is desirable to reduce the overpotentials associated with electrolysis.

It has been proposed to use an electrolyser comprising porous walls to avoid use of a PEM membrane. WO 2022/106874 proposes the use of porous walls to divide a central first fluidic channel 7 which receives a fluid from outer second and third channels 9, 11 , with electrodes 3, 5 provided within the second and third channels 9, 11. The porous walls are porous by virtue of having inclined canals which connect the first fluid channel 7 with the second and third channels 9, 11 respectively, the inclined canals having an opening width or diameter of between 50pm and 200pm.

Summary

According to a first aspect there is disclosed flow arrangement for an electrolyser, comprising: first and second porous walls corresponding to first and second electrodes of the electrolyser; an inlet chamber disposed between the first and second porous walls and configured to receive a fluid through an inlet; first and second outlet chambers for retaining respective fluid reaction products of electrolysis, separated from the inlet chamber by the first and second porous walls respectively; wherein one of, or each of, the first and second porous walls has a discontinuous porous structure, wherein the or each porous wall having the discontinuous porous structure: comprises a body having an inlet side adjacent to the inlet chamber and an outlet side adjacent to the respective outlet chamber, wherein the body is elongate along a longitudinal direction, and has a thickness direction from the inlet side to the outlet side; comprises a plurality of porous regions extending through the body at discrete locations to permit the fluid to flow from the inlet chamber to the respective outlet chamber, wherein each porous region defines a respective network of flow paths through the body

It may be that for the or each porous wall having the discontinuous porous structure, each porous region is elongate along a path through the body having a longitudinal component. It may be that for the or each porous wall having the discontinuous porous structure, each porous region is elongate along a path through the body defining a path angle relative to the longitudinal direction of between 20°-80°.

The path angle may be between 25°-75°, for example between 30°-70°, between 35°-70°, between 40°-70°, for example between 50°-70°.

The expression “elongate” relates to the property of the porous region being long in relation to its width, such that the direction along which the porous region is elongate defines the path (having the longitudinal component). The longitudinal component is a non-zero component of the path which corresponds to (e.g. is parallel with) the longitudinal direction of the body. It should be appreciated that an elongate object (e.g. structure, region, member) has a single elongate path (which may be a bi-directional path). Accordingly, a path along which the porous region has an extent, but which does not correspond to the path along which the porous region is elongate, is not a path meeting the above definition. The path may be linear or substantially linear. The path may be curved.

Definitions with respect to the elongate nature of each porous region and/or its orientation (e.g. angle of inclination) may be interchangeably expressed with respect to the porous region itself, or a boundary of the porous region at which it interfaces with the body. In particular, for each porous region as defined above, there is such a boundary which is elongate along a path through the body having a longitudinal component (e.g. the same path as that of the porous region). Each of the definitions above with respect to the path of the porous region are equally and interchangeably applicable with respect to the boundary. Each boundary may surround the respective porous region to form a closed boundary with open ends corresponding to the inlet side and the outlet side of the body. A path of the porous region or boundary may be defined by reference to a centerline path through the porous region or boundary (respectively), which may be a path through centroids of a plurality of cross-sections of the porous region or boundary (respectively) at successive positions along the thickness direction (and which may be understood as or referred to as a centroidal axis).

The flow arrangement may be for a continuous electrolysis electrolyser (i.e. an electrolyser configured to simultaneously receive an inlet flow of an electrolyte fluid and to discharge outlet flows of fluid reaction products generated by respective electrolysis reactions).

The body of the or each porous wall may be defined as having an anisotropic porous structure provided by the arrangement of the porous regions within the body (i.e. by virtue of the porous regions being discretely arranged, and being elongate along paths which have a longitudinal component, thereby providing for non-uniform flow through the body as a whole). The porous regions themselves may be defined as having an isotropic structure, for example as provided by a generally uniform distribution of a porous medium. The porous medium may be isotropic in that it is generally uniformly distributed within the confines of the porous region, with only the boundaries of the porous regions providing anisotropy to the porous wall.

The flow arrangement may be configured for installation in an installed orientation in which, for the or each porous wall having the discontinuous porous structure, the paths along which each of the respective plurality of porous regions are elongate, have an upwards component towards the outlet chamber. Each porous region may therefore be configured to inhibit a return flow from the respective outlet chamber to the inlet chamber when there is a prevailing buoyancy driven flow through the porous region having an upward component.

The flow arrangement may be configured so that, when the longitudinal direction of the or each porous wall having the discontinuous porous structure is vertically upward, each of the respective plurality of porous regions is elongate along a path towards the respective outlet chamber having an upward component.

The thickness direction may be a direction corresponding to a shortest distance from the inlet side to the outlet side. The thickness direction may be orthogonal to the longitudinal direction of the respective porous wall. The porous walls may each be elongate with respect to a common longitudinal direction (i.e. they may each be elongate along parallel directions), and may have respective thickness directions orthogonal to the common longitudinal direction.

The or each porous wall may be axisymmetric about its longitudinal direction, such that the thickness direction at any angular location around the longitudinal direction is a local thickness direction corresponding to a radial direction about the longitudinal direction. When both porous walls are axisymmetric, they may also be coaxial with each other. The longitudinal direction may pass through a centroid of the cross-sections of the or each porous wall.

Otherwise, the flow arrangement may be implemented with one or more non- axisymmetric porous walls, such as planar porous walls extending linearly along both a lateral direction and a longitudinal direction. The extent of the porous walls along the longitudinal direction may be relatively greater than the extent along the lateral direction, such that it is elongate along the longitudinal direction. It may be that both of the first and second porous walls have the discontinuous porous structure, and at least one property of the discontinuous porous structure differs between the first and second porous walls by a respective minimum offset, selected from the group consisting of: a porosity of the respective porous walls, with an associated minimum offset of 0.01 ; a macro porosity of the respective porous walls, for each porous wall defined as the porosity of the body of the porous wall in the absence of the porous region, with an associated minimum offset of 0.01 ; a micro porosity of the respective porous walls, for each porous wall defined as the porosity of the porous regions, with an associated minimum offset of 0.05; a pitch by which the respective porous regions are spaced apart, with a minimum offset of 10% relative to a smallest of respective pitches of the porous walls; an average cross-sectional area of the respective porous regions, each cross- sectional area being determined by a volume of the porous region divided by an extent of the porous region along the thickness direction, with an associated minimum offset of 10%; an average diameter of the respective porous regions, when each porous region has a circular cross-section normal to the path along which the porous region is elongate, with an associated minimum offset of 10%; a path angle of the respective porous regions, determined as the angle between the paths along which the porous regions are elongate and the respective longitudinal direction, with an associated minimum offset of 5°;and a thickness of the porous wall along the respective thickness direction, with an associated minimum offset of 10% relative to a thinnest one of the porous walls.

A minimum offset as defined above is a minimum amount by which the respective values of the parameter may differ, for example they may differ by at least this amount (and so may differ by more than that amount).

A minimum offset associated with the porosity of the respective porous walls may be at least 0.01 , for example at least 0.02, at least 0.05, at least 0.1 , at least 0.2 or at least 0.3. The porosity of the respective porous walls may differ by an offset of between 0.01-0.5, for example 0.01-0.3, 0.02-0.3, 0.1-0.3, 0.2-0.3.

A minimum offset associated with the macro porosity of the respective porous walls may be at least 0.01 , for example at least 0.02, at least 0.05, at least 0.1 , at least 0.2 or at least 0.3. The porosity of the respective porous walls may differ by an offset of between 0.01-0.5, for example 0.01-0.3, 0.02-0.3, 0.1-0.3, 0.2-0.3.

A minimum offset associated with the micro porosity of the respective porous walls may be at least 0.05, for example at least 0.1 , at least 0.2, at least 0.3 or at least 0.5. The porosity of the respective porous walls may differ by an offset of between 0.05-0.8, for example 0.1-0.8, 0.2-0.8, 0.3-0.8, 0.5-0.8.

A macro porosity as defined above may otherwise be referred to as a primary porosity of the or each porous wall. A micro porosity as defined above may otherwise be referred to as a secondary porosity of the or each porous wall.

A minimum offset associated with an average cross-sectional area of the respective porous regions may be at least 10%, for example at least 20%, at least 30%, at least 50%, or at least 100%. The average cross-sectional areas may differ by an offset of between 10%-200%, for example between 10%-200%, 20%-200%, 30%-200%, 50%- 200% or 100%-200%. The average cross-sectional areas may differ by an offset of between 10%-100%, for example between 10%-100%, 20%-100%, 30%-100%, or 50%- 100%.

A minimum offset associated with an average diameter of the respective porous regions may be at least 10%, for example at least 20%, at least 30%, at least 50%, or at least 100%. The average diameters may differ by an offset of between 10%-200%, for example between 10%-200%, 20%-200%, 30%-200%, 50%-200% or 100%-200%. The average diameters may differ by an offset of between 10%- 100%, for example between 10%-100%, 20%-100%, 30%-100%, or 50%-100%.

A minimum offset associated with the path angle of the respective porous regions may be 5° or more, for example 10° or more, 15° or more, 20° or more. The path angles may differ by an offset of 5°-60°, for example 10°-60°, 15°-60°, 20°-60°, 5°-40°, 10°-40°, 15°-40°, 20°-40°, 5°-20°, or 10°-20°.

A minimum offset associated with a thickness of the respective porous regions may be at least 10%, for example at least 20%, at least 30%, at least 50%, or at least 100%. The thickness may differ by an offset of between 10%-200%, for example between 10%- 200%, 20%-200%, 30%-200%, 50%-200% or 100%-200%. The average cross-sectional areas may differ by an offset of between 10%-100%, for example between 10%-100%, 20%-100%, 30%-100%, or 50%-100%.

Where the parameters above relate to averages, these may be defined as number- weighted averages for the respective plurality of porous regions. Porosity as defined herein relates to an open porosity of the respective component or structure.

Porosity of a porous wall is a total porosity of the porous wall. As described elsewhere herein, the porosity is to be evaluated over a longitudinal extent of the porous wall which is configured to be porous (e.g. excluding non-porous proximal or distal locations, for example for electrical connections).

A micro porosity of a porous wall relates only to the porosity of the porous regions of the porous wall. The micro porosity of the porous wall may therefore relate to the porosity of a porous medium located within the porous regions. Micro porosity of the porous medium (for example, if a sample of the porous medium is isolated from the porous wall or manufactured separately), and therefore micro porosity of the porous wall, may be measured directly by mercury porosimetry according to ASTM standards D4284 and D6761 using an AutoPore V device (available from Micromeritics Instrument Corporation, USA).

A macro porosity of the porous wall relates to the porosity of the body of the porous wall considered alone (e.g. with the entirety of the porous regions being considered to be open, that is to say, not containing porous medium). Macro porosity of the porous wall can be calculated by reference to the design and/or measured dimensions of the porous wall (including the dimensions of the open regions). Alternatively, if the open regions are not or yet to be filled with porous medium, macro porosity of the porous wall can be measured directly by mercury porosimetry according to ASTM standards D4284 and D6761 using an AutoPore V device (available from Micromeritics Instrument Corporation, USA).

A total porosity of a porous wall can be determined as a product of macro and micro porosity. Alternatively, if the open regions of a porous wall already contain porous medium, a total porosity of the porous wall can be measured directly by mercury porosimetry according to ASTM standards D4284 and D6761 using an AutoPore V device (available from Micromeritics Instrument Corporation, USA). The micro porosity of the porous wall may also be determined (e.g. without isolating the porous medium from the porous wall) by dividing the measured total porosity of the porous wall by the measured or calculated macro porosity of the porous wall.

The pitch may be uniform or may be determined by reference to the closest spacing in a non-uniform distribution of porous regions. The path angle of the respective porous regions may be uniform, or may be determined as an average of the angles of the respective porous regions. A diameter of a porous region may be determined as an average diameter of the porous region measured at ends of porous region at inlet and outlet sides of the respective wall. Considering that the porous regions may have a longitudinal extent, they may present a substantially elliptical profile at the inlet and outlet sides, in which case the diameter is to be measured as the circumferential extent of the respective end of the porous region.

A path angle between a path along which a porous region is elongate and the longitudinal direction may be determined as the angle between a path along which a boundary of the porous region extends and the longitudinal direction, evaluated in a plane which includes the longitudinal direction and the local thickness direction (i.e. a plane in which those directions lie). Similarly, an angle between paths along which porous regions are elongate and the respective longitudinal direction may be determined as the angle between paths along which boundaries of the porous regions extend and the longitudinal direction, evaluated in respective planes which include the longitudinal direction and the respective local thickness direction. The local thickness direction is a thickness direction local to the respective porous region. For example, for an annular porous wall, the thickness direction varies around the porous wall, and may correspond to a radial direction of the porous wall. The angle of inclination of the respective porous regions may be uniform, or may be determined as an average of the angles of the respective boundaries of the porous regions.

The expression “pitch” as used above takes the usual meaning of the art, relating to the distance between centres of the porous regions. The disclosure envisages that, when the first and second porous walls are non-planar (for example annular), the pitch is still determined by reference to the absolute distance between centres of the porous regions, but evaluated at the radially-inner side of the respective porous wall when annular (e.g. where the porous region terminates at the radially-inner side of the respective porous wall, or where the radially-inner side has an opening to the respective porous region). Accordingly, for concentric and annular first and second annular walls, an equal angular distribution of porous regions would nevertheless correspond to a different pitch.

It may be that for the or each porous wall having the discontinuous porous structure, the porous regions each have a porosity (e.g. microporosity) of between 0.2- 0.9, for example, between 0.2-0.8, or between 0.3-0.8, or between 0.5-0.8, or between 0.6-0.8, or between 0.3-0.7. The porosity of the porous regions may be determined by mercury porosimetry as described elsewhere herein (e.g., according to ASTM standards D4284 and D6761 using an AutoPore V device (available from Micromeritics Instrument Corporation, USA).

It may be that the or each porous wall having the discontinuous porous structure has a porosity (e.g. total porosity) of between 0.03-0.5, for example, between 0.03-0.3, or between 0.03-0.2, or between 0.05-0.15, or between 0.05-0.1. The porosity of the porous walls may be determined by mercury porosimetry as described elsewhere herein (e.g., according to ASTM standards D4284 and D6761 using an AutoPore V device (available from Micromeritics Instrument Corporation, USA).

It may be that the or each porous wall having the discontinuous porous structure has a (e.g. total) porosity of between 1% and 20%, for example between 2% and 20%, or between 3% and 20%, or between 4% and 20%, or between 5% and 20%, or between 6% and 20%, or between 7% and 20%, or between 8% and 20%, or between 9% and 20%, or between 10% and 20%, or between 1% and 19%, or between 2% and 19%, or between 3% and 19%, or between 4% and 19%, or between 5% and 19%, or between 6% and 19%, or between 7% and 19%, or between 8% and 19%, or between 9% and 19%, or between 10% and 19%, or between 1% and 18%, or between 2% and 18%, or between 3% and 18%, or between 4% and 18%, or between 5% and 18%, or between 6% and 18%, or between 7% and 18%, or between 8% and 18%, or between 9% and 18%, or between 10% and 18%, or between 1% and 17%, or between 2% and 17%, or between 3% and 17%, or between 4% and 17%, or between 5% and 17%, or between 6% and 17%, or between 7% and 17%, or between 8% and 17%, or between 9% and 17%, or between 10% and 17%, or between 1% and 16%, or between 2% and 16%, or between 3% and 16%, or between 4% and 16%, or between 5% and 16%, or between 6% and 16%, or between 7% and 16%, or between 8% and 16%, or between 9% and 16%, or between 10% and 16%, or between 1% and 15%, or between 2% and 15%, or between 3% and 15%, or between 4% and 15%, or between 5% and 15%, or between 6% and 15%, or between 7% and 15%, or between 8% and 15%, or between 9% and 15%, or between 10% and 15%, or between 1% and 14%, or between 2% and 14%, or between 3% and 14%, or between 4% and 14%, or between 5% and 14%, or between 6% and 14%, or between 7% and 14%, or between 8% and 14%, or between 9% and 14%, or between 10% and 14%, or between 1% and 13%, or between 2% and 13%, or between 3% and 13%, or between 4% and 13%, or between 5% and 13%, or between 6% and 13%, or between 7% and 13%, or between 8% and 13%, or between 9% and 13%, or between 10% and 13%, or between 1% and 12%, or between 2% and 12%, or between 3% and 12%, or between 4% and 12%, or between 5% and 12%, or between 6% and 12%, or between 7% and 12%, or between 8% and 12%, or between 9% and 12%, or between 10% and 12%, or between 1% and 11 %, or between 2% and 11%, or between 3% and 11%, or between 4% and 11%, or between 5% and 11%, or between 6% and 11%, or between 7% and 11 %, or between 8% and 11%, or between 9% and 11 %, or between 10% and 11 %, or between 1% and 10%, or between 2% and 10%, or between 3% and 10%, or between 4% and 10%, or between 5% and 10%, or between 6% and 10%, or between 7% and 10%, or between 8% and 10%, or between 9% and 10%. The porosity of the porous walls may be determined by mercury porosimetry as described elsewhere herein (e.g., according to ASTM standards D4284 and D6761 using an AutoPore V device (available from Micromeritics Instrument Corporation, USA).

It may be that for the or each porous wall having the discontinuous porous structure, the porous regions each have a median pore diameter, as determined by mercury porosimetry as described elsewhere herein (e.g., according to ASTM standards D4284 and D6761 using an AutoPore V device (available from Micromeritics Instrument Corporation, USA), of between about 10 to about 50 pm, for example, about 20 to about 40 pm.

It may be that the or each porous wall having the discontinuous porous structure has a permeability, as determined by mercury porosimetry as described elsewhere herein (e.g., according to ASTM standards D4284 and D6761 using an AutoPore V device (available from Micromeritics Instrument Corporation, USA), of between about 10 to about 400 mdarcy, or between about 10 to about 50 mdarcy, or from about 250 to about 400 mdarcy, or from about 100 to about 200 mdarcy.

It may be that for the or each porous wall having the discontinuous porous structure, the porous regions each have a characteristic length, as determined by mercury porosimetry as described elsewhere herein (e.g., according to ASTM standards D4284 and D6761 using an AutoPore V device (available from Micromeritics Instrument Corporation, USA), of between about 5 to about 60 pm, or between about 5 to about 20 pm, or between about 40 to about 60 pm, or between about 20 to about 50 pm.

It may be that the or each porous wall having the discontinuous porous structure has a tortuosity, as determined by mercury porosimetry as described elsewhere herein (e.g., according to ASTM standards D4284 and D6761 using an AutoPore V device (available from Micromeritics Instrument Corporation, USA), of between about 10 to about 80, or between about 10 to about 30, or between about 30 to about 50, or between about 50 to 80.

It may be that for the or each porous wall having the discontinuous porous structure: each porous region has a cross-sectional area of between 10,000- 250,000pm ² , each cross-sectional area being determined as a volume of the porous region divided by an extent of the porous region along the thickness direction; and/or each porous region has an average diameter of between 25-250pm.

It may be that each porous region has a generally circular (e.g. circular) crosssection normal to the path along which it is elongate.

Each porous region may have an average diameter of 25-250pm, for example 50- 250 pm, 50-150 pm, 70-150pm, or approximately 120 pm. The average diameter may be an average cross-sectional diameter along the length of the respective porous region.

Each porous region may have a midpoint diameter of 25-250 pm, for example 25- 100 pm, 25-80 pm, or 25-50 pm. The midpoint diameter may be the diameter half-way along the length of the respective porous region.

Each porous region may have an inlet diameter of 25-250 pm, for example 50-250 pm, 50-150 pm, 70-150 pm, or approximately 120 pm. The inlet diameter may be the diameter of the porous region at the inlet side of the respective wall.

Each porous region may have an outlet diameter of 25-250 pm, for example 50- 250 pm, 50-150 pm, 70-150 pm, or approximately 120 pm. The outlet diameter may be the diameter of the porous region at the outlet side of the respective wall.

It may be that each porous region has a cross-sectional area of between 10,000- 250,000pm ² , for example 15, 000-250, 000pm ² , 15, 000-150, 000pm ² , 20,000- 150,000pm ² , 50, 000-150, 000pm ² , or approximately 100,000 pm ² .

The volume and extent of a porous region are both determined by reference to boundaries of the porous region which do not extend beyond a region between the inlet side and the outlet side of the body. Accordingly, even if the porous region is continuous with a further region of porous material (e.g. extending over an outlet side of the body), that further region is not to be considered when evaluating the volume and extent of the porous region, which is defined as a porous region extending through the body. The porous regions may each be defined as extending through the body between, or within a zone between, the inlet side and the outlet side of the body.

The body may be configured to prevent fluid flow therethrough, except for flow through the porous regions. The body may be substantially non-porous, for example it may be configured so that there is no flow path through the body except for through the discrete porous regions. The body may have a porosity (open porosity) of 0..

It may be that one of the first and second outlet chambers is an outer annular chamber and the other is an inner central chamber surrounded by the inlet chamber having an annular configuration.

The first and second porous walls separating the respective first and second outlet chambers from the inlet chamber may be coaxial with each other. The flow arrangement may be configured so that each of the first and second outlet chambers is configured to only receive fluid flow via the respective porous walls.

The flow arrangement may be configured so that the inlet chamber is only configured to receive fluid from outside of the flow arrangement through the inlet (e.g. the inlet being a single inlet or opening into the inlet chamber). Accordingly, all flow entering an outlet chamber passes through the porous regions of the respective porous wall, and as such the flow regime into the outlet chamber (e.g. flow rate) may be reliably controlled in the design of the flow arrangement by controlling the properties of the porous regions. Further, when the porous wall provides an electrode for electrolysis, all flow entering an outlet chamber passes through the electrode (e.g. through an electrocatalytic region of the electrode) rather than bypassing active regions of the electrode. An electrocatalytic region may be interchangeably referred to as an electrocatalytically active region.

It may be that for the or each porous wall having the discontinuous porous structure a material composition of the porous regions differs from a material composition of the body.

Each porous region may interface with the body at a respective boundary which surrounds the porous region. Each such boundary may correspond to an internal wall of the body.

It may be that for the or each porous wall having the discontinuous porous structure: the body is integrally formed with the plurality of porous regions; and each porous region interfaces with the body at a respective boundary which surrounds the porous region and is defined by a change in porosity between the body and the porous region.

For the or each porous wall having the discontinuous porous structure, the body may be integrally formed with the respective plurality of porous regions by an additive manufacturing process.

It may be that for the or each porous wall having the discontinuous porous structure a material composition of the body is the same as the material composition of the respective porous regions.

The inlet sides of the first and second porous walls may provide opposing surfaces delimiting the inlet chamber. As above, the porous walls correspond to electrodes for an electrolyser (e.g. they comprise or define the electrodes when the flow arrangement is implemented in an electrolyser).

It may be that for at least 50% of a surface area of one of the inlet sides, there is a substantially constant shortest distance of separation to the opposing inlet side. It may be that the inlet sides are substantially locally parallel with each other.

It is desirable to both minimise a separation distance between opposing electrodes, and for it to be substantially constant so as to provide for a relatively uniform rate of reaction along the extent of the electrodes. Although the inlet side may not be electrocatalytically active in use (for example if it is provided with a passivating layer), any electrocatalytic region of a porous wall may terminate at or be at a substantially constant depth behind the inlet side (e.g. in the porous regions of the porous wall), and as such a separation between the opposing inlet sides of the porous walls is considered representative of a separation between regions of the respective porous walls for ion exchange (e.g. electrocatalytic regions of the respective porous walls).

The expression "locally parallel" is intended to mean that, where one or both of the porous walls is non-planar, a plane aligned with a local shortest vector of separation from any point on one of the inlet sides to the other intersects the opposing inlet side to define substantially parallel lines. For example, the opposing inlet sides may be cylindrical or conical and concentric, such that the inlet sides may not be globally parallel with each other around the common axis, but a plane locally intersecting the two boundaries along a local vector of shortest separation defines two respective lines which are parallel with one another.

The first and second porous walls may each extend linearly, parallel to the longitudinal direction (or axis) and may each be elongate along the longitudinal direction. Along a longitudinal extent of the porous walls where they are co-extensive for ion exchange, a cross-section of the flow arrangement may be substantially constant.

By configuring the flow arrangement to have a constant cross-section where the porous walls are co-extensive for ion exchange, the configuration of the flow arrangement is adaptable to easily increase or decrease the elongate length of the porous walls (and/or the flow arrangement, electrolyser). For example, this may be suitable for varying the capacity of an electrolyser (e.g. as may be measured by a reaction rate, a flow-rate for a given reaction efficiency (e.g. the proportion of an electrolyte fluid which is reacted to produce the respective reaction products), or power input measured in kW).

The first and second porous walls may oppose each other, for ion exchange, along a co-extensive longitudinal extent over which there is an average shortest distance of separation between the respective inlet sides. A ratio the co-extensive elongate extent to the average shortest distance of separation may be no less than 5, for example no less than 10.

The ratio corresponds to the slender configuration of the flow arrangement, and in particular the slender configuration of a separation gap between the opposing porous walls. While it is generally known to be desirable to reduce the separation distance between opposing electrodes, the elongate extent of the porous walls along which this is possible in previously considered arrangements may be limited by the means for keeping the reaction products generated at the respective electrodes separate (e.g. by way of an ion-exchange membrane or a laminar-flow buffer - i.e. a flow between the electrodes which maintains laminar conditions to prevent migration of reaction products from one side to the other). In previously-considered arrangements, those means for keeping the reaction products separate tend to either involve an ion-exchange membrane or are hard to maintain over any appreciable longitudinal extent. The invention provides for the porous walls (and thereby the electrodes) to be separated by a slender separation gap - i.e. one that has an elongate/longitudinal extent (along which the electrolyte flows) that is far greater than the separation distance between the porous walls. Without wishing to be bound by theory, it is thought that this may be enabled by the use of a porous wall which provides a resistance to flow reversal from an outlet chamber to the inlet chamber, thereby preventing flow reversal without relying on a high flow inertia from the inlet chamber to the outlet chamber. A high flow inertia may be achieved by providing a relatively small flow area, which would tend to reduce the surface area of a porous wall. By using a porous wall that inhibits flow reversal, rather than relying on high flow inertia, the flow area for a given mass flow through a porous wall can be increased. This permits the separation gap between the porous walls to become slender in nature, thereby increasing a ratio between a surface area of the porous walls for an electrolysis reaction to the volume of the fluid within the separation gap. A rate of electrolysis (e.g. as may be measured by reference to the proportion of a volume of electrolyte fluid which is reacted as it passes through the electrolyser) tends to increase as these ratios increase (i.e. the ratios between the respective surface areas of the electrodes to the volume of the electrolyte fluid within the separation gap between them).

A flow arrangement in accordance with the first aspect may comprise or be defined by an electrolytic cell of an electrolyser. According to a second aspect there is disclosed an electrolyser for performing continuous electrolysis of an electrolyte fluid, wherein the electrolyser comprises a flow arrangement according to the first aspect or according to a seventh aspect (described below) for receiving the electrolyte fluid at the inlet, wherein the first and second porous walls provide first and second electrodes of the electrolyser respectively.

It may be that the electrolyser comprises a controller, wherein the controller is configured to maintain supercritical conditions for the electrolyte fluid at the first and/or second porous walls.

The controller may be configured to maintain supercritical conditions for the electrolyte fluid at the first and/or second porous walls by: controlling flow control equipment to maintain a target inlet pressure and a target inlet temperature of electrolyte fluid at the inlet; and/or controlling a current through, and/or a voltage applied between, the first and second electrodes.

It may be that heating of the electrolyte fluid, for example to a critical temperature corresponding to supercritical conditions for the electrolyte fluid, is provided at the or each respective porous wall (electrode) of the electrolyser.

The controller may be configured to control the flow control equipment, for example a heater, so that the electrolyte fluid is provided to the inlet at a temperature within 50°C (for example within 30°C or within 20°C) of a critical temperature, for example a critical temperature for an aqueous electrolyte fluid of 374°C.

The controller may be configured to control flow control equipment, for example a compressor and/or one or more discharge valves associated with the electrolyser, to maintain a target inlet pressure. The target inlet pressure may be at least a critical pressure for the respective electrolyte fluid. For example, the target inlet pressure may be at least 22MPa for an aqueous electrolyte fluid.

It may be that the controller is configured to control flow control equipment to maintain supercritical conditions for the electrolyte fluid within the inlet chamber and in the first and second outlet chambers.

It may be that the controller is configured to control the flow control equipment so that the electrolyte fluid is provided to the inlet at supercritical conditions.

The flow control equipment may comprise a heater configured to heat electrolyte fluid upstream of the inlet chamber. The heater may be part of the electrolyser or may be installed together with it at (or in) an electrolysis installation. The flow control equipment may comprise a compressor configured to compress the electrolyte fluid upstream of the inlet chamber. The flow control equipment may be part of the electrolyser or may be installed together with it at (or in) an electrolysis installation.

Supercritical conditions for the electrolyte fluid may be supercritical pressure and temperature conditions for the electrolyte fluid of at least 22 MPa pressure and at least 374°C temperature for an aqueous electrolyte fluid.

Supercritical conditions may be between 22 and 27 MPa pressure and between 374°C and 550°C temperature., for example between 374 and 400°C. By maintaining supercritical conditions at the first and/or second porous walls, losses associated with an electrolysis reaction may be reduced by (i) inhibiting formation of bubbles of reaction products on surfaces of the electrodes and/or (ii) increasing the conductivity of an electrolyte fluid for a given electrolyte concentration for a given amount of electrolyte (or vice versa, achieving a suitable conductivity using a relatively lower concentration of electrolyte).

Although some examples discussed herein relate to operation at supercritical conditions and the appended claims refer to an electrolyser or electrolyser installation in which a controller is configured to maintain supercritical conditions, the disclosure envisages electrolysers and electrolyser installations and methods of operation according as described herein (e.g. according to any combination of features envisaged in the present disclosure) and in which there is no such controller or control to maintain supercritical conditions for the electrolyte fluid.

It may be that the electrolyser comprises a controller configured to control flow control equipment to provide the electrolyte fluid to the inlet chamber at an inlet temperature of at least 320°C, for example at least 350°C. The inlet temperature may be below a critical temperature for the respective electrolyte fluid, for example a critical temperature of 374°C for an aqueous electrolyte fluid. The inlet temperature may be within 50°C (for example within 30°C or within 20°C) of a critical temperature for the electrolyte fluid. The inlet temperature may be greater than or equal to 320° and less than 374°C, for example 350°C-370°C or 350°C-360°C. It may be that the controller is configured to control the flow control equipment to provide the electrolyte fluid to the inlet chamber at a pressure which is less than or greater than a critical pressure for the respective electrolyte fluid (for example a critical temperature of 22MPa for an aqueous electrolyte fluid). It may be that the controller is configured to control the flow equipment to provide the electrolyte fluid to the inlet chamber at a pressure of at least 22MPa.

It may be that the inlet temperature and inlet pressure are controlled (e.g., by control of the flow control equipment) so that subcritical conditions are maintained throughout the inlet chamber and outlet chambers.

The flow control equipment may comprise first and second discharge valves as defined below. Such valves may be part of the electrolyser or may be installed together with the electrolyser at (or in) an electrolysis installation. The first discharge valve and/or the second discharge valve may be variable control valves.

The electrolyser may have a first outlet associated with the first outlet chamber for discharging a fluid reaction product generated at the first electrode; and a second outlet associated with the second outlet chamber for discharging a fluid reaction product generated at the second electrode. The flow control equipment may comprise a first discharge valve and a second discharge valve in fluid communication with the first and second outlets respectively.

The first discharge valve may be configured to maintain a first target pressure upstream of the valve and the second discharge valve may be configured to maintain a second target pressure upstream of the valve.

It may be that the first discharge valve and the second discharge valve are configured to maintain different target pressures to cause a pressure drop over each of the respective porous walls, to control respective branch flows of electrolyte fluid driven through the respective porous walls from the inlet chamber into the respective outlet chambers. It may be that the controller is configured to control the first discharge valve and/or the second discharge valve to maintain the respective target pressure(s).

It may be that the controller is configured to control the first and/or second discharge valves to: maintain a target flow rate or a target composition out of one, or each of, the first and second outlets, based on flow rate data, upstream pressure data and/or composition data received by the controller; and/or to maintain a target flow rate ratio between flow out of the first outlet and flow out of the second outlet, based on flow rate data, upstream pressure data and/or composition data received by the controller. The target flow rate ratio may correspond to: a ratio of a total flow rate out of the first outlet and a total flow rate out of the second outlet; or a ratio of a flow rate of a first fluid reaction product out of the first outlet and a flow rate of a second fluid reaction product out of the second outlet. The controller may be configured to control the first and/or second discharge valves to maintain a respective target flow rate or a respective target composition of flow out of each of the first and second outlets, based on flow rate data and/or composition data received by the controller for a respective discharge flow.

A target flow rate as referred to above may be a total flow rate out of the respective outlet, or it may be a flow rate of a particular component fluid of a mixture flowing out of the respective outlet, for example a flow rate of oxygen within a mixture comprising oxygen and electrolyte fluid, or a flow rate of hydrogen within a mixture comprising oxygen and electrolyte fluid. The target flow rate may be a mass flow rate.

Flow rate data may be determined by a sensor of the electrolyser or of an installation in which the electrolyser is installed. For example, there may be a flowmeter downstream of each respective discharge valve. There may be one or more flowmeters downstream of a separation apparatus within an installation in which the electrolyser is installed. For example, a separation apparatus may separate flow discharged through each outlet into a flow of the respective reaction product and a flow of the electrolyte fluid (or water), and a flowmeter may be installed on the or each outlet of the separator to monitor a flow rate of each component (i.e. the reaction product and the electrolyte fluid). A combination of a discharge valve, a flowmeter for monitoring a flow rate through the respective valve, and a controller which controls the discharge valve based on a signal from the flowmeter may provide a mass flow controller, with the controller controlling the discharge valve based on the signal from the flowmeter to maintain a target flow rate.

Upstream pressure data may be determined by one or more pressures sensors of the electrolyser or of an installation in which the electrolyser is installed, to monitor a pressure of the electrolyte fluid upstream of the valve. For example, a pressure sensor may extend into one or each of the outlet chambers which retain the respective reaction products and be configured to send respective pressure signals to the controller, or there may be a differential pressure sensor (which may be a differential pressure transducer) responsive to a pressure difference between the chambers and configured to send a differential pressure signal to the controller. For example, the controller may be calibrated to maintain the or each respective pressure within a respective target range, or to maintain the differential pressure within a target range, in order to maintain target operating conditions (e.g. for example, a target flow rate of one or each respective reaction product through the outlets).

A target composition may be a target proportion of a fluid reaction product within flow discharged through a respective outlet. The composition may be a mass fraction. Composition data may be obtained based on monitoring flow rates of each component of a flow discharged through a respective outlet (i.e. the respective reaction product and the electrolyte fluid).

It may be that the controller does not receive flow rate data or composition data from which a numerical value of a flow rate or component mass fraction can be obtained, but may receive data relating to a flow rate or composition of a flow which may be used in order to control the or each discharge valve such that, with suitable calibration of the controller, a target flow rate, composition or flow rate ratio can be maintained.

It may be that the controller is configured to determine whether there is an excessive amount of the second fluid reaction product in an outlet flow through the first outlet; and/or configured to determine whether there is an excessive amount of the first fluid reaction product in an outlet flow through the second outlet, based on composition data received at the controller for the respective outlet flow. The controller may be configured to control the first discharge valve and/or the second discharge valve to vary a flow rate through a porous wall of the electrolyser, based on the determination.

For example, the controller may control the first discharge valve and/or the second discharge valve to increase a pressure drop from the inlet chamber to the respective outlet to increase a flow rate through the respective outlet. For example, when the outlet is downstream of a porous wall, it may be that increasing the flow rate through that porous wall inhibits return flow of a reaction product through that porous wall (e.g. owing to an inertia of flow through the porous wall).

The controller may be configured to maintain thermodynamic and/or flow rate conditions of the electrolyte fluid through the electrolyser corresponding to a Reynolds number in the inlet chamber being no more than 4000, for example no more than 2300 (for example by controlling a flow rate through the electrolyser for determined thermodynamic conditions of the electrolyte fluid). Controlling the conditions of the electrolyte fluid to meet the above Reynolds number criterion may prevent turbulent and/or transitional mixing within the inlet chamber that may otherwise promote migration of a reaction product between the electrodes across a separation gap between the electrodes.

It may be that for the or each porous wall having the discontinuous porous structure and providing an electrode of the electrolyser, the porous regions comprise an electrocatalyst and thereby define an electrocatalytic region of the respective electrode for an electrolysis half-reaction. The expression “electrocatalyst” as used herein refers to an electrocatalyst (e.g. an electrocatalytically active material) suitable for catalysing a half-reaction of electrolysis of an electrolyte fluid. Provision of such an electrocatalyst is considered to provide an associated electrocatalytic region of the porous wall which is electrocatalytically active (i.e. for the respective half-reaction of electrolysis). The electrolysis reaction may be electrolysis of water or an aqueous (water-based) electrolyte fluid. Accordingly, the electrocatalyst may be an electrocatalyst suitable for catalysing a half-reaction of electrolysis of water.

The electrocatalyst may comprise (e.g. consist essentially of, consist of, or be) one or more elements selected from: noble metals (i.e. Ru, Rh, Pd, Os, Ir, Pt, Au, Ag, Re), d-block transition metals (i.e. Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Y, Zr, Nb, Mo, Tc, Ru, Rh, Pd, Ag, Cd, Hf, Ta, W, R, Os, Ir, Pt, Au, Hg, Rf, Db, S, Rh, Hs, Mt, Ds, Rg, Cn), f- block lanthanides (i.e. La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), rare earth metals (i.e. Sc, Y, La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), platinum group metals (i.e. Ru, Rh, Pd, Os, Ir, Pt), P, S, C and/or combinations thereof. Said one or more elements may be present in elemental form (e.g. as (substantially) pure metals), in mixtures (e.g. alloys), in compounds (e.g. oxides, hydroxides, nitrides, borides, sulfides, phosphides, carbonates and/or organic compounds such as metalorganic frameworks) and/or incorporated into nanomaterials such as nanoparticles, nanotubes (e.g. carbon nanotubes), nanocages (e.g. fullerenes) and/or nanosheets (e.g. graphene).

The electrocatalyst may comprise (e.g. consist essentially of, consist of or be) Co, Fe, Ir, Li, Ni, P, S, Ti, Zn, and/or combinations thereof.

The electrocatalyst may comprise (e.g. consist essentially of, consist of or be) Ce, Co, Fe, Ho, Ir, Mo, Ni, Pd, Pt, Ru, Sm, W, and/or combinations thereof.

The electrocatalyst may comprise (e.g. consist essentially of, consist of or be) Ni, Fe, Co, P, S, and/or combinations thereof.

The electrocatalyst may comprise (e.g. consist essentially of, consist of or be) Pt, Ir, Pd, Ni, Mo, and/or combinations thereof.

In some examples, the electrocatalyst comprises (e.g. consists essentially of, consists of or is) Ni (e.g. elemental Ni) or Pt (e.g. elemental Pt).

In some examples, the electrocatalyst comprises (e.g. consists essentially of, consists of or is) Ni (e.g. elemental Ni).

In some examples, the electrocatalyst comprises (e.g. consists essentially of, consists of or is) Pt (e.g. elemental Pt). The electrocatalyst may be provided in the form of, or may be formed from, an electrocatalyst-containing particulate. Particles of the electrocatalyst-containing particulate may comprise (e.g. consist essentially of, or consist of) the electrocatalyst. The electrocatalyst-containing particulate may comprise (e.g. consist essentially of, or consist of) particles of the electrocatalyst.

For example, the electrocatalyst-containing particulate may comprise (e.g. consist essentially of, or consist of) particles comprising (e.g. consisting essentially of, or consisting of) one or more elements selected from: noble metals (i.e. Ru, Rh, Pd, Os, Ir, Pt, Au, Ag, Re), d-block transition metals (i.e. Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Y, Zr, Nb, Mo, Tc, Ru, Rh, Pd, Ag, Cd, Hf, Ta, W, R, Os, Ir, Pt, Au, Hg, Rf, Db, S, Rh, Hs, Mt, Ds, Rg, Cn), f-block lanthanides (i.e. La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), rare earth metals (i.e. Sc, Y, La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), platinum group metals (i.e. Ru, Rh, Pd, Os, Ir, Pt), P, S, C and/or combinations thereof. Said one or more elements may be present in elemental form (e.g. as (substantially) pure metals), in mixtures (e.g. alloys), in compounds (e.g. oxides, hydroxides, nitrides, borides, sulfides, phosphides, carbonates and/or organic compounds such as metal-organic frameworks) and/or incorporated into nanomaterials such as nanoparticles, nanotubes (e.g. carbon nanotubes), nanocages (e.g. fullerenes) and/or nanosheets (e.g. graphene).

The electrocatalyst-containing particulate may comprise (e.g. consist essentially of, or consist of) particles comprising (e.g. consisting essentially of, or consisting of) Co, Fe, Ir, Li, Ni, P, S, Ti, Zn, and/or combinations thereof.

The electrocatalyst-containing particulate may comprise (e.g. consist essentially of, or consist of) particles comprising (e.g. consisting essentially of, or consisting of) Ce, Co, Fe, Ho, Ir, Mo, Ni, Pd, Pt, Ru, Sm, W, and/or combinations thereof.

The electrocatalyst-containing particulate may comprise (e.g. consist essentially of, or consist of) particles comprising (e.g. consisting essentially of, or consisting of) Ni, Fe, Co, P, S, and/or combinations thereof.

The electrocatalyst-containing particulate may comprise (e.g. consist essentially of, or consist of) particles comprising (e.g. consisting essentially of, or consisting of) Pt, Ir, Pd, Ni, Mo, and/or combinations thereof.

In some examples, the electrocatalyst-containing particulate comprises (e.g. consists essentially of or consists of) particles comprising (e.g. consisting essentially of, or consisting of) Ni (e.g. elemental Ni) or Pt (e.g. elemental Pt). In some examples, the electrocatalyst-containing particulate comprises (e.g. consists essentially of or consists of) particles comprising (e.g. consisting essentially of, or consisting of) Ni (e.g. elemental Ni).

In some examples, the electrocatalyst-containing particulate comprises (e.g. consists essentially of or consists of) particles comprising (e.g. consisting essentially of, or consisting of) Pt (e.g. elemental Pt).

In some examples, the electrocatalyst-containing particulate is a Ni-containing particulate. The Ni-containing particulate may comprise (e.g. consist essentially of, or consist of) particles of Ni (e.g. elemental Ni) or an Ni-based alloy.

In some examples, the electrocatalyst-containing particulate is an Ni particulate consisting essentially of particles of Ni.

It may be that for the or each porous wall having the discontinuous porous structure and providing an electrode of the electrolyser, the respective porous regions each comprise a porous medium formed from an electrocatalyst-containing particulate.

The porous medium may be formed by consolidating the electrocatalyst-containing particulate. In other words, the respective porous regions may each comprise the electrocatalyst which is provided in the form of a consolidated porous medium.

The term “consolidating” used herein refers to a process of physically and/or chemically adhering particles of the electrocatalyst-containing particulate to one another to form the porous medium. Consolidation of the electrocatalyst-containing particulate may comprise sintering the electrocatalyst-containing particulate, fusing the electrocatalyst-containing particulate and/or bonding particles of the electrocatalystcontaining particulate to one another by way of a binder and/or a chemical reaction.

For example, the porous medium may be formed by sintering the electrocatalystcontaining particulate. It will be appreciated that sintering is a process of forming a solid mass of material from a particulate by application of heat and/or pressure without melting the particulate to the point of liquefaction. During sintering, particles of the particulate may be bonded to one another by diffusion of atoms and/or molecules between neighbouring particles at temperatures below the melting point of the material. Plastic deformation of the particles may also occur.

Accordingly, the respective porous regions may each comprise the electrocatalyst which is provided in the form of a sintered porous medium.

The porous medium may be formed by fusing the electrocatalyst-containing particulate. It will be appreciated that fusion is a process involving melting of solid material to liquid. A porous medium may be formed from a particulate by heating the particulate to a temperature at which local melting (for example, melting at particle surfaces) occurs, leading to bonding of adjacent particles, while avoiding total liquefaction of the particulate.

Accordingly, the respective porous regions may each comprise the electrocatalyst which is provided in the form of a fused porous medium.

The porous medium may be formed by bonding particles of the electrocatalystcontaining particulate to one another using a binder. The binder may be a polymeric binder such as a thermoplastic binder or a thermosetting binder (e.g. a resin). A thermoplastic binder is a polymeric binder which melts or becomes pliable at elevated temperatures and which solidifies upon cooling. A thermosetting binder (e.g. a resin) is a polymeric binder which hardens irreversibly (i.e. cures) by heating, exposure to radiation and/or exposure to a suitable catalyst.

The porous medium may be formed by bonding particles of the electrocatalystcontaining particulate to one another by a chemical reaction. For example, a chemical reaction may take place which forms a reaction product which bonds adjacent particles to one another. The chemical reaction may be an oxidation reaction, for example, the growth of an oxide layer at surfaces of the particles, which oxide layer bonds adjacent particles to one another. The chemical reaction may take place at elevated temperatures, for example, on exposure to a suitable (e.g. oxidising) atmosphere.

Accordingly, the respective porous regions may each comprise the electrocatalyst which is provided in the form of a bonded porous medium.

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise (e.g. consist essentially of, or consist of) one or more elements selected from: noble metals (i.e. Ru, Rh, Pd, Os, Ir, Pt, Au, Ag, Re), d-block transition metals (i.e. Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Y, Zr, Nb, Mo, Tc, Ru, Rh, Pd, Ag, Cd, Hf, Ta, W, R, Os, Ir, Pt, Au, Hg, Rf, Db, S, Rh, Hs, Mt, Ds, Rg, Cn), f-block lanthanides (i.e. La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), rare earth metals (i.e. Sc, Y, La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), platinum group metals (i.e. Ru, Rh, Pd, Os, Ir, Pt), P, S, C and/or combinations thereof. Said one or more elements may be present in elemental form (e.g. as (substantially) pure metals), in mixtures (e.g. alloys), in compounds (e.g. oxides, hydroxides, nitrides, borides, sulfides, phosphides, carbonates and/or organic compounds such as metal-organic frameworks) and/or incorporated into nanomaterials such as nanoparticles, nanotubes (e.g. carbon nanotubes), nanocages (e.g. fullerenes) and/or nanosheets (e.g. graphene). The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise (e.g. consist essentially of, or consist of) Co, Fe, Ir, Li, Ni, P, S, Ti, Zn, and/or combinations thereof.

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise (e.g. consist essentially of, or consist of) Ce, Co, Fe, Ho, Ir, Mo, Ni, Pd, Pt, Ru, Sm, W, and/or combinations thereof.

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise (e.g. consist essentially of, or consist of) Ni, Fe, Co, P, S, and/or combinations thereof.

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise (e.g. consist essentially of, or consist of) Pt, Ir, Pd, Ni, Mo, and/or combinations thereof.

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise (e.g. consist essentially of, or consist of) Ni (e.g. elemental Ni) or Pt (e.g. elemental Pt).

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise (e.g. consist essentially of, or consist of) Ni (e.g. elemental Ni).

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise (e.g. consist essentially of, or consist of) Pt (e.g. elemental Pt).

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no less than about 50 wt. %, for example, no less than about 60 wt. %, or no less than about 70 wt. %, or no less than about 80 wt. %, or no less than about 90 wt. %, or no less than about 95 wt. %, or no less than about 99 wt. %, electrocatalyst. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no more than about 100 wt. %, for example, no more than about 99 wt. %, or no more than about 95 wt. %, or no more than about 90 wt. %, electrocatalyst. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise from about 50 wt. % to about 100 wt. %, for example, from about 60 wt. % to about 100 wt. %, or from about 70 wt. % to about 100 wt. %, or from about 80 wt. % to about 100 wt. %, or from about 90 wt. % to about 100 wt. %, or from about 95 wt. % to about 100 wt. %, or from about 99 wt. % to about 100 wt. %, electrocatalyst.

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no less than about 50 wt. %, for example, no less than about 60 wt. %, or no less than about 70 wt. %, or no less than about 80 wt. %, or no less than about 90 wt. %, or no less than about 95 wt. %, or no less than about 99 wt. %, noble metal (i.e. Ru, Rh, Pd, Os, Ir, Pt, Au, Ag, Re), d-block transition metal (i.e. Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Y, Zr, Nb, Mo, Tc, Ru, Rh, Pd, Ag, Cd, Hf, Ta, W, R, Os, Ir, Pt, Au, Hg, Rf, Db, S, Rh, Hs, Mt, Ds, Rg, Cn), f-block lanthanide (i.e. La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), rare earth metal (i.e. Sc, Y, La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), platinum group metal (i.e. Ru, Rh, Pd, Os, Ir, Pt), P, S, C and/or combinations thereof. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no more than about 100 wt. %, for example, no more than about 99 wt. %, or no more than about 95 wt. %, or no more than about 90 wt. %, noble metal (i.e. Ru, Rh, Pd, Os, Ir, Pt, Au, Ag, Re), d-block transition metal (i.e. Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Y, Zr, Nb, Mo, Tc, Ru, Rh, Pd, Ag, Cd, Hf, Ta, W, R, Os, Ir, Pt, Au, Hg, Rf, Db, S, Rh, Hs, Mt, Ds, Rg, Cn), f-block lanthanide (i.e. La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), rare earth metal (i.e. Sc, Y, La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), platinum group metal (i.e. Ru, Rh, Pd, Os, Ir, Pt), P, S, C and/or combinations thereof. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise from about 50 wt. % to about 100 wt. %, for example, from about 60 wt. % to about 100 wt. %, or from about 70 wt. % to about 100 wt. %, or from about 80 wt. % to about 100 wt. %, or from about 90 wt. % to about 100 wt. %, or from about 95 wt. % to about 100 wt. %, or from about 99 wt. % to about 100 wt. %, noble metal (i.e. Ru, Rh, Pd, Os, Ir, Pt, Au, Ag, Re), d-block transition metal (i.e.

Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Y, Zr, Nb, Mo, Tc, Ru, Rh, Pd, Ag, Cd, Hf, Ta, W,

R, Os, Ir, Pt, Au, Hg, Rf, Db, S, Rh, Hs, Mt, Ds, Rg, Cn), f-block lanthanide (i.e. La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), rare earth metal (i.e. Sc, Y, La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), platinum group metal (i.e. Ru, Rh, Pd, Os, Ir, Pt), P, S, C and/or combinations thereof.

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no less than about 50 wt. %, for example, no less than about 60 wt. %, or no less than about 70 wt. %, or no less than about 80 wt. %, or no less than about 90 wt. %, or no less than about 95 wt. %, or no less than about 99 wt. %, Co, Fe, Ir, Li, Ni, P,

S, Ti, Zn, and/or combinations thereof. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no more than about 100 wt. %, for example, no more than about 99 wt. %, or no more than about 95 wt. %, or no more than about 90 wt. %, Co, Fe, Ir, Li, Ni, P, S, Ti, Zn, and/or combinations thereof. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise from about 50 wt. % to about 100 wt. %, for example, from about 60 wt. % to about 100 wt. %, or from about 70 wt. % to about 100 wt. %, or from about 80 wt. % to about 100 wt. %, or from about 90 wt. % to about 100 wt. %, or from about 95 wt. % to about 100 wt. %, or from about 99 wt. % to about 100 wt. %, Co, Fe, Ir, Li, Ni, P, S, Ti, Zn, and/or combinations thereof. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no less than about 50 wt. %, for example, no less than about 60 wt. %, or no less than about 70 wt. %, or no less than about 80 wt. %, or no less than about 90 wt. %, or no less than about 95 wt. %, or no less than about 99 wt. %, Ce, Co, Fe, Ho, Ir, Mo, Ni, Pd, Pt, Ru, Sm, W, and/or combinations thereof. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no more than about 100 wt. %, for example, no more than about 99 wt. %, or no more than about 95 wt. %, or no more than about 90 wt. %, Ce, Co, Fe, Ho, Ir, Mo, Ni, Pd, Pt, Ru, Sm, W, and/or combinations thereof. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise from about 50 wt. % to about 100 wt. %, for example, from about 60 wt. % to about 100 wt. %, or from about 70 wt. % to about 100 wt. %, or from about 80 wt. % to about 100 wt. %, or from about 90 wt. % to about 100 wt. %, or from about 95 wt. % to about 100 wt. %, or from about 99 wt. % to about 100 wt. %, Ce, Co, Fe, Ho, Ir, Mo, Ni, Pd, Pt, Ru, Sm, W, and/or combinations thereof.

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no less than about 50 wt. %, for example, no less than about 60 wt. %, or no less than about 70 wt. %, or no less than about 80 wt. %, or no less than about 90 wt. %, or no less than about 95 wt. %, or no less than about 99 wt. %, Ni, Fe, Co, P, S, and/or combinations thereof. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no more than about 100 wt. %, for example, no more than about 99 wt. %, or no more than about 95 wt. %, or no more than about 90 wt. %, Ni, Fe, Co, P, S, and/or combinations thereof. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise from about 50 wt. % to about 100 wt. %, for example, from about 60 wt. % to about 100 wt. %, or from about 70 wt. % to about 100 wt. %, or from about 80 wt. % to about 100 wt. %, or from about 90 wt. % to about 100 wt. %, or from about 95 wt. % to about 100 wt. %, or from about 99 wt. % to about 100 wt. %, Ni, Fe, Co, P, S, and/or combinations thereof.

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no less than about 50 wt. %, for example, no less than about 60 wt. %, or no less than about 70 wt. %, or no less than about 80 wt. %, or no less than about 90 wt. %, or no less than about 95 wt. %, or no less than about 99 wt. %, Pt, Ir, Pd, Ni, Mo, and/or combinations thereof. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no more than about 100 wt. %, for example, no more than about 99 wt. %, or no more than about 95 wt. %, or no more than about 90 wt. %, Pt, Ir, Pd, Ni, Mo, and/or combinations thereof. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise from about 50 wt. % to about 100 wt. %, for example, from about 60 wt. % to about 100 wt. %, or from about 70 wt. % to about 100 wt. %, or from about 80 wt. % to about 100 wt. %, or from about 90 wt. % to about 100 wt. %, or from about 95 wt. % to about 100 wt. %, or from about 99 wt. % to about 100 wt. %, Pt, Ir, Pd, Ni, Mo, and/or combinations thereof.

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no less than about 50 wt. %, for example, no less than about 60 wt. %, or no less than about 70 wt. %, or no less than about 80 wt. %, or no less than about 90 wt. %, or no less than about 95 wt. %, or no less than about 99 wt. %, Ni and/or Pt. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no more than about 100 wt. %, for example, no more than about 99 wt. %, or no more than about 95 wt. %, or no more than about 90 wt. %, Ni and/or Pt. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise from about 50 wt. % to about 100 wt. %, for example, from about 60 wt. % to about 100 wt. %, or from about 70 wt. % to about 100 wt. %, or from about 80 wt. % to about 100 wt. %, or from about 90 wt. % to about 100 wt. %, or from about 95 wt. % to about 100 wt. %, or from about 99 wt. % to about 100 wt. %, Ni and/or Pt.

The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no less than about 50 wt. %, for example, no less than about 60 wt. %, or no less than about 70 wt. %, or no less than about 80 wt. %, or no less than about 90 wt. %, or no less than about 95 wt. %, or no less than about 99 wt. %, Ni. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise no more than about 100 wt. %, for example, no more than about 99 wt. %, or no more than about 95 wt. %, or no more than about 90 wt. %, Ni. The (e.g. consolidated, sintered, fused and/or bonded) porous medium may comprise from about 50 wt. % to about 100 wt. %, for example, from about 60 wt. % to about 100 wt. %, or from about 70 wt. % to about 100 wt. %, or from about 80 wt. % to about 100 wt. %, or from about 90 wt. % to about 100 wt. %, or from about 95 wt. % to about 100 wt. %, or from about 99 wt. % to about 100 wt. %, Ni.

It may be that for the or each porous wall having the discontinuous porous structure: a material composition of the porous regions differs from a material composition of the body. It may be that the body comprises a passive region of the respective electrode to inhibit electrolysis.

It may be that for the or each porous wall having the discontinuous porous structure: the body is integrally formed with the plurality of porous regions; each porous region interfaces with the body at a respective boundary which surrounds the porous region and is defined by a change in porosity between the body and the porous region; the porous regions comprise an electrocatalyst, thereby belonging to an electrocatalytic region of the respective electrode for an electrolysis half-reaction; and the porous regions and the body have a common material composition comprising the electrocatalyst.

The material composition of the body which is in common with the material composition of the respective porous regions may be a material composition of a core portion of the body which surrounds and is adjacent to each of the porous regions. The material composition of the body which is in common with the material composition of the respective porous regions may be an uncoated material composition of the body (i.e. a material coating of the body excluding any coating). For example, the body may comprise a coating (e.g. a passivating coating) which defines the inlet side of the body, with the passivating coating having a different composition to a core portion of the body. It may be that, for the or each porous wall having the discontinuous structure, the porous wall is configured for an electrolysis reaction additionally on the outlet side of the porous wall. For example, the outlet side of the porous wall may comprise the electrocatalyst and may be configured to react with electrolyte fluid.

It may be that the porous wall is formed by an additive manufacturing process to provide the integrally formed body and porous regions.

It may be that for the or each porous wall having the discontinuous porous structure, the inlet side of the body is defined by a passive region which is configured to be less electrocatalytically active than the electrocatalytic region. It may be that the passive region comprises a passivating coating defining the inlet side of the body to inhibit electrolysis.

The passive region as defined with respect to any of the statements above may be configured to inhibit a respective half-reaction of electrolysis. By providing the passive region (e.g. as provided by the body as a whole, a core portion or uncoated portion of the body, or at an inlet side of the body - for example by a passivating coating), the respective half-reaction of electrolysis may be inhibited or prevented from occurring on the inlet side of the porous wall, such that the reaction may only take place or may predominantly take place once the electrolyte fluid (and/or respective ions) has passed into one of the porous regions (and/or passed through the porous regions to reach an electrocatalytic region defining the outlet side of the porous wall). The passive region (and any passive region as described herein) may be configured to inhibit electrolysis by being less electrocatalytically active than the electrocatalytic region (for the respective half-reaction of electrolysis of the electrolyte fluid). A rate of the respective half-reaction in the passive region (as a whole) may be lower than a rate of the respective half-reaction in the electrocatalytic region (as a whole), and this may make the passive region less electrocatalytically active than the electrocatalytic region. The rate of the respective half-reaction may be: a rate of generation of the respective fluid reaction product (e.g. hydrogen or oxygen depending on the electrode) in the respective region, for example a mass of the fluid reaction product generated per unit time (e.g. g/s of hydrogen); or the current density integrated over an electrolyte interface of the respective region (e.g. in A) - the electrolyte interface being the surface of the respective region configured to contact the electrolyte fluid. The rate for the respective region as used herein is the total rate for the respective region, rather than a specific rate per unit volume, thickness, mass, or surface area (e.g. of the porous wall or of the electrolyte interface of the respective region of the porous wall) etc. The passive region and the electrocatalytic region may be configured so that the rate of the respective half-reaction in the passive region (as a whole) may be lower than a rate of the respective half-reaction in the electrocatalytic region (as a whole). For example, the rate in the passive region may be no more than 50% of the rate in electrocatalytic region, no more than 20%, nor more than 10%, no more than 5% or no more than 1%. The rate in the passive region may be substantially zero.

It may be that the passive region comprises less of an electrocatalyst (for electrocatalysing the respective half-reaction of electrolysis of the electrolyte fluid) than the electrocatalytic region, for example less of the same electrocatalyst of the electrocatalytic region; and/or any electrocatalyst of the passive region may be less electrocatalytically active for the respective half-reaction than the electrocatalyst of the electrocatalytic region.

The passive region may comprise less of the electrocatalyst of the electrocatalytic region at an electrolyte interface of the passive region with the electrolyte fluid (i.e. in a surface region of the passive region which contacts the electrolyte fluid in use) than there is at the electrolyte interface of the electrocatalytic region (i.e. in the surface region of the electrocatalytic region). In other words, a material composition of the passive region at the electrolyte interface of the passive region (i.e. in the surface region) may comprise less of the electrocatalyst than a material composition of the electrocatalytic region at the respective electrolyte interface. For example, it may comprise less than 5 wt.% of the electrocatalyst, less than 1 wt.% of the electrocatalyst, of less than 0.1 wt.% of the electrocatalyst. The composition of the passive region at the electrolyte interface may be assessed by reference to a surface region of the passive region adjacent to the electrolyte fluid in use, for example having a thickness of 1 pm.

An electrocatalytic region as referred to herein may have a first composition. A passive region as referred to herein may have a second composition. A nominal electrode for conducting the respective half-reaction of electrolysis of the electrolyte fluid in a nominal cell may have a surface region which is adjacent the electrolyte fluid at an electrolyte interface of the nominal electrode. The surface region may have a thickness of 1 pm from the electrolyte interface.

The first and second compositions may be selected so that, when the surface region of the nominal electrode has the first composition, the overpotential required to reach a current density of 10 mA cm ^-2 at the nominal electrode in the nominal cell is no more than half, no more than one third, or no more than one fifth of the overpotential required to reach the current density of 10 mA cm ^-2 at the nominal electrode in the nominal cell when the surface region of the nominal electrode has the second composition, at a nominal pressure and temperature, for example 22.5 MPa and 375°C.

A respective half-reaction may be inhibited by the passive region being provided as a passivating layer or coating as discussed elsewhere herein.

Any of the above statements relating to the electrocatalytic region, the passive region, and the properties of those regions as electrocatalytically active or less electrocatalytically active regions may be applied to any of the aspects or statements provided above and below and as discussed in the detailed description.

When the passive region comprises a passivating coating defining the inlet side of a porous wall, the passivating coating may be a dielectric coating, for example an oxide coating (e.g. an inorganic metal oxide coating), such as silica (SiC>2), zinc oxide (ZnO), or zirconia (ZrC>2)). All references herein to passivating or providing a passivating coating may comprise applying a dielectric coating, such as a coating of the example materials described in the preceding sentence. A passivating coating may be provided, for example by CVD (chemical vapour deposition) of a passivating substance, such as Alumina (AI2O3), Zirconia (ZrCh) or Titania (TiCh).

By passivating the inlet side of a porous wall, the respective fluid reaction product generated at the electrode may only be generated at locations which are downstream from the inlet chamber (e.g. within the porous regions of the porous or on the outlet side of the porous wall), rather than on the inlet side which delimits the inlet chamber. This may inhibit the respective fluid reaction product being generated in or migrating to the inlet chamber. Irrespective of the provision of a passive region, it may be that any fluid reaction product generated within or provided to the outlet chamber (for example by reaction with an electrocatalytic region at or defining the outlet side of the compound porous electrode) is inhibited from flowing back through the porous wall, as this would require the lower- density fluid reaction product to flow against buoyancy forces downwardly through the porous regions (which flow may also be against a pressure gradient over the porous wall). This may be referred to as inhibiting return flow by downstream-biased buoyancy effects, since buoyancy forces within the porous regions promote downstream flow to the outlet chamber rather than upstream flow to the inlet chamber.

According to a third aspect there is disclosed an electrolysis installation comprising: a source of electrolyte fluid (for example an aqueous electrolyte fluid); and an electrolyser in accordance with the second aspect for performing continuous electrolysis of the electrolyte fluid.

The electrolysis installation may comprise a plurality of electrolysers in accordance with the first aspect of the invention. The electrolysis installation may comprise flow control equipment as defined herein.

According to a fourth aspect there is disclosed a method of operating an electrolyser in accordance with the second aspect or an electrolysis installation in accordance with the third aspect, comprising: providing an inlet flow of electrolyte fluid to the inlet chamber via the inlet to conduct electrolysis half-reactions at the first and second electrodes provided by the first and second porous walls, to generate respective fluid reaction products; wherein the electrolyte fluid and/or associated ions flow into the porous regions of the or each electrode having the discontinuous porous structure to react with the respective electrode; wherein each of the first and second outlet chambers retains the respective fluid reaction product for discharge, and the respective electrode inhibits return flow of the fluid reaction product from the outlet chamber to the inlet chamber.

It may be that, for the or each porous wall (electrode) having the discontinuous porous structure and which has a passive region and electrocatalytic region, the rate of the respective half-reaction of electrolysis at the respective electrocatalytic region is greater than the rate of the half-reaction of electrolysis at the respective passive region. The method may comprise controlling inlet and/or outlet conditions of the electrolyser so that there is a pressure drop over each porous wall from the inlet chamber to the respective outlet chamber.

The electrolyte fluid may be received into each outlet chamber only through the respective porous wall.

The method may further comprise controlling operation of the electrolyser to maintain supercritical temperature and pressure conditions for the electrolyte fluid at the first and/or second porous walls.

Controlling operation of the electrolyser may comprise controlling thermodynamic and/or flow rate of the electrolyte fluid.

Controlling thermodynamic and/or flow rate conditions may comprise controlling flow control equipment to maintain a target inlet pressure and a target inlet temperature of electrolyte fluid at the inlet; and/or controlling a current through, and/or a voltage applied between, the first and second electrodes.

It may be that the electrolyte fluid is heated, for example to a critical temperature corresponding to supercritical conditions for the electrolyte fluid, is provided at the ot each respective porous wall (electrode) of the electrolyser.

Controlling thermodynamic and/or flow rate conditions may comprise controlling flow control equipment, for example a heater, so that the electrolyte fluid is provided to the inlet at a temperature within 50°C (for example within 30°C or within 20°C) of a critical temperature, for example a critical temperature for an aqueous electrolyte fluid of 374°C.

Controlling thermodynamic and/or flow rate conditions may comprise controlling flow control equipment, for example a compressor and/or one or more discharge valves associated with the electrolyser, to maintain a target inlet pressure. The target inlet pressure may be at least a critical pressure for the respective electrolyte fluid. For example, the target inlet pressure may be at least 22MPa for an aqueous electrolyte fluid.

It may be that the supercritical temperature and pressure conditions for the electrolyte fluid are maintained within the inlet chamber and in the first and second outlet chambers. It may be that the thermodynamic and/or flow rate conditions are controlled so that the electrolyte fluid is provided to the inlet at supercritical conditions.

It may be that the supercritical temperature and pressure conditions are at least 22 MPa pressure and at least 374°C temperature for an aqueous electrolyte fluid. Supercritical conditions may be between 22 and 27 MPa pressure and between 374°C and 550°C temperature., for example between 374 and 400°C.

The electrolyser may comprise first and second discharge valves as defined above. It may be that the first discharge valve maintains a first target pressure upstream of the valve; and that the second discharge valve maintains a second target pressure upstream of the valve. The first and second target pressures may differ such that there is a pressure drop over the porous wall driving a branch flow of electrolyte fluid through the porous wall from the inlet chamber into the outlet chamber.

The controller may control the first and/or second discharge valve to maintain a target flow rate or a target composition out of one of, or each of, the first and second outlets, based on flow rate data, upstream pressure data and/or composition data received by the controller. The controller may control the first and/or second discharge valve to maintain a target flow rate ratio between flow out of the first outlet and flow out of the second outlet, based on flow rate data and/or composition data received by the controller. The target flow rate ratio may correspond to a ratio of a total flow rate out of the first outlet and a total flow rate out of the second outlet; or a ratio of a flow rate of a first fluid reaction product out of the first outlet and a flow rate of a second fluid reaction product out of the second outlet.

The controller may determine whether there is an excessive amount of the second fluid reaction product in an outlet flow through the first outlet, or determining whether there is an excessive amount of the first fluid reaction product in an outlet flow through the second outlet, for example based on composition data received at the controller for the respective outlet flow. The controller may control the first discharge valve and/or the second discharge valve to vary a flow rate through a porous wall of the electrolyser, based on the determination.

The electrolyser may be operated to maintain thermodynamic and/or flow rate conditions such that a Reynolds number in the inlet chamber is no more than 4000, for example no more than 2300.

According to a fifth aspect there is disclosed a method of manufacturing a porous wall having a discontinuous porous structure, for an electrolyser, comprising: providing a body for the porous wall, wherein the body is elongate along a longitudinal direction, and has a thickness direction from a first side to a second side; removing material from the body to form a plurality of open regions, the open regions extending through the body at discrete locations, wherein each open region is elongate along a path through the body having a longitudinal component; applying an electrocatalyst composition to the body so that it flows into the open regions; heating the body to perform a heat treatment operation in which an electrocatalyst component of the electrocatalyst composition forms a porous region at each location of the open regions, wherein each porous region defines a respective network of flow paths through the body to permit fluid to flow from the first side of the body to the second side of the body.

The porous wall may have any combination of the features disclosed above with respect to the porous walls of the first and second aspects.

The method may further comprise a drying operation to vaporise a component of the electrocatalyst composition, conducted after applying the electrocatalyst composition and before the heat treatment operation.

The drying operation may be conducted to reduce a mass of the electrocatalyst composition retained on the body by 5-60%, for example 10-60%, 20-60%, 30-60%, or 30-50%. The drying operation may be conducted to reduce a mass of the electrocatalyst composition corresponding to a mass of one or more solvents in the electrocatalyst composition as applied. The drying operation may be carried out at a temperature between 100°C and 300°C, for example between 150°C and 250°C, for example approximately 200°C. The drying operation may be carried out in an oven. The drying operation may be carried out for between 1 and 4 hours, for example between 1 and 3 hours, for example approximately 2 hours.

It may be that the heat treatment operation comprises heating the body to a target temperature of between 150-1000°C;

The target temperature may be between 150-1000°C, for example between 250- 800°C, between 300-600°C, between 300-450°C, for example approximately 350°C.

Alternatively, the target temperature may be between 150-1500°C, for example between 150-1000°C, or between 600-1000°C, or between 800-1000°C, or between 900-1000°C, for example approximately 930°C.

The heat treatment operation may be a sintering operation in which the electrocatalyst component of the electrocatalyst composition is sintered to form a porous region (e.g. a sintered porous region) at each location of the open regions. The heat treatment operation may be a fusing operation in which the electrocatalyst component of the electrocatalyst composition is fused to form a porous region (e.g. a fused porous region) at each location of the open regions.

The heat treatment operation may be a bonding operation in which electrocatalystcontaining particles (e.g. electrocatalyst particles) of the electrocatalyst composition are bonded to one another to form a porous region (e.g. a bonded porous region) at each location of the open regions. Bonding the particles may comprise bonding the particle by way of a binder (e.g. polymeric binder). Bonding the particles may comprise melting and/or curing the binder (e.g. polymeric binder).

The heat treatment operation may be a chemical reaction operation in which electrocatalyst-containing particles (e.g. electrocatalyst particles) of the electrocatalyst composition are bonded to one another by way of a chemical reaction to form a porous region (e.g. a bonded porous region) at each location of the open regions. For example, the chemical reaction operation may comprise forming a reaction product (e.g. oxide) by a chemical reaction, the reaction product bonding electrocatalyst-containing particles (e.g. electrocatalyst particles) to one another.

The heat treatment operation may comprise a ramp phase in which the temperature ramps to the target temperature. In the ramp phase a ramp rate may be between 0.5 and 5°C per minute, for example between 0.5°C and 2°C per minute, such as approximately 1 °C per minute. The heat treatment operation may comprise a holding phase in which the temperature is maintained at the target temperature. The holding phase may be between 30 and 300 minutes, for example between 30 and 90 minutes, such as approximately 60 minutes.

The heat treatment operation may be a two-stage heat treatment operation comprising heating the body to a first target temperature of between 150-500°C and then to a second target temperature of between 500-1000°C. The first target temperature may be between 200-500°C, for example between 250-450°C, or between 300-500°C, for example approximately 300C°C or approximately 350°C. The second target temperature may be between 600-1000°C, for example between 700-1000°C, or between 800-1000°C, or between 900-1000°C, or between 900-950°C, for example approximately 930°C. The first stage of the heat treatment operation may comprise holding the body at the first target temperature for 2-10 hours, for example 4-8 hours, for example approximately 6 hours. The second stage of the heat treatment operation may comprise holding the body at the second target temperature for between 30 and 300 minutes, for example between 30 and 90 minutes, such as approximately 60 minutes. The first stage may be carried out to pyrolyze or burn off non-electrocatalyst components of the electrocatalyst composition such as binder components of the electrocatalyst composition. The first stage may therefore be carried out in an oxidising atmosphere. The first stage of the heat treatment operation may therefore be considered to be a pyrolysis operation.

The second stage may be carried out to consolidate (e.g. sinter) electrocatalystcontaining particles (e.g. electrocatalyst particles) to one another. The second stage may therefore be carried out in an inert (or reducing) atmosphere such as an argon atmosphere or a nitrogen/hydrogen atmosphere. The second stage of the heat treatment operation may therefore be considered to be a consolidating (e.g. sintering) operation.

The second stage of the heat treatment operation may be conducted at the same time as, comprise or be another heat treatment operation, such as a post-weld heat treatment operation (e.g. a stress-relief heat treatment operation).

The method may be a method of manufacturing a porous wall for an electrolyser in accordance with the second aspect.

It may be that the electrocatalyst composition has a viscosity of from about 1 Pa s to about 30 Pa s when applied to the body.

The electrocatalyst composition may have a (i.e. dynamic) viscosity no less than about 1 Pa s, for example, no less than about 2 Pa s, or no less than about 4 Pa s, or no less than about 5 Pa s, or no less than about 6 Pa s, or no less than about 8 Pa s, or no less than about 10 Pa s, or no less than about 12 Pa s, or no less than about 14 Pa s, or no less than about 16 Pa s, or no less than about 18 Pa s, or no less than about 20 Pa s, or no less than about 21 Pa s, or no less than about 22 Pa s, or no less than about 23 Pa s, or no less than about 24 Pa s, or no less than about 25 Pa s. The electrocatalyst composition may have a (i.e. dynamic) viscosity no greater than about 30 Pa s, for example, no greater than about 28 Pa s, or no greater than about 26 Pa s, or no greater than about 25 Pa s, or no greater than about 24 Pa s, or no greater than about 23 Pa s, or no greater than about 22 Pa s, or no greater than about 21 Pa s, or no greater than about 20 Pa s, or no greater than about 19 Pa s, or no greater than about 18 Pa s, or no greater than about 17 Pa s, or no greater than about 16 Pa s, or no greater than about 14 Pa s, or no greater than about 12 Pa s, or no greater than about 10 Pa s, or no greater than about 8 Pa s. The electrocatalyst composition may have a (i.e. dynamic) viscosity from about 1 Pa s to about 30 Pa s, for example, from about 1 Pa s to about 28 Pa s, or from about 1 Pa s to about 26 Pa s, or from about 1 Pa s to about 25 Pa s, or from about 1 Pa s to about 24 Pa s, or from about 1 Pa s to about 23 Pa s, or from about 1 Pa s to about 22 Pa s, or from about 1 Pa s to about 21 Pa s, or from about 1 Pa s to about 20 Pa s, or from about 1 Pa s to about 18 Pa s, or from about 1 Pa s to about 16 Pa s, or from about 1 Pa s to about 14 Pa s, or from about 1 Pa s to about 12 Pa s, or from about 1 Pa s to about 10 Pa s, or from about 1 Pa s to about 8 Pa s, or from about 2 Pa s to about 30 Pa s, or from about 2 Pa s to about 28 Pa s, or from about 2 Pa s to about 26 Pa s, or from about 2 Pa s to about 25 Pa s, or from about 2 Pa s to about 24 Pa s, or from about 2 Pa s to about 23 Pa s, or from about 2 Pa s to about 22 Pa s, or from about 2 Pa s to about 21 Pa s, or from about 2 Pa s to about 20 Pa s, or from about 2 Pa s to about 18 Pa s, or from about 2 Pa s to about 16 Pa s, or from about 2 Pa s to about 14 Pa s, or from about 2 Pa s to about 12 Pa s, or from about 2 Pa s to about 10 Pa s, or from about 2 Pa s to about 8 Pa s, or from about 4 Pa s to about 30 Pa s, or from about 4 Pa s to about 28 Pa s, or from about 4 Pa s to about 26 Pa s, or from about 4 Pa s to about 25 Pa s, or from about 4 Pa s to about 24 Pa s, or from about 4 Pa s to about 23 Pa s, or from about 4 Pa s to about 22 Pa s, or from about 4 Pa s to about 21 Pa s, or from about 4 Pa s to about 20 Pa s, or from about 4 Pa s to about 18 Pa s, or from about 4 Pa s to about 16 Pa s, or from about 4 Pa s to about 14 Pa s, or from about 4 Pa s to about 12 Pa s, or from about 4 Pa s to about 10 Pa s, or from about

4 Pa s to about 8 Pa s, or from about 5 Pa s to about 30 Pa s, or from about 5 Pa s to about 28 Pa s, or from about 5 Pa s to about 26 Pa s, or from about 5 Pa s to about 25 Pa s, or from about 5 Pa s to about 24 Pa s, or from about 5 Pa s to about 23 Pa s, or from about 5 Pa s to about 22 Pa s, or from about 5 Pa s to about 21 Pa s, or from about

5 Pa s to about 20 Pa s, or from about 5 Pa s to about 18 Pa s, or from about 5 Pa s to about 16 Pa s, or from about 5 Pa s to about 14 Pa s, or from about 5 Pa s to about 12 Pa s, or from about 5 Pa s to about 10 Pa s, or from about 5 Pa s to about 8 Pa s, or from about 6 Pa s to about 30 Pa s, or from about 6 Pa s to about 28 Pa s, or from about

6 Pa s to about 26 Pa s, or from about 6 Pa s to about 25 Pa s, or from about 6 Pa s to about 24 Pa s, or from about 6 Pa s to about 23 Pa s, or from about 6 Pa s to about 22 Pa s, or from about 6 Pa s to about 21 Pa s, or from about 6 Pa s to about 20 Pa s, or from about 6 Pa s to about 18 Pa s, or from about 6 Pa s to about 16 Pa s, or from about 6 Pa s to about 14 Pa s, or from about 6 Pa s to about 12 Pa s, or from about 6 Pa s to about 10 Pa s, or from about 6 Pa s to about 8 Pa s, or from about 8 Pa s to about 30 Pa s, or from about 8 Pa s to about 28 Pa s, or from about 8 Pa s to about 26 Pa s, or from about 8 Pa s to about 25 Pa s, or from about 8 Pa s to about 24 Pa s, or from about 8 Pa s to about 23 Pa s, or from about 8 Pa s to about 22 Pa s, or from about 8 Pa s to about 21 Pa s, or from about 8 Pa s to about 20 Pa s, or from about 8 Pa s to about 18 Pa s, or from about 8 Pa s to about 16 Pa s, or from about 8 Pa s to about 14 Pa s, or from about 8 Pa s to about 12 Pa s, or from about 8 Pa s to about 10 Pa s, or from about 8 Pa s to about 8 Pa s, or from about 10 Pa s to about 30 Pa s, or from about 10 Pa s to about 28 Pa s, or from about 20 Pa s to about 26 Pa s, or from about 10 Pa s to about 25 Pa s, or from about 10 Pa s to about 24 Pa s, or from about 10 Pa s to about 23 Pa s, or from about 10 Pa s to about 22 Pa s, or from about 10 Pa s to about 21 Pa s, or from about 10 Pa s to about 20 Pa s, or from about 10 Pa s to about 18 Pa s, or from about 10 Pa s to about 16 Pa s, or from about 10 Pa s to about 14 Pa s, or from about 10 Pa s to about 12 Pa s, or from about 12 Pa s to about 30 Pa s, or from about 14 Pa s to about 30 Pa s, or from about 16 Pa s to about 30 Pa s, or from about 18 Pa s to about 30 Pa s, or from about 20 Pa s to about 30 Pa s, or from about 21 Pa s to about 30 Pa s, or from about 22 Pa s to about 30 Pa s, or from about 23 Pa s to about 30 Pa s, or from about

24 Pa s to about 30 Pa s, or from about 25 Pa s to about 30 Pa s, or from about 16 Pa s to about 25 Pa s, or from about 18 Pa s to about 25 Pa s, or from about 20 Pa s to about

25 Pa s, or from about 16 Pa s to about 23 Pa s, or from about 18 Pa s to about 23 Pa s, or from about 20 Pa s to about 23 Pa s, or from about 16 Pa s to about 20 Pa s, or from about 18 Pa s to about 20 Pa s. The (i.e. dynamic) viscosity of the electrocatalyst composition may be measured at 25 °C, for example, according to ISO 3104:2020.

It may be that the electrocatalyst composition comprises a mixture of an electrocatalyst and liquid when applied to the body.

The electrocatalyst composition is a composition which comprises (e.g. consists essentially of, consists of, or is) an electrocatalyst for the respective half-reaction of electrolysis. For example, the electrocatalyst composition may comprise (e.g. be, consist essentially of, or consist of) a mixture of the electrocatalyst for the respective halfreaction of electrolysis and liquid.

The electrocatalyst composition may comprise (e.g. be, consist essentially of, or consist of) a slurry comprising the electrocatalyst for the respective half-reaction of electrolysis and liquid.

The electrocatalyst composition may comprise (e.g. be, consist essentially of, or consist of) a suspension of the electrocatalyst for the respective half-reaction of electrolysis in liquid.

The electrocatalyst may be provided in the mixture in the form of a particulate.

The electrocatalyst composition may comprise (e.g. be, consist essentially of, or consist of) a colloidal suspension (e.g. of the particulate in liquid). For example, the electrocatalyst composition may comprise (e.g. be, consist essentially of, or consist of) a sol (i.e. a solid-in-liquid colloidal suspension), a conductive ink (i.e. a suspension of electrically conductive electrocatalyst particles in liquid), or a paste (i.e. a solid-in-liquid suspension having a sufficiently high solids contents that the suspension behaves as a solid in response to low applied stresses) comprising the electrocatalyst for the respective half-reaction of electrolysis.

The electrocatalyst may be any suitable electrocatalyst described herein.

Particle size properties of the electrocatalyst particulate may be measured in a well-known manner by a laser scattering technique (e.g. standard ISO 13320-1). In this technique, the size of particles in powders, suspensions and emulsions is measured using the diffraction of a laser beam, based on an application of Mie theory. A machine, for example available from Microtrac MRB, provides measurements and a plot of the cumulative percentage by volume of particles having a size, referred to in the art as the ‘equivalent spherical diameter’ (e.s.d), less than given e.s.d values. The mean particle size dso is the value determined in this way of the particle e.s.d at which there are 50% by volume of the particles which have an equivalent spherical diameter less than that dso value. The dgg value is the value determined in the same way of the particle e.s.d. at which there is 99% by volume of the particles which have an equivalent spherical diameter less than the dgg value.

The electrocatalyst particulate may have a dso no less than about 100 nm, for example, no less than about 500 nm, or no less than about 1 pm, or no less than about 4 pm, or no less than about 5 pm, or no less than about 8 pm, or no less than about 10 pm. The electrocatalyst particulate may have a dso no more than about 100 pm, for example, no more than about 75 pm, or no more than about 50 pm, or no more than about 25 pm, or no more than about 20 pm. The electrocatalyst particulate may have a dso from about 100 nm to about 100 pm, for example, from about 500 nm to about 100 pm, or from about 750 nm to about 100 pm, or from about 1 pm to about 100 pm, or from about 10 pm to about 100 pm, or from about 500 nm to about 100 pm, or from about 500 nm to about 50 pm, or from about 750 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 50 pm, or from about 500 nm to about 20 pm, or from about 750 nm to about 20 pm, or from about 1 pm to about 20 pm, or from about 10 pm to about 20 pm, or from about 500 nm to about 50 pm, or from about 750 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 50 pm, or from about 500 nm to about 25 pm, or from about 750 nm to about 25 pm, or from about 1 pm to about 25 pm, or from about 10 pm to about 25 pm, or from about 500 nm to about 20 pm, or from about 750 nm to about 20 pm, or from about 1 pm to about 20 pm, or from about 10 pm to about 20 pm, or from about 500 nm to about 15 pm, or from about 750 nm to about 15 pm, or from about 1 pm to about 15 pm, or from about 10 pm to about 15 pm.

The electrocatalyst particulate may have a dgg no less than about 1 pm, for example, no less than about 10 pm, or no less than about 20 pm, or no less than about 30 pm, or no less than about 35 pm. The electrocatalyst particulate may have a dgg no more than about 150 pm, for example, no more than about 100 pm, or no more than about 75 pm, or no more than about 50 pm, or no more than about 45 pm, or no more than about 40 pm. The electrocatalyst particulate may have a dgg from about 1 pm to about 150 pm, for example, from about 1 pm to about 75 pm, or from about 1 pm to about 50 pm, or from about 1 pm to about 45 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 150 pm, or from about 10 pm to about 100 pm, or from about 10 pm to about 75 pm, or from about 10 pm to about 50 pm, or from about 10 pm to about 45 pm, or from about 10 pm to about 40 pm, or from about 20 pm to about 150 pm, or from about 20 pm to about 100 pm, or from about 20 pm to about 75 pm, or from about 20 pm to about 50 pm, or from about 20 pm to about 45 pm, or from about 20 pm to about 40 pm, or from about 30 pm to about 150 pm, or from about 30 pm to about 100 pm, or from about 30 pm to about 75 pm, or from about 30 pm to about 50 pm, or from about 30 pm to about 45 pm, or from about 30 pm to about 40 pm.

Smaller particle sizes, as evidenced by lower values of dso and/or dgg, may facilitate better penetration of the electrocatalyst composition into open regions.

For example, the electrocatalyst particulate may comprise (e.g. consist essentially of, or consist of) particles comprising (e.g. consisting essentially of, or consisting of) one or more elements selected from: noble metals (i.e. Ru, Rh, Pd, Os, Ir, Pt, Au, Ag, Re), d-block transition metals (i.e. Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Y, Zr, Nb, Mo, Tc, Ru, Rh, Pd, Ag, Cd, Hf, Ta, W, R, Os, Ir, Pt, Au, Hg, Rf, Db, S, Rh, Hs, Mt, Ds, Rg, Cn), f- block lanthanides (i.e. La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), rare earth metals (i.e. Sc, Y, La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu), platinum group metals (i.e. Ru, Rh, Pd, Os, Ir, Pt), P, S, C and/or combinations thereof; wherein the electrocatalyst particulate has: a dso from about 100 nm to about 100 pm, for example, from about 500 nm to about 100 pm, or from about 500 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 1 pm to about 25 pm; and a dgg from about 1 pm to about 150 pm, for example, from about 1 pm to about 75 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 75 pm, or from about 10 pm to about 75 pm, or from about 20 pm to about 75 pm, or from about 20 pm to about 50 pm, or from about 30 pm to about 75 pm, or from about 30 pm to about 45 pm, or from about 30 pm to about 40 pm. The electrocatalyst particulate may comprise (e.g. consist essentially of, or consist of) particles comprising (e.g. consisting essentially of, or consisting of) Co, Fe, Ir, Li, Ni, P, S, Ti, Zn, and/or combinations thereof, wherein the electrocatalyst particulate has: a dso from about 100 nm to about 100 pm, for example, from about 500 nm to about 100 pm, or from about 500 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 1 pm to about 25 pm, or from about 8 pm to about 16 pm; and a dgg from about 1 pm to about 150 pm, for example, from about 1 pm to about 75 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 75 pm, or from about 10 pm to about 75 pm, or from about 20 pm to about 75 pm, or from about 20 pm to about 50 pm, or from about 30 pm to about 75 pm, or from about 30 pm to about 45 pm, or from about 30 pm to about 40 pm.

The electrocatalyst particulate may comprise (e.g. consist essentially of, or consist of) particles comprising (e.g. consisting essentially of, or consisting of) Ce, Co, Fe, Ho, Ir, Mo, Ni, Pd, Pt, Ru, Sm, W, and/or combinations thereof, wherein the electrocatalyst particulate has: a dso from about 100 nm to about 100 pm, for example, from about 500 nm to about 100 pm, or from about 500 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 1 pm to about 25 pm; and a dgg from about 1 pm to about 150 pm, for example, from about 1 pm to about 75 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 75 pm, or from about 10 pm to about 75 pm, or from about 20 pm to about 75 pm, or from about 20 pm to about 50 pm, or from about 30 pm to about 75 pm, or from about 30 pm to about 45 pm, or from about 30 pm to about 40 pm.

The electrocatalyst particulate may comprise (e.g. consist essentially of, or consist of) particles comprising (e.g. consisting essentially of, or consisting of) Ni, Fe, Co, P, S, and/or combinations thereof, wherein the electrocatalyst particulate has: a dso from about 100 nm to about 100 pm, for example, from about 500 nm to about 100 pm, or from about 500 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 1 pm to about 25 pm, or from about 4 pm to about 10 pm; and a dgg from about 1 pm to about 150 pm, for example, from about 1 pm to about 75 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 75 pm, or from about 10 pm to about 75 pm, or from about 20 pm to about 75 pm, or from about 20 pm to about 50 pm, or from about 30 pm to about 75 pm, or from about 30 pm to about 45 pm, or from about 30 pm to about 40 pm.

The electrocatalyst particulate may comprise (e.g. consist essentially of, or consist of) particles comprising (e.g. consisting essentially of, or consisting of) Pt, Ir, Pd, Ni, Mo, and/or combinations thereof, wherein the electrocatalyst particulate has: a dso from about 100 nm to about 100 pm, for example, from about 500 nm to about 100 pm, or from about 500 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 1 pm to about 25 pm; and a dgg from about 1 pm to about 150 pm, for example, from about 1 pm to about 75 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 75 pm, or from about 10 pm to about 75 pm, or from about 20 pm to about 75 pm, or from about 20 pm to about 50 pm, or from about 30 pm to about 75 pm, or from about 30 pm to about 45 pm, or from about 30 pm to about 40 pm.

In some examples, the electrocatalyst particulate comprises (e.g. consists essentially of or consists of) particles comprising (e.g. consisting essentially of, or consisting of) Ni (e.g. elemental Ni) or Pt (e.g. elemental Pt), wherein the electrocatalyst particulate has: a dso from about 100 nm to about 100 pm, for example, from about 500 nm to about 100 pm, or from about 500 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 1 pm to about 25 pm; and a dgg from about 1 pm to about 150 pm, for example, from about 1 pm to about 75 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 75 pm, or from about 10 pm to about 75 pm, or from about 20 pm to about 75 pm, or from about 20 pm to about 50 pm, or from about 30 pm to about 75 pm, or from about 30 pm to about 45 pm, or from about 30 pm to about 40 pm.

In some examples, the electrocatalyst particulate comprises (e.g. consists essentially of or consists of) particles comprising (e.g. consisting essentially of, or consisting of) Ni (e.g. elemental Ni), wherein the electrocatalyst particulate has: a dso from about 100 nm to about 100 pm, for example, from about 500 nm to about 100 pm, or from about 500 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 1 pm to about 25 pm; and a dgg from about 1 pm to about 150 pm, for example, from about 1 pm to about 75 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 75 pm, or from about 10 pm to about 75 pm, or from about 20 pm to about 75 pm, or from about 20 pm to about 50 pm, or from about 30 pm to about 75 pm, or from about 30 pm to about 45 pm, or from about 30 pm to about 40 pm.

In some examples, the electrocatalyst particulate comprises (e.g. consists essentially of or consists of) particles comprising (e.g. consisting essentially of, or consisting of) Pt (e.g. elemental Pt), wherein the electrocatalyst particulate has: a dso from about 100 nm to about 100 pm, for example, from about 500 nm to about 100 pm, or from about 500 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 1 pm to about 25 pm; and a dgg from about 1 pm to about 150 pm, for example, from about 1 pm to about 75 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 75 pm, or from about 10 pm to about 75 pm, or from about 20 pm to about 75 pm, or from about 20 pm to about 50 pm, or from about 30 pm to about 75 pm, or from about 30 pm to about 45 pm, or from about 30 pm to about 40 pm. In some examples, the electrocatalyst particulate is a Ni-containing particulate, wherein the electrocatalyst particulate has: a dso from about 100 nm to about 100 pm, for example, from about 500 nm to about 100 pm, or from about 500 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 1 pm to about 25 pm; and a dgg from about 1 pm to about 150 pm, for example, from about 1 pm to about 75 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 75 pm, or from about 10 pm to about 75 pm, or from about 20 pm to about 75 pm, or from about 20 pm to about 50 pm, or from about 30 pm to about 75 pm, or from about 30 pm to about 45 pm, or from about 30 pm to about 40 pm. The Ni-containing particulate may comprise (e.g. consist essentially of, or consist of) particles of Ni (e.g. elemental Ni) or an Ni-based alloy.

In some examples, the electrocatalyst-containing particulate is an Ni particulate consisting essentially of particles of Ni, wherein the electrocatalyst particulate has: a dso from about 100 nm to about 100 pm, for example, from about 500 nm to about 100 pm, or from about 500 nm to about 50 pm, or from about 1 pm to about 50 pm, or from about 1 pm to about 25 pm; and a dgg from about 1 pm to about 150 pm, for example, from about 1 pm to about 75 pm, or from about 1 pm to about 50 pm, or from about 10 pm to about 75 pm, or from about 10 pm to about 75 pm, or from about 20 pm to about 75 pm, or from about 20 pm to about 50 pm, or from about 30 pm to about 75 pm, or from about 30 pm to about 45 pm, or from about 30 pm to about 40 pm.

The electrocatalyst-containing particulate may have any suitable particle shape (i.e. morphology), such as (e.g. substantially) spherical, spheroidal, acicular or needlelike, fibrous, platy, flake-like, etc. In some examples, the electrocatalyst-containing particulate has a flake-like particle shape. In some examples, the electrocatalystcontaining particulate comprises (e.g. consists essentially or, or consists of) flakes of electrocatalyst, for example, flakes of metal or metal alloy (such as flakes of Ni or Ni alloy).

A suitable liquid may be selected based on the nature of the electrocatalyst particulate, the dimensions of the open regions, and/or manufacturing constraints such as heating temperatures.

The liquid may be polar or non-polar.

The liquid may be aqueous (e.g. water-based). That is to say, the liquid may comprise (e.g. consist essentially of, consist of, or be) water.

The liquid may be organic. The liquid may comprise (e.g. consist essentially of, consist of, or be) one or more organic species, e.g., hydrocarbons.

The liquid may comprise (e.g. consist essentially of, consist of, or be) one or more alcohols, esters, acetates, acetate esters, ether acetates, and/or acids. The liquid may comprise (e.g. consist essentially of, consist of, or be) one or more solvents, e.g., one or more polar solvents or one or more non-polar solvents, for example, one or more organic solvents.

In some examples, the liquid comprises (e.g. consists essentially of, consists of, or is) a non-polar ester (e.g. a non-polar ester solvent).

In some examples, the liquid comprises (e.g. consists essentially of, consists of, or is) a reaction product of a carboxylic acid and an alcohol, for example, wherein the alcohol comprises an alkoxyalcohol. In some examples, the liquid comprises (e.g. consists essentially of, consists of, or is) an acetate ester or an ether acetate. In some examples, the liquid comprises (e.g. consists essentially of, consists of, or is) 2- butoxyethyl acetate.

The electrocatalyst composition may comprise a binder, for example, a polymeric binder._The polymeric binder may be a thermoplastic binder or a thermosetting binder (e.g. a resin). A thermoplastic binder is a polymeric binder which melts or becomes pliable at elevated temperatures and which solidifies upon cooling. A thermosetting binder (e.g. a resin) is a polymeric binder which hardens irreversibly (i.e. cures) by heating, exposure to radiation and/or exposure to a suitable catalyst. The binder may be dispersed or dissolved in the liquid.

The electrocatalyst composition may comprise one or more additives such as viscosity or rheology modifiers, tackifiers, plasticizers, pigments, etc. For example, the electrocatalyst composition may comprise carbon black.

A suitable electrocatalyst composition may be conductive ink 116-25 available from Creative Materials Inc (of Massachusetts, USA). The viscosity of the conductive ink l ie- 25 may be adjusted as required by dilution with solvent 112-19, solvent 102-03, or solvent 113-12, also available from Creative Materials Inc (of Massachusetts, USA).

The method according to the fifth aspect may be for providing a porous wall having a discontinuous porous structure for a flow arrangement in accordance with the first aspect, or for an electrolyser in accordance with the second aspect.

According to a sixth aspect there is disclosed a method of manufacturing a porous wall having a discontinuous porous structure, for an electrolyser, comprising: forming, by an additive manufacturing process, a porous wall comprising: a body having a first side and a second side, wherein the body is elongate along a longitudinal direction, and has a thickness direction from the first side to the second side; a plurality of porous regions extending through the body at discrete locations to permit fluid to flow from the first side to the second side; wherein each porous region defines a respective network of flow paths through the body; and wherein each porous region is elongate along a path through the body having a longitudinal component; wherein the additive manufacturing process is controlled to vary a porosity of the porous wall during forming, so that the porous regions are formed with a higher open porosity than the body.

Accordingly, each porous region may interface with the body at a respective boundary which surrounds the porous region and is defined by a change in porosity between the body and the porous region.

It may be that the porous regions and the body have a common material composition.

It may be that the material composition comprises an electrocatalyst, whereby the porous wall comprises an electrocatalytic region which includes at least the porous regions. The electrocatalyst may be or have any of the properties as discussed above with respect to the fifth aspect. Although regions of the body away from the porous region would also comprise electrocatalyst, they may not be configured to come into contact with an electrolyte fluid in use (e.g. embedded parts of the wall not exposed to fluid flow).

The method may further comprise providing the first side of the body with a passivating coating which is configured to be less electrocatalytically active than an electrocatalytic region of the porous wall which includes the porous regions.

The method according to sixth aspect may be to provide a porous wall having a discontinuous porous structure for a flow arrangement in accordance with the first aspect, or for an electrolyser in accordance with the second aspect.

According to a seventh aspect there is disclosed a flow arrangement for an electrolyser, comprising: first and second porous walls corresponding to first and second electrodes of the electrolyser; an inlet chamber disposed between the first and second porous walls and configured to receive electrolyte fluid through an inlet; and first and second outlet chambers for retaining respective fluid reaction products of electrolysis, separated from the inlet chamber by the first and second porous walls respectively. The or each porous wall may comprise a body having an inlet side adjacent to the inlet chamber and an outlet side adjacent to the respective outlet chamber, wherein the body is elongate along a longitudinal direction, and has a thickness direction from the inlet side to the outlet side.

In a flow arrangement according to the seventh aspect, the porous walls may not have a discontinuous porous structure as described elsewhere herein.

For example, the or each porous wall may comprise a plurality of channels extending through the body.

It may be that each channel is elongate along a path through the body having a longitudinal component. It may be that each channel is elongate along a path through the body defining a path angle relative to the longitudinal direction of between 20°-80°.

The path angle may be between 25°-75°, for example between 30°-70°, between 35°-70°, between 40°-70°, for example between 50°-70°.

Each channel may have an average path diameter of 25-250pm, for example 50- 250 pm, 50-150 pm, 70-150pm, or approximately 120 pm. The average diameter may be an average cross-sectional diameter along the length of the channel.

Each channel may have a midpoint diameter of 25-250 pm, for example 25-100 pm, 25-80 pm, or 25-50 pm. The midpoint diameter may be the diameter half-way along the length of the channel.

Each channel may have an inlet diameter of 25-250 pm, for example 50-250 pm, 50-150 pm, 70-150 pm, or approximately 120 pm. The inlet diameter may be the diameter of the channel at the inlet side of the respective wall.

Each channel may have an outlet diameter of 25-250 pm, for example 50-250 pm, 50-150 pm, 70-150 pm, or approximately 120 pm. The outlet diameter may be the diameter of the channel at the outlet side of the respective wall.

It may be that each porous region has a cross-sectional area of between 10,000- 250,000pm ² , for example 15, 000-250, 000pm ² , 15, 000-150, 000pm ² , 20,000- 150,000pm ² , 50, 000-150, 000pm ² , or approximately 100,000 pm ² . Each cross-sectional area may be determined as a volume of the channel divided by an extent of the channel along the thickness direction.

The body of the or each porous wall may comprise or be defined by a porous medium. For example, the porous medium may define the inlet and outlet sides of the body over a continuous longitudinal extent of the body over which the porous wall is porous for fluid flow between the inlet chamber and a respective outlet chamber. The flow arrangement according to the seventh aspect may have any combination of the features as described above with respect to the first aspect and/or with respect to the second to sixth aspects, except such features as are mutually exclusive.

The controller(s) described herein may comprise a processor. The controller and/or the processor may comprise any suitable circuity to cause performance of the methods described herein and as illustrated in the drawings. The controller or processor may comprise: at least one application specific integrated circuit (ASIC); and/or at least one field programmable gate array (FPGA); and/or single or multi-processor architectures; and/or sequential (Von Neumann)/parallel architectures; and/or at least one programmable logic controllers (PLCs); and/or at least one microprocessor; and/or at least one microcontroller; and/or a central processing unit (CPU), to perform the methods and or stated functions for which the controller or processor is configured.

The controller may comprise or the processor may comprise or be in communication with one or more memories that store that data described herein, and/or that store machine readable instructions (e.g. software) for performing the processes and functions described herein (e.g. determinations of parameters and execution of control routines).

The memory may be any suitable non-transitory computer readable storage medium, data storage device or devices, and may comprise a hard disk and/or solid state memory (such as flash memory). In some examples, the computer readable instructions may be transferred to the memory via a wireless signal or via a wired signal. The memory may be permanent non-removable memory, or may be removable memory (such as a universal serial bus (USB) flash drive). The memory may store a computer program comprising computer readable instructions that, when read by a processor or controller, causes performance of the methods described herein, and/or as illustrated in the Figures. The computer program may be software or firmware, or be a combination of software and firmware.

Except where mutually exclusive, a feature described in relation to any one of the above aspects may be applied mutatis mutandis to any other aspect. Furthermore, except where mutually exclusive, any feature described herein may be applied to any aspect and/or combined with any other feature described herein. Brief Description of Figures

The invention is described, by way of example only, with reference to the accompanying drawings in which:

Figure 1a shows a cross-sectional view of an example flow arrangement for an electrolyser;

Figures 1 b-1 d show cross-sectional views of example configurations of a porous wall for the flow arrangement;

Figures 2a-2c show cross-sectional views of further example configurations of a porous wall, comprising a passive region;

Figure 3a is a flow diagram of a method of manufacturing a porous wall;

Figure 3b schematically illustrates applicators for an electrocatalyst composition;

Figures 4a-4d show SEM (scanning electron microscopy) images of a side of the porous wall at various levels of magnification

Figure 5 is a flow diagram of a method of manufacturing a porous wall;

Figure 6 is an image of a cross section of an example porous wall as manufactured by the method of Figure 5;

Figure 7 schematically shows a geometry of an axisymmetric flow simulation model;

Figures 8a-8c show contour plots of fluid reaction product and electrolyte fluid concentration in a flow simulation;

Figures 8d-8e show mass flow and species crossover results for flow simulations corresponding to Figures 8a-8c, at selected flow rates;

Figures 9a-f show mass flow and species crossover results from selected parametric studies;

Figures 10a-10d are matrices of flow simulation results from a parametric study of independent variables of porosity and pore angle;

Figure 11 schematically shows example an electrolysis installation comprising an electrolyser according to Figure 1a;

Figure 12 is a flow diagram of a method of controlling an electrolyser;

Figure 13 is an optical micrograph (at 377.5x magnification) of a surface of a porous electrode formed by laser drilling channels into the wall of an Inconel® alloy 625 tube coated with alumina;

Figure 14 is an image of a cross-section of the porous electrode shown in Figure 12 obtained by X-ray Computed Tomography (XCT);

Figure 15 (a) is a magnified view of a portion of the image shown in Figure 13; Figure 15 (b) is an optical micrograph (at 377.5x magnification) of a surface of a porous electrode formed by laser drilling channels into the wall of an Inconel® alloy 625 tube coated with alumina;

Figure 16 (a) is a representative gas chromatogram of cathode gas obtained from an electrolyser according to the invention featuring three peaks relating to (going from left to right) hydrogen, oxygen and nitrogen content;

Figure 16 (b) is a representative gas chromatogram of anode gas obtained from an electrolyser according to the invention featuring three peaks relating to (going from left to right) hydrogen, oxygen and nitrogen content;

Figure 17 (a) is a representative gas chromatogram of oxygen calibration gas;

Figure 17 (b) is a representative gas chromatogram of hydrogen and nitrogen calibration gas;

Figure 18 (a) is a plot of the cumulative intrusion versus pore size diameter obtained by mercury porosimetry for three sample porous electrodes;

Figure 18 (b) is a plot of the cumulative volume versus pore size diameter obtained by mercury porosimetry for three sample porous electrodes;

Figure 18 (c) is a plot of the log differential intrusion versus pore size diameter obtained by mercury porosimetry for three sample porous electrodes;

Figure 18 (d) is a plot of the cumulative intrusion versus pore size diameter obtained by mercury porosimetry for five sample porous electrodes;

Figure 18 (e) is a plot of the cumulative volume versus pore size diameter obtained by mercury porosimetry for five sample porous electrodes;

Figure 18 (f) is a plot of the log differential intrusion versus pore size diameter obtained by mercury porosimetry for five sample porous electrodes; and

Figure 19 shows three SEM images (a), (b) and (c) of different regions of sintered electrocatalyst material obtained at xIOOO magnification.

Detailed Description

Figure 1a schematically shows an example flow arrangement 100 for an electrolyser having first and second porous walls. Figures 1 b to 1d show example configurations of the porous walls.

The flow arrangement 100 comprises an inlet chamber 102 which in this example is an annular chamber 102 having a central longitudinal direction A, and is elongate along the longitudinal direction A. The inlet chamber 102 is configured to receive an annular inlet flow at an annular inlet 101. In this example, various structures are axisymmetric with respect to the longitudinal direction A, which may be referred to as a longitudinal direction A of the flow arrangement.

The flow arrangement 100 further comprises first and second outlet chambers 130, 140 which are separated from the inlet chamber by the first and second porous walls 110, 120 respectively. In this example, the first outlet chamber 130 is an inner central chamber radially within and surrounded by the inlet chamber 102, and the second outlet chamber 140 is an outer annular chamber radially outside of and surrounding the inlet chamber 102. Each of the first and second outlet chambers are concentric with and elongate along the longitudinal direction A.

The flow arrangement is generally elongate and has a proximal end corresponding to the inlet 101 which is to receive a fluid flow, and a distal end corresponding to outlets for discharging flow (to be described). Various components of the flow arrangement (and of related apparatus such as an electrolyser) may be described with reference to their proximal and distal ends according to the same frame of reference.

The inlet chamber 102 is open at its proximal end at the inlet 101 , and is closed at its distal end 104. Accordingly, the flow arrangement 10 is configured so that the only inlet to the inlet chamber 102 is the inlet 101 , and the only routes for flow out of the inlet chamber 102 is via one or both of the porous walls 110, 120.

The first outlet chamber 130 is closed at a proximal end and defines a first outlet 132 at or towards its distal end. The first outlet chamber 130 may extend beyond a longitudinal extent of the first porous wall 110 (or beyond a longitudinal extent of the first porous wall 110 which is configured to provide flow to the first outlet chamber), as schematically shown with extension lines in Figure 1. The first outlet chamber 130 is configured to only receive flow via the first porous wall 110, and to only discharge flow through the first outlet 132.

While the first outlet chamber 130 is shown as cylindrical in this example, it may have any other suitable configuration, for example it may be annular or conical, or non- axisymmetric as further described below.

The second outlet chamber 140 is closed at a proximal end and defines a second outlet 142 at or towards its distal end. The second outlet chamber 140 may extend beyond a longitudinal extent of the second porous wall 120 (or beyond a longitudinal extent of the second porous wall which is configured to provide flow to the second outlet chamber), as schematically shown with extension lines in Figure 1.

While the second outlet chamber 140 is shown as annular in this example, it may have any other suitable configuration, for example having a non-cylindrical radially-outer wall, or a non-axisymmetric configuration as described below. The example flow arrangement 10 further comprises an inlet flow distributor 150 configured to provide an annular flow of fluid to the annular inlet 101 of the inlet chamber 102. In this example, the flow distributor 150 has a diverging conical profile configured to direct the fluid around a flow diverter 152. The flow distributor has a port 154 for coupling to a source of fluid. In variant examples, the flow distributor 150 may take any suitable shape or configuration, for example it may have cylindrical outer walls and there may be a conical flow diverter within the flow distributor to direct flow towards the annular inlet 101.

The first and second porous walls 110, 120 are each elongate along a respective longitudinal direction, which in this example is the longitudinal direction A of the flow arrangement 100. The first and second porous walls 110, 120 each have a thickness direction from an inlet side adjacent to the inlet chamber 102 and an outlet side adjacent to the respective outlet chamber 130, 140.

An example of flow through the flow arrangement will now be described. A fluid (for example an electrolyte fluid) is received from a fluid source (e.g. a source of an electrolyte fluid) at the port 154 and flows through the flow distributor 150 to the inlet 101 to the inlet chamber 102. The inlet chamber 102 has no flow outlets other than through the porous walls 110, 120. In use, branch flows from the inlet chamber 102 to the respective outlet chambers 130, 140 become established, each branch flow passing through a respective porous wall 110, 120. Each outlet chamber 130, 140 only has a single inlet in the form of the respective porous wall 110, 120, and has an outlet 132, 142 (e.g. a single outlet). Accordingly, in each outlet chamber flow passes from the respective porous wall 110, 120 along the outlet chamber in a generally longitudinal direction, and is discharged at the outlet.

Further features regarding mechanisms for inhibiting reverse flows through the porous walls (i.e. from an outlet chamber 130, 140 to the inlet chamber 102) are discussed below. An example use of the flow arrangement in an electrolyser is discussed below, following discussion relating to an electrolysis reaction.

As shown in Figure 1b, in a first example, the first and second porous walls 110, 120 each have an isotropic porous configuration, being formed of an isotropic porous medium having an porosity to permit fluid to flow therethrough, from the inlet side to the outlet side. The expression “porosity” as used herein refers to an open porosity of a component (e.g. the volume fraction that is open for fluid flow therethrough). A suitable test or procedure for determining a porosity is described elsewhere herein. By way of example, Figure 1 b shows the example of flow through the first porous wall 110 from the inlet chamber 102 to the first outlet chamber 130, but is equally applicable and representative of flow through the second porous wall 120 from the inlet chamber 102 to the second outlet chamber 140.

As used herein, the expression “isotropic porous medium” relates to a porous medium in which the porosity does not significantly vary by an order of magnitude according to the direction of measurement. The porosity may locally vary, for example it may be defined by a structure of fused granular or particulate elements having a distribution of characteristic dimensions and defining open spaces between them, but without a directional configuration as to their arrangement which significantly biases flow in a particular direction (e.g. by an order of magnitude) within the medium.

In isolation (e.g. when not provided in a flow arrangement which may bias flow in a particular direction), the isotropic porous wall is configured to permit flow in any direction therethrough. Accordingly, the local direction of flow at any point in the isotropic porous wall is determined by local pressure gradients, and the relative flow resistance offered by different routes through a network of flow paths defined by the isotropic porous wall.

In the example flow arrangement 100, fluid is caused to flow across (i.e. through) each porous wall 110, 120 by virtue of a pressure difference between the fluid in the inlet chamber 102 and the fluid in a respective outlet chamber 130, 140. The pressure difference acts to drive flow through the porous wall 110, 120 and this may generally be along the thickness direction 112, since the resistance offered by any particular path through a porous medium is a function of the length of the path, and the thickness direction 112 provides the shortest distance between the inlet and outlet sides of the porous wall. However, at each point along the isotropic porous wall, a particular local route taken through the network of flow paths may depend on local variations in the porosity, such that there are local routes through the isotropic porous wall which are tortuous, as shown by example routes 113.

As shown in Figure 1c, in a second example, the first and second porous walls 110, 120 each have an anisotropic porous configuration provided by a plurality of inclined channels 114 which extend through the respective wall from the inlet side to the outlet side. For the purposes of illustration, only the first porous wall 110 is shown, but the following description applies equally to both porous walls 110, 120. Each of the channels 114 is elongate along a path which is inclined with respect to both the longitudinal direction A and the thickness direction 112 of the respective porous wall, such that flow along the channels from the inlet chamber 102 to the respective outlet chamber has a longitudinal component. As used herein, the expression “longitudinal component” refers to a component of a path being along the longitudinal direction A (i.e. parallel with it and of the same sign/direction). In the examples described herein, the longitudinal direction extends from the proximal end of the flow arrangement to the distal end, and so a definition that a path along which something is elongate has a longitudinal component requires that the path has a positive longitudinal component in the proximal to distal direction. As shown in Figure 1b, in this example the longitudinal direction A corresponds to a vertically upward direction, and each of the porous walls 112, 122 is inclined so that flow passing therethrough from the inlet chamber 102 to the respective outlet chamber 130, 140 flows along a path 115 having an upward component.

The expression “channel” as used herein relates to an open path for flow which is free of obstruction, delimited by bounding walls of the channel.

As shown in Figure 1d, in a third example, the first and second porous walls 110, 120 each have a discontinuous porous structure, by which each porous wall comprises a body 116 and a plurality of porous regions 118 extending through the body from the inlet side of the body (adjacent to the inlet chamber) to an outlet side of the body (adjacent to the respective outlet chamber). For the purposes of illustration, only the first porous wall 110 is shown, but the following description applies equally to both porous walls 110, 120. Similarly, a flow arrangement may be implemented in which only one of the porous walls has a discontinuous porous structure as described herein.

The body 116 defines the overall profile of the porous wall 110, 120, being elongate along the longitudinal direction A, and having a thickness direction 112 from the inlet side to the outlet side.

The plurality of porous regions 118 extend through the body 116 at discrete locations to permit fluid to flow from the inlet chamber 102 to the respective outlet chamber 130, 140. Each porous region 118 defines a respective network of flow paths through the body (such that there is a plurality of discrete networks of flow paths through the body). In this example, each porous region 118 is elongate along a path 119 through the body having a longitudinal component (which may be referred to herein as an “inclined discontinuous porous structure”). However, it is also envisaged that the porous regions may extend along a path through the body which does not have a longitudinal component (e.g. is orthogonal to the longitudinal direction).

Each porous region 118 is defined by a porous medium (i.e. a medium having open porosity). As in the example of Figure 1c, a local route through the network of flow paths provided by a particular porous region 118 may be tortuous, and may depend on variations in the local porosity and the relative resistance to flow offered by different routes. Example tortuous routes are indicated by arrows 119.

Any of the flow arrangements described above are suitable for implementing in an electrolyser, with each porous wall corresponding to (e.g. providing) a respective electrode of the electrolyser. The present disclosure envisages that two porous walls of the same flow arrangement may have any combination of the three configurations described above with respect to Figures 1b, 1c and 1d. It may be that at least one of the porous walls has a discontinuous porous structure (e.g. in accordance with the example of Figure 1d). It may be that at least one of the porous walls has an anisotropic porous structure defined by a plurality of inclined channels (e.g. in accordance with the example of Figure 1c). It may be that at least one of the porous walls has an isotropic porous wall (e.g. in accordance with the example of Figure 1b). When such a flow arrangement is implemented in an electrolyser, fluid reaction products such as hydrogen or oxygen may be generated at respective electrodes defined by the porous walls.

By separating the inlet chamber 102 and the outlet chambers 130, 140 with respective porous walls 110, 120 as disclosed herein, each porous wall serves to (e.g. is configured to) inhibit a return flow of a respective fluid reaction product through the respective porous wall towards the inlet chamber, without relying on an ion exchange membrane. Such a return flow may also be described as a reverse flow herein. The porous walls are porous to permit fluid to flow therethrough, and also permit ion exchange in either direction to enable respective half-reactions of electrolysis. Accordingly, a flow arrangement as disclosed herein may be implemented in an electrolyser without use of an ion exchange membrane between respective electrodes, such as a polymer-electrolyte membrane, PEM.

By permitting fluid flow (e.g. of an electrolyte fluid) between the inlet chamber 102 and an outlet chamber 130, 140, the respective porous wall 110, 120 places the inlet chamber and the respective outlet chamber in fluid communication with each other, such that the pressures in the respective chambers are linked. As compared with arrangements which prevent fluid communication between such chambers, this may prevent an excessive pressure difference becoming established over any intervening wall/electrode/membrane, which may deform or otherwise damage such a wall/electrode/membrane and the integrity of the associated electrolyser. Such a risk of damage may be more pronounced when operating at high pressure, such as high pressure for operation at supercritical conditions (i.e. temperature and pressure at or above the supercritical point for the respective fluid). Accordingly, the provision of a porous wall to separate each outlet chamber from the inlet chamber may protect or improve the structural integrity of an electrolyser, particularly for high pressure operation.

The provision of a porous wall 110, 120 between the inlet chamber 102 and each outlet chamber 130, 140 provides a flow restriction to flow therethrough, such that fluid flow through the wall is associated with a pressure drop being established over the porous wall. A pressure drop and a flow rate of a fluid through a component are related to each other, and may also be dependent on properties of the component (e.g. its flow factor Kv - which may be considered a constant) and properties of the fluid (e.g. the specific gravity SG). A pressure drop over the porous wall inhibits a return flow of a respective fluid reaction product through the porous wall towards the inlet chamber, because such flow would be against the pressure gradient over the wall.

An additional mechanism to inhibiting return flow may be provided when the flow arrangement comprises one or more porous walls in accordance with the second and third examples (described above with respect to Figures 1c and 1d respectively).

In particular, in such arrangements, there may be a prevailing buoyancy-driven flow towards the or each outlet chamber. The buoyancy-driven flow may be biased to flow upwards owing to the relatively lower density of fluid reaction products. Accordingly, when an inclined channel 114 (second example) or a porous region 117 (third example) is elongate along a path having an upward component, a return flow therethrough may be against the direction of a prevailing buoyancy-driven flow, and therefore inhibited.

By way of example, in an electrolysis reaction involving an aqueous electrolyte fluid, the fluid reaction products (oxygen and hydrogen) have a lower density that water (or any suitable electrolyte fluid), such that buoyancy forces on those fluid reaction products will tend to drive those fluid reaction products upward. When the respective reaction takes place within the inclined channels 114 of a porous wall having inclined channels (the second example of Figure 1c) or within the porous regions 117 of a porous wall having an inclined discontinuous porous structure (the third example of Figure 1 d), or when the reaction products are otherwise provided at such locations, the buoyancy forces acting owing to the lower density of the respective fluid reaction product may drive the flow upward and thereby along the respective inclined channel 114 or porous region 117 towards the respective outlet chamber.

In the third example porous wall as shown in Figure 1d, there are no inclined channels through the porous wall. There are a plurality of discrete porous regions which each extend along a path having a longitudinal component, to permit fluid to flow through the body. By providing the flow arrangement 100 with a porous wall according to the third example of Figure 1d, flow effects relating to (i) flow rate and (ii) reverse flow may depend on respective properties of the porous region. Firstly, a flow rate through the or each porous wall may be at least partly determined by properties of the porous regions 117 other than the orientation of the path along which it extends, such as the porosity, cross- sectional area or diameter, and the length of the porous region. Such properties may be referred to herein as properties of the porous region 117 relating to flow rate.

Secondly, by providing the porous regions 117 so that they are elongate along a path having a longitudinal component (e.g. a vertical component in use) and therefore at an angle relative to the longitudinal direction, the routes through each respective network of flow paths are effectively constrained to extend along a path having a longitudinal component (e.g. within boundaries of the porous region which are oriented at an angle relative to the longitudinal direction). The porous regions 117 may be configured so that a majority and optionally all (each and every) routes through the network of flow paths have a longitudinal component (e.g. irrespective of whether the route has a tortuous pathway through the porous region 117 having local upward and/or downward components, it may have a net positive longitudinal component corresponding to the longitudinal direction of the respective porous wall). Accordingly, without having to specifically configure each flow path of the porous region to provide such a longitudinal component, this can nevertheless be achieved for the majority or optionally all (each and every) route through the porous region. Accordingly, a return flow through the porous region is inhibited as it would be against the direction of the prevailing buoyancy-driven flow. For example, each porous region 117 may be configured so that, with respect to the longitudinal position, an end of the porous region 117 at the outlet side is longitudinally spaced apart an end of the porous region 117 at the inlet side, with no overlapping longitudinal extend of the ends.

The flow effect of inhibiting a reverse-flow owing to buoyancy effects is related to properties of the paths along which the porous regions extend, for example the angle of the path relative to the longitudinal direction.

Accordingly, in the design and configuration of a flow arrangement having one or more porous walls according to the third example of Figure 1 d, the above flow effects may be separately specified and controlled by reference to the respective properties of the porous regions 117. For example, one or more properties of the porous regions 117 relating to flow rate (e.g. a porosity, cross-sectional area or diameter, and/or length of the porous region) may be selected or configured to target a performance property relating to flow rate. Further and separately, properties of the paths along which the porous regions 117 extend (e.g. an angle of inclination of the path) may be selected or configured to target a performance property relating to reverse flow (e.g. controlling or inhibiting a crossover of reaction products between the outlet chambers). It will be appreciated that a length of the porous region may depend on an angle of inclination, and so to an extent there may be a link between flow effects relating to flow rate, and flow effects relating to buoyancy.

As an example, flow effects relating to the flow rate through a porous wall may include a resistance to flow presented by the porous wall. A resistance to flow may otherwise be expressed as its efficiency at allowing fluid flow (e.g. corresponding to the flow factor Kv, whether of the porous wall as a whole, the porous regions as a whole, or an individual porous region). The resistance/efficiency to fluid flow (for example as expressed by a flow factor Kv) relates the flow rate to the pressure drop. In various implementations of an electrolyser, it may be that the pressure drop is fixed and the flow rate is variable, or that the flow rate is fixed (e.g. the total inlet flow rate) and the pressure drop is variable. Another example flow effect relating to the flow rate is a balance of flow rates through the first and second porous walls. For example, when one of the porous walls offers less resistance to flow than the other, flow may be biased through the respective wall. As will be discussed herein, properties relating to the flow rate through a porous wall may therefore be selected to control (e.g. target, or mitigate against) a flow bias. When a flow arrangement as described above is implemented in an electrolyser with the porous walls corresponding to (e.g. providing) electrodes of the electrolyser, a flow bias towards one porous wall or the other may influence diffusion and/or species crossover effects in the electrolyser, as further discussed below.

The porous medium of the porous regions 117 may be formed or provided without control as to a directional orientation of the porous medium, and as such it may that the porous medium itself has a structure or porosity which does not significantly vary by an order of magnitude according to the direction of measurement. Within the boundaries of each porous region 117, the porous medium may be substantially isotropically arranged, or may not be isotropic. In either case, it is considered that the porous medium of each porous region 117 itself is not provided for the purpose of providing a directional bias itself, and instead it is considered that the profile or path of the porous region itself provides anisotropy to the porous wall, for example to provide any buoyancy-related effect for inhibiting a reverse flow. The flow effects and example properties of the porous regions 117 and body 116 will be further described below, in relation to simulation results.

Although the third example porous wall 110 of Figure 1d is shown with the porous regions 117 extending throughout the full extent of the body along the thickness direction, in variant implementations the porous regions 117 may have an extent along the thickness direction which is less than the thickness of the body.

In variants of any of the above examples, a flow arrangement as described above may not be axisymmetric. For example, a flow arrangement may comprise a rectilinear duct partitioned by porous walls into side-by-side inlet and outlet chambers (i.e. with a central inlet chambers and two outer outlet chambers).

Although the example flow arrangement of Figures 1a schematically shows first and second outlets 132, 142 at distal ends of the first and second outlet chambers, it will be appreciated that outlets from the outlet chambers may be provided at any suitable location. Nevertheless, the outlets may generally be longitudinally separated the portion of the outlet chamber which receives flow through the respective porous wall.

An electrolyser may comprise a flow arrangement 100 as described above with reference to Figure 1a, with one or both of the porous walls 110, 120 having a configuration as described above with reference to any of Figures 1b-1d, or a configuration as described below with reference to any of Figures 2a-2c. In such an electrolyser, each of the porous walls 110, 120 may provide a respective electrode for the electrolyser - i.e. an anode or a cathode. A configuration of an electrolyser comprising the example flow arrangement 100 substantially corresponds to the configuration of the flow arrangement as described above, and as such Figure 1a is considered to show a configuration of an electrolyser 100.

While the description above primarily relates to the flow effects in the flow arrangement, the following further description of the electrolyser 100 of Figure 1a is provided in relation to the functionality and configuration of the porous walls 110, 120 as electrodes, with particular reference to any electrocatalytic regions and/or passive regions of the electrodes (to be described with reference to Figures 2a-2c).

It will be appreciated that the electrolyser 100 may be provided with further components in addition to those shown in Figure 1a, for example electrical connections (also known as electrical feed-throughs) to the electrodes provided by the porous walls. Each of the porous walls 110, 120 may have an electrocatalytic region comprising an electrocatalyst for a respective half-reaction of electrolysis. In use, respective fluid reaction products of the electrolysis reaction are generated at the electrocatalytic regions. For example, in electrolysis of an aqueous electrolyte fluid to generate hydrogen and oxygen, hydrogen is generated at an electrocatalytic region of the cathode, and oxygen is generated at an electrocatalytic region of the anode. A wide range of materials are suitable as electrocatalysts, and a particular choice of electrocatalyst may depend on a selected electrolyte fluid. Example electrocatalysts and electrolyte fluids are discussed elsewhere herein.

The following disclosure relates to example configurations of the porous walls 110, 120 as described above to provide respective electrocatalytic regions and/or passive regions. The following disclosure refers to Figures 1 b-1 d and Figures 2a-2c, which each show a first porous wall 110. However, it will be appreciated that the following disclosure is equally applicable to the second porous wall 120.

When a porous wall 110 (whether as anode or cathode) has an isotropic porous configuration according to the first example of Figure 1 b, an electrocatalytic region may be defined by the porous medium, and may effectively correspond to the entirety of the porous wall except for any passivating coatings (to be described below). For example, the porous medium may have a material composition comprising (e.g. consisting essentially of, consisting of, or which is) a suitable electrocatalyst for the respective halfreaction of electrolysis. The porous medium may be formed from a material comprising the electrocatalyst. Otherwise, the porous medium may be defined by a matrix having a coating, the matrix having a first material composition and defining a network of flow paths through the porous wall, and the coating being provided on the matrix and having a second composition comprising (e.g. consisting essentially of, consisting of, or which is) the electrocatalyst for the respective half-reaction of electrolysis. The coating may be provided over a surface of the matrix which defines a network of flow paths through the porous medium (e.g. including internal surfaces within the porous wall), and may be provided at a coating thickness which retains a porosity through the porous wall to permit fluid flow therethrough (with “porosity” referring to an open porosity as discussed above).

Figure 2a shows a fourth example configuration of the porous wall which corresponds to the first example of Figure 1 b, but differs in the provision of a passive region 202 on the inlet side of the porous wall adjacent to the inlet chamber 102. As discussed elsewhere herein, a passive region is configured to inhibit a respective halfreaction of electrolysis. The passive region may inhibit the respective half-reaction of electrolysis by virtue of being less electrocatalytically active than the electrocatalytic region for the respective half-reaction of electrolysis reaction, as described elsewhere herein.

The provision of a passive region (which may be formed by a passivating coating) defining the inlet side of the porous wall has the effect that the respective half-reaction of electrolysis is inhibited from occurring within the inlet chamber 102. Accordingly, the respective-half reaction of electrolysis may primarily occur when electrolyte fluid has passed through the passive region to reach the electrocatalytic region, thereby generating a respective fluid reaction product downstream of the passive region, with respect to a direction of flow through the respective porous wall. As discussed above, the porous walls disclosed herein have the effect of inhibiting a reverse flow of fluid reaction products, and as such the fluid reaction products are inhibited from flowing upstream to return to the inlet chamber 102.

When a porous wall 110 (whether as anode or cathode) has an anisotropic porous configuration according to the second example of Figure 1c, an electrocatalytic region may be defined by the material of the porous wall itself, or (for example) by an electrocatalytic coating on at least the internal walls of the channels 114. For example, the porous wall 110 may be formed by providing a body having a first material composition, and forming the channels 114 to extend through the body. When the first material composition comprises a suitable electrocatalyst for the respective half-reaction of electrolysis, the body defines an electrocatalytic region of the porous wall.

Alternatively, the porous wall 110 may be formed by providing a body, forming the channels 114 to extend through the body, and coating surfaces of the body including at least the internal walls of the channels 114 with a coating comprising an electrocatalyst for the respective half-reaction of electrolysis, thereby defining a respective electrocatalytic region of the porous wall. The body may define a passive region of the porous wall as discussed herein, for example the body may comprise (e.g. consist of) a material which is not electrocatalytically active for the respective half-reaction of electrolysis, or is less electrocatalytically active than the electrocatalytic region, such as a body of stainless steel 316 or titanium (e.g. to be coated with an electrocatalytic coating for example comprising nickel or platinum).

Figure 2b shows a fifth example configuration of the porous wall which corresponds to the second example of Figure 1b, but differs in the provision of a passive region 202 on the inlet side of the porous wall adjacent to the inlet chamber 102. For example, the passive region may be provided as a passivating coating on the inlet side of the porous wall. A passivating coating may be provided when the body has the first material composition comprising the electrocatalyst (such that the electrocatalytic region corresponds to the extent of the body), or when the body is provided with a coating comprising the electrocatalyst. The passivating coating may be provided before or after formation of the channels 114. Application before or after formation of the channels may be selected depending on a manner of application. For example, it may be that when using a submersion-based technique (e.g. dip coating), the passivating coating is applied before formation of the channels 114 in order to prevent the passivating coating flowing into the channels. However, for other techniques, such as CVD (chemical vapour deposition), it may be that a depletion effect associated with deposition of the passivating coating is such that there is only a limited penetration of the passivating coating onto walls of pre-formed channels, such that application after forming of the channels may be selected.

When a porous wall 110 (whether as anode or cathode) has a discontinuous porous structure according to the third example of Figure 1 d, an electrocatalytic region may be defined by (i) the porous regions 117 (e.g. only) or (ii) the body 116 and the porous regions 117. In the first alternative, the body may have a first material composition and may define a passive region of the porous wall (e.g. may comprise (e.g. consist essentially, consist of, or be) stainless steel 316 or titanium), whereas the porous regions 117 may have a second material composition including an electrocatalyst for the respective half-reaction of electrolysis (e.g. nickel or platinum, with further suitable electrocatalysts described elsewhere herein), and thereby define an electrocatalytic region of the porous wall. An example method for forming the porous regions 117 with a different material composition is described below with reference to Figure 3a.

In the second alternative, the body may have a first material composition which comprises an electrocatalyst for the respective half-reaction of electrolysis (e.g. nickel), and the porous regions may be integrally formed with the body and have the same first material composition, or may have a second material composition also including an electrocatalyst for the respective half-reaction of electrolysis, as discussed above. When the porous region is integrally formed with the body and has the same first material composition, each porous region 117 interfaces with the body at a respective boundary which surrounds the porous region 117 and is defined by a change in porosity between the body and the porous region. An example method for integrally forming the body 116 and porous regions 117 is described below with reference to Figure 5. Figure 2c shows a sixth example configuration of the porous wall which corresponds to the third example of Figure 1c, but differs in the provision of a passive region 202 on the inlet side of the porous wall adjacent to the inlet chamber 102. The passive region 202 may be provided as a passivating coating as described elsewhere herein, and may be applied before or after formation of the porous regions as described above. As noted above, example methods for forming a porous wall having a discontinuous porous structure are described below with reference to Figures 3a and 5, and such methods describe formation of the passive region (which may comprise a passivating coating).

To further illustrate the flow paths and locations of electrolysis reactions in an electrolyser 100 having the flow arrangement of Figure 1a and one or more porous walls according to any of the first to sixth examples of Figures 1 b-2c, an example of use will now be described. The example of use will be described with reference to electrolysis with an electrolyte fluid which is a water-based (aqueous) electrolyte fluid (e.g. electrolyte solution), at supercritical conditions for flow through the respective electrodes, for example at pressure of 22.5 MPa and an inlet temperature of 375°C.

At supercritical conditions, a fluid behaves such that it does not have distinct liquid and gas phases. Supercritical electrolyte fluids may be advantageous for use in electrolysis, as compared with subcritical electrolyte fluids. In particular, supercritical fluids are completely miscible with each other, such that mixtures of fluids form a single phase with no phase boundary or surface tension between them. Therefore, when the reaction products are supercritical rather than gaseous as they are generated (e.g. by having a lower critical temperature and pressure than the electrolyte fluid, or the temperature and pressure otherwise being higher than the respective critical point), those reaction products are completely miscible with the electrolyte fluid. Accordingly, bubbles of reaction products do not accumulate on the surfaces of the electrodes. Such accumulation may otherwise inhibit the reaction by preventing local interaction between the electrolyte fluid and the electrode.

Further, it is known that a conductivity of an electrolyte fluid tends to be higher at elevated temperature and pressure (see for example the paper "High Pressure Electrolyte Conductivity of the Homogeneous, Fluid Water-Sodium Hydroxide System to 400°C and 3000bar", A. Eberz and E.U. Franck, Ber. Bunsenges. Phys. Chem. 99, 1091- 1103 (1995) No. 9, in particular Table 2). Without wishing to be bound by theory, it is thought that a dissociation constant and conductivity of water increases as temperature and pressure rises. Further, a conductivity of an electrolyte or an electrolyte fluid (e.g. an electrolyte solution) is a function of pressure and temperature, with increasing conductivity being observed as pressure and temperature rise towards the critical point. Within the supercritical range (i.e. in the range of pressure and temperature conditions in which the temperature is at least the critical temperature for the fluid, and the pressure is at least the critical pressure for the fluid), conductivity may decrease with rising temperature at constant pressure. However, conductivity may increase within the supercritical range when pressure is increased. Accordingly, a concentration of electrolyte within an electrolyte solution may be reduced by operating at relatively high temperatures and pressures. The supercritical range is an example of relatively high pressure and temperature conditions, and it is considered that operation in the supercritical range may provide a relatively higher conductivity of an electrolyte fluid as compared with operation within the subcritical range at temperatures and pressures significantly below the critical temperature and critical pressure. By operating at conditions such that the conductivity of an electrolyte fluid is higher, ohmic losses associated with the electrolyte may be reduced, or otherwise a different electrolyte that is associated with reduced ohmic losses may be used, while still providing a suitable conductivity.

By way of example only, a suitable electrolyte fluid for use with the example electrolyser 100 described above is an aqueous electrolyte solution comprising NaOH as electrolyte (with further suitable electrolyte fluids being disclosed elsewhere herein). As stated above, conductivity of an electrolyte fluid tends to increase with temperature and pressure

To illustrate the above trend of conductivity increasing with pressure and temperature, reference can be made to the Eberz and Franck paper referred to above, according to which at 0.1 MPa and 25 °C an electrolyte composed of NaOH with a concentration of 17 wt% exhibits a conductivity of 0.4 S cm-1 whereas at 30 MPa and 400 °C, the electrolyte exhibits a conductivity of 1.3 S cm-1. It is estimated that, for a 0.5 M NaOH solution the conductivity at 22.5 MPa and 375 °C would be approximately 150-200 mS cm-1 , whereas at 0.1 MPa and 25°C the conductivity may be approximately 100 mS cm-1.

As an example of a suitable electrolyte fluid for use with the example electrolyser 100, the electrolyte fluid may comprise an aqueous solution comprising NaOH at a concentration of 0.5 M NaOH. In this example, the first porous wall 110 is configured as a cathode (for hydrogen generation) and the second porous wall is configured as an anode (for oxygen generation), (by virtue of respective electrical configuration of the electrolyser).

As discussed above with respect to the general flow paths through the flow arrangement 100 of Figure 1 , when the electrolyte fluid is provided to the inlet chamber 102 via the inlet 101 , branch flows through the respective porous walls (electrodes) 110, 120 to the respective outlet chambers 130, 140 become established.

As the flow passes through an electrocatalytic region of each porous wall, a respective half-reaction of electrolysis is conducted. In particular, the electrolyte fluid undergoes a reduction reaction as it passes through the electrocatalytic region of the first porous wall 110 (cathode) to generate a fluid reaction product which in this example is supercritical hydrogen, and the electrolyte fluid undergoes an oxidation reaction at the second porous wall 120 (anode) to generate a fluid reaction product which in this example is supercritical oxygen. Ions are transferred between the anode and cathode, for example by electromotive force. In this particular example, hydroxide ions (OH-) are transferred through the electrolyte fluid to the anode for generation of a fluid reaction product (oxygen) at the anode (along with transferring electrons to the anode).

In this example, the pressure and temperature conditions of the electrolyser are controlled such that both the electrolyte fluid and the respective fluid reaction products are in a supercritical state at the porous walls. As such, the fluid reaction products are considered to be completely miscible with the electrolyte fluid within the porous walls with no interface (e.g. bubble interface) forming therebetween, as discussed above.

When the electrocatalytic region of a respective porous wall does not define the inlet side of a respective porous wall, the respective half-reaction of electrolysis does not tend to take place at the inlet side of the porous wall and so the respective fluid reaction product does not tend to be generated within the inlet chamber.

For an isotropic porous wall, the electrocatalytic region may not define the inlet side of the porous wall when there is a passive region as described above with respect to the fourth example of Figure 2a. For an anisotropic porous wall comprising inclined channels 114 as in the second and fifth examples of Figures 1c and 2b respectively, the electrocatalytic region may not define the inlet side of the porous wall when the body of the porous wall defines a passive region of the porous wall with an electrocatalytic region defined at the walls of the channels 114 (for example by a coating comprising an electrocatalyst). Otherwise, the anisotropic porous wall may have a passive region defining the inlet side as in the fifth example of Figure 2b. Similarly, for a porous wall having a discontinuous porous structure as discussed herein, the electrocatalytic region may not define the inlet side of the porous wall by virtue of the electrocatalytic region being limited to the porous regions 117 and the body 116 defining a passive region of the porous wall (as in the third example of Figure 1 d), and/or by virtue of there being a passive region defining the inlet side as in the sixth example of Figure 2c.

Accordingly, when the porous walls are in accordance with any of the second to sixth examples of Figures 1c-2c, the respective half-reactions of electrolysis do not tend to occur in the inlet chamber, but instead occur as the flow passes through the porous wall itself, and in particular through the electrocatalytic region. The location of the respective half-reaction of electrolysis is therefore downstream of the inlet chamber, with respect to the direction of the branch flows of fluid from the inlet chamber to the outlet chambers as described above.

For each of the example configurations of the porous walls described herein, a reverse flow of fluid reaction products generated within the porous wall is inhibited by virtue of being against a pressure gradient acting through the wall. This effect may be stronger with increasing pressure difference acting over the respective porous wall. The fluid reaction products are effectively entrained in a pressure-driven flow through the respective porous wall, which limits or mitigates migration or diffusion of the respective fluid reaction product in the reverse direction.

When the porous wall has an anisotropic configuration according to the second, third, fifth and sixth examples (i.e. with inclined channels 114 or with an inclined discontinuous porous structure having porous regions 117), the flow paths through the porous wall (e.g. a majority or all of them) may be effectively constrained to have a longitudinal component, which in this example corresponds to a vertically upward component. Accordingly, buoyancy forces are considered to influence the flow as discussed above. The fluid reaction products each have a lower density than the electrolyte fluid, and therefore tend to establish a buoyancy driven flow through the respective porous wall and outlet chamber by which the flow preferentially flows upward. A reverse flow of the fluid reaction products is inhibited by virtue of the buoyancy effect, because flow along the reverse direction toward the inlet chamber would require the fluid reaction product to move against the prevailing buoyancy-driven flow, to flow downwards towards the inlet chamber.

In use at the supercritical conditions described above (22.5 MPa and 375 °C), the thermoneutral voltage for electrolysis is approximately 1 ,3V, which corresponds to 35.62 kWH per kg of hydrogen reaction product that is generated. This is 110% of the higher heating value of hydrogen (39.4 kWh per kg), and 93.5% of the lower heating value of hydrogen (33.3 kWh), and so is representative of efficient electrolysis. To maintain the electrolysis reaction at these conditions, the ohmic losses of the electrolyser should be no more than 0.120 V. In this example, the ohmic losses of the electrolyser (associated with ion exchange through the example electrolyte fluid of 0.5 M NaOH, having a predicted conductivity of about 150 mS cm-1) is approximately 0.11 V when the electrode spacing is 1mm, whereas ohmic losses associated with bubble formation at the electrodes is eliminated owing to operation at supercritical conditions. The electrode spacing may be less than 1mm, for example 0.5mm or between 0.5mm and 1mm. Suitable electrocatalysts may be determined by the skilled person and are discussed elsewhere herein, but merely as an example suitable electrocatalysts may comprise nickel or a nickel alloy at the anode (which may be the first electrode), and nickel, a nickel alloy or platinum for the cathode (which may be the second electrode).

The ohmic losses are a function of the electrolyte conductivity and the electrode spacing. For example, if the electrodes are spaced apart by 3mm then it may be that a higher electrolyte conductivity is required, such as 2000 mS cm-1.

While example supercritical conditions have been discussed above, the conditions may be varied. For example, supercritical operation of the electrolyser may correspond to operation so that the pressure of the electrolyte fluid is between 22 MPa and 27 MPa at the porous walls for an aqueous electrolyte fluid, and so that the temperature of the electrolyte fluid is at least 374°C at the porous walls for an aqueous electrolyte fluid, for example between 374-550°C or 374-400°C.

It may generally be desirable to operate to achieve lower temperatures and pressures towards the critical point at the porous walls, which may minimise energy resources to pressurise and to heat the electrolyte fluid. Ohmic losses may be lower at higher temperatures and pressures, but it is thought that efficiency advantages to be gained at elevated temperature and pressure conditions may be offset by the additional energy required to heat and/or pressurise the electrolyte fluid, unless such additional energy is available as surplus energy (e.g. as excess heat from a heating source that would otherwise be discharged without thermal recovery).

Although the above discussion focusses on examples in which each porous wall 110, 120 provides an electrocatalytic region to react with an electrolyte fluid as it passes through the respective porous wall for a respective half-reaction of electrolysis, the disclosure envisages implementations of the flow arrangement 100 in an electrolyser in which an electrode and/or an electrocatalytic region of a porous wall is only provided in an outlet chamber associated with a porous wall. For example, an electrode may be disposed in an outlet chamber and separated from the respective porous wall, or an electrocatalytic region may be disposed only on the outlet side of a respective porous wall to form an electrode. In such arrangements, the electrolyser is configured so that the respective half-reaction of electrolysis takes place in the outlet chamber (e.g. only). Such arrangements may be particularly suitable for use at non-supercritical conditions at which the fluid reaction products of the respective electrolysis reaction are non- miscible (e.g. immiscible) with the electrolyte fluid at the electrocatalytic region and/or as they pass through the respective porous wall. When subcritical conditions prevail at the electrode, it may be that, if a respective half-reaction of electrolysis were to take place on an inlet side or within a flow path through a porous wall, a bubble may form having a bubble interface between the respective fluid reaction product and the electrolyte fluid. The bubble may lead to disadvantageous flow effects, for example the respective bubble may block a portion of a surface of an electrocatalytic region of the respective electrode (e.g. porous wall), and thereby inhibit the continuing reaction. Further, the bubble may block a flow path through the porous wall, thereby inhibiting both the continuing reaction and flow through the porous wall.

However, when an electrode or an electrocatalytic region of a respective porous wall is provided in the outlet chamber as noted above, such effects relating to interaction between the bubble and the porous wall may be avoided. In such implementations, a separation between the electrodes or electrocatalytic regions of the respective electrodes may be relatively large, as compared to implementations in which the or each porous wall provides a respective electrocatalytic region within the thickness of the porous wall (e.g. defining a flow path through the porous wall). This may increase ohmic losses.

By inhibiting reverse flows of the fluid reaction products, for example by any of the mechanisms described above, the inventors have found that the electrodes (e.g. porous walls) can be arranged with a relatively low separation gap between them (and without use of a membrane), which reduces ohmic losses in the electrolyser.

In particular, by inhibiting reverse flows, a flow through a porous walls as described herein may be substantially one-way, even if the area of the porous wall is relatively large for a given amount of flow (i.e. even when the flow rate per unit area through the porous wall is relatively low), as compared with previously-considered arrangements. In previously-considered arrangements, a flow-through electrode may be provided to span an elongate duct such that the area of the electrode is relatively low in relation to the flow rate, or such that the flow rate per unit area through the flow-through electrode is relatively high. By providing a relatively high flow rate through the flow-through electrode in previously-considered arrangements, it is thought that a reverse flow may be prevented primarily owing to the inertia of the flow.

As the area of an electrode is increased relative to such previously-considered arrangements for a given flow rate of electrolyte fluid, such inertial effects may reduce and there may be a risk of local flow reversal that could permit mixing of reaction products. However, by providing a porous wall as disclosed herein, the flow rate per unit area can be maintained relatively low for a given reaction area of the electrodes (or conversely the reaction area can be relatively high for a given flow rate), while relying on the porous wall to prevent flow reversal as described herein.

Accordingly, the inventors have found that an arrangement for an electrolyser can be provided in which a porous wall has a relatively high area for a given flow rate. Consequently the electrodes can have relatively higher flow-through areas for a given flow rate of electrolyte fluid into and/or reaction products generated by the electrolyser, enabling a relatively higher surface area for conducting the electrolysis reaction, which may enable a higher reaction rate for a given flow rate of electrolyte fluid. These advantages apply equally to all example arrangements herein. The expression “flow- through area” as used herein with reference to the porous walls (electrodes) is intended to refer to the area of a substantially continuous surface extending over a respective boundary of the porous wall of the electrode (e.g. the inlet side of the porous wall), rather than to the total surface area of an electrolyte interface defined by a porous medium (e.g. a surface defining a tortuous pathway through a porous medium within an electrocatalytic region of the porous wall).

Consequently the separation gap between the electrodes can be made relatively slender, as the size of the electrodes can be relatively large compared to a cross- sectional area and/or separation gap defined by the inlet chamber. For example, the first and second porous walls may oppose each other for ion exchange along a co-extensive longitudinal extent over which there is an average shortest distance of separation between opposing ion exchange surfaces of the electrodes. A ratio of the co-extensive longitudinal extent (i.e. a distance along the longitudinal axis A over which the electrodes oppose one another for ion exchange) to the average shortest distance of separation may be at least 5, for example at least 20 or at least 50 (such as 5-200, 20-200, 50-150).

In the particular example of the electrolyser 100 of Figure 1a, the co-extensive elongate extent of the electrodes is 100 mm, whereas the average shortest distance of separation is approximately 0.7mm, resulting in a ratio of approximately 140. Figure 3a is a flow diagram of a method of manufacturing a porous wall for an electrolyser having a discontinuous porous structure, for example a porous wall according to the third example of Figure 1 d, or according to the sixth example of Figure 2c. The porous wall comprises a body and a plurality of porous regions. The method will be described by way of example with reference to the sixth example porous wall of Figure 2c (using reference numerals associated with Figure 2c).

In block 302, a body 116 for the porous wall is provided. In this example, the body 116 has a material composition corresponding to an electrocatalytic region of the porous wall, but in variant examples the body may have a material composition corresponding to passive region of the porous wall.

The body may have a shape and size corresponding to shape of the porous wall to be manufactured. It may be machined to conform to the shape and size. The dimensions of the porous wall may correspond to the example dimensions provided for (either of) the inner and outer electrode as specified in Table 2 below.

In this example, the body comprises (e.g. consists of) nickel or a nickel alloy. In this particular example, the body consists of a nickel-chromium alloy (in particular an Inconel® alloy, Inconel® 600).

In a variant example in which the body has a material composition corresponding to a passive region of the porous wall, the material composition of the body may be such that the body is less electrocatalytically active than the porous regions according to any definition herein, and as such the body may form a passive region of the porous wall once manufactured. For example, the body may have a material comprising (e.g. essentially consisting of, consisting of, or which is) stainless steel 316 or titanium.

In any case, he body has a material composition for conducting electrical current, such that the body is configured to conduct a current between an electrocatalytic region of the porous wall (as will be described below) and an electrical connection of an electrolyser, to thereby serve as an electrode of the electrolyser (e.g. an anode or cathode).

Optionally in block 304, a passivating coating (e.g. a dielectric coating) is applied to a side of the body to define a passive region 202 of the porous wall. In this example, the body material composition is electrocatalytic and so a passivating coating is applied. In a variant example in which the body has a material composition corresponding to a passive region of the porous wall, the porous region corresponding to the body and a passive region corresponding to the passivating coating may be considered as first and second passive regions, each less electrocatalytically active than the porous regions according to any definition herein.

The passivating coating may be applied by any suitable process, for example a sputtering process such as magnetron sputtering, a chemical vapour deposition (CVD) process, or a dipping process. A side of the body may be masked to prevent coating with the passivating coating (e.g. the outlet side of the body). Optionally, the passivating coating may be applied to both sides of the body (i.e. both the inlet and outlet sides, without masking of either side).

In block 306, open regions are formed at discrete locations extending through the body to provide the body with an anisotropic porous structure. The open regions correspond to the porous regions 117 to be formed in the body 116. The open regions are therefore formed so that they are elongate along a path through the body having a longitudinal component, as described above with respect to the porous regions 117. The open regions are formed by removing material from the body, for example using a laser drilling process.

The open regions may be formed with an average cross-sectional diameter along their length of 25-250 pm, for example 50-250 pm, 50-150 pm, 70-150pm, or approximately 120 pm, and may have a generally circular cross section normal to the paths along which they are each respectively elongate.

The open regions may be formed to have a midpoint diameter of 25-250 pm, for example 25-100 pm, 25-80 pm, or 25-50 pm.

The open regions may be formed to have an inlet diameter of 25-250 pm, for example 50-250 pm, 50-150 pm, 70-150 pm, or approximately 120 pm.

The open regions may be formed to have an outlet diameter of 25-250 pm, for example 50-250 pm, 50-150 pm, 70-150 pm, or approximately 120 pm.

The open regions may be formed to have an average cross-sectional area of 10, 000-250, 000pm ² , for example 15, 000-250, 000pm ² , 15, 000-150, 000pm ² , 20,000- 150,000pm ² , 50, 000-150, 000pm ² , or approximately 100,000 pm ² . The average cross- sectional area is determined as the volume of the open region divided by the extent of the open region along the thickness direction of the body.

An example of suitable laser drilling equipment is a "Lasertec 50 Powerdrill" available from DMG Mori Seiki Co., Ltd of Japan, or a high-power ultra-short pulse laser available from Amphos GmbH of Germany (e.g. the "Amphos 2302" or "Amphos 3000" series laser). A further example of suitable laser drilling equipment is the series of machines available from IPG Photonics of the USA referred to as QCW Fiber Lasers (Quasi Continuous Wave Laser) - operable in both pulsed and continuous wave modes, for example to generate pulses from 4-joule to 200-joule at various pulse frequencies e.g. from 100Hz to 5kHz).

A suitable laser drilling process may be ultrashort-pulse laser drilling (also known as laser micromachining), with ultra-short laser pulses being in the order of pico or femtoseconds, such as 10ps, or from 0.1 ps to 10ps. Such machining processes are discussed in the publication with reference Aizawa, Tatsuhiko & Inohara, Tadahiko; (2019); Pico- and Femtosecond Laser Micromachining for Surface Texturing; 10.5772/intechopen.83741 , also available at: htps://www.intechopen.com/books/micromachining/pico-and-femt osecond-laser- micromachining-for-surface-texturinq

In block 308, the method comprises applying an electrocatalyst composition to the body so that it flows into the open regions.

The electrocatalyst composition is a composition which comprises (e.g. consists essentially of, consists of, or is) an electrocatalyst for the respective half-reaction of electrolysis. For example, the electrocatalyst composition may comprise (e.g. be, consist essentially of, or consist of) a mixture of the electrocatalyst for the respective halfreaction of electrolysis and a liquid. The electrocatalyst composition may comprise (e.g. be, consist essentially of, or consist of) a slurry comprising the electrocatalyst for the respective half-reaction of electrolysis and a liquid. The electrocatalyst composition may comprise (e.g. be, consist essentially of, or consist of) a suspension of the electrocatalyst for the respective half-reaction of electrolysis in the liquid. The electrocatalyst may be provided in the form of a particulate. The suspension may be a colloidal suspension (e.g. of the particulate in the liquid). For example, the electrocatalyst composition may comprise (e.g. be, consist essentially of, or consist of) a sol (i.e. a solid-in-liquid colloidal suspension), a conductive ink (i.e. a suspension of electrically conductive electrocatalyst particles in liquid), or paste (i.e. a solid-in-liquid suspension having a sufficiently high solids contents that the suspension behaves as a solid in response to low applied stresses) comprising the electrocatalyst for the respective half-reaction of electrolysis. The liquid, for example a solvent, may influence a viscosity of the electrocatalyst composition. In addition to the electrocatalyst and the liquid, the electrocatalyst composition may comprise binder (e.g. polymeric binder such as a resin binder) and/or one or more additives. Example electrocatalyst compositions are discussed below and elsewhere herein.

The electrocatalyst composition may be applied to the body by any suitable method. By way of example, the electrocatalyst composition may be applied on to a surface of the body and scraped (e.g. dragged over) over the surface in an application direction parallel with the longitudinal direction of the body. The application direction (i.e. forward or backward along the longitudinal direction) may be selected so as to promote directing the electrocatalyst composition into the open region, which may differ depending on the side of the body to which the electrocatalyst composition is applied. For example, when the open regions are elongate along a path towards a side of the body corresponding to the outlet side of the porous wall, and the path is at an angle of 70° relative to the longitudinal direction of the body, the application direction would be forward along the longitudinal direction if applying the electrocatalyst composition on a side of the body corresponding to the inlet side of the porous wall, and backward along the longitudinal direction if applying the electrocatalyst composition on a side corresponding to the outlet side of the porous wall.

The electrocatalyst composition may be scraped (e.g. dragged over) the surface by an applicator, which may be a flexible applicator having a shape corresponding to a shape of the respective side of the body. When the body is substantially planar, the applicator may have a substantially rectilinear profile terminating at a linear applicator lip. When the body is substantially annular, the applicator may have a corresponding annular profile having a substantially circular lip. It will be appreciated that, for an annular body, the electrocatalyst composition may be applied from a radially outer or a radially inner direction. An applicator for radially outer application may comprise a frustoconical applicator inwardly tapering towards a lip, and configured to slide over the body along the longitudinal direction. An applicator for radially inner application may comprise a frustoconical applicator outwardly tapering towards a lip and configured to slide through the body along the longitudinal direction, in a plunger action.

Application of a vacuum to one side of the body may assist penetration of the electrocatalyst composition into the open regions at the opposing side of the body. For example, when the electrocatalyst composition is applied from a radially outer direction to an annular body, a vacuum may be applied to the interior of the body to draw the electrocatalyst composition into the open regions at the exterior of the body. Alternatively, when the electrocatalyst composition is applied from a radially inner direction to an annular body, a vacuum may be applied to the exterior of the body to draw the electrocatalyst composition into the open regions at the interior of the body.

Excess electrocatalyst composition may be removed by a scraper blade, which may have a similar configuration to the applicator but differ by virtue of being relatively less flexible (i.e. having a higher flexural rigidity), such that it is configured to remove excess electrocatalyst rather than flex to direct it into to the open regions.

Figure 3b shows an example applicator for a cylindrical porous wall 110 having a porous longitudinal extend 111. The applicator comprises two opposing plungers 309, 309’ which are configured to be slidingly and sealingly inserted into opposing ends of the porous wall 110 to compress an electrocatalyst composition therebetween, thereby driving the electrocatalyst composition under pressure to flow into the open regions of the porous wall 110 from the radially inner side (e.g. the outlet side in the case of the first porous wall 110) to the radially out outer side (e.g. the inlet side in the case of the first porous wall 110). The applicator may be used for either porous walls to ensure that the electrocatalyst flows through and occupies the open regions throughout the extent of the body along the thickness direction. One of the applicators 309, 309’ may be removed and the other driven through the longitudinal extent of the porous wall 110 to remove excess electrocatalyst composition from an interior of the porous wall. As shown in Figure 3b, it may be that at least one of the applicators 309, 309’ has a conical end configured to mate (e.g. abut with) the opposing applicator. It may be that an amount of electrocatalyst composition is dosed into the porous wall such that mating between the applicators corresponds to having applied sufficient relative movement to drive the electrocatalyst composition through each of the open regions of the porous wall 110.

Referring back to Figure 3a, at block 310, the method optionally comprises a drying operation in which the electrocatalyst composition as applied to the body is dried to vaporise (e.g. partially) a liquid component of the electrocatalyst composition. Parameters of a drying operation may be suitably adjusted to achieve a suitable consistency of the electrocatalyst composition for heat treatment (e.g. sintering) at block 312. Such parameters may depend on ambient and conditions (e.g. humidity). By way of example, a drying operation may be conducted at a temperature between 80°C - 120°C for a period of 1-4 hours, for example for 2 hours at 120°C. By way of another example, the drying operation may be conducted at a temperature between 100°C and 300°C for a period of 1-4 hours, for example for 2 hours at 200°C. The drying operation may be conducted by placing the body in a drying oven or on a hot plate, for example.

At block 312, a heat treatment (e.g. sintering) operation is conducted in which the body and the applied electrocatalyst composition is subjected to heat, thereby vaporising a liquid component of the electrocatalyst composition and causing a solid (e.g. particulate) component of the electrocatalyst composition to form a porous region of material (e.g. comprising connected (e.g. sintered or partially fused) granular or particulate elements) comprising an electrocatalyst for the respective half-reaction of electrolysis. The porous region thereby occupies the open region, forming the porous region 117 of the body as described above with respect to the third and sixth examples of Figures 1d and 2c respectively.

It has been observed that such heat treatment results in the solid electrocatalyst component occupying and bridging across (i.e. extending diametrically across) the open region to form a porous medium.

It may be that the porous medium has an extent along the thickness direction which is less than an extent of the body. For example, this may result from the electrocatalyst composition penetrating part way through a respective open region during application.

The heat treatment (e.g. sintering) operation may be conducted by placing the body (with applied electrocatalyst composition) in a temperature controlled environment to subject the body to a treatment temperature over treatment time. The temperature controlled environment may be an interior chamber of a heater, such as an oven, or a muffle furnace (e.g. as available from Carbolite Gero, a group company of the Verder Group (Verder International B.V. of Germany)).

The heat treatment (e.g. sintering) operation may be conducted by ramping up a treatment temperature over a ramp phase of the treatment time. For example, the treatment temperature may be ramped to a peak temperature of between 150°C-1000°C. For example, the peak temperature may be between 250°C and 800°C, between 300°C- 600°C, between 300°C-450°C, for example approximately 350°C. In another example, the peak temperature may be between 600-1000°C, for example between 700-1000°C, or between 800-1000°C, or between 900-1000°C, or between 900-950°C, for example approximately 930°C.

The treatment temperature may be ramped up to the peak temperature from a starting temperature, which may be an ambient (e.g. room) temperature, for example 20°C, or a starting temperature of the heater, such as 30°C. The ramp rate may be between 0.5°C-2°C per minute, for example approximately 1 °C per minute. The heat treatment (e.g. sintering) operation may comprise maintaining the treatment temperature at the peak temperature for a dwell phase of the treatment time, for example a dwell phase of between 1-10 hours, for example between 1-8 hours, or 1-6 hours, or 1-4 hours, or approximately 1 hour, or approximately 2 hours, or approximately 4 hours, or approximately 6 hours, or approximately 8 hours.

It will be appreciated that, in some embodiments, sufficient drying of the electrocatalyst composition may take place during the heat treatment (e.g. sintering) operation itself, such that a separate drying operation (i.e. block 310) is not necessary. In such embodiments, the body may be subjected to one unitary heating operation which combines drying and heat treatment of the electrocatalyst composition.

The mechanism for forming the porous region of material during the heat treatment operation may depend on the nature of the electrocatalyst composition, the treatment temperature and the treatment time. It may be that particles of electrocatalyst sinter or fuse with one another during the heat treatment operation, for example, by diffusive processes. It may be that a component (e.g. a polymeric binder) of the electrocatalyst composition melts and/or cures during the heat treatment operation, thereby binding particles of electrocatalyst to one another. It may be that a chemical reaction (e.g. oxidation) takes places during the heat treatment operation, thereby forming a reaction product (e.g. oxide) which binds particles of electrocatalyst to one another.

In trials of the method using an electrocatalyst composition comprising particulate nickel, it was found that peak treatment temperatures in excess of 400°C, when the heat treatment is carried out in an oxidising atmosphere, may result in disadvantageous oxidation of the electrocatalyst to generate excess quantities of nickel oxide. However, such oxidation effects may be avoided by performing the heat treatment operation in a controlled atmosphere substantially free of oxygen (e.g. an inert or reducing atmosphere), for example an atmosphere comprising (e.g. consisting of) argon or nitrogen and/or hydrogen (e.g. a mix of 95 wt% nitrogen and 5 wt% hydrogen). It will be appreciated that the level of oxidation for a given electrocatalyst composition could be controlled by varying the treatment temperature, treatment time and/or the atmospheric composition.

The method may include one or more secondary heat treatment operations which are carried out subsequent to the primary heat treatment (e.g. sintering) operation at block 312. For example, the body may be placed in a temperature controlled environment to subject the body to a secondary treatment temperature over a secondary treatment time. The temperature controlled environment may be an interior chamber of a heater, such as an oven, or a muffle furnace (e.g. as available from Carbolite Gero, a group company of the Verder Group (Verder International B.V. of Germany)).

The secondary heat treatment operation may be conducted by ramping up a secondary treatment temperature over a ramp phase of the secondary treatment time. For example, the secondary treatment temperature may be ramped to a peak temperature of between 500°C-1500°C. For example, the peak temperature may be between 750°C and 1200°C, between 800°C-1000°C, between 850°C-950°C, for example approximately 930°C. The secondary heat treatment operation may be conducted in a controlled atmosphere substantially free of oxygen (e.g. an inert or reducing atmosphere), for example an atmosphere comprising (e.g. consisting of) argon or nitrogen and/or hydrogen (e.g. a mix of 95 wt% nitrogen and 5 wt% hydrogen).

The secondary heat treatment operation may comprise (e.g. be) a stress-relief heat treatment operation intended to relieved weld stresses in the body. Additionally or alternatively, the secondary heat treatment operation may be a secondary sintering operation during which densification of the porous medium takes place, for example by grain boundary diffusion.

In some examples, the heat treatment operation is considered to comprise a first stage and a second stage. The first stage may be carried out to pyrolyze or burn off nonelectrocatalyst components of the electrocatalyst composition such as binder components of the electrocatalyst composition, and may therefore be carried out in an oxidising atmosphere. The second stage may be carried out to sinter electrocatalystcontaining particles to one another, and may therefore be carried out in an inert (or reducing) atmosphere such as an atmosphere comprising (e.g. consisting of) argon or nitrogen and/or hydrogen. The first stage may be carried out with a peak temperature of between 150-500°C, for example between 200-500°C, or between 250-450°C, or between 300-500°C, for example approximately 300°C or approximately 350°C. The second stage may be carried out with a peak temperature of between 500-1000°C, for example between 600-1000°C, or between 700-1000°C, or between 800-1000°C, or between 900-1000°C, or between 900-950°C, for example approximately 930°C. The first stage may last for 2-10 hours, for example 4-8 hours, for example approximately 6 hours. The second stage may last for between 30 and 300 minutes, for example between 30 and 90 minutes, such as approximately 60 minutes.

An example electrocatalyst composition is a mixture of a conductive ink available from Creative Materials Inc (of Massachusetts, USA) under the product name 116-25, with a solvent available from Creative Materials Inc under the product name 112-19.

Conductive ink 116-25 contains from 50 wt. % to 70 wt. % nickel particulate (CAS number 7440-02-0), from 30 wt. % to 50 wt. % non-polar ester solvent, from 5 wt. % to about 10 wt. % polymer resin and less than 2 wt. % carbon black (CAS number 1333- 86-4). Conductive ink 116-25 contains greater than about 84 wt. % nickel following curing for 5 minutes at 125 °C. Conductive ink 116-25 has a specific gravity of about 2.21 (measured relative to water), a viscosity of about 25 Pa s, a boiling point greater than about 196 °C and a flash point greater than about 100 °C.

Solvent 112-19 is the same non-polar ester solvent found in the conductive ink 116-25. It has a boiling point greater than about 196 °C, a flash point of about 100 °C, an upper flammability limit of about 8 vol. % and a lower flammability limit of about 0.9 vol. %, a vapour pressure at 20 °C of about 0.3 hPa, a specific gravity of about 1.092 (measured relative to water) and an auto-ignition temperature of 370 °C.

By way of example, electrocatalyst compositions comprising variable amounts of the solvent (e.g. solvent 112-19) have been tested and found to provide a suitable viscosity for flowing into the open regions, with the remainder of the electrocatalyst compositions consisting of the conductive ink (e.g. 116-25).

Thermal Gravimetric Analysis (TGA) was performed on three such electrocatalyst compositions comprising 5 wt. %, 10 wt. % and 15 wt. % of solvent (112-19), with the remainder consisting of the conductive ink (116-25). The conductive ink and solvent were mixed by dual axial centrifugation (DAC) for 2 minutes at 2000 rpm to form the electrocatalyst compositions. The solids content of the samples ranged from approximately 60 wt. % (5 wt. % solvent) to approximately 40 wt. % (15 wt. % solvent). The viscosity for the respective samples ranged from approximately 5 Pa s (15 wt. % solvent) to approximately 13 Pa s (5 wt. % solvent). The solids content and viscosity values were determined by a rheometry evaluation method conducted at 25 °C using 40 mm parallel plates at a gap of 500 pm, at a shear rate of 1 1/s. Values relating to the solids content and viscosity of the example electrocatalyst compositions are reported in Table 1 below.

Table 1.

As an alternative to solvent 112-19, solvents 102-03 and/or 113-12 (also available from Creative Materials Inc) could be used to dilute the conductive ink.

Solvent 102-03 contains from about 90 wt. % to about 100 wt. % of a proprietary diluent and from about 0 wt. % to about 10 wt. % of 2-butoxyethyl acetate. Solvent 102- 03 has a boiling point greater than about 190 °C, a flash point of about 90 °C (in a closed cup), a specific gravity of about 1.13 (measured relative to water), an auto-ignition temperature of greater than 400 °C, and a viscosity at 25 °C from about 12 Pa s to about 18 Pa s.

Solvent 113-12 contains greater than about 95 wt. % 2-butoxyethyl acetate. It has a boiling point of about 192 °C, a flash point of about 76 °C, an upper flammability limit of about 8.54 vol. % and a lower flammability limit of about 0.88 vol. %, a vapour pressure at 20 °C of about 0.29 mmHg, and a specific gravity of about 0.94 (measured relative to water).

Example Heat Treatment Results

Heat treatment (e.g. sintering) performance was tested by manufacturing a representative porous wall according to the manufacturing method described above with reference to Figure 3a.

In block 302, a body for the porous wall was provided, comprising a square planar portion of a nickel-chromium alloy (in particular an Inconel® alloy, Inconel® 600) having a longitudinal dimension of 50mm, a lateral dimension of 50mm, and a thickness dimension of 1 mm.

In block 306, a pattern of open regions were formed in the body by a laser drilling process as described above. The open regions were formed through the full thickness of the body using a millisecond laser operating at 500 Hz. The open regions were formed so that they are elongate along a path which is at an angle of inclination relative to the longitudinal direction of 70°C (and thereby at an angle of 20°C relative to a thickness direction of the planar body). The open regions were drilled in laterally-offset rows to form a regular grid pattern with equal lateral and longitudinal pitch.

In block 308, an electrocatalyst composition corresponding to the example electrocatalyst composition described above (5 wt. % solvent variant) was applied using an applicator as described above.

In block 310, a drying operation was conducted, in which the body (and applied electrocatalyst composition) was maintained at 120°C for 2 hours on a high precision hot plate.

In block 312, a heat treatment operation was conducted to form a porous wall with porous regions extending through the body. The heat treatment operation was conducted using a muffle furnace, and consisted of a ramp phase at 1 °C per minute to raise a treatment temperature from a starting temperature of 30°C to a peak temperature of 350°C, at which it was maintained for a dwell time of 3 hours.

Figures 4a-4d show SEM (scanning electron microscopy) images of a side of the porous wall at various levels of magnification, with scales indicated on the drawings ranging from 100 pm (Figure 4a) down to 10 pm (Figure 4d). Figure 4a shows ends of four porous regions terminating at a side of the body. Figures 4b-4d show the porous medium within the porous region, showing a generally a porous structure formed by partially sintered (e.g. fused) granular or particulate elements (in this example, comprising nickel) defining open spaces between them. As can be seen from Figures 4b-4d, the spacing between granular or particulate elements appears to vary, but without a directional configuration of the various that might significantly bias flow in a particular direction.

Figure 5 shows a flow diagram of a method 500 of manufacturing a porous wall according to the third example of Figure 1d by additive manufacture, and will be described in relation to features of the porous wall of that example (using reference numerals associated with that example). The method is for manufacturing the porous wall in a layer-by-layer (or slice-by-slice) manner, in which a build material (e.g. a powder) is deposited on a bed or platform in successive layers which are progressively selectively heat treated (e.g. sintered or fused) to form slices of the porous wall.

In block 502, model data for additive manufacture of the porous wall is provided to control an additive manufacturing apparatus, for example by selective laser sintering (SLS), in particular Laser-Powder Bed Fusion (L-PBF). The model data corresponds to the example porous wall 110 described above with respect to Figure 1 d, including a body 116 and a plurality of discrete porous regions 117 as described above. For example, the model data may define a geometry of the body 116 and a geometry of the porous regions, and may define a porosity (or a parameter relating to a porosity) which is variable between the body 116 and the porous regions 117. For example, the model data may be defined so that a parameter (e.g. porosity or a parameter relating to porosity) is assigned to discrete regions of the model (e.g. regions assigned an identity for parameter attribution), or such a parameter may be applied to sub-regions or points of the model belonging to a continuous distribution or matrix of such sub-regions or points, which are not particularly assigned to a region for parameter attribution (e.g. a voxel of the model, or a pixel of slices of the model). The parameter may be defined as a porosity, or a related parameter (e.g. an energy density) having a relationship with a formed porosity (i.e. a porosity manifested in the manufactured component).

In block 504, a controller of an additive manufacturing apparatus controls the additive manufacturing apparatus to form the porous wall based on the model data. Additive manufacturing apparatus of suitable types are well understood in the art, and it is considered that a schematic depiction of such an apparatus in the drawings is not required to aid an understanding of an additive manufacturing process. Merely by way of example, the apparatus may be configured to conduct a L-PBF process. An example commercially-available product is an EOS M400-4 additive manufacturing machine available from EOS GmbH of Germany. A further example of a commercially-available product is a Renishaw AM400 or AM500 machine, a pulsed laser machine available from Renishaw pic of the UK. Either machine may be used, for example, with any suitable electrocatalyst as described elsewhere herein, for example a nickel or nickel alloy build material (such as a nickel-chromium alloy, for example Inconel® 600). A suitable particle size distribution for the build material may be 15-45pm.

The model data may be pre-processed to provide layer-by-layer instructions for the additive manufacturing apparatus. The pre-processing may be performed by a controller for the additive manufacturing apparatus or a separate controller. For example, the layer-by-layer instructions may define, for each layer of a build material to be heat treated (e.g. sintered) to form the porous wall, one or more laser control parameters including any or all of (i) a path of a laser over the layer to form the respective slice of the porous wall, (ii) a scanning speed of the laser, (iii) a scan spacing (also known as a hatch distance) between adjacent (side-by-side) segments of the path of the laser, (iv) an energy of the laser and (v) a spot size of the laser. Any or all of the laser control parameters may be controlled to vary an energy density applied to the layer and thereby control an amount of fusing of the build material. In particular, a degree to which particles of build material are fused depends on the energy density applied, and a fusing depth may depend on the energy density (i.e. the depth in and below the active layer which is subject to fusing by action of the laser, which may be greater than the height of a layer of build material as deposited on the bed or platform for the active layer).

The pre-processing may be conducted to vary the or each laser control parameter to compensate for laser penetration and/or thermal migration that may otherwise occur to cause excess fusing in a region intended to be porous. For example, when fusing build material in an active layer to form a solid (e.g. non-porous) portion corresponding to the body 116 of the porous wall which overlies a portion of a preceding layer corresponding to a porous region 117 of the porous wall (which may be partially fused or unfused), the laser may penetrate through the active layer and/or heat generated by the laser at the active layer may transfer to the preceding layer to cause excess fusing. The pre-processing may define the layer-by-layer instructions to compensate for (e.g. mitigate) such excess fusing effects, for example by determining or reducing an energy density for forming a solid portion (corresponding to the body 116) in a layer n, based on a determination that the respective portion in layer n overlies a partially fused or unfused portion (corresponding to the porous region 117) in a layer n-1 (with the notation “n”, “n- 1” etc denoting successive layers). The pre-processing may be conducted to determine whether the energy density for forming the solid portion in the layer n may be reduced (e.g. relative to a typical or standard energy density for forming a solid portion) based on a determination that laser penetration and/or thermal migration from fusing a solid portion in a higher layer n+1 would influence fusing of build material in the layer n to provide a target degree of fusing. For example, the pre-processing may be conducted to graduate an energy density over multiple layers local to a transition between regions having different parameters relating to porosity (e.g. different target porosities corresponding to the body and porous regions respectively).

In addition or as an alternative to any pre-processing based on a layer-by-analysis of laser penetration and/or heat migration, pre-processing may be conducted to determine a buffer region of the body adjacent to a porous region, and to apply a parameter for control of the additive manufacturing process (e.g. a porosity or related parameter as discussed above), to reduce an energy density applied in the buffer region as compared with an energy density with a remainder of the body. Such a parameter may be assigned or attributed to discrete regions identified within the model data or instructions, or to respective sub-regions (e.g. voxels or pixels) as discussed above..

As shown in Figure 5, block 504 comprises a repeating loop of distributing a layer of build material (block 506) and controlling selective fusing of build material based on the model data and/or based on layer-by-layer instructions (block 508).

Optionally in block 510, the method comprises a heat treatment process conducted after the additive manufacturing process, by which the porous wall formed by the additive manufacturing process is heated in a temperature-controlled environment. The heating may be conducted to cause fusing of unfused or partially fused build material in the model, and a temperature and treatment time of the heat treatment process may be determined to cause a desired degree of such fusing, for example based on understood best practices in the technical field and suitable trials. Such a heat treatment process may be alternatively referred to or understood as a curing process in the art.

Optionally in block 512, the method comprises a finishing process. This may include machining to remove excess build material from the porous wall. The finishing process may be conducted to achieve a target size and shape of the porous wall. The finishing process may be conducted to selective remove material, for example on an external surface of the porous wall.

Any suitable build material can be used for the additive manufacturing process. Suitable build materials comprising an electrocatalyst for a respective half-reaction of electrolysis comprise (e.g. consist of) nickel or a nickel-based alloy such as a nickel- based superalloy. A suitable example is a nickel-chromium alloy such as an Inconel® alloy (e.g. Inconel® 600). Such a material may be a suitable electrocatalyst for catalysing a half-reaction of electrolysis at the anode, and at the cathode, for an aqueous electrolyte fluid.

Figure 6 shows an image of a cross-section of a sample porous wall produced by an additive manufacturing process as discussed above. The sample porous wall was cut to expose a cross-section through a plurality of porous regions. The inclined nature of the porous regions can be observed within the body (manifesting substantially no porosity). The porosity of the porous regions can be observed by the distribution of the porous medium in the cross-section.

The image shows bowing of the cut porous wall, which results from pressure applied on a mounting press for the imaging process.

Heat treatment (e.g. sintering) performance was tested in a further example. In this example, a square planar portion of a nickel-chromium alloy (in particular an Inconel® alloy, Inconel® 600) was coated with an electrocatalyst composition corresponding to the example electrocatalyst composition described above (5 wt. % solvent variant). A drying operation was conducted in which the plate (and applied electrocatalyst composition) was maintained at 120°C for 2 hours on a high precision hot plate. A two-stage heat treatment operation was then conducted as described above using a muffle furnace. In the first stage, conducted in an oxidising atmosphere, the plate was heated from room temperature at a rate of 5°C per minute to a peak temperature of 300°C, at which the plate was held for 6 hours. In the second stage, conducted in an argon atmosphere, the temperature was increased to 930°C, at which temperature the plate was held for a dwell time of 15 minutes. Figure 19 shows three subsequent SEM (scanning electron microscopy) images (a)-(c), obtained at 1000x magnification, of different regions of the porous medium formed on sintering of the electrocatalyst composition. The porous, sintered structure is evident.

Simulations

Figure 7 shows an axisymmetric model geometry 700 for a flow simulation model of an electrolyser having the flow arrangement described above with respect to the example flow arrangement 100 of Figure 1a. Reference numerals corresponding to the flow arrangement 100 as described with respect to Figure 1a are also provided for corresponding features shown in Figure 7, including for the annular inlet chamber 102 and inlet 101 , first and second porous walls 110, 120, inner and outer outlet chambers 130, 140.

The flow simulation model has a vertical orientation so that its longitudinal axis A is oriented vertically, and gravity is downward in the orientation of the drawing.

The geometry of an upstream inflow manifold 702 and downstream outflows 730, 740 as implemented in the flow simulation model slightly differ from comparable physical features shown in Figure 1a in order to reduce any flow effects local to inlet and outlet boundary conditions in the model adversely influencing flow through the inlet and outlet chambers 102, 130, 140 of the flow simulation model.

The inflow manifold 702 is upstream of and extends below the annular inlet 101 to the inlet chamber 102 by a length Inflow and is provided to permit development of the inlet flow towards the inlet flow, from an inflow boundary 701 at its upstream end. The inflow manifold 702 includes a conical diverting wall having an angle relative to the longitudinal direction A of 30° to provide flow to the annular inlet 101 of the inlet chamber 102.

The downstream outflows 730, 740 are downstream of and extend longitudinally beyond (e.g. above) the respective longitudinal portions of the outlet chambers 130, 140 which are laterally/radially adjacent to the porous walls 110, 120. The downstream outflows 730, 740 each have a length within the flow simulation model l _ou tfiow and terminate at respective outflow boundaries 732, 742. The outflows are provided to separate the outflow boundaries 732, 742 of the model from the portion of the model including the porous walls, for example to reduce any impact of the particular outflow geometry on that portion of the model.

The flow simulation model conforms with the set of electrolyser dimensions set out in Table 2 below.

Table 2

Simulations of flow in the flow simulation model have been conducted, including simulations with an electrolysis reaction. Simulations of various sets of conditions using the flow simulation model are referred to herein as “simulation(s)” or “simulation case(s)” interchangeably. The simulations are conducted to evaluate flow patterns that relate to diffusion and crossover of species (e.g., electrolyte fluid and respective reaction products).

The simulations were conducted using simulation software provided by COMSOL Multiphysics (RTM) 5.6, developed by COMSOL AB of Sweden.

The flow simulation model is defined to reflect the following assumptions. A hydrogen-oxygen-water mixture with NaOH electrolyte is a fully miscible and supercritical single phase flow at all conditions of interest (e.g., pressure >22MPa, temperature > 380°C). An inlet flow of up to 40 g/minute to the electrolyser remains laminar within the electrolyser at the operating conditions.

The flow simulation model models diffusion of the components of a hydrogen- oxygen-water mixture. Density and viscosity are modelled as functions of the local composition, and pressure, for a reference temperature of 400°C. The flow simulation model is defined to model fluid flow using the weakly compressible Navier-Stokes equations, including gravity (for modelling buoyancy forces). The flow simulation model is defined to solve a steady-state flow condition. The simulations are isothermal.

The flow simulation model models the inflow boundary 701 as a fully developed (parabolic) laminar flow profile at a specified mass flow rate. The simulation cases flow simulation model models model the outflow boundaries 732, 742 as constant outlet pressure boundaries (at 23 MPa). The no slip condition is applied for walls of the chambers of the model. In each of the simulation cases discussed below, the inlet flow is modelled at a pressure of 23 MPa, and all species are modelled with properties corresponding to that pressure and a temperature of 400°C (corresponding to water density of 134kg/rrr ³ , viscosity of 27.4 pPa). The flow simulation model models the porous walls 110, 120 as a porous medium with porosity parameters applied to reflect the configuration of the specific flow simulation case (e.g. whether it is an isotropic porous wall as in the examples of Figures 1 b and 2a, an anisotropic porous wall with open channels as in the examples of Figures 1c and 2b, or a porous wall with a discontinuous porous structure, optionally extending along a path having a longitudinal component, as in the examples Figures 1d and 2c).

With reference to flow simulation cases in which the or each porous wall has an inclined channel or an inclined discontinuous porous structure, porosity parameters are applied to reflect the directionality of the porous medium. In particular, the flow simulation model implements a (variably-oriented) structured mesh to simulate flow through the electrode, and this mesh is aligned with the path angle of the respective channel or porous region. In each model, porous regions are simulated as homogenised porous media by means of application (e.g. simulation) of a Brinkman force. The associated permeability tensor is adjusted to correspond to the path angle of the respective simulation case. In particular, the associated permeability vale was set along the path of the respective channel or porous region (i.e. the path along which the respective channel or porous region is elongate), and is set to a null (zero) value in orthogonal directions to that path. Accordingly, the model provides constraints to the porous flow corresponding to the path of the respective channels or porous regions. Porosity parameters are set as 1 for simulation cases for configurations having open channels, and with a value between 0 and 1 (as described below) for simulation cases for porous regions (i.e. the discontinuous porous structure).

The flow simulation model is defined so that the respective electrolysis (e.g. hydrolysis) half-reactions are simulated to occur at a variable rate within the porous walls 110, 120, corresponding to the availability of the electrolyte fluid at that location, and to the applied current. Accordingly, the flow simulation model is capable of simulating flow patterns and species distributions corresponding to electrode water exhaustion effects as described below. In particular, electrode water exhaustion effects may occur when the configuration of the porous walls is such that an inlet electrolyte fluid is biased to flow through proximal portions of the porous walls (i.e. longitudinal portions of the porous walls which are relatively closer to the inlet 101) such that there is a relatively lower proportion of electrolyte fluid flowing through distal portions of the porous walls, with the consequence that a rate of the respective half-reactions is commensurately lower at the distal portions, as is a rate at which the respective reaction products are generated in the model. This may serve to limit a sustainable current for the electrolysis reaction. The flow simulation model simulates the reaction using sources and sinks for the respective fluid reaction products in the porous walls 110, 120.

The simulation cases are defined to reflect an operating current of 250A, which may correspond to a target hydrogen production rate of 1.42kg.tr ¹ for 150 cells in parallel.

The inventors have conducted simulations for four baseline configurations to evaluate selected flow phenomena and effects as discussed herein. The four baseline configurations are referred to herein as follows:

Configuration A: an isotropic porous wall corresponding to the example of Figure 1 b (or the example of Figure 2a);

Configuration B: a porous wall having open channels, corresponding to the examples of Figure 1c or 2b, for providing an anisotropic porous structure. The channels may be oriented perpendicular to the longitudinal direction of the porous wall, or the channels may extend along paths which are elongate the longitudinal direction (i.e. so as to have a longitudinal component;

Configuration C; a porous wall having a discontinuous porous structure corresponding to the example of Figure 1d.

To permit comparative discussion of the simulations of each configuration A-C, various parameters are defined for the simulations as set out below. Before that discussion, the relevant parameters are discussed.

Porosity is a measure of the open fraction of the porous wall that permits flow therethrough, and complies with the standard usage of the expression in the art.

In the flow simulation model, the entire longitudinal extent of the porous is configured for flow therethrough. It will be appreciated that in other implementations, a porous wall may one or more substantially non-porous longitudinal portions which are not intended for flow therethrough, for example at a proximal or distal end of the porous wall (e.g. for coupling to respective electrical terminals of an electrolyser cell), and a longitudinal porous portion which corresponds to the porous wall of the flow simulation model. Accordingly, the expression porosity and related expression “macro porosity” as used herein corresponds to the respective property as measured/defined over the longitudinal porous portion of the porous wall.

Macro porosity is a measure of the open fraction of a porous wall defined by channels therethrough (e.g. configuration B), and also to the open fraction of a body of a porous wall with discrete porous regions (e.g. configuration C). With reference to configuration C, the macro porosity is the porosity of the porous wall if the porous medium within the porous regions were to be removed, or the porosity of a precursor/intermediate product before a porous medium is introduced into corresponding open regions).

Macro pattern corresponds to the arrangement of channels/porous regions in a porous wall, which for example might be a regular matrix of channels/porous regions arranged in columns and rows, or with laterally/circumferentially offset rows. In a hexagonal arrangement, for example, the rows are laterally offset so that for each channel/porous region there are six surrounding channels/porous regions porous angularly distributed in the local plane/surface of the respective wall of 60°, with a substantially uniform separation/pitch to each surrounding hole. In a regular grid pattern, channels/porous regions of successive rows are not laterally/circumferentially offset.

Pitch corresponds to the separation between centres of the channels (configuration B) and porous regions (configuration C), as evaluated at the inlet side of the respective porous wall. For porosity patterns having unequal separation in different directions (e.g. in a regular grid having columns and rows, the diagonal distance being longer), the pitch is the shortest such separation. In each of the simulations discussed below, the circumferential pitch is the shortest separation, and is generally referred to as a circumferential pitch (or circ. pitch). It may also be defined by reference to the number of paths per circumference (“Paths per circ.”)

Pitch height corresponds to the separation between respective rows of channels (configuration B) and of porous regions (configuration C).

Micro porosity is a measure of the porosity within the porous regions 117 in configuration C (e.g. Figures 1d and 2c). The porosity of the porous wall as a whole can be calculated as the product of the macro porosity and the micro porosity.

Path refers to the path along which a channel (configuration B) or porous region (configuration C) is elongate through the body of a porous wall.

Path diameter is the diameter of the path of a channel (configuration B) or a porous region (configuration C), corresponding to a substantially circular profile of the respective channel or porous region.

Path number defines the number of channels (configuration B) or porous regions (configuration C) through a respective body of a porous wall. This does not relate to the network of flow paths within each porous region in configuration C, in contrast - each porous region in configuration C corresponds to a single path.

Path angle is the angle at which the path is inclined from the longitudinal axis. For example a path angle of 70° corresponds to the path extending along an elongate direction offset from the longitudinal direction by 70° and from an orthogonal (radial or thickness) direction of 20°. For each of the simulation cases discussed below, the inner (first) porous wall serves as the cathode (for hydrogen generation) and the outer (second) porous wall serves as the anode (for oxygen generation.

The inventors have defined simulation cases for the simulation model to evaluate various flow effects, phenomena and trade-offs associated with the baseline configurations, and an initial discussion of these and associated terminology aids interpretation of the simulation results.

The inventors have observed that in a flow arrangement having an axisymmetric arrangement (i.e. of concentrically arranged porous walls), there tends to be a flow bias towards the outer chamber when the porous walls have a similar construction - with the expression “flow bias” referring to a preference for flow in one direction (i.e. relatively higher mass flow in one direction than another). This is thought to be because the outer porous wall has a higher radius and therefore a higher surface area/volume for flow to pass through, and thereby presents less resistance to flow if porous features are provided in a similar way (e.g. an equal distribution of similarly-sized channels).

In the context of a flow arrangement for an electrolyser, it may not be that an equal flow distribution is desired, considering the stoichiometry of the respective half-reactions of electrolysis at the two electrodes. For example, it may be that a greater proportion of the electrolyte fluid is required at one of the electrodes than the other to perform the respective half-reactions of electrolysis at mutually sustainable rates. Accordingly, a mass flow bias is not necessarily undesired, but a flow bias may be excessive or undesirable in either direction if it has the effect of promoting crossover of species (i.e. flow of a fluid reaction product over the separation gap between two electrodes).

The inventors have determined that an undesirable flow bias can be mitigated by providing the porous walls with different properties that influence their resistance to flow (e.g. porosity, cross-sectional area or diameter, and length of path through a porous medium). For example, the outer wall can be made relatively less porous (e.g. by control of porosity, or by reducing a path number, path diameter, or increasing a path angle), or the inner wall can be made relatively more porous.

The inventors have also observed that the fluid reaction products tend to diffuse across the inlet chamber so that, for example, a fluid reaction product generated at the cathode arrives at the anode and is discharged through the respective outlet chamber. A flow bias within the flow arrangement can act together with (e.g. to accentuate) or against (e.g. to mitigate) diffusion of a species.

Fluid reaction products may have different diffusivities. The inventors have found that, for electrolysis of an aqueous electrolyte fluid for hydrogen and oxygen generation, hydrogen is more diffusive than oxygen. When the inner porous wall serves as the cathode, hydrogen generation is at the inner porous wall and so diffusion of hydrogen in the radially outer direction towards the outer porous wall is to be mitigated against.

One way of mitigating species diffusion in a particular direction is to influence or provide a neutral flow bias, or to bias flow in the opposing direction. However, this may have the result of compounding any diffusion of another fluid reaction product (e.g. promoting diffusion of oxygen generated at a radially outer anode to towards the radially inner cathode), and may deprive the opposing electrode of electrolyte fluid for the respective half-reaction of electrolysis.

Separately to flow biasing effects, the inventors have observed that the flow resistance provided by a porous wall can affect the distribution of flow of an electrolyte fluid through in the inlet chamber and through the respective porous walls. As discussed elsewhere herein, a porous wall provides resistance to flow such that a pressure difference over the wall is required (or becomes established) for there to be a flow therethrough. If the porous wall provides relatively low flow resistance (e.g. a relatively high flow factor Kv), then a flow rate of fluid along any particular pathway from the inlet to an outlet and going through the porous wall may be dominated by (i.e. largely determined by) properties of the flow path elsewhere and away from the wall. For example, when there is a relatively low separation gap between the electrodes, it may be that frictional (e.g. viscous drag) forces within the inlet chamber are relatively higher than frictional (e.g. viscous drag) forces within the respective outlet chamber. Accordingly, when naturally following a path of lowest resistance through the electrolyser, an electrolyte fluid may tend to preferentially flow through a proximal portion of the respective porous wall (i.e. a portion relatively closer to the inlet) and benefit from the lower frictional (e.g. viscous drag) forces in the outlet chamber, with consequentially higher mass and velocity flow rates through the proximal portion of the respective porous wall. This may result in “exhaustion” of the electrolyte fluid at the respective porous wall, corresponding to the distal portions of the porous wall being deprived of electrolyte fluid to conduct the respective half-reaction of electrolysis with. As noted above, such non- uniform flow of the electrolyte fluid may arise when the porous wall offers a relatively low flow resistance. In contrast, when a porous wall offers a relatively higher flow resistance, the frictional (e.g. viscous drag) forces within the porous wall itself become more dominant in relation to the total forces experienced from inlet to outlet of the flow arrangement/electrolyser, and any localised peaks of flow velocity (and mass flow) along a porous wall are reduced to result in a more uniform flow (particularly with drag forces being to the second power of velocity).

It may also be that flow bias effects are compounded by a relatively low flow resistance. In particular, it is considered when the porous walls generally offer a relatively low flow resistance, a relatively change in the flow resistance offered by one of the walls may lead to a significant change in the flow bias. Accordingly, it is considered that it may be relatively more challenging to select conditions to maintain a desired flow bias when one or more of the porous walls offers a relatively low flow resistance.

With regards to providing an efficient flow arrangement for electrolysis, the inventors have determined that it may be desirable to maximise a surface area of porous walls that are exposed to the electrolyte fluid and electrocatalytically active for the respective half-reaction of electrolysis. The inventors have determined that it is also desirable to reduce a separation gap between the electrocatalytic regions of the porous walls.

The inventors have considered that, instead of providing a porous wall as an isotropic porous medium (i.e. configuration A defined above), one or more channels can be provided through the porous wall to provide porosity to the wall (e.g. configuration B). By selecting parameters of the channels (e.g. path diameter, path angle, thickness), properties relating to the flow bias and flow resistance may be controlled.

The inventors have also considered that such channels can be provided as inclined channels as described herein. In particular, it is thought that since the fluid reaction products of the respective half-reactions of electrolysis are of a lower density than the electrolyte fluid, their generation within the respective channels can establish a buoyancy driven flow directed upwardly towards the respective outlet chambers 130, 140, such that any reverse flow of a fluid reaction product generated within a channel (configuration B) is inhibited, as such flow would be against the direction of a prevailing buoyancy- driven flow through the respective porous wall. Accordingly, the inclination is thought to prevent crossover of species (in either direction). However, there may be a trade-off between various factors including manufacturability, reaction surface area, flow resistance to prevent exhaustion, and providing flow resistance for a suitable flow bias.

For example, it is considered that open channels according to configuration B may be formed with a relatively small path diameter in order to reduce porosity and provide a relatively large flow resistance to avoid exhaustion effects and sensitivity to flow bias. However, it is also considered that channels having a relatively larger path diameter are more practical and inexpensive to form.

In contrast, while channels having a relatively larger path diameter may be more practical and inexpensive to form, they are associated with a relatively lower flow resistance. Flow resistance of a porous wall may be improved by providing relatively fewer channels, but this is associated with a reduction in the surface area available for the electrolysis reaction within the channels.

The inventors have determined that good performance with respect to each of the factors of the trade-offs discussed above may be achieved by providing the discontinuous porous structure of configuration C (e.g. as per the examples of Figures 1d and 2c). In particular, rather than forming channels, discrete porous regions to permit flow through the porous wall, each porous region defining a network of flow paths through a porous medium. The porous regions can be formed with a relatively large path diameter (e.g. corresponding to a diameter of an open region as discussed above, prior to providing the porous medium), while adverse effects such as low flow resistance and low reaction surface area can be avoided since the porous medium is filled with the porous medium. Further, the porous regions can be inclined (e.g. to have a path angle less than 90° relative to the longitudinal direction) to provide a buoyancy-driven flow upwards through the porous regions which may inhibit a reverse flow of the fluid reaction products.

Table 3 below defines simulation settings for a first set of simulations to evaluate the impact of providing an angle of inclination to the porous regions of configuration C. The notation for the simulation ID SxRy-Z corresponds to Set x, Run y- Configuration Z. For example, the second run of this first set of simulations has configuration C, and therefore has the ID S1 R2-C. In the tables and discussion below, some values are provided for both the inner and outer porous walls using a dividing e.g. 70770° to indicate a 70° path angle for each porous wall

Table 3

In Table 3, molar concentrations of crossover are reported with respect to the molar concentration of the respective fluid reaction (e.g. hydrogen) product in a resultant mixture of the fluid reaction products (e.g. hydrogen and oxygen), excluding any residual electrolyte fluid (e.g. water). This reflects the molar concentration within a mixture of fluid reaction products that may be separated from the electrolyte fluid at the respective outlets (as will be described further below). The same is true of other values of crossover by molar concentration as reported herein. In contrast, molar concentration as shown in the drawings are local molar concentration (e.g. in the hydrogen-oxygen-electrolyte fluid mixture).

As shown in Table 3 by comparison of the results for S1 R1-C and S1 R2-C, the effect of providing the porous regions at an angle of 70° relative to the longitudinal direction, as opposed to an orthogonal direction (of 90°) is that the crossover performance improves for both the hydrogen (H2) crossover into the outer outlet chamber, and for the oxygen (02) crossover into the inner outlet chamber.

Figures 8a-8c show contour plots of hydrogen molar concentration (Figure 8a), oxygen molar concentration (Figure 8b) and water molar concentration (Figure 8c), for simulation case S1 R2-C, which corresponds to both porous walls having the inclined discontinuous porous structure, with a path angle of 70° from the longitudinal direction and porosity values for the first (inner) and second (outer) porous walls of 0.1/0.11.

As shown in Figure 8a, contours at increments of 0.001 molar concentration of oxygen show a growing concentration of oxygen in the outer porous wall 120 where the respective half-reaction of electrolysis takes place, but the absence of any contours for oxygen in the inlet chamber 102 and inner porous wall 110 indicates a very low crossover of oxygen.

Similarly, as shown in Figure 8b, contours at increments of 0.005 molar concentration of hydrogen show a growing concentration of hydrogen in the inner porous wall 110 where the respective half-reaction of electrolysis takes place, with only the lowest contour extending through the inlet chamber 102 and the outer porous wall 120 (towards an upper end), thereby indicating a low crossover of hydrogen.

In Figure 8c, contours at increments of 0.05 molar concentration of water show a single contour in the first (inner) porous wall 110. This indicates that there is no exhaustion of electrolyte fluid along the longitudinal extend of the porous walls.

A simulation case for configuration B with the same path diameter, path angle and porosity as simulation case S1 R2-C was conducted. The simulation case matches the overall porosity by reducing the path number in each electrode. For the particular combination of parameters selected, it was predicted that exhaustion effects would occur, which would limit the electrolysis reaction (e.g. limit the current below the 250A target). This may demonstrate that, despite a similar overall porosity, a porous wall configuration with open channels may provide a lower flow resistance effect than a porous wall configuration having a discontinuous porous structure as described herein, which may result in an undesirable flow bias and/or exhaustion effects.

The inventors have found that it is possible to adjust a flow bias by varying one or more parameters (as further discussed herein) and this has been done to limit exhaustion effects and permit simulations of a flow arrangement with porous walls according to configuration B at the 250A target current. Merely as an example, in a second set of simulations (“Set 2”), the porosity of the inner wall was increased and the porosity of the outer wall was decreased, in order to inhibit a flow bias towards the outer wall. Model settings for Set 2 are shown in Table 4 below, for simulations in both configuration B and configuration C. Flow results are shown in accompanying drawings

Table 4

Figure 8d shows flow split between the inner and outer outflows of the model, for both simulation models over a range of flow rates 20-40 g/min. Figure 8e shows hydrogen and oxygen crossover (molar %) at the same flow rates.

The simulation results indicate that, by increasing the porosity of the inner porous wall and lowering the porosity of the outer porous wall, a flow bias is towards the inner outflow of the flow simulation model at all flow rates (see Figure 8d). This is more pronounced in the simulations for configuration B, with a significant oxygen crossover as shown in Figure 8e and a low hydrogen crossover. In contrast, the simulations for configuration C show a slightly higher hydrogen crossover, and a very low oxygen crossover.

Although the porosity of the inner porous wall is higher in S2R2-B than in S2R1-C which may drive a greater flow bias towards the inner outlet chamber, it is thought that the porous walls according to configuration B generally offer a lower flow resistance, such that the crossover effect in the direction of the flow bias is significantly more pronounced. Accordingly, it may be that a flow arrangement with open channels as in configuration B is more sensitive to flow bias, as compared with a flow arrangement with the discontinuous porous structure of configuration C. In addition to comparing flow results and phenomena with different configurations of the porous walls, the inventors have conducted parametric studies by simulation to evaluate the impact of varying selected parameters of porous walls having the discontinuous porous structure.

These parametric studies include: i. a parametric study of dissimilar porosities for the inner and outer porous walls; a. in conjunction with a further sweep of dissimilar path angles; ii. a parametric study of path diameters; iii. a parametric sweep of microporosities (which may be considered to be equivalent to different particle sizes); iv. a comparison of results at two different electrode lengths.

As discussed above, for a flow arrangement having an angular configuration as described herein, if the porous walls each have a similar configuration, then the outer porous wall will naturally provide less resistance to flow as it has a larger area. The inventors have run the flow simulation model with dissimilar porosities for the inner and outer porous walls, as implemented by varying the path number (i.e. the number of porous regions) in the respective walls. The particular implementation varies the path number by varying the circumferential pitch of the porous regions, but is considered to be representative of varying a pitch height or uniform pitch.

The geometry and configuration of the flow simulation model is as described above with respect to Figure 7, with the porous walls having the discontinuous porous configuration (configuration C) and a path angle relative to the longitudinal direction of less than 90°.

For (i) the sweep of dissimilar porosities, the porosities of the respective walls were varied by adjusting the path number for the respective porous walls, without varying the path diameter (120pm) or microporosity associated with each porous region. Figures 9a-9d show four matrices of sweep results at respective combinations of angles. The path angles are relative to the longitudinal direction.

The results are presented in numbered categories: 1 , 2, 3 and X. These categories relate to compliance with a threshold amount of crossover, known in the art as the lower explosion limit (LEL). For each combination of porosities shown in each matrix, the flow simulation was conducted at a number of different inlet mass flow rates, between 20- 40g/min.

Category 1 corresponds to both outflows having crossover corresponding with 50% or less of the LEL, for at least one flow rate in the set of flow rates. Category corresponds to both outflows having crossover corresponding to 100% or less of the LEL. Category 3 indicates that a simulation was conducted but did not meet the requirements of category 1 or 2. Category X indicates that no simulation was run.

As can be seen, in this particular example (i.e. with the particular definition of the conditions of the flow simulation model), the results indicate that the porosity of the outer porous wall is to be reduced relative to that of the inner porous wall in order to provide a flow bias that leads to suitable results, with poorer results being achieved as the porosity of the outer porous wall increases. Decreasing the angle for the inner porous wall from 70° to 50° had no effect on the results as reported (by comparing the matrices of Figures 9a and 9b). However, reducing the angle for the outer porous wall from 70° to 50° and then to 30° tends to result in more cases in categories 1 and 2.

It may be considered that, in addition to any buoyancy-related crossover-reduction effect of the reduced angle relative to the longitudinal direction, the reduced angle increases the path length through the respective electrode and therefore increases the resistance to flow, thereby further limiting or reversing a flow bias towards the outer porous wall.

The above results illustrate how flow resistance can be varied on the porous walls independently, for example by varying a path number and/or by varying a path angle. The results are presented as an example only, and it is to be appreciated that any optimal combination of parameters may be specific to a particular porous wall geometry, with a different optimal combination being suitable for other geometries. For example, a flow bias may be influenced in a flow arrangement in further different ways, and may be mitigated or controlled by greater or lesser variation of the porosity and/or path angles. In this context, it is to be appreciated that further factors that may influence flow bias may include a difference in outlet pressures at the outflow, a difference in wall thickness of the porous walls, differing geometries of the outlet chambers etc.

Figures 9a-9d show results for a further parametric study evaluating the effects of both porosity and path angle variations for porous walls having the discontinuous porous structure. The parametric study includes a baseline configuration in which the inner and outer porous walls have a porosity of 0.1 and 0.11 respectively, with the porous regions having a path angle of 70° (relative to the longitudinal direction). In the parametric study, the porosity is varied to 0.1 and 0.05 respectively, and the path angle for the outer porous wall is varied to 50°. Table 5 below shows settings for the simulations of the parametric study.

Table 5

Figures 10a and 10b show, for each simulation case at a variety of flow rates, a split between mass flow through the inner and outer outflows of the flow simulation model. As shown, the effect of varying the path angle of the outer porous wall on the flow rate split between the inner and outer outflow is that the flow is more biased to the inner outflow (at each flow rate). This indicates that the reduction in the path angle from 70° to 50° increases a flow resistance at the outer porous wall. A similar effect is observed in relation to reducing the porosity of the outer porous wall from 0.11 to 0.05, by comparison of Figures 10a and 10b. These results further demonstrate how a flow bias can be adjusted by varying such parameters.

Figures 10c and 10d show trends of species crossover (hydrogen and oxygen, molar fraction excluding water) at flow rates of 20, 30 and 40 g/minute, for each of the simulation cases defined in Table 5 above. There is a general trend that the crossover reduces as the flow rate increases. It is considered that this may result from an increased inertia of flow through the porous walls, thereby preventing reverse flow of the fluid reaction products. Figure 10c shows hydrogen crossover. It can be observed from Figure 10c that the variations which tend to bias the flow towards the inner porous wall, which is the hydrogen-generating cathode, tend to reduce the hydrogen crossover the outer outlet chamber and outflow. In particular, the hydrogen crossover is reduced by the path angle variation to 50° at the outer porous wall, and then further reduced by reducing the porosity of the outer porous wall.

Figure 10d shows corresponding results for oxygen crossover. The variation of the path angle at the outer porous wall to 50° causes oxygen crossover to increase, which is considered to be caused by the flow bias towards the inner porous wall as discussed above. However, it is noted that the increase in oxygen crossover is small compared to the decrease in hydrogen crossover, and this may be related to the buoyancy-driven flow at the outer porous wall serving to limit oxygen crossover, with the path angle of 50° bringing the angle of the porous regions closer to the longitudinal direction. The reduction in the porosity of the outer porous wall is also seen to increase the oxygen crossover.

By suitable variation of the flow bias, a configuration of the porous walls can be selected which provides a desired balance of hydrogen and oxygen crossover. In the examples of Figure 10a-10d, hydrogen crossover is generally higher, and so it may be that variations which tend to drive a flow bias towards the inner porous wall which serves as the cathode are desirable. In other configurations, a different flow bias may be established or desired, depending on the configuration of the flow arrangement (e.g. electrolyser) and which of the porous walls serves as anode and cathode respectively.

In a parametric study of the effect of path diameter on a flow patterns in the flow arrangement (e.g. electrolyser), simulation cases for porous walls having an inclined discontinuous porous structure were defined. In the parametric study, simulation cases with a range of path diameters between 50pm and 500pm were run, while maintaining other parameters including (i) a fixed porosity of 0.25 on both the inner and outer porous walls, (ii) a path angle of 70° for both the inner and outer porous walls and (iii) a micro porosity of 0.5 for both the inner and outer porous walls. It is considered that a micro porosity of 0.5 may be representative of a particle size of approximately 5pm. The result of the parametric study was that a flow bias between the inner and outer electrodes was substantially the same for each path diameter tested, as was crossover performance.

In a parametric study of the effect of the length of the porous walls (and in particular, a longitudinal extent of a porous portion of the porous walls), simulation cases corresponding to S3R3-C and S3R4-C defined in Table 5 above were defined and run, but differing in that the length l _ma in of the porous walls is 150mm instead of 100mm. Figures 10e and 10f show results for hydrogen and oxygen crossover, illustrating that generally the same trend is observed but with a slightly higher crossover of species in each case.

In a parametric study of the effect of microporosity, a set of simulations having a baseline configuration as defined in Table 6 below were defined and run, but differing in that the micro porosity of the porous regions was varied between 0.3 and 0.9 at 0.1 increments, with consequential variation of the total porosity of the respective porous walls

Table 6

Plots of hydrogen and oxygen crossover are provided in Figures 10g and 10h. These show the same general trend of decreasing crossover with flow rate, with both the hydrogen and oxygen crossover generally increasing with increasing porosity, corresponding to there being a progressively lower resistance to flow through each of the porous walls. With the particular settings of this set of simulation results, the oxygen crossover significantly increases above 0.8 porosity and the hydrogen crossover reduces, which may be indicative of a flow bias towards the inner porous wall.

Figure 11 schematically shows an example electrolysis installation 10 comprising an electrolyser 100. The electrolyser comprises a flow arrangement as described above with reference to Figure 1 , for example with one or both of the porous walls having a configuration according to any of the first, second and third examples described with respect to Figures 1 b-1d. In sequential flow order as illustrated, the example electrolysis installation 10 comprises a source of electrolyte fluid 12, a compressor 14 to compress (pressurise) the electrolyte fluid, a heater 16 to heat the electrolyte fluid, an inlet conduit 18 leading to an inlet manifold 20, the electrolyser 100, a first discharge manifold 24 and a second discharge manifold 30..

In this example, the electrolysis installation is for operation at supercritical conditions at the porous walls of the electrolyser, with an electrolyte fluid which comprises an aqueous electrolyte solution, for example as described above with respect to operation of the electrolyser 100 of Figure 1a. In this example, the compressor 14 is configured to compress the electrolyte fluid to a pressure of at least 22 MPa, for example between 22 MPa and 27 MPa. Further, the heater 16 is configured to heat the electrolyte fluid to a temperature for supercritical conditions at the porous walls of the electrolyser (e.g. at least 374°C, for example 374°C-550°C or 374°C-400°C). Heating may also occur within the electrolyser, for example at the porous walls which define the electrodes. It may be that the heater 16 is configured and/or controlled to heat the electrolyte fluid to a temperature within 50°C of a critical temperature for the respective electrolyte fluid, for example within 30°C of the critical temperature or within 20°C of the critical temperature. It will be appreciated that the temperature of the electrolyte fluid may be a function of both the compression by the compressor 14 and heating by the heater, and the compressor and heater may be configured and/or controlled so that the electrolyte fluid has the stated conditions after having passed through both. The compressor and the heater may be provided in any order (e.g. a reversed order relative to that illustrated).

The inlet conduit 18 directs the heated and pressurised electrolyte fluid to the inlet manifold 20, which supports a temperature sensor 22 for monitoring a temperature of electrolyte in the inlet manifold, the temperature sensor being coupled to a controller 56 via a connection 23. Although the connection 23 is illustrated in Figure 11 separately from the temperature sensor 22, it will be appreciated that the connection 21 may be any suitable form of connection, for example a wired or wireless link.

The inlet manifold 20 is coupled to the electrolyser 100 to provide the electrolyte fluid into the electrolyser 100 for an electrolysis reaction as described above with respect to operation of the example electrolyser 100 of Figure 1a. There are separate first and second outlet pathways 23, 29 coupled with the first and second outlets 132, 142 (as best shown in Figure 1a) of the electrolyser respectively (the first and second outlets being configured to discharge first and second fluid reaction products respectively). Figure 11 schematically shows the first outlet pathway 23 coupled to a centrally located first outlet of the electrolyser 100, whereas the second outlet pathway 29 is coupled to a radially-outwardly (e.g. annular) second outlet of the electrolyser 100.

The first outlet pathway 23 is in fluid communication with a first discharge manifold 24 which receives a flow of electrolyte fluid and a first reaction product from the first outlet of the electrolyser 100. The first discharge manifold 24 is fluidically coupled to a first discharge valve 34 via a discharge line 25. The first discharge valve 34 may be a control valve that provides a variable restriction to flow therethrough.

The second outlet pathway 29 is in fluid communication with a second discharge manifold 30. As schematically shown in Figure 11 , in this example the first outlet pathway 23 extends through the second discharge manifold 30 while functionally bypassing it (i.e. such that flow within the first outlet pathway 23 does not mix with fluid in the second discharge manifold 30), but in other implementations may be configured differently. As with the first discharge manifold 24, the second discharge manifold 30 is fluidically coupled to a second discharge valve 46 via a discharge line 31 , which may be of a similar type as the first discharge valve 34.

Optionally each of the first discharge manifold 24 and the second discharge manifold 30 is provided with respective a pressure sensor 26, 32 having a pressure sensor element in communication with the fluid discharged from the respective (first or second) outlet of the electrolyser or in the respective chamber immediately upstream of the outlet. Each pressure sensor 26, 32 is coupled to a controller 56 via a respective connection 27, 33 to provide a respective pressure signal to the controller 56 (as above, any suitable form of connection may be provided despite the connections 27, 33 being shown as separate from the respective pressure sensors 26, 32 in the drawing). Alternatively or additionally, a differential pressure sensor may be provided in communication with the respective discharge manifolds or outlet pathways, and connected to the controller 56 to provide a differential pressure signal to the controller 56.

The first discharge valve 34 and/or the second discharge valve 46 may be a control valve. For example, the first discharge valve 34 and/or the second discharge valve 46 may be a controllable pressure-maintaining valve configured to maintain a target pressure upstream of the respective valve corresponding to target operating conditions in the electrolyser 100 (e.g. supercritical pressure conditions for the electrolyte fluid at the respective porous walls and/or throughout the inlet and outlet chambers of the electrolyser). The first discharge valve 34 and/or the second discharge valve 46 may expand the flow to a lower pressure, for example a pressure at which the respective reaction product is gaseous and the residual electrolyte fluid is liquid, whereby the respective reaction product may be separated from the electrolyte fluid relatively easily (for example by phase separation in an accumulator tank). The residual electrolyte fluid may be recirculated to the source 12 of electrolyte fluid.

Optionally the electrolysis installation comprises separators 36, 48 downstream of the respective discharge valves 34, 46 for separating the respective fluid reaction product from the electrolyte fluid. Each separator 36, 48 has a respective return line 38, 50 for a flow of discharged electrolyte fluid from the separator, which may be returned to the source of electrolyte fluid for re-use. Each separator 36, 48 further comprises an outlet line for 40, 52 for discharging the respective fluid reaction product. The fluid reaction product may be in gaseous form within the respective separator and discharged through the outlet line as a gas. Optionally, the monitoring apparatus comprises flowmeters 42, 54 on the respective outlet lines for monitoring a flow rate of the respective fluid reaction product, each outputting a respective signal to a controller 56 as described below.

As shown in Figure 11 , there is a controller 56 which is coupled to flow control apparatus and monitoring apparatus in order to control operation of the electrolyser installation 10. The monitoring apparatus may comprise the temperature sensor 22 for monitoring an inlet temperature of electrolyte fluid, the pressure sensors 26, 32 (or differential pressure sensor) for monitoring pressures of fluid within or discharged from the first and second outlets of the electrolyser 100 respectively, and the flowmeters 42, 54 on the outlet lines 40, 52 for monitoring the outlet flows of the reaction products. The flow control apparatus may comprise the monitoring apparatus.

The flow control apparatus includes one or more components that determine (i.e. influence or affect) conditions within the electrolyser, such as the thermodynamic and/or flow rate conditions within the electrolyser. The flow control apparatus (or equipment) may therefore comprise any or all of the heater 16, the compressor 14, the first and second discharge valves 34, 46, and a cell controller configured to control a current through, and/or a voltage applied between, the first and second electrodes. The controller 56 may comprise the cell controller.

The thermodynamic conditions relate to the pressure and temperature of fluid within the electrolyser, for example the pressure and temperature of electrolyte fluid provided to the electrolyser, or the pressure and temperature of the electrolyte fluid in combination with the first and/or second reaction products in the respective chambers in which they are retained within the electrolyser. In this example the thermodynamic conditions are a function of a pressure to which the electrolyte fluid is pressurised at the compressor 14, a temperature to which the electrolyte fluid is heated at the heater 16, any heating of the electrolyte fluid at the porous walls of the electrolyser (e.g. as controlled by a cell controller), and optionally the operation of the first and second discharge valves 34, 46 (e.g. a target back pressure which the valves are configured to maintain upstream of the valves).

The flow rate conditions relate to the flow rate of electrolyte fluid provided to the electrolyser, and optionally to the or each flow rate of a branch flow of electrolyte fluid that passes from the inlet chamber to the or each outlet chamber via a respective porous wall.

In steady state conditions, the flow rate into the electrolyser is equivalent to the total of first and second flow rates out of the respective first and second outlets. Each outlet flow may depend on the pressure difference between the inlet chamber of the electrolyser (e.g. the pressure to which the compressor 14 compresses the electrolyte fluid) and the respective one of the first and second discharge valves 34, 46, and any flow resistance along the respective flow path (e.g. primarily any associated porous wall, but also any other features of the flow path which may effect a pressure drop, such as bends and flow restrictions).

Example methods of controlling an electrolysis reaction will be described below merely by way of example with reference to the electrolysis installation 10 of Figure 11 , and with reference to the flow diagram of Figure 12.

Figure 12 is a flow diagram of a method of controlling the or each discharge valve 34, 46 associated with the outlet flows comprising, respectively a mixture of the first fluid reaction product and the electrolyte fluid, and a mixture of the second fluid reaction product and electrolyte fluid

Block 1202 represents monitoring data received at the controller 56 from associated sensors of the monitoring apparatus.

In block 1204, monitoring data is received at the controller from associated sensors of the monitoring apparatus. For example, the data may be received continuously, periodically, on demand, or at the initiative of the respective sensor (e.g. when it detects a predetermined condition).

In block 1206, the controller evaluates a target criterion relating to the monitoring data.

In block 1208, the controller determines a setting for one or more of the discharge valves 34, 46 to control operation of the electrolysis installation, based on the evaluation of the target criterion in block 1206. The method continuously repeats by returning to block 1204. In a first example of the method 1200, the monitoring data is upstream pressure data received from pressure sensors 26, 32 configured to monitor a pressure of fluid discharged from the respective first and second outlets of the electrolyser (i.e. outlets of the respective chambers of the electrolyser), or from a differential pressure sensor.

Evaluating the target criterion may comprise determining a difference between a pressure of fluid discharged through a first outlet 132 (e.g. associated with a first outlet chamber 130) and a pressure fluid discharged through a second outlet 142 (e.g. associated with a second outlet chamber 140), or comparing a pressure of fluid discharged from a respective outlet with a target range of pressures for the fluid.

In block 1208, a setting for one or more of the discharge valves may be determined based on the evaluation target criterion. For example, there may be a predetermined target range for the pressure difference, and the setting may be determined to maintain or return the pressure difference to the target range. When the electrolyser is to be operated with no pressure difference between the outlets, the target range may include a zero pressure difference. When the electrolyser is to be operated with a pressure difference between the outlets, the target may range may not include a zero pressure difference. The target range (or respective target ranges for each pressure) may be predetermined, for example based on commissioning of the electrolysis installation to determine pressure differences that correspond to satisfactory performance of the electrolysis installation, with a suitable balance of flow rates between the outlets. Accordingly, maintenance of the pressures or pressure difference within respective target ranges may result in maintaining a target flow rate ratio between flow out of the first outlet and flow out of the second outlet.

In a second example of the method 1200, the monitoring data may be flow rate data, for example as received from flowmeters configured to monitor a flow rate of a mixture of flow a fluid reaction product and electrolyte fluid discharged from a respective outlet 132, 142 of the electrolyser (whether upstream of the respective discharge valve 34, 46, such as along the respective discharge line 25, 31 ; or downstream of the respective discharge valve). The flow rate data may be received from flowmeters configured to monitor a component flow rate of a respective fluid reaction product separated from the electrolyte fluid, for example from a flowmeter 42, 54 installed on an outlet line for discharging the respective fluid reaction product from a separator 36, 48 downstream of the discharge valves 34, 46.

The target criterion may correspond to maintaining a target flow rate out of one or each of the outlets, or maintaining a target flow rate ratio between flow out of the first outlet and flow out of the second outlet. Evaluating the target criterion may comprise comparing the associated flow rate data with predetermined targets, or determining a ratio of the respective flow rates out of the outlets. The controller may determine a control setting of one or both the valves in order to maintain the flow rate or flow rate ratio within a target range.

In a third example of the method 1200, the monitoring data may be composition data relating to a composition of the mixture of fluid out of the respective outlets. For example, the composition data may be determined by comparing a flow rate of a component fluid reaction product out of a separator and a total flow rate of fluid discharged from the respective outlet, for example to determine a mass or molar fraction of the respective fluid reaction product in the fluid mixture discharged from the respective chamber of the electrolyser. Alternatively, the composition data may be determined as a flow rate of the respective fluid reaction product, irrespective of the flow rate of the electrolyte fluid with which it was discharged from the respective outlet.

The target criterion may correspond to maintaining a target composition out of the respective outlet, for example a target mass or molar fraction, or target flow rate of the fluid reaction product which may be considered to correspond to satisfactory performance of the electrolyser. The controller may determine a setting for the respective discharge valve based on the composition data. For example, if it is determined that the mass or molar fraction of a reaction product is below a target range, this may correspond to too much electrolyte fluid flowing through the respective chamber. Accordingly, the controller may determine to adjust a setting of the discharge valve to further restrict flow through the discharge valve, which may raise the mass or molar fraction of the respective fluid reaction product.

Alternatively, the target criterion may correspond to maintaining a target ratio of the flow rates of the respective fluid reaction products out of respective outlets of the electrolyser. For example, there may be a target range for the ratio corresponding to satisfactory performance of the electrolyser, for example with a flow rate of a first reaction product being within a target range of a desired ratio to a flow rate of the second reaction product. The desired ratio may correspond to the sustainable performance of the respective electrolysis reaction (i.e. with the respective half-reactions in balance).

Flow rates as discussed above may be mass flow rates. A flowmeter may monitor a volume flow rate of a fluid, and the flowmeter or a controller may be provided with information corresponding to a density of the respective fluid in order to determine a parameter corresponding to the mass flow rate.

In a fourth example of the method, the monitoring data may be composition data relating to an amount of a contaminant fluid reaction product in a fluid flow out of a respective outlet of the reaction chamber. For example, as discussed above the first fluid reaction product is to be retained in a first outlet chamber for discharge through a first outlet of the electrolyser, whereas the second fluid reaction product is to be retained in a second outlet chamber for discharge through a second outlet of the electrolyser. Discharge of the second fluid reaction product through the first outlet corresponds to the presence of a contaminant fluid reaction product in the respective outlet flow, and vice versa. The monitoring data may be received from a component gas sensor configured to monitor a composition of a gas, for example downstream of a separator for the respective fluid reaction product (e.g. at the location of the respective flowmeter 42, 54 as described above). Any suitable sensor may be selected, but merely as an example a thermal conductivity sensor may be used, configured to determine a composition of a flow stream based on a monitored amount of energy (e.g. electrical power) to maintain a temperature probe within the flow stream at a target level. When calibrated based on the respective fluid reaction product that is expected to be discharged (e.g. hydrogen), and the contaminant fluid reaction product (e.g. oxygen), for example based on the respective specific heat capacities of the fluid reaction products, the sensor or a controller to which it is coupled is configured to determine when the flow stream comprises an amount of the contaminant fluid reaction product. For example it may be calibrated to indicate that the amount is above a threshold, or may be configured to estimate or calculate the mass fraction of the respective fluid reaction products (e.g. hydrogen and oxygen).

The presence of a contaminant fluid reaction product may be indicative of a bias of flow towards the respective chamber of the electrolyser from which the outlet flow comprising the contaminant fluid reaction product is discharged. Accordingly, the controller may determine to adjust a setting of one or more of the discharge valves so as to reduce or control the bias. For example, if an excess amount of a contaminant fluid reaction product is determined in an outlet flow associated with the first discharge valve 34, the controller may control the first discharge valve 34 to reduce a flow rate through the first discharge valve 34 (for example, by partially closing the valve or setting a higher back pressure to be maintained).

While the above examples discuss reactive control methods to maintain target operating parameters of the electrolyser, such control methods may be optional and the electrolyser may be configured such that the electrolysis reaction proceeds without such interventions.

Example Electrolyser and Test 1 A test electrolysis installation was arranged as illustrated schematically in Figure 11 , including a source of electrolyte fluid 12, a compressor 14, a heater 16, an inlet conduit 18 leading to an inlet/monitoring manifold 20, an electrolyser 100, a first discharge manifold 24, and a second discharge manifold 30.

The electrolyser 100 was constructed in the form of an annular electrolyser as illustrated schematically in Figure 1a. The first (inner) porous wall 110 and the second (outer) porous wall 120 were each connected to a power supply biased such that the first (inner) porous wall 110 functioned as a cathode and the second (outer) porous wall 120 functioned as an anode.

The first porous wall was prepared by laser drilling channels into a first tube formed from Inconel® alloy 625 (a nickel-chromium alloy also containing iron, molybdenum, niobium and other alloying elements). The first tube had an outer diameter of 0.25 inches (6.35 mm), a tube wall thickness of 0.889 mm and a longitudinal length of 215 mm. Prior to laser drilling, the outer and inner surfaces of the first tube were passivated by coating with alumina (AI2O3) by chemical vapour deposition (CVD). The coating of alumina was about 1 pm thick, as determined by gravimetric estimation. The channels were drilled through the full thickness of the first tube wall (including the alumina coating) using a millisecond laser operating at 500 Hz. The channels were inclined at 20° to the horizontal when the longitudinal axis of the tube was aligned with the vertical (i.e. 70° from the longitudinal axis). The channels were drilled in rows to form a pattern as shown in Figure 13. The pattern was characterised by a channel-to-channel offset of 0.2 mm, a row-to- row offset of 0.2 mm (such that there were 5 rows per mm) and a circumferential channel number density of 99 holes per circumference of the tube. Each channel was formed using 30 laser pulses and requiring 0.2 J of energy. The dimensions of the channels were assessed using X-ray Computed Tomography (XCT) and optical microscopy and analysis of the resultant images, the results of which are shown in Figures 14, 15(a) and 15(b). The channels were found to have diameters from about 60 pm to about 80 pm at the inlets and outlets of the channels (i.e. at the outer and inner external surfaces of the tube). Figure 15 (a) illustrates a channel found to have an internal radius of about 40 pm using XCT. Figure 15 (b) illustrates channels found to have inlet diameters of about 69.6 pm, about 74.6 pm and about 65.4 pm by optical microscopy.

The second porous wall 120 was prepared by laser drilling channels into a second tube formed from Inconel® alloy 625. The second tube had an outer diameter of 0.375 inches (9.525 mm), a tube wall thickness of 0.889 mm and a longitudinal length of 142 mm. Prior to laser drilling, the outer and inner surfaces of the second tube were passivated by coating with alumina (AI2O3) by chemical vapour deposition (CVD). The coating of alumina was about 1 pm thick, as determined by gravimetric estimation. The channels were drilled through the full thickness of the second tube wall (including the alumina coating) using a millisecond laser operating at 500 Hz. The channels were inclined at 20° to the horizontal when the longitudinal axis of the tube was aligned with the vertical (i.e. 70° to the longitudinal direction). The channels were drilled in rows to form a pattern as shown in Figure 13. The pattern was characterised by a channel-to-channel offset of 0.152 mm, a row-to-row offset of 0.2 mm (such that there were 5 rows per mm) and a circumferential channel number density of 196 holes per circumference of the tube. Each channel was formed using 20 laser pulses and requiring 0.18 J of energy. The dimensions of the channels were assessed using XCT and optical microscopy and analysis of the resultant images. The channels were found to have diameters from about 60 pm to about 80 pm at the inlets and outlets of the channels (i.e. at the inner and outer external surfaces of the tube).

The first and second porous walls 110, 120 were arranged as illustrated in Figure 1a with a separation gap of about 0.692 mm. The first and second walls were surrounded by an outer tubular housing having a diameter of 19.05 mm. The first, inner outlet chamber 130 was defined as the internal space surrounded by the first porous wall 110. The second, outer outlet chamber 140 was defined as the space between the second porous wall 120 and the outer tubular housing. The inlet chamber 102 was defined as the space between the first and second porous walls 110, 120. The first and second tubes are installed so that in the assembled configuration, the effective lengths of the electrodes that are exposed to the inlet chamber for radial flow and ion transfer are equal and approximately 40mm.

The electrolyser was operated using an electrolyte consisting of a 0.5 mol (1 .2 wt. %) solution of lithium hydroxide (LiOH) in water. The electrolyte was prepared by diluting 98% reagent grade lithium hydroxide (available from Sigma-Aldrich) with deionised water to obtain the desired concentration.

The heater and the compressor were operated to hold the system at a pressure of 230 bar (23MPa) and a temperature of 385°C.

The flow rate of electrolyte through the electrolyser was controlled to be 10 ml/minute.

Following an initial pressurisation and heating of the system using deionised water, once the operating pressure and temperature were reached, the electrolyte was pumped through the electrolyser. The electrolyser cell was monitored until a voltage drop was observed on the power supply and then the current supplied to the cell was gradually increased to reach operating conditions of 500 mA at 1.56 V (corresponding to 80 % of the lower heating value of hydrogen). Electrolysis of the electrolyte was then carried out for 15 minutes.

Gas output from the first and second outlet chambers into the first and second discharge manifolds during electrolysis was collected in corresponding glass vessels filled with deionised water (the product gases displacing the water in the vessels as it was collected). The gases collected in the glass vessels were analysed by gas chromatography using an Agilent 6890 Gas Chromatograph (available from Agilent Technologies, Inc., USA) equipped with a Shincarbon ST column (available from Shinwa Chemical Industries Ltd., Japan) supplied with argon as the carrier gas. Gases were detected using a thermal conductivity detector built into the Agilent 6890 Gas Chromatograph. Gas chromatography runs were performed isothermally at 35°C. 100 pL samples of gas were injected into the column at time. Calibration plots were also produced by injecting various quantities of oxygen and hydrogen into the gas chromatograph and analysing the results.

Representative gas chromatography results for the gases obtained from running the electrolyser are shown in Figures 16 and 17. Figure 16 (a) is a representative chromatogram of gas obtained from the cathode (cathode gas) featuring three peaks relating to (going from left to right) hydrogen, oxygen and nitrogen content. Figure 16 (b) is a representative chromatogram of gas obtained from the anode (anode gas) also featuring three peaks relating to (going from left to right) hydrogen, oxygen and nitrogen content. Figure 17 (a) is a representative chromatogram of oxygen calibration gas. Figure 17 (b) is a representative chromatogram of hydrogen and nitrogen calibration gas.

The composition of the cathode gas and anode gas, as obtained by the gas chromatography, is provided in Table 7. These results were obtained by averaging measurement values across three different injected samples. The corresponding standard deviations in the measurement values are also provided.

Table 7 The values in Table 7 have been corrected to account for air ingress into the gas storage vessels by quantifying the amount of nitrogen present in each sample and subtracting the equivalent amount of oxygen according to the ratio of nitrogen to oxygen in air. For example, assuming that the ratio of N2:O2 in air is 3.7, the volume of oxygen in the sample arising from air contamination can be calculated as the measured volume of nitrogen in the sample divided by 3.7.

Test 2

A test electrolysis installation was arranged as in Test 1 , except that the laser- drilled channels in the first (inner) porous wall 110 and the second (outer) porous wall 120 of the electrolyser were filled with a porous Ni-based electrocatalytic material prior to use.

Filling of the first (inner) porous wall 110 was achieved by drawing a vacuum on the inside of the tube and applying a nickel-based conductive ink (available from Creative Materials Inc (of Massachusetts, USA) under the product name 116-25) to the outside of the tube, allowing the vacuum to draw the ink into the channels. Filling of the second (outer) porous wall 120 was achieved by drawing a vacuum on the outside of the tube and forcing the nickel-based conductive ink (available from Creative Materials Inc (of Massachusetts, USA) under the product name 116-25) through the inside of the tube, allowing the vacuum to draw the ink into the channels. In both cases, excess ink was removed using a dry cloth, following which the porous walls were left to dry in air for 10 minutes before being cleaned with a small amount of I PA. The tubes were then dried fully in a 200°C oven for 2 hours. Binder material was then removed from the ink during a burn-off stage, whereby the material was pyrolysed at 300°C for 6 hours, leaving behind only metallic ink particles. Finally, the particles were partially sintered together using a high temperature heating and cooling cycle with a target temperature of 930°C.

The electrolyser was operated using an electrolyte consisting of a 0.5 mol (1 .2 wt. %) solution of lithium hydroxide (LiOH) in water. The electrolyte was prepared by diluting 98% reagent grade lithium hydroxide (available from Sigma-Aldrich) with deionised water to obtain the desired concentration.

Two test runs of the electrolyser were carried out. In a supercritical run of the electrolyser, the heater and the compressor were operated to hold the inlet electrolyte fluid at a pressure of 230 bar (23MPa) and a temperature of 385°C, thus achieving supercritical conditions. In a subcritical run of the electrolyser, the heater and the compressor were operated to hold the inlet electrolyte fluid at a pressure of 228 bar (22.8MPa) and a temperature of between 350°C and 360°C, thus achieving subcritical conditions.

The flow rate of electrolyte through the electrolyser was controlled to be 10 ml/minute for each test run.

Following an initial pressurisation and heating of the system using deionised water, once the operating pressure and temperature were reached, the electrolyte was pumped through the electrolyser. The electrolyser cell was monitored until a voltage drop was observed on the power supply and then the current supplied to the cell was gradually increased to reach operating conditions of 2 A at 1.31-1.37 V (for the supercritical test run) or 3.8 A at 1 .6 V (for the subcritical test run). Electrolysis of the electrolyte was then carried out for 8 hours.

Gas output from the first and second outlet chambers into the first and second discharge manifolds during electrolysis was collected in corresponding glass vessels filled with deionised water (the product gases displacing the water in the vessels as it was collected). The gases collected in the glass vessels were analysed by gas chromatography using an Agilent 6890 Gas Chromatograph (available from Agilent Technologies, Inc., USA) equipped with a Shincarbon ST column (available from Shinwa Chemical Industries Ltd., Japan) supplied with argon as the carrier gas. Gases were detected using a thermal conductivity detector built into the Agilent 6890 Gas Chromatograph. Gas chromatography runs were performed isothermally at 35°C. 100 pL samples of gas were injected into the column at time. Calibration plots were also produced by injecting various quantities of oxygen and hydrogen into the gas chromatograph and analysing the results.

The composition of the cathode gas and anode gas, as obtained by the gas chromatography, is provided in Table 8. These results were obtained by averaging measurement values across three different injected samples. Table 8

The values in Table 8 have been corrected to account for air ingress into the gas storage vessels by quantifying the amount of nitrogen present in each sample and subtracting the equivalent amount of oxygen according to the ratio of nitrogen to oxygen in air. For example, assuming that the ratio of N2:O2 in air is 3.7, the volume of oxygen in the sample arising from air contamination can be calculated as the measured volume of nitrogen in the sample divided by 3.7.

Further Simulation Cases

Table 9 below defines simulation settings and results for a set of simulation cases to correspond to example manufactured electrolysers and tests, including as described above with reference to Table 7 and Figures 13-17.

The simulations are conducted using the flow simulation model described above with reference to Figure 7 and Table 3, with all geometry and parameter settings for the simulations being equal to the description above unless otherwise specified below.

The simulation set includes two simulation cases S5R1-B and S5R2-C. For simulation case S5R1-B the porous walls are defined with open inclined channels corresponding to configuration B as described above (e.g. corresponding to the second example of Figure 1c or the fifth example of Figure 2b as described above). For simulation case S5R2-C the porous walls are defined to have an inclined discontinuous porous structure corresponding to configuration C as described above (e.g. corresponding to the third example of Figure 1d or the sixth example of Figure 2c as described above).

As discussed above, the simulation cases are each defined to simulate conditions at a pressure of 23MPa and for a reference temperature of 400°C.

The simulation cases area each defined to simulate a current through the porous walls of 1 A, which is considered to be suitable for a comparison of the overall flow bias and crossover effects with respect to the experimental example discussed above.

Table 9

The definition of S5R1-B corresponds to the example electrolyser manufactured and tested as described above with reference to Figures 12-16, with experimental testing results reported in Table 7.

The results of simulation case S5R1-B demonstrate that the flow simulation model reflects the flow patterns and results observed with the experimental testing. In particular, the flow simulation model appears to simulate the flow through the porous walls such that, when the resultant crossover of the electrolysis reaction products are compared, they represent the same direction of flow bias. The experimental testing result shows a significantly higher crossover of hydrogen to the outer outlet (molar concentration of 57% ± 0.17 in the outer outflow gases) as compared with crossover of oxygen to the inner outlet (molar concentration of 1.54% ±0.94 in the inner outflow cases). The same trend is reflected in simulation case S5R1-B, with both the simulation and experimental test predicting hydrogen crossover (mol %) which is an order of magnitude greater than oxygen crossover. This indicates that the flow patterns modelled by the flow simulation model are representative of those arising in experimental tests.

Three sample porous walls were prepared by laser drilling channels into three hollow tubes formed from Inconel® alloy 600. The tubes each had an outer diameter of 0.375 inches (9.525 mm), a tube wall thickness of 0.9 mm and a longitudinal length of 3 cm. The channels were drilled through the full thickness of each tube using a millisecond laser operating at 500 Hz configured to generate a channel diameter of about 70 pm. The channels were inclined at 20° to the horizontal when the longitudinal axis of each tube was aligned with the vertical (i.e. 70° from the longitudinal axis). The channels were drilled over a 1 cm length section of each tube, with a density of 150 channels per circumference.

The porosity of the three samples was studied using mercury porosimetry according to ASTM standards D4284 and D6761 using an AutoPore V device (available from Micromeritics Instrument Corporation, USA). The measured median pore diameter, bulk density, skeletal density, permeability, characteristic length and tortuosity obtained by the mercury porosimetry for each sample are shown in Table 10.

Table 10.

Samples 1 and 2 exhibited greater median pore diameters and permeability than Sample 3, although the permeability of all three samples was relatively high. All three samples exhibited a relatively low tortuosity and a similar characteristic length value. Plots of the cumulative intrusion versus pore size diameter obtained by the mercury porosimetry for each sample are shown in Figure 18(a). Plots of the cumulative volume versus pore size diameter obtained by the mercury porosimetry for each sample are shown in Figure 18(b). Plots of the log differential intrusion versus pore size diameter obtained by the mercury porosimetry for each sample are shown in Figure 18(c).

All three samples showed a large peak in the plot of the log differential intrusion versus pore size at pore size diameter of about 50 pm. However, while Samples 1 and 2 exhibited a similar distribution of pore sizes, with a relatively large number of pores having diameters between 100 pm and 200 pm, Sample 3 displayed pores which are smaller than in Samples 1 and 2 (between 10 pm and 20 pm) and had relatively few pores having diameters between 100 pm and 200 pm. The inventors posit that peaks in the data above 500 pm are indicative of surface roughness rather than porosity.

Two further sample porous walls Samples 4 and 5 were prepared by laser drilling channels into two hollow tubes formed from Inconel® alloy 600. The tubes each had an outer diameter of 0.375 inches (9.525 mm), a tube wall thickness of 0.9 mm and a longitudinal length of 5 cm. The channels were drilled through the full thickness of each tube using a millisecond laser operating at 500 Hz configured to generate a channel diameter of about 70 pm. The channels were inclined at 20° to the horizontal when the longitudinal axis of each tube was aligned with the vertical (i.e. 70° from the longitudinal axis). The channels were drilled over a 1 cm length section of each tube, with a density of 150 channels per circumference.

Samples 4 and 5 were filled with a nickel-based conductive ink available from Creative Materials Inc (of Massachusetts, USA) under the product name 116-25. Sample 4 was filled with a version of the ink having a higher polymer binder content and Sample 5 was filled with a version of the ink having a lower polymer binder content.

Samples 4 and 5 were filled with the respective inks by drawing a vacuum on the inside of each tube and applying the ink filling to the outside of each tube, allowing the vacuum to draw the ink into the channels. Excess ink was then removed using a dry cloth. The samples were left to dry in air for 10 minutes before being cleaned with a small amount of I PA. The samples were then dried fully in a 200°C oven for 2 hours. The binder material was then removed during a burn-off stage, whereby the material was pyrolysed at 300°C for 6 hours, leaving behind only metallic ink particles. Finally, the particles were partially sintered together using a high temperature heating and cooling cycle with a target temperature of 930°C. Solid (i.e. non-laser drilled) ends of the samples were removed by cutting and grinding. Both samples were then placed together inside the AutoPore V device and mercury porosimetry results were obtained as an aggregate of both Samples 4 and 5.

Plots of the cumulative intrusion versus pore size diameter obtained by the mercury porosimetry for the aggregate of Samples 4 and 5 are shown in Figure 18(d) compared against Samples 1 , 2 and 3. Corresponding plots of the cumulative volume versus pore size diameter are shown in Figure 18(e). Corresponding plots of the log differential intrusion versus pore size diameter are shown in Figure 18(f).

From these plots, it is evident that Samples 4 and 5 exhibited a bimodal distribution of pore sizes grouped around 5 pm to 15 pm and around 100 pm. The peak at 100 pm generally aligned with that of Samples 1 , 2 and 3, suggesting that it corresponds to the channel “macroporosity” - i.e. the presence of unfilled or partially filled channels. The peak at around 5 pm to 15 pm appears to correspond to the filled channel “microporosity” - i.e. the contribution from the filled and sintered channels.

The measured median pore diameter, bulk density, skeletal density, permeability, characteristic length and tortuosity obtained by the mercury porosimetry for the aggregate of Samples 4 and 5 are shown in Table 11 , where the results were obtained separately from the portions of the curves corresponding to the 5 pm to 15 pm peak (“microporosity”) and the 100 pm peak ("macroporosity”), and then also averaged, as compared to unfilled Samples 1 , 2 and 3 and also the average thereof.

Table 11.

It can be seen from the data that the addition of the sintered nickel based ink introduced a large number of 1 to 20 pm sized pores to the channel structure. This significantly reduced the permeability and the characteristic length but increased the tortuosity.