Electrolyzer Systems and Operation Thereof

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is a continuation of PCT Application No. PCT/US24/24908, filed Apr. 17, 2024, which claims priority to U.S. provisional patent application No. 63/460,270, filed Apr. 18, 2023, U.S. provisional patent application No. 63/469,906, filed May 31, 2023, and U.S. provisional patent application No. 63/521,381, filed Jun. 16, 2023, each of which are hereby incorporated by reference in their entirety.

BACKGROUND

Hydrogen has a wide variety of applications and is used as an input for many industrial processes including, for example, ammonia production for fertilizers and oil refining. Currently, the majority of hydrogen produced in the United States is created through natural gas reforming. Because natural gas reforming releases air pollutants that contribute to global warming, there is a desire to replace natural gas reforming (and other forms of hydrogen production that emit undesirable gasses) with more environmentally-friendly hydrogen production methods where an electrolyzer is used to split water into hydrogen and oxygen through electrolysis using electricity.

BRIEF DESCRIPTION OF THE DRAWINGS

Features, aspects, and advantages of the presently disclosed technology may be better understood with regard to the following description, appended claims, and accompanying drawings.

FIG. 1A is a diagram of an example electrolyzer system of an embodiment.

FIG. 1B is a diagram showing an example implementation with multiple cell stacks (#1, #2, . . . , #n) that can be independently controlled.

FIG. 1C is a diagram of an example electrolyzer system of an embodiment.

FIG. 1D is a diagram of an example electrolyzer device of an embodiment.

FIG. 2 is a graph showing a relationship between operating temperatures and levelized cost of hydrogen.

FIG. 3A is a high-level diagram of an example high-temperature alkaline electrolyzer of an embodiment.

FIG. 3B is a more-detailed diagram of an example high-temperature alkaline electrolyzer of an embodiment.

FIG. 4 is a front view of an example diaphragm of an embodiment.

FIG. 5 is a cross-sectional view of an example diaphragm of an embodiment having a central support structure with ceramic material on either side of the support structure.

FIG. 6 is a cross-sectional view of an example diaphragm of an embodiment having a ceramic material sandwiched between layers of support structure.

FIG. 7 is a cross-sectional view of an example diaphragm of an embodiment having a support structure with ceramic material on only one side.

FIG. 8 is a cross-sectional view of an example diaphragm of an embodiment having a support structure with ceramic material on only one side.

FIG. 9 is a cross-sectional view of an example diaphragm of an embodiment having n layer(s) of a polymeric material.

FIG. 10 is a cross-sectional view of an example diaphragm of an embodiment having ceramic material alone.

FIG. 11A is an example diagram of a pump of an embodiment.

FIG. 11B is an example diagram of a pump cooling system of an embodiment.

FIG. 12 is a flow chart illustrating example high-level operational phases of an embodiment.

FIG. 13 is a flow chart of an example startup phase of an embodiment.

FIG. 14 is a flow chart of an example normal operation phase of an embodiment.

FIG. 15 is a flow chart of an example standby phase of an embodiment.

FIG. 16 is a flow chart of an example shutdown phase of an embodiment.

FIG. 17 is a flow chart illustrating an example operation of a controller of an embodiment.

FIG. 18 is a diagram of an example high-temperature alkaline electrolyzer of an embodiment where heat is transferred from a first stream.

FIG. 19 is a diagram of an example high-temperature alkaline electrolyzer of an embodiment where heat is transferred from a second stream.

FIG. 20 is a diagram of an example high-temperature alkaline electrolyzer of an embodiment where heat is transferred from both first and second streams.

FIG. 21 is an illustration of an example electrolyzer cell of an embodiment.

FIG. 22 is an illustration of an example electrolyzer cell stack of an embodiment.

FIG. 23 is a diagram showing an impedance of an example electrolyzer cell Z_cell of an embodiment seen by a power supply.

FIG. 24 is a Nyquist plot of impedance seen by a power supply in an example electrolyzer of an embodiment.

FIG. 25 is a graph of a magnitude of impedance seen by a power supply versus frequency in an example electrolyzer of an embodiment.

FIGS. 26A, 26B, and 26C are flowcharts showing example processes that can be executed by a controller in an example electrolyzer of an embodiment.

FIG. 27 is a graph showing an example current output of a power supply of an embodiment.

FIG. 28 is a graph showing another example current output of a power supply of an embodiment.

FIG. 29 is a graph of an example embodiment showing multiple frequencies provided in a current output with a lower-frequency pulse that, during the “on-time,” has an embedded higher frequency pulse.

FIG. 30 is a graph showing an improvement in electrical efficiency offered by current pulsing at different operating temperatures and using different diaphragm materials in an example electrolyzer of an embodiment.

FIGS. 31A-C are graphs of an example embodiment showing power availability and current magnitude over time.

FIGS. 32A-B are graphs of an example embodiment showing power availability and number of active stacks over time.

FIG. 33A depicts an example implementation of an embodiment that employs pulsing techniques over a range of hydrogen production rates from a minimum production capacity to a maximum production capacity.

FIG. 33B depicts an example implementation of an embodiment that dynamically adjusts a number of active cell stacks over a range of hydrogen production rates from a minimum production capacity to a maximum production capacity.

FIG. 34 is a graph of another example embodiment showing power availability and current magnitude over time.

FIG. 35 is a diagram of an example electrolyzer facility of an embodiment.

FIG. 36 is an example levelized cost of hydrogen (LCOH) formula of an embodiment.

FIG. 37 is a diagram of an example cost-based control system of an electrolyzer of an embodiment.

FIG. 38 is an example formula for calculating transient LCOH₂.

FIG. 39 is a flow chart of an example method of an embodiment for providing cost-based control of an electrolyzer of an embodiment.

FIG. 40 is an example LCOH₂ matrix of an embodiment.

FIG. 41 is an example of LCOH₂ matrices of different time periods of an embodiment.

FIG. 42 is a graph of an embodiment showing example parameter values that are targeted on an instantaneous basis.

FIG. 43 is a graph of an embodiment showing example parameter values that have been smoothed.

FIG. 44A is a graph of an embodiment showing example parameter values that are targeted on an instantaneous basis.

FIG. 44B is a graph of an embodiment showing example parameter values that are targeted over an entire length of time that encompasses all three time periods.

FIG. 45 is a diagram of an example cost-based control system of an electrolyzer of an embodiment with a separate safety system.

The drawings are for purposes of illustrating example embodiments, but it should be understood that the inventions are not limited to the arrangements and instrumentality shown in the drawings.

DETAILED DESCRIPTION

1. Introduction

a. Brief Overview of Hydrogen Production Processes

As mentioned above, hydrogen has a wide variety of applications and is used as an input for many industrial processes including, for example, ammonia production for fertilizers and oil refining. Currently, the majority of hydrogen produced in the United States is created through natural gas reforming. In natural gas reforming, air pollutants (e.g., carbon monoxide) are undesirably released that contribute to global warming. There is a trend to replace natural gas reforming (and other forms of hydrogen production that emit undesirable gasses) with green hydrogen production methods where no greenhouse gas emissions are produced.

Green hydrogen production, which makes up only about 1% of the hydrogen produced in the United States today, typically involves the generation of electrical energy through renewable sources (e.g., solar, wind, etc.), and that energy is used to split water into hydrogen and oxygen through electrolysis in an electrolyzer. Other examples of green production methods include pyrolysis, steam methane reforming (SMR) with carbon capture, thermolysis, and radiolysis.

Today, electrolyzer designers are often focused on achieving the highest electrical efficiency possible. The electrical efficiency of an electrolyzer describes the relationship between the electrical energy consumed and the hydrogen produced. A higher electrical efficiency means that less electrical energy needs to be consumed in order to create a given amount of hydrogen (and less electrical energy is lost as waste heat). Such a focus has led to the design of complex Proton Exchange Membrane (PEM) and Anion Exchange Membrane (AEM) electrolyzers with high electrical efficiencies (e.g., above 80%). One problem with such electrolyzer designs is that, despite high electrical efficiency, they are particularly expensive to construct and operate and therefore are still expensive in terms of levelized cost of hydrogen (LCoH2) (i.e., the levelized cost per kg of hydrogen produced from a system (ignoring real estate cost and certain other external variables)) when used in green hydrogen production applications. This high levelized cost of hydrogen (LCoH2) for such complex electrolyzer designs is driven by the large capital expenditure (capex) and operational expenditure (opex) costs associated with these designs, which outweigh the benefits yielded from their higher electrical efficiency. The following equation shows a standard levelized cost of hydrogen formula, where “O&M” refers to operation and maintenance cost, “overnight capital cost” refers to overnight capital cost to build the electrolyzer system, “capital recovery factor” is a value calculated based on the lifespan of the electrolyzer and an interest rate, a “feedstock cost” can refer to the cost of electricity, and “capacity factor” refers to the ratio of the output to the maximum output of the electrolyzer.

LCoX = { ( overnight ⁢ capital ⁢ cost * capital ⁢ recovery ⁢ factor ) + fixed ⁢ O & ⁢ M ⁢ cost 8760 * capacity ⁢ factor } * ( feedstock ⁢ cost * use ⁢ rate ) * variable ⁢ O & ⁢ M

Because of their high levelized costs of hydrogen, adoption of PEM and AEM electrolyzers has been slow in industrial applications, and most electrolyzers installed today are older, traditional alkaline electrolyzer designs that are cheaper to install and have a lower electrical efficiency.

Another feature of traditional alkaline electrolyzer designs is a cooling system that cools the electrolyzer to remove heat generated by the electrolysis reaction. This keeps the operating temperature of the electrolyzer below a predefined maximum operating temperature, which is typically at or below 100 degrees Centigrade (C). Keeping the maximum temperature low (e.g., below 100 degrees C.) avoids a number of challenging design problems that come with higher-temperature operation. For instance, the materials typically employed in the construction of key electrolyzer components (e.g., a diaphragm that facilitates separation of the hydrogen and oxygen gases) are not capable of operating at higher temperatures, as described in greater detail below. Further, increasing the temperature of the electrolyte solution (i) increases the corrosiveness of the electrolyte solution so as to damage components that come in contact with it and (ii) increases the pressure required to keep the electrolyte solution in a liquid state, which is desired to avoid the complicated dynamics of a multi-phase or vapor system that may require expensive gas diffusion electrodes.

b. High-Temperature Alkaline Electrolyzer Introduction

Traditional alkaline electrolyzers use a cooling system to remove heat generated by electrolysis and keep the operating temperature of the electrolyzer below 100 degrees Centigrade (C). Keeping the electrolyzer at that relatively-low temperature helps avoid complications that can arise with higher temperatures, such as a failure of the diaphragm of the electrolyzer, corrosion of the electrolyzer components, and complications with the use of the electrolyte solution, among other potential complications. However, such benefits of operating at relatively-low temperatures come at the expense of desirable electrolyzer characteristics that are available at higher temperatures. Some examples of such benefits include greater efficiency, higher throughput, dynamic operability (particularly when coupled with a variable supply like solar/wind and/or to serve demand-response applications with highly-variable demand for hydrogen), lower cost, and the ability to reduce or eliminate parasitic losses to cooling systems (which is common in today's electrolyzers), among other examples.

In contrast to existing solutions that operate at relatively-low temperatures, embodiments described herein include a high-temperature alkaline electrolyzer that allows a relatively-high-temperature operation. These embodiments also present various features that can be used to address the complications associated with running an electrolyzer at a relatively-high temperature. The use of these embodiments can result in lower hydrogen production costs and provide a substantial savings relative to conventional, low-temperature electrolyzers.

In one embodiment, an alkaline electrolyzer system is provided comprising a power input to receive power from an external power source, and an electrolyzer cell. The electrolyzer cell comprises a cathode coupled to the power input, an anode coupled to the power input, and a diaphragm at least partially disposed between the anode and the cathode. The electrolyzer cell is configured to (i) use power from the power input to produce a first stream comprising oxygen gas and a second stream comprising hydrogen gas from an alkaline solution and (ii) operate at a temperature in excess of 100 degrees Centigrade (C). The alkaline electrolyzer system also comprises thermal insulation at least partially covering the electrolyzer cell and a pump configured to circulate the alkaline solution through the electrolyzer cell.

In another embodiment, a method is provided of operating an alkaline electrolyzer system. The method comprises: circulating an alkaline solution through an electrolyzer cell comprising a cathode, an anode, and a diaphragm at least partially disposed between the anode and cathode; producing a first stream comprising oxygen gas and a second stream comprising hydrogen gas at least in part by inducing a current between the cathode and the anode in the electrolyzer cell; while producing the first stream and the second stream, operating the electrolyzer at a temperature that exceeds 100 degrees Centigrade (C); separating the hydrogen gas from the second stream; and outputting and/or collecting the hydrogen gas (the oxygen gas may also be collected).

In yet another embodiment, a diaphragm for an alkaline electrolyzer is provided. The diaphragm comprises a support structure comprising at least one of polyetheretherketone (PEEK) or polytetrafluoroethylene (PTFE), the support structure having a thickness of no more than, for example, 350 microns between a first side and a second side that is opposite the first side. The diaphragm can also comprise a stabilization additive (e.g., Yttria to form Yttria-Stabilized Zirconia (YSZ)) or stabilized zirconium dioxide (ZrO2) disposed on each of the first side and the second side of the support structure.

Other embodiments are disclosed, and the disclosed embodiments can be used alone or in combination.

These and other aspects are discussed in more detail in the passages that follow.

c. Electrode Pulsing Techniques Introduction

Additional aspects of the present disclosure relate to novel techniques for delivering current (e.g., a regulated pulsed current) to an electrolyzer cell in an electrolyzer system so as to: (i) improve the electrical efficiency of the electrolyzer system by reducing activation and ohmic overpotentials, (ii) reduce degradation of components in the electrolyzer system by minimizing side-reactions, and/or (iii) facilitate further improvements in dynamic operation of the electrolyzer system, among other functions and benefits.

Conventional techniques involving applying a constant direct current to electrolyzer cell stacks require operation above a particular minimum voltage (e.g., a reversible cell potential, an equilibrium potential, a thermoneutral potential, and/or a thermally balanced potential, lower heating value potential, and higher heating value potential). Due to the thermodynamics of electrolysis, operating at or above a particular minimum voltage requires the corresponding delivery of a particular current, and, thus, there is a minimum power input required for electrolyzer operation. This minimum power input fundamentally limits the dynamic operating range of existing electrolyzers, which has negative impacts especially in renewable-tied deployments. Further, constant current operation results in undesirable overpotentials-specifically from activation and diffusion-that can be avoided.

The pulse electrolysis capability described herein was demonstrated on a ZIRFON UTP 500 membrane, and the performance of pulse operation versus constant current operation was benchmarked using a power supply operating in current-control mode. Results indicate that the system can operate at substantially lower ranges of available power than constant-current electrolyzers and improve electrical efficiency. Further, a modular stack design allows for pulses to be delivered to cell groups with lower inductance than longer stacks powered by a single supply.

In one embodiment, an electrolyzer system is provided comprising a fluid input; an electrolyzer cell comprising a cathode and an anode; a pump coupled between the fluid input and the electrolyzer cell and configured to circulate fluid received from the fluid input through the electrolyzer cell; a power supply coupled the electrolyzer cell and configured to induce a pulsed current between the cathode and the anode of the electrolyzer cell that varies based on a control signal; at least one processor; and at least one non-transitory computer-readable medium comprising program instructions. The program instructions are executable by the at least one processor such that the at least one processor is configured to: determine at least one characteristic for the pulsed current to be induced between the cathode and the anode, the at least one characteristic for the pulsed current including a frequency of pulses; and generate the control signal for the power supply based on the determined at least one characteristic for the pulsed current.

In another embodiment, a method of operating an electrolyzer system is provided. The method comprises: dynamically determining a frequency of pulses of current that will reduce an impedance of an electrolyzer cell in the electrolyzer system; causing the pulses of current at the determined frequency to be induced between a cathode and an anode of the electrolyzer cell as solution is circulating through the electrolyzer cell; and producing a first stream comprising oxygen gas and a second stream comprising hydrogen gas at least in part due to electrolysis of the solution caused by the current induced between the cathode and the anode in the electrolyzer cell.

In yet another embodiment, an electrolyzer system is provided comprising: an electrolyzer cell comprising a cathode and an anode, wherein an impedance characteristic of an electrical path through the electrolyzer cell varies based on at least one operating parameter of the electrolyzer system; at least one processor; and at least one non-transitory computer-readable medium comprising program instructions. The program instructions are executable by the at least one processor such that the at least one processor is configured to cause a pulsed current to be generated across the cathode and the anode at a frequency that reduces the impedance characteristic of the electrical path through the electrolyzer cell.

In another embodiment, an electrolyzer system is provided comprising: a power supply; an electrolyzer cell comprising a cathode coupled to the power supply, an anode coupled to the power supply, and a diaphragm at least partially disposed between the anode and the cathode, wherein the electrolyzer cell is configured to (i) use power from the power supply to produce a first stream comprising oxygen gas and a second stream comprising hydrogen gas from an alkaline solution, and (ii) operate at a temperature in excess of 100 degrees Centigrade (C); thermal insulation at least partially covering the electrolyzer cell; a pump configured to circulate the alkaline solution through the electrolyzer cell; at least one processor; and at least one non-transitory computer-readable medium comprising program instructions. The program instructions are executable by the at least one processor such that the at least one processor is configured to: determine at least one characteristic for pulsed current to be induced between the cathode and the anode, the at least one characteristic for the pulsed current including a frequency of pulses; and generate a control signal for the power supply based on the determined at least one characteristic for the pulsed current.

Other embodiments are disclosed, and the disclosed embodiments can be used alone or in combination.

These and other aspects are discussed in more detail in the passages that follow.

d. Electrolyzer Control System Introduction

Still yet further aspects of the present disclosure relate to a control system for an electrolyzer. Conventional control schemes for electrolyzers focus on maximizing electrical efficiency, which describes the relationship between the electrical energy consumed and the hydrogen produced. One problem with such conventional control schemes is that, despite optimizing for high electrical efficiency, the electrolyzer system may still be unnecessarily expensive to operate in terms of levelized Cost of Hydrogen (LCoH₂) (i.e., the levelized cost per kg of hydrogen produced from a system (ignoring real estate cost and certain other external variables)) at least because such a scheme fails to take into account key factors (e.g., the constant fluctuations of feedstock prices) that impact cost. A control scheme that is entirely focused on electrical efficiency (and that does not consider the changing cost of electricity) may result in an unnecessarily high LCoH₂ at least in part because of: (i) consumption of a significant amount of feedstocks (e.g., electricity) when the feedstock prices are very high and/or (ii) underutilization of feedstocks (e.g., operate below a maximum hydrogen production capacity) when feedstock prices are near zero (or negative).

To address these and other similar issues, a control system of the electrolyzer can determine the cost and/or profit associated with producing a gas at each of a plurality of operating conditions. The control system can select the operating condition to use based on which operating condition is associated with the lowest cost and/or highest profit, even though that operating condition may not be associated with the highest electrical efficiency.

These embodiments can provide advantages over conventional electrolyzer designs that only focus on electrical efficiency. For example, a control system of an embodiment can use a cost model to generate estimated cost values that are used to determine target setpoints for the electrolyzer. In this way, the electrolyzer can be controlled to account for cost instead of electrical efficiency. The cost model can consider inputs such as the current state of process variable(s) of the electrolyzer, fixed cost(s), variable cost(s) (e.g., operation and maintenance costs and energy or other feedstock costs), and/or a capital recovery factor. Basing the control of the electrolyzer on the output of such a cost model or other mathematical relationships can result in an improved operation of the electrolyzer, as the cost of its output can be reduced or minimized.

In one disclosed example embodiment, a non-transitory computer-readable medium is provided that stores program instructions. The program instructions, when executed by one or more processors, cause the one or more processors to perform functions comprising: (a) calculating a cost or profit associated with producing a gas using one or more electrolyzers for each of a plurality of operating states of the one or more electrolyzers; (b) selecting one of the plurality of operating states based on a cost or profit consideration; and (c) causing the one or more electrolyzers to be configured according to the selected one of the plurality of operating states.

In another disclosed example embodiment, a method is provided for controlling an electrolyzer system. In this method, a stream is produced comprising hydrogen gas at least in part by (i) circulating a fluid through an electrolyzer cell that comprises two electrodes; and (ii) inducing a current between the two electrodes of the electrolyzer cell. While producing the stream, the following acts are performed: (i) identifying a cost associated with at least one feedstock of the electrolyzer system; (ii) for each of a plurality of operating states for the electrolyzer system, estimating a cost associated with the hydrogen gas based on the identified cost associated with the at least one feedstock; (iii) identifying a target operating state for the electrolyzer system based on the estimated cost associated with the hydrogen gas for each of the plurality of operating states; and (iv) controlling at least one controllable component of the electrolyzer system based on the target operating state.

In yet another disclosed example embodiment, an electrolyzer is provided comprising: an electrolyzer cell comprising a cathode and an anode, wherein the electrolyzer cell is configured to operate at a temperature in excess of 100 degrees Centigrade (C); thermal insulation at least partially covering the electrolyzer cell; a pump configured to circulate an alkaline solution through the electrolyzer cell; at least one processor; and at least one non-transitory computer-readable medium. The at least one non-transitory computer-readable medium comprise program instructions that are executable by the at least one processor such that the at least one processor is configured to: generate a cost estimate for hydrogen gas produced by the electrolyzer for each of a plurality of different operating states of the electrolyzer; select one of the plurality of different operating states based on the generated cost estimates; and control the electrolyzer to operate according to the selected operating state.

Other embodiments are disclosed, and the disclosed embodiments can be used alone or in combination.

These and other aspects are discussed in more detail in the passages that follow. It should be noted that the techniques described herein may be beneficial for electrolyzers that have dynamic setpoints and/or otherwise operate in a dynamic fashion. Some electrolyzers are grid-tied, get electricity at a fixed price as part of a contract irrespective of consumption, and operate at a constant setpoint 24/7. For those specific electrolyzer installations, the techniques described herein may not be usable in at least some respects.

2. Example Electrolyzer Systems

As mentioned above, hydrogen has a wide variety of applications and is used as an input for many industrial processes including, for example, ammonia production for fertilizers and oil refining. Currently, the majority of hydrogen produced in the United States is created through natural gas reforming. In natural gas reforming, air pollutants (e.g., carbon monoxide) are undesirably released that contribute to global warming. There is a trend to replace natural gas reforming (and other forms of hydrogen production that emit undesirable gasses) with green hydrogen production methods where no greenhouse gas emissions are produced.

Green hydrogen production, which makes up only about 1% of the hydrogen produced in the United States today, typically involves the generation of electrical energy through renewable sources (e.g., solar, wind, etc.), and that energy is used to split water into hydrogen and oxygen through electrolysis in an electrolyzer. Other examples of green production methods include pyrolysis, steam methane reforming (SMR) with carbon capture, thermolysis, and radiolysis.

FIG. 1A shows an example electrolyzer system 10 of an embodiment. As shown in FIG. 1A, the electrolyzer system of this embodiment comprises a stack 100 of one or more electrolyzer cells, each having an anode 110 and a cathode 120 (the electrodes) separated by a diaphragm or membrane (not shown). The electrolyzer cells in the stack 100 can be arranged in any of a variety of configurations. For instance, the electrolyzer cells can be arranged in parallel (e.g., where each cell has input(s) coupled to pump(s) and output(s)) or in series (e.g., where the cells are connected to each other and only the first cell has input(s) coupled to the pump(s) and only the last cell has output(s)). Another alternative arrangement is to have multiple smaller cell stacks where the cells within a cell stack are in series while the cell stacks themselves may be in series, parallel, or in an array of series and parallel connections.

As used herein, the term cell stack may refer to a set of multiple cells (e.g., electrolyzer cells) that are fluidly coupled together (e.g., share one or more common ports for circulating a fluid such as an electrolyte solution) and/or electrically coupled (e.g., electrically connected to a common set of one or more terminals of a power supply). In some instances, a given cell stack may comprise multiple sub-stacks. For example, a cell stack may comprise a set of cells that are fluidly coupled together and the cell stack may be divided into a first sub-stack that is electrically coupled together and a second sub-stack that is electrically coupled together. Similarly, a cell stack may comprise a set of cells that are electrically coupled together and the cell stack may be divided into a first sub-stack that is fluidly coupled together and a second sub-stack that is fluidly coupled together.

FIG. 1B shows an example implementation with multiple cell stacks (#1, #2, . . . , #n) that can be independently controlled. The ability to independently control the cell stacks can increase the dynamic range of the electrolyzer system 10. For instance, the number of active cell stacks may be adjusted in concert with adjustments to target production capacity (e.g., more cell stacks may be active for higher hydrogen production values and fewer cell stacks may be active for lower hydrogen production values). Such an increase in the dynamic range may advantageously provide a larger range of possible operating points that the control system described herein can optimize over. For illustration, 100 cells can be divided into 20x stacks connected in parallel where each of the 20x stacks has five cells connected in series. Other arrangements are also possible. Other arrangements are also possible.

Returning to FIG. 1A, the anodes 110 and cathodes 120 of the electrolyzer cell stack 100 are electrically connected to a power supply 150, which is controlled by a controller 160. The controller 160 can be in communication with computing device(s) 260 via a network 280. The controller 160 is used in this embodiment as part of a feedback system, where the controller 160 receives information on temperature(s), pressure(s), and/or tank level(s) and uses that information to control the output of the power supply 150 and/or to control the operation of pump(s) 170 (and/or other components of the system).

The controller 160 can take any suitable form. In one embodiment, the controller 160 comprises one or more processors, a non-transitory computer-readable medium, and program instructions stored on the non-transitory computer-readable medium that, when executed by the one or more processors, cause the one or more processors to perform the functions ascribed herein to the controller 160, as well as other/different functions, if desired. In other embodiments, the controller 160 takes the form of a pure-hardware implementation, including, for example, logic gates, switches, an application specific integrated circuit (ASIC), a programmable logic controller, or an embedded microcontroller. In yet other embodiments, the controller 160 comprises a pure-hardware implementation for some of its functions and a software-based implementation for other of its functions. Further, some or all of the controller 160 may be implemented in a server located remote from the electrolyzer and may be communicatively coupled to the electrolyzer (for control) and/or sensors (for receipt of input information) over a wide-area network (WAN), such as Internet, etc.

Pump(s) 170 circulate fluid from a fluid inlet 190 to the electrolyzer cell stack 100. The type of fluid (and the state-liquid phase or vapor phase) provided by the fluid inlet 190 can depend on, for example, the type of electrolyzer (e.g., alkaline electrolyzer, proton exchange membrane (PEM) electrolyzer, solid oxide electrolyzer, etc.) and/or the application for the electrolyzer. In some types of electrolyzers (e.g., an alkaline electrolyzer), the fluid inlet 190 may receive an alkaline solution (e.g., potassium hydroxide (KOH) or sodium hydroxide (NAOH)). In other types of electrolyzers (e.g., a PEM electrolyzer), the fluid inlet 190 may receive water (e.g., regular water or seawater).

In operation, the pump(s) 170, as controlled by the controller 160, circulate the electrolyte solution through the electrolyzer cell stack 100. As the fluid is circulated through the electrolyzer cell(s) in the stack 100, a current is induced between the anode 110 and the cathode 120 of the electrolyzer cell(s) by the power supply 150, which applies a voltage across the anode 110 and the cathode 120. The power supply 150 behaves like a current source in that the power supply 150 applies a sufficient voltage in order to achieve a particular target current level through the anode 110 and cathode 120 of the electrolyzer cell(s).

The current induced between the anode 110 and the cathode 120 of the electrolyzer cell(s) triggers an electrolysis reaction that generates first and second output streams 186, 188. The content of the output streams 186, 188 can vary based on the particular application and construction of the electrolyzer system 10. In water electrolysis applications, the first output stream 186 can comprise hydrogen gas while the second output stream 188 can comprise oxygen gas. In seawater electrolysis, the first output stream 186 can comprise hydrogen gas while the second output stream 188 can comprise oxygen and chlorine gasses. It should be understood that the two streams 186, 188 can contain impurities (e.g., some hydrogen in a stream that is primarily oxygen and some oxygen in a stream that is primarily hydrogen, which is common in existing electrolyzers used in industry). Accordingly, the first stream 186 can be said to be “substantially hydrogen gas,” and the second stream 188 can be said to be “substantially oxygen gas,” for example.

In an alternate design shown in FIG. 1C, the two streams 186, 188 are sent to gas separators 182, 184. In the case of water electrolysis applications, the gas separators 182, 184 separate out the hydrogen gas and the oxygen gas, respectively. The hydrogen gas can be collected, and the oxygen gas can also be collected (e.g., for use by other processes) or vented into the atmosphere.

In this design, the remaining electrolyte solution is recirculated through the electrolyzer cell by the pump(s) 170, thereby creating two closed loops. The electrolyzer system 10 can be constructed with a balanced pressure loop or with separate pressure loops to achieve a desired pressurization (e.g., 10 bar). In a balanced pressure loop design, the pressure in each side of an electrolyzer cell (e.g., the first half with the cathode and the second half with the anode) is balanced such that there is not a pressure differential (or such that there is a minimized pressure differential) across the diaphragm. Further, the flow rate of the electrolyte solution into each side of the electrolyzer cell may be independently controllable. In a separate pressure loop design, the pressure in each side of the electrolyzer cell may be independently controllable so as to create a pressure difference across the diaphragm to force gas from one side to the other (e.g., have hydrogen cross to the oxygen side or vice versa).

With separate pressure loops (and independent flow rate control of the electrolyte solution), there may be instances where the electrolyte solution is depleted in one side of the loop while the opposite side of the loop has extra electrolyte solution. With certain diaphragms (particularly those that are more porous), there can be migration of electrolyte solution from one loop to the other in the electrolyzer cells when the flow rates are made the same. Accordingly, this flow can be counteracted from one side of the loop to the other of electrolyte solution by adjusting the flow rates to counteract the flow of electrolyte from one side of the diaphragm to the other.

In some embodiments, the electrolyzer system 10 uses a cooling system (not shown) that removes heat generated by the electrolysis reaction to keep the operating temperature of the electrolyzer system 100 below a predefined maximum operating temperature, such as at or below 100 degrees Centigrade (C). Keeping the maximum temperature low (e.g., below 100 degrees C.) avoids a number of challenging design problems that come with higher-temperature operation. In other embodiments, instead of this low-temperature approach, steps can be taken to allow and/or facilitate high-temperature operation (e.g., by placing thermal insulation around the electrolyzer cell stack 100 and/or other components of the electrolyzer system 10 and/or by using thermally-insulated materials). Using a high-temperature alkaline electrolyzer designed to operate at temperatures above 100 degrees C. (e.g., 200 degrees C.) can offers a number of advantages including, for example, (1) better reaction kinetics (higher efficiency), (2) greater production density (better utilization of capex), (3) a reduction in parasitic losses associated with the eliminated cooling system (yet higher efficiency), (4) better dynamic operation capability (across a wider range of current densities) and (5) a cost and energy savings by removing the cooling system. Taken together, these advantages decrease the levelized cost of hydrogen produced and offer a substantial savings relative to lower-temperature designs.

FIG. 1D is a diagram of an alternative design in which multiple components of an electrolyzer system are integrated into an electrolyzer device 151 with a single housing 180 to simplify manufacture, customization, and maintenance.

In some implementations, the housing 180 may be 3D printed from various materials (e.g., polymeric materials such as PEEK, metallic materials such as stainless steel, ceramic materials such as zirconia and alumina, or any combination thereof). Utilization of a corrosion resistant material (e.g., PEEK) provides for high-temperature applications and simplifies maintenance and service in light of the removal of separate pipes and components relative to other implementations described above.

In some implementations, the housing 180 may be formed by multiple separate pieces that are bolted together or otherwise joined and/or fused together.

In addition to components described above with respect to other embodiments and/or implementations of an electrolyzer device, the electrolyzer device 151 as illustrated in FIG. 1D may also include channels 152, 156; a cavity 158; ports 162, 164, 166; terminals 172, 174, 176, conductors 163, 165, 167; and sensors 169.

Channels 152, 156 are channels in the housing 180 that may function as fluid conduits through which fluids can pass. It will be appreciated that channels 152, 156 replace pipes, hoses, and/or tubes in other electrolyzer designs.

Cavity 158 is an open space within the electrolyzer device 151 that can be used to form an electrolyzer cell when anodes 110 and/or cathodes 120 are incorporated.

Ports 162, 164, 166 refer to ports that allow fluid to pass into and out of the electrolyzer device 151. Such ports may include fittings or other structure to which external components may be removably coupled or otherwise attached.

Terminals 172, 174, 176 refer to electrical connection points that external electronics can be connected to the electrolyzer device 151. Examples of such external electronics include the controller 160 and/or the power supply 150.

Conductors 163, 165, 167 refer to metal conductors that may be integrated into the housing 180 to carry electrical signals and/or power to components within the electrolyzer device 151.

Sensors 169 refer to temperature, pressure, and/or level sensors that may be integrated into the housing 180. Sensors 169 may be electrically coupled to, for example, the controller 160. For instance, sensors 169 may be coupled to terminal 176 by conductor 167. In turn, the controller 160 may be coupled to the terminal 176.

The electrolyzer device 151 of FIG. 1D may operate similar to the electrolyzer device systems described above. However, fluid can be circulated through the electrolyzer device 151 of FIG. 1D in any of a variety of ways. In some implementations, the pump 170 can be integrated into the device itself. Alternatively, the pump 170 may be separate from the electrolyzer device 151. For example, the pump 170 may be coupled between the port 164 and the fluid inlet 190.

In yet other implementations, the pump 170 may be omitted altogether by leverage temperature gradients in the electrolyzer device 151. In these implementations, an electrolysis reaction may generate heat and/or pressure gradients that cause fluid to move from port 164 to ports 162, 166.

It should be noted that details of the electrolyzer system 10 and electrolyzer device 151 described above and below are merely examples and that different/additional components can be used. For example, an alternate electrolyzer system can have additional components between the outlets of the first and second streams 186, 188 and any industrial process that consumes a particular desirable gas that is generated (e.g., hydrogen and/or oxygen). For instance, there could be one or more components that purify and/or compress the desirable gasses (e.g., oxygen/hydrogen) leaving the outlets.

The following section provides an example high-temperature alkaline electrolyzer of an embodiment.

3. Example High-Temperature Alkaline Electrolyzer

a. Example System Design

Contrary to the low-temperature approach of existing solutions, the following embodiments take the approach of allowing and/or facilitating high-temperature operation. Modifying the design of the electrolyzer to accommodate higher temperatures offers a number of advantages including, for example, (1) better reaction kinetics (higher efficiency), (2) greater production density (better utilization of capex), (3) a reduction in parasitic losses associated with the eliminated cooling system (yet higher efficiency), (4) better dynamic operation capability (across a wider range of current densities) and (5) a cost and energy savings by removing the cooling system. Taken together, these advantages decrease the levelized cost of hydrogen produced and offer a substantial savings relative to conventional designs.

This countertrend approach is particularly suited for alkaline electrolyzers due to their abundance in the marketplace and their lower initial cost relative to other designs. (It should be noted that while these embodiments will be described in terms of alkaline electrolysis, it may also be possible to use these embodiments with other types of electrolysis, including, for example, salt water brine electrolysis (e.g. for producing hydrogen from seawater). While alkaline electrolyzers may have a lower electrical efficiency than other types of electrolyzers (e.g., PEM and/or AEM electrolyzers), the typical measure of electrical efficiency omits the power consumption of the cooling system. In many electrolyzers, as much as 50% of the power consumption of the electrolyzer system is consumed by the cooling system alone. Accordingly, the slightly-lower energy efficiency is more than entirely offset by the energy savings from not having a cooling system to yield a lower total electrical energy consumption (taking into account the consumption of the cooling system) per unit of hydrogen produced.

FIG. 2 shows an example relationship between levelized cost of hydrogen and operation temperature for an alkaline electrolyzer that was calculated using data from a publication (“High temperature and pressure alkaline electrolysis,” JC Ganley, Int. J. Hydrogen Energy 2009, volume 34, Issue 9, pages 3604-3611). In FIG. 2, the Y-axis shows the percentage change in levelized cost of hydrogen (lower is better) relative to a benchmark that is an alkaline electrolyzer operating at the normal operating temperature of 80 degrees C., the X-axis shows the current density in milliamperes per centimeter squared, and the various lines show the relationship between levelized cost of hydrogen and current density at different operating temperatures in degrees Centigrade (35, 80, 200, 250, 350, 400, and 450).

As shown in FIG. 2, there is a significant reduction (˜25%) in the levelized cost of hydrogen when operating at 200 degrees C. relative to operating at 80 degrees C., and there is a diminishing benefit from further increases in operating temperature. Given that the challenges posed by higher operating temperatures scale with the temperature (i.e., the higher the temperature, the more complex it is to design the electrolyzer), there are tradeoffs that come with operation at these successively-higher temperatures. Accordingly, the target for the design of one embodiment is to achieve operation at or around 200 degrees C. in order to capture the sizable levelized cost of hydrogen (LCoH₂) benefit.

FIG. 3A is a high-level diagram of an example high-temperature alkaline electrolyzer of an embodiment that is designed to operate at temperatures above 100 degrees C. (e.g., 200 degrees C.). As shown in FIG. 3A, the high-temperature alkaline electrolyzer of this embodiment comprises a stack 100 of one or more electrolyzer cells, each having an anode 110 and a cathode 120 separated by a diaphragm 130. The electrolyzer cell stack 100 is surrounded by thermal insulation 140 to reduce thermal losses and help increase the temperature and achieve the benefits discussed above. To simplify the drawings, the thermal insulation 140 is depicted as a relatively-small area compared to the electrolyzer cell stack 100. However, it should be understood that the thermal insulation 140 can cover all or substantially all of the electrolyzer cell stack 100. Further, instead of or in addition to using thermal insulation for the electrolyzer cell stack 100 and/or other components of the electrolyzer system, various components in the system can be made of a thermally-insulating material.

The electrolyzer cells in the stack 100 can be arranged in any of a variety of configurations. For instance, the electrolyzer cells can be arranged in parallel (e.g., where each cell has input(s) coupled to pump(s) and output(s) connected to gas separators) or in series (e.g., where the cells are connected to each other and only the first cell has input(s) coupled to the pump(s) and only the last cell has output(s) coupled to the gas separators). Another alternative arrangement is to have multiple smaller cell stacks where the cells within a cell stack are in series while the cell stacks themselves may be in series, parallel, or in an array of series and parallel connections. Other arrangements are also possible. For illustration, 100 cells can be divided into 20x stacks connected in parallel where each of the 20x stacks has five cells connected in series.

Any suitable design can be used for the anode 110 and cathode 120 of an electrolyzer cell. For example, a porous anode can be used that is chemically compatible with KOH at elevated temperatures and offers high electrical performance. The anode can be made of sheet metal, metal foam or wire mesh, in one of the following example materials: (1) Inconel, (2) porous nickel-cobalt, (3) cobalt oxide, (4) a perovskite, such as La-Sr-CoO3 on porous nickel, or (5) silver nanowires on a metal foam, such as NiFeCrAl, porous nickel, or Raney nickel. Any of the above materials can be coated as a catalyst onto another conductive substrate (e.g., nickel, copper, stainless steel, graphite) in porous or non-porous form.

Regarding the cathode, a porous cathode can be used that is chemically compatible with KOH at elevated temperatures and offers high electrical performance. The cathode can be based from sheet metal, metal foam or wire mesh, in one of the following example materials: (1) Inconel®, (2) Raney nickel, (3) titanium or molybdenum-doped porous nickel, or (4) Raney nickel-cobalt. Any of the above materials can be coated as a catalyst onto another conductive substrate (e.g., nickel, copper, stainless steel, graphite).

As shown in FIG. 3A, the anodes 110 and cathodes 120 of the electrolyzer cell stack 100 are electrically connected to a power supply 150, which is controlled by a controller 160.

The controller 160 is used in this embodiment as part of a feedback system, where the controller 160 receives information on temperature(s), pressure(s), and tank level(s) and uses that information to control the output of the power supply 150 and also to control the operation of pump(s) 170 (and, optionally, other components of the system). The electrolyzer of this embodiment also comprises an inlet 190 for an electrolyte solution (e.g., potassium hydroxide (KOH) or sodium hydroxide (NaOH)), and first and second gas separators 182, 184 with respective outlets 186, 188. The fluid conduits (e.g., pipes) that circulate the electrolyte solution can also be insulated with thermal insulation 142, 144, 146, and 148 (or made from a thermally-insulated material) to achieve the high-temperature benefits discussed above. To simplify the drawings, the thermal insulation 142, 144, 146, and 148 is depicted as relatively-small areas compared to the conduits, but it should be understood that the thermal insulation 142, 144, 146, and 148 can cover all or substantially all of the conduits.

In operation, the pump(s) 170, as controlled by the controller 160, circulate the electrolyte solution through the electrolyzer cell stack 100 and the gas separators 182, 184. For each of the cells in the electrolyzer cell stack 100, a current is driven through the anode and cathode by the power supply 150 (e.g., a DC power supply, a pulsed power supply, etc.) to drive the electrolysis reaction and create hydrogen gas at the cathode and oxygen gas at the anode. In this embodiment, the heat from the electrolysis reaction is trapped in the system using the thermal insulation 140-148 to achieve a higher operating temperature and its associated benefits. The hydrogen gas is separated from the electrolyte solution using the first gas separator 182 and collected while the electrolyte solution is recirculated through the electrolyzer cell by the pump(s) 170. Similarly, the oxygen gas is separated from the electrolyte solution using the second gas separator 184, and the electrolyte solution is recirculated. The oxygen gas may be collected (e.g., for use by other processes) or vented into the atmosphere.

It should be noted that the two streams can be “substantially oxygen gas” and “substantially hydrogen gas,” as impurities can be present, including some hydrogen in oxygen and some oxygen in hydrogen, which is common in existing electrolysers used in industry. Also, in some embodiments, the inlet stream can be purified (or distilled/deionized) water rather than an electrolyte solution. The salt (e.g. KOH) can circulate around the two halves of the system, and as water is removed from that solution in the form of gas production from the electrolytic cell, that water is replenished at the same rate at the inlet. So, a circuit can be formed that processes the high-concentration electrolyte from the “outlet” (downstream of the cell (e.g. >45% wt. KOH) in a black box that returns set-concentration electrolyte (e.g. =45% wt. KOH) to the circuit.

In this embodiment, the electrolyzer is pressurized to, for example, 10 bar, to keep the electrolyte solution in a liquid state at these higher temperatures. Additionally (or alternatively), the concentration of the electrolyte solution may be selected to further assist in keeping the electrolyte solution in a liquid state at higher temperatures (e.g., higher concentrations of the electrolyte solution remain a liquid at higher temperatures than lower concentrations). For instance, the electrolyte solution may be 45% KOH instead of 30% KOH as might typically be used in other applications. Keeping the electrolyte solution in a liquid state can help avoid the following complications: (1) dealing with the more-complicated dynamics involved in a multiphase or vapor system and (2) needing to use more-expensive and fouling-prone components, such as gas diffusion electrodes.

The electrolyzer system can be constructed with a balanced pressure loop or with separate pressure loops to achieve the desired pressurization (e.g., 10 bar). In a balanced pressure loop design, the pressure in each side of an electrolyzer cell (e.g., the first half with the cathode and the second half with the anode) is balanced such that there is not a pressure differential (or such that there is a minimized pressure differential) across the diaphragm. Further, the flow rate of the electrolyte solution into each side of the electrolyzer cell may be independently controllable. In a separate pressure loop design, the pressure in each side of the electrolyzer cell may be independently controllable so as to create a pressure difference across the diaphragm to force gas from one side to the other (e.g., have hydrogen cross to the oxygen side or vice versa).

With separate pressure loops (and independent flow rate control of the electrolyte solution), there were instances where the electrolyte solution would be depleted in one side of the loop while the opposite side of the loop had extra electrolyte solution. With certain diaphragms (particularly those that are more porous), there was migration of electrolyte solution from one loop to the other in the electrolyzer cells when the flow rates were made the same. Accordingly, this flow can be counteracted from one side of the loop to the other of electrolyte solution by adjusting the flow rates to counteract the flow of electrolyte from one side of the diaphragm to the other.

In some implementations, filters 290, 292 may be integrated into the loops to remove particulate caused by, for example, corrosion of components within the electrolyzer system. The loops may, in some instances, include a bypass for the filters 290, 292 (as shown in FIG. 3A) to, for example, facilitate service and/or replacement of the filters during operation. In other instances, bypasses for the filters 290, 292 may be omitted. The filters 290, 292 may be specially constructed from materials that can withstand highly caustic environment. For instance, the filters 290, 292 may be constructed from polymeric material (e.g., PEEK) mesh with pore sizes that go from large to small (e.g., larger near the entrance to the filter and smaller near the exit of the filter).

It should be noted that FIG. 3A is a high-level diagram and that other implementations can be used. An example of another, more-detailed implementation is shown in FIG. 3B. It should be understood that this is merely an example, and that other/different components can be used. As shown in FIG. 3B, this embodiment contains several components not shown in the embodiment in FIG. 3A. For example, in this embodiment, the streams output from the gas separators 182, 184 are passed through condensers 202, 204 and tanks 206, 208 to purify and/or compress the desirable gasses (e.g., oxygen/hydrogen). This embodiment also comprises a bath heater 210 to heat the electrolyte solution prior to the electrolyte solution entering the electrolyzer stack 100, and a pump head cooling system 220 (connected with water chillers and pumps 261) to help cool the circulation pumps 170. FIG. 3B also shows that a peristaltic pump 230 is used to pump the electrolyte solution from the reservoir inlet 190, and that a pressure pump 240 is used to pump argon gas from a tank 250 into the electrolyte solution stream. These features will be discussed in more detail below. Lastly, FIG. 3B shows a data logger 270 that can be used to collect various data from the system and provide it to a network 280 for analysis. Also, this network 280 can be used, as noted above, when some or all of the controller 160 may be implemented in a server, located remote from the electrolyzer, and may be communicatively coupled to the electrolyzer (for control) and/or sensors (for receipt of input information) over the network 280.

b. Insulation/Heating Examples

The electrolysis reaction is endothermic, and the overpotential above the thermoneutral voltage (e.g., losses from operating at higher current densities due to impedance (not the electrolysis reaction)) generates heat. Conventional systems are generally designed to disseminate this heat (not trap/preserve heat as in this embodiment) using a cooling system. In contrast, as mentioned above, the electrolyzer of this embodiment omits a cooling system and, instead, employs thermal insulation around components (or thermally-insulated components) to enable increased operating temperatures. So, in this embodiment, heat is kept within the electrolyzer as a passive heating source to achieve higher operating temperatures. For instance, the electrolysis cells may be insulated in addition to any pipes and/or tanks that carry the electrolyte solution to trap heat. The operating temperature of the electrolyzer cell can be adjusted by modifying the output of the power supply 150 to the electrolyzer cell (e.g., changing the current being applied to the anode/cathode to reduce or increase the temperature).

In some instances, the insulation may be constructed to be chemically resistant to spills of the electrolyte solution that may occur through normal operation. In one example, the insulation itself may be constructed from a material that is chemically resistant (e.g., a ceramic fiber). In another example, the insulation itself may be constructed from an insulating material that is not chemically resistant to the electrolyte solution (e.g., fiberglass, etc.) but that is wrapped in another material that is chemically resistant (and may or may not be insulating).

Optionally, the alkaline electrolyzer may additionally include an active heat source (e.g., the bath heater 210) that may be used when warranted in conjunction with the passive heat source. For instance, the active heat source may be used during startup periods to achieve a desired operating temperature at greater speed. The active heat source may be constructed using any of a variety of available technologies including, for example, a resistive heater, an electrical heat pump, recovered waste heat from another system or process, or a solar thermal source.

In designs that employ an active heat source, the active heat source may be advantageously placed proximate to an inlet of the electrolyzer cells to: (1) enable tighter control of the temperature of the electrolyte solution entering the electrolyzer cells (e.g., so as to better control the electrolysis reaction in the electrolyzer cells themselves) and (2) minimize the thermal losses between the heater and the electrolyzer cell. For instance, as shown in FIG. 3B, a bath heater 210 can be placed between the pump(s) 170 and the inlet to the electrolyzer cell stack 100.

Optionally, the alkaline electrolyzer may include a heat storage device. The heat storage device may be employed to store excess heat when operating at high temperatures and provide heat to the electrolyte solution when needed. The heat storage device can be used in combination with an active heat source or in place of an active heat source.

c. Example Diaphragm Designs

A conventional diaphragm for an alkaline electrolyzer is designed for continuous operation at only 80 degrees C. and typically has a maximum operating temperature around 100 degrees C. A Zirfon® diaphragm is an example of a conventional alkaline electrolyzer diaphragm. That diaphragm is 500 microns thick, highly wettable (due to its 85% Zirconia construction) which ensures ionic permeability, and uses a polysulfone composite to improve mechanical properties (making it feel a bit like rubber). It is constructed of a woven polymer support, with the ZrO2-polysulfone composite material film-cast symmetrically to both sides of the weave. The composite membrane “core” has manufactured pores using self-assembly methods (i.e. chemical processing) and features pores of around 100 nanometers (nm) in diameter with industry-leading tolerances of +/−15%.

It has been observed in laboratory testing that operating a conventional Zirfon® diaphragm at temperatures above the design limits caused the diaphragm to deform through contraction. The contraction is caused by polysulfone in the Zirfon® diaphragm that shrinks when exposed to high temperatures. Accordingly, a different type of diaphragm may be needed to operate in a more-corrosive environment at the higher temperatures and pressures used in a high-temperature alkaline electrolyzer.

Various features can be considered in designing a diaphragm for a high-temperature alkaline electrolyzer. For example, it may be desired for the diaphragm to be chemically stable for a chosen electrolyte (e.g., 45% aqueous KOH) at up to a maximum operating temperature (e.g., 200 degrees C., 250 degrees C., etc.). Performance can be measured (e.g., after some period of time) as a maximum percentage degradation of the nominal parameters described below. It may also be desired for the diaphragm to be mechanically stable (e.g., so it does not shrink, melt, or otherwise degrade) at up to the maximum operating temperature (e.g., 200 degrees C.), which can be measured in a shrinking/swelling tolerance and under a pressure differential within the design limits (if any) (which can be measured in a yield strength). Further, it may be desired for the diaphragm to have high ionic permeability (for OH− and H+ ions to pass through) (which can be measured in F/m) and have low gas permeability to keep oxygen and hydrogen gas products on their respective “sides” (which can be measured in L/min.cm^2 at, e.g., 5 bar). As yet another consideration, it may be desired for the diaphragm to be as thin as possible (e.g., 500 microns thick or less with a 50 micron tolerance) to keep the electrode gap as small as possible, which minimizes the ohmic part of cell impedance.

Polytetrafluoroethylene (PTFE) may be a suitable material to satisfy at least some of these considerations for up to around 150 degrees C., and polyetheretherketone (PEEK) may be a suitable material up to 200 degrees C. Many implementations are possible, including using polyetheretherketone (PEEK)/polytetrafluoroethylene (PTFE) (woven (e.g., with ˜356 threads per inch (tpi)), or a porous PTFE film) that is coated in zirconium dioxide (ZrO2) as the diaphragm. As another example, just a polymeric PEEK weave/porous PTFE film is used with no Zirconia approach (e.g., (1) amending woven PEEK or a PTFE film by “squishing” it at temperature to deform the material and create finer pores, (2) layering woven PEEK to reduce the effective pore size, and/or (3) using a single layer without any modification like layering or squishing). As yet another example, a composite ceramic can be used that is a combination of a Zirconia-like flexible ceramic fiber (FCF) construction with a polymeric additive to improve mechanical properties of the diaphragm. That composite can then be applied to a weave or other structural component (e.g., constructed from polyetheretherketone (PEEK)/polytetrafluoroethylene (PTFE)). Other examples include sandwiching a ceramic-like pure Zirconia (or Yttria-Stabilized Zirconia (YSZ)), Alumina, or other ceramic felt between layers of PEEK, as well as sandwiching a ceramic felt between porous mesh or gas diffusion electrodes in a “zero-gap configuration” (i.e., no gap between the electrodes and the diaphragm).

In other implementations, a diaphragm may comprise woven ceramic fibers, which may provide for enhanced performance relative to PEEK alone in terms of ohmic losses along with reducing and/or limiting undesirable gas crossover in the system. Example ceramic materials from which a diaphragm can be constructed include alumina oxide ceramic (e.g., Nextel 610 DF-11 by 3M®) and a zirconia oxide ceramic with yttrium (e.g., Zircar Zirconia woven cloth). In some implementations, ceramic fibers could be combined with a structural component like PEEK or other polymeric material. However, in other implementations, the fabric could be used alone.

Further, it should be appreciated that the alkaline electrolyzer can be constructed without a diaphragm altogether. There are a number of designs that exist to remove the need for a diaphragm to facilitate gas separation. For instance, one can use a Pressure Swing Adsorption system on the output streams to separate the oxygen from the hydrogen. The levelized cost of the Pressure Swing Adsorption system to separate oxygen gas from hydrogen gas may be significantly less than the levelized cost added by a diaphragm. Another approach is described in US Patent Publication No. 2016/0312370 titled “Electrochemical Cell Without an Electrolyte-Impermeable Barrier.” Accordingly, there are a variety of approaches to achieve a diaphragm-less design, and the design of the electrolyzer cells is not limited in this respect.

Returning to the drawings, FIGS. 4-10 are illustrations of various example diaphragms of an embodiment. FIG. 4 is a front view of one example diaphragm, but it should be noted that the diaphragm can have any suitable shape (e.g., circular, rectangular, square, etc.) depending on the construction of the cell. FIG. 5 is a cross-sectional view of an example diaphragm 130 of an embodiment having a central support structure 310 (e.g., comprising a polymeric material) with ceramic material 300 (e.g., alone or with a polymeric material) on either side of the support structure 310. (The ceramic material on either side may be bonded or otherwise joined together through the pores of the support structure). The thickness of the support structure 310 may vary according to the thickness of a polymeric material and/or the number of layers of polymeric material that are used. In one example, the thickness of the support structure 310 is less than 350 microns. In another example, less thickness may be sufficient to provide structural support. According to such an example, the thickness of the support structure 310 may be approximately 50 microns, or may be somewhere within the range of between 50 microns and 350 microns. In yet another example, the thickness of the support structure 310 may be greater than 350 microns. Other examples may exist.

In the variation shown in FIG. 6, the ceramic material 300 is sandwiched between layers of the support structure 310. In FIGS. 7 and 8, the ceramic material 300 is on only one side of the support structure 310. In yet another variation, FIG. 9 shows the diaphragm 130 only having a support structure 310 with n layer(s) of polymeric material (represented by the three dots in FIG. 9), whereas FIG. 10 shows the diaphragm 130 only having the ceramic material 300 (alone or with polymeric material). In FIG. 10, the ceramic material (e.g., a ceramic felt) can be sandwiched between the electrodes in a zero-gap configuration, such that there is no gap between the diaphragm 130 and the electrodes, as discussed above.

Regarding material selection, a polymeric material can be used, and thermoplastics are a type of polymeric material. Polyetheretherketone (PEEK) is a type of thermoplastic, and polytetrafluoroethylene (PTFE) is another type of thermoplastic. Example polymeric materials include thermoplastics such as polyetheretherketone (PEEK) and polytetrafluoroethylene (PTFE). These polymeric materials may be constructed in any of a variety of forms such as a weave (e.g., having a certain number of thread per inch (tpi)), a film, or a sheet. Similarly, an example ceramic material includes zirconium dioxide (ZrO2).

In some implementations, an average pore size of a diaphragm may be dynamically adjustable. For example, the diaphragm may have two or more pieces that can be rotated relative to each other to change the average pore size (e.g., two diaphragms each made from a woven material that can be aligned to achieve a first average pore size or rotated by a certain angle (e.g., 45 degrees) to achieve a second average pore size). In another example, the diaphragm may be mechanically stretched or otherwise elongated to achieve different average pore sizes (e.g., a smaller average pore size in a relaxed state and a larger average pore size in an elongated state). In yet another example, the diaphragm may be constructed from an electroactive material (e.g., electroactive polymeric material) that exhibits a change in physical characteristics based on electrical stimulus.

Such adjustments to the average pore size may be made during operation to minimize the gas crossover from the oxygen side to the hydrogen side (e.g., during transient events or changes in operating conditions). For instance, the average pore size may be reduced during transient power events (e.g., brief substantially decreased in available power) to mitigate the risk of undesirable gas crossover.

d. Example Material Selection for Pipes/Tanks/Gaskets

In a conventional alkaline electrolyzer, the electrolyte solution is at a relatively-low temperature (e.g., 80 degrees C.) and at a relatively-low pressure (e.g., ambient pressure). In such an environment, pipes and tanks may be constructed from a relatively-low-grade stainless steel (e.g., 304 stainless steel) while gaskets may be constructed from rubber. With the high-temperature alkaline electrolyzer of these embodiments, these components will be exposed to an electrolyte solution that is at a significantly-higher temperature (e.g., 200 degrees C.) and significantly-higher pressure to keep the electrolyte solution in a liquid state (e.g., 10 bar). Further, the corrosiveness of the electrolyte solution increases with temperature, which results in a significantly-more-corrosive environment. Given these harsh conditions, typical materials employed for construction of pipes, tanks, and gaskets may not be suitable for use in a high-temperature alkaline electrolyzer. As such, different materials can be employed for piping, tanks, and/or gaskets to operate in a more-corrosive environment at higher temperatures and pressures.

For example, for piping and tanks, 316 or 316L stainless steel can be used, as this grade of stainless steel offers enhanced corrosion resistance relative to the more-typical 304 stainless steel. In another example, piping may be constructed with a metal casing (e.g., 316 stainless steel) that has an inner lining made from a corrosion resistant material such as PTFE. Pipe fittings may be constructed similarly (e.g., metal such as 316 stainless steel that is lined with PTFE or other corrosion resistant material). In yet another example, Inconel is utilized for piping.

Similarly, in another example, tanks may include specially treated, coated, and/or lined metals such as a black nitride treated carbon steel and carbon steel with a PTFE or paraformaldehyde (PFA) coating/lining.

Any of a variety of other materials can be employed to construct the pipes and tanks that can handle the temperature, pressure, and corrosiveness of the electrolyte solution. Other suitable materials include, but are not limited to, high-end metals (e.g., Inconel®), polymers (e.g., polyetheretherketone (PEEK), polytetrafluoroethylene (PTFE), etc.), and/or ceramics. Also, although not shown in FIG. 3B to simplify the drawings, a filter can be used in the electrolyte loop to remove corrosion products (e.g., powder/particles/ions) that result from the high corrosivity.

For gaskets, virgin Teflon may be utilized. However, polytetrafluoroethylene (PTFE) or polyetheretherketone (PEEK) can also be used. Performance of a gasket in this harsh environment can be further improved by evenly torquing down the gasket (e.g., in a crosswise pattern, avoiding torquing corner bolts) and applying high-vacuum grease that is inert (e.g., as a sealant). Any of a variety of other materials can be employed to construct the gaskets that can handle the temperature, pressure, and corrosiveness of the electrolyte solution in addition to being non-conductive and compressible.

In one example implementation, the main components (e.g., fluid conduits such as piping, cell body, tanks, valves, etc., or a subset thereof) of the system are made from PEEK or similar polymer that (a) is electrically insulated, (b) corrodes negligibly in the presence of concentrated electrolyte (e.g., 45% KOH) as well as hydrogen and oxygen, and (c) is a better thermal insulator than steel, thus minimizing or avoiding the need for external insulation to be applied. Alternatively, the material used may be a ceramic with low brittleness or metals that have low corrosion rates, such as, for example, Alloy 400, Alloy 690, Alloy 800, or Pure Nickel (the latter in the presence of iron or silicon in the electrolyte, which reduces corrosion rate of nickel to 1/100th). Another option is to use PEEK or ceramic components encased in or supported by an outer metal frame. The frame can be low grade steel that will account for mechanical strength and the PEEK will account for corrosion protection.

In certain examples, one or more components of the electrolyzer system may be advantageously constructed from a non-conducting material to minimize (or altogether eliminate) undesirable electrical characteristics such as shunt currents (e.g., arising when an electrolyte solution is fed to an electrolyzer cell stack comprising multiple cells connected in series). For instance, the issue of shunt current can be mitigated (or avoided altogether) through the use of non-conducting materials (e.g., polymeric materials such as PEEK and/or ceramic materials) for the conduits leading the electrolyte solution to/from the individual cells within the cell stack (alone or in combination with with separation of the electrolyte solution circuits for the anode and cathode sides of each cell).

e. Example Pump and Pump Cooling Designs

FIG. 11A is a diagram of an example magnetic drive pump that may be utilized with, for example, alkaline electrolyzers. It will be appreciated that pumps (see, e.g., pump 170) may need to operate in a harsh environment that is highly caustic for long periods of time. Such a challenging environment causes pumps using conventional constructions to prematurely fail. One design that lasts longer in such a caustic environment is to employ a specially constructed magnetic drive pump.

A magnetic drive pump may include a housing 1104; a pump motor 1102; a rotating member/shaft 1106 that is rotated by the pump motor 1102; and magnets 1108 positioned on the rotating member/shaft 1106. The magnetic drive pump further includes an impeller 1112 and magnets 1110 positioned on the impeller. In some implementations, the housing 1104 of the magnetic pump and the impeller 1112 may be constructed from corrosion resistant materials such as PEEK, nickel alloys, and 316 stainless steel.

During operation, the pump motor 1102 rotates the rotating member/shaft 1106 along with the magnets 1108 positioned on the rotating member/shaft 1106. The magnets 1108 of the rotating member/shaft 1106 complement the magnets 1110 of the impeller 1112 such that as the pump motor 1102 rotates the magnets 1108 of the rotating member/shaft 1106, the magnets 1110 of the impeller 1112 also rotate. The rotation of the magnets 1110 of the impeller 1112 causes the impeller 1112 to rotate and move fluid through the magnetic drive pump.

Conventional alkaline electrolyzers typically employ passively-cooled (i.e., ambient-air cooled) pumps to circulate the electrolyte solution though the electrolyzer. Passively-cooled pumps may be appropriate for these conventional alkaline electrolyzers because the temperature of the electrolyte solution is kept low with an active cooling system. With the high-temperature alkaline electrolyzer of these embodiments, the temperature of the electrolyte solution is significantly higher and exceeds the maximum operating temperatures of typical passively-cooled pumps. Accordingly, in some embodiments, the pump may have an active cooling system to remove excess heat from the pump to avoid overheating the motor and internal components.

FIG. 11B is a diagram of an example pump cooling system 220 of an embodiment that may be suitable for use with an electrolyte solution at about 200 degrees C. As shown in FIG. 11B, in this embodiment, the pump head 175 is submerged in a water bath 405 of a heat exchanger 400, where the water 405 is circulated using a recirculation pump 420 to prevent stratification. A coil of (e.g., quarter-inch) copper tubing surrounds and circulates chilled water around the pump head 175. The chilled water may be input feedwater to the system, where the cooling of the pump performs the double function of preheating the feedwater, thus acting as a heat recovery system. Cool air (e.g., via a compressed air line) is also directed at the pump 170 to provide further cooling to the motor. As illustrated in FIG. 11B, the effect of this pump cooling system 220 is to lower the temperature of the electrolyte solution as it passes through the pump 170. Even though the electrolyte solution is cooled as it passes through the pump 170, it can still be at a relatively-high temperature during electrolysis to achieve the benefits discussed above.

It should be understood that the pump cooling system 220 in FIG. 11B is merely one example and that other ways can be used to provide active cooling for a pump. For example, a pump can be designed with a housing having cavities through which a cool liquid can be circulated to remove excess heat.

f. Example Operational Phases

As shown in the flow chart of FIG. 12, in one embodiment, the operation of a high-temperature alkaline electrolyzer can be divided into four operational phases: a startup phase (1210) where the electrolyzer is getting to temperature and pressure, a normal operation phase (1220) where the electrolyzer is operating at temperature and pressure, a standby phase (1230) where the electrolyzer stops producing hydrogen temporarily and maintains pressure and temperature so as to be able to return to normal operation phase expeditiously, and a shutdown phase (1240) where the electrolyzer shuts down. These phases will now be discussed in conjunction with the flow charts of FIGS. 13-16.

Turning first to the startup phase, since the high-temperature electrolyzer is operating at a significantly-higher pressure (e.g., 10 bar) than a typical electrolyzer operating at ambient pressure, the startup procedure may be different so as to create the requisite pressure in the electrolyzer cell. As shown in FIG. 13, the controller 160 can cause the pressure to rise to a target value (1310). For instance, the electrolyzer may have a valve (e.g., a check valve) that allows the pressure to build in the electrolyzer cells during startup until they reach the target pressure (e.g., approximately 9.8 bar, approximately 10 bar, etc.). In another instance, the electrolyzer may expedite the process of pressurizing the electrolyzer cells. One way to increase the pressure during startup is to mix an inert gas (e.g., argon) in with the electrolyte going into the electrolyzer cell, as shown in FIG. 3B. In such an instance, all of the valves (except the inert gas feed-in valve(s)) may be shut until the target pressure is reached. After the target pressure is reached, the inert gas feed-in valve(s) may be closed and the inert gas may be purged from the high-temperature electrolyzer (e.g., to avoid mixing the inert gas into the valuable hydrogen output). The inert gas may be purged in any of a variety of ways. For instance, the high-temperature electrolyzer may (e.g., using relief valves set to the target operating pressure) vent one or more output streams until a concentration of the inert gas falls below a threshold (e.g., as measured by one or more sensors).

Another way to pressurize the high-temperature electrolyzer, which is also shown in FIG. 3B, is to pressurize the water going into the oxygen/hydrogen collection tanks to compress the air in the tanks and, as a result, the entire loop. Further, pressure can be allowed to build from the generation of hydrogen/oxygen gasses from normal operation (e.g., circulating the electrolyte and producing hydrogen gas).

Another aspect of startup operation that may be different from conventional electrolyzers is the benefit of rapidly increasing the temperature during startup. The LCoH2 of operating at higher temperatures may be lower (better) than at lower temperatures. Accordingly, there may be benefits to getting the electrolyte solution to temperature quickly. As shown in FIG. 13, the controller 160 can cause the temperature to rise to a target value (1320). This can involve, for example, using heat from an active heat source, a heat storage device, and/or heat exchanger to increase the temperature of the electrolyte and/or allow heat to build from the electrolysis reaction in normal operation (e.g., circulate the electrolyte and produce hydrogen gas). In designs that implement a heat storage system (or an active heating system), additional heat may be applied to the electrolyte solution to rapidly increase the temperature of the electrolyte solution (e.g., faster than using passive heat sources alone). Conversely, the active heat source and/or heat store can be turned off when the electrolyte solution has reached the desired operating temperature.

Turning now to the normal operation phase (see FIG. 14), the electrolyte solution is circulated through electrolyzer cell(s) using the pump (1410), which produces a stream comprising hydrogen gas (1420). The hydrogen gas is then separated from the stream, and the electrolyte is recirculated (1430). (Optionally, water can be separated out using a condenser.) Then, hydrogen gas is outputted for storage and/or use by another process (1440).

In the standby phase (see FIG. 15), the electrolysis reaction is stopped, and heat is retained (1510). The electrolysis reaction can be stopped, for example, when the thermoneutral potential is reached (i.e., the minimum voltage across the electrodes where the electrolysis reaction is endothermic, and the overpotential above the thermoneutral voltage heats up the electrolyte solution). This keeps the heat in the electrolyte solution instead of allowing operation to continue while the voltage across the electrodes is below the thermoneutral potential and above the equilibrium potential (i.e., the minimum voltage across the electrodes to trigger the electrolysis reaction), where the electrolyte solution temperature drops (i.e., heat is removed). If a heat storage element is included in the design, the heat from the electrolyte solution could be stored in the heat storage element and returned just prior to returning to the normal operation process.

Finally, in the shutdown phase (see FIG. 16), heat is removed (1610), and the electrolysis reaction is stopped (1620). Heat can be removed from the electrolyte solution, for example, by operating the electrolyzer cell below the thermally-balanced voltage to cool the electrolyte solution

Additionally, as mentioned above and as shown in FIG. 17, the controller 160 of this embodiment can implement a control loop (e.g., a feedback control loop, a feedforward control loop, or a combination thereof) where process variable(s) (and, optionally, other inputs) are monitored (1710), a setpoint is determined for the process variable(s) (and, optionally, other inputs) (1720), and control signal(s) are generated for the controllable elements in the system (1730). The values for the setpoints may be determined in accordance with a higher-level control scheme.

g. Example Heat Exchangers

In one embodiment, one or more heat exchangers are used to transfer heat from one or both of the output streams to the feedwater. By using the feedwater to perform cooling in the condenser (e.g., by preheating the feedwater with heat from the electrolyzer output streams) via the condenser coils acting as a heat exchanger, there is a vast reduction in waste heat and the need for external cooling is vastly reduced or eliminated. Additionally or alternatively, the heat exchanger(s) can be used to transfer heat from one or both of the output streams to the electrolyte solution that is being fed into the electrolyzer. This allows the output streams to be cooled simultaneously with heating the electrolyte solution. Any suitable configuration can be used. For example, the heat exchanger 1800 can be positioned to transfer heat from a first stream (comprising hydrogen) to the electrolyte solution (FIG. 18), from the second stream (comprising oxygen) to the electrolyte solution (FIG. 19), or from both the first and second streams to the electrolyte solution (FIG. 20).

h. Example Electrolyzer Cell and Cell Stack Designs

The electrolyzer cell can be configured not in a “zero gap” configuration (as is commonly pursued) but with a small gap (e.g., 0.2 mm) between the diaphragm and each electrode to avoid the issue of reducing the electrode active area from contact with a hydrophobic diaphragm material. The gap on the cathode side can be larger (e.g., double) than what it is on the anode side to reduce the differential pressure across the separator from the 2x production of hydrogen versus oxygen in the cell. FIG. 21 depicts an example of gaps between the anode/cathode and a diaphragm of an embodiment.

In one embodiment, a gas separation diaphragm can be made of a single layer or multiple layers of microporous PEEK or a similar chemically-compatible material that is mechanically stable and not electrically conductive, wherein the layers are stacked randomly so as to misalign the pores and create a smaller effective pore size. The diaphragm may be a woven material (e.g., a 356 tpi woven PEEK material). The material can be mechanically pressed at a temperature above the material's glass transition temperature but below its melting temperature to reduce the pore size of the material and/or add rigidity by bonding its fibers together at the weave crossover points.

This may be done in single or multi-layer configurations. The stacking may be random or controlled so as to misalign the fibers (e.g., by microscopic computer vision or a similar optical process combined with a mechanical positioning system). The stacking may also alternate fiber orientation (e.g., the second layer at a 45 degree angle to the first layer in a two-layer configuration, two subsequent layers at a 30 degree angle to the previous layer in a three-layer configuration, etc.) to better control effective pore size reduction. The layers can be spot-bonded across the surface to fix their position relative to each other and create a uniform diaphragm structure (without the layers sliding or peeling apart). The bonds can be small enough and spread out enough so as to avoid a significant reduction in overall porosity.

In another embodiment, a diaphragm-less design is used. In this embodiment, the electrolyzer cell may have no diaphragm at all or replaced with a highly-porous separator (e.g., a single layer of 86 tpi PEEK that allows gas and electrolyte solution to easily pass) solely for the purpose of preventing an electrical short-circuit between the electrodes. In a diaphragm-less design, there is a greater opportunity for the generated hydrogen and oxygen gasses to mix. As a result, the oxygen output stream may have a higher concentration of hydrogen than would otherwise be the case for designs that employ a diaphragm. Given the greater degree of mixing in these designs, the output streams may be directly connected (e.g., with short connection piping and without intermediate storage for safety reasons) to a device or system that directly consumes both hydrogen and oxygen, such as a boiler that makes steam from hydrogen and oxygen. Alternatively, the outputs can be directly connected to a gas purification system, which could be a Pressure Swing Adsorber, Cryogenic Separation Device, or Proton-Exchange Membrane (PEM) device with the oxygen output stream connected to the anode side and the hydrogen output stream connected to the cathode side, wherein the anode oxidizes H2 into H+ ions, conducts them through the PEM to the cathode, reduces them back to H2, and releases the H2 back into the H2 stream, thus recovering otherwise-wasted H2 from the O2 stream. The heat exchange(s) discussed above may be advantageous here to keep the PEM feed temperature below its operating temperature limit, which may be 80 degrees C.

FIG. 22 illustrates a cross section of an example electrolyzer cell stack 2202 that includes five cells 2204 bounded on either side by end plates 2206. However, any number of cells may be included in the cell stack.

Each cell 2204 includes a positive electrode 2208 and a negative electrode 2210. The positive and negative electrodes 2208, 2210 may be constructed, at least in part, from one or more metallic materials. The positive and negative electrodes 2208, 2210 may (or may not) be solid depending on the particular implementation. For example, in some implementations, the positive and negative electrodes 2208, 2210 may be perforated.

Each cell (except for the cell furthest to the right) additionally includes a gas separating plate 2211. As shown in FIG. 22, the cell furthest to the right is bounded by the end plate 2206 rather than a gas separating plate 2211. The gas separating plate 2211 may, in some instances, have a solid construction. The gas separating plate 2211 may serve to separate the fluid streams from two adjoining cells (e.g., separate a first stream from a first cell containing hydrogen from a second stream from a second cell containing oxygen).

Each cell defines a first channel 2212, a second channel 2214 and a third channel 2216. The first channel 2212 is positioned adjacent to the positive electrode. In some implementations, the first channel 2212 is approximately 1 mm wide and is configured to allow a first stream (e.g., comprising oxygen and/or electrolyte solution) to flow through the cell.

The second channel 2214 is positioned between the positive and negative electrodes 2208, 2210. A diaphragm 2218 such as those described in the present application is positioned in the second channel 2214 between the positive and negative electrodes 2208, 2210. In some implementations, the diaphragm 2218 is positioned approximately 0.2 mm from the positive electrode 2208 and positioned approximately 0.2 mm from the negative electrode 2210. The exact separation distance between the positive and negative electrodes 2208, 2210 may vary based on any of a variety of factors (e.g., construction of the diaphragm, construction of the electrodes, etc.). The second channel 2214 is configured to allow a first stream (e.g., comprising oxygen and/or electrolyte solution) to flow through the cell between the positive electrode 2208 and the diaphragm 2218 and to allow a second stream (e.g., comprising hydrogen and/or electrolyzer solution) to flow through the cell between the diaphragm 2218 and the negative electrode 2210.

The third channel 2216 is positioned between the negative electrode 2210 and the gas separating plate 2211. In some implementations, the third channel 2216 is approximately 1 mm wide and is configured to allow a second stream (e.g., comprising hydrogen and/or electrolyte solution) to flow through the cell.

It will be appreciated that the gas separating plate 2211 separates a third channel 2216 for a first cell where a first stream (e.g., comprising hydrogen and/or electrolyte solution) is present from a first channel 2212 of an adjacent second cell where a second stream (e.g., comprising oxygen and/or electrolyte solution) is present.

The cells 2204 of the cell stack are electrically connected in series with each other and the series of cells are electrically connected to a positive terminal 2220 and a negative terminal 2222. The positive and negative terminals 2220, 2222 may then be connected to a power source (e.g., a DC power source).

Due to the gas separating plate 2211 between two adjacent cells, one or more electrical connectors 2224 is positioned to cross the gas separating plate 2211 and provide a path for current to flow from a negative electrode 2210 of a first cell to a positive electrode 2212 of an adjacent second cell. In some implementations the gas separating plate 2211 comprises a polymeric material (e.g., PEEK).

In some implementations, the cell stack is cylindrical. A cylindrical electrolyzer cell stack configured as described above provides for improved flow through the electrolyzer cells. In particular, the electrolyzer cell stack provides less resistance to a pump relative to conventional designs and advantageously allows a smaller pump to achieve a given circulation of fluid through the electrolyzer.

This improvement in flow is additionally achieved in-part by the 1 mm separation of the third channel 2216 between the negative electrode 2210 and the gas separating plate 2211 separating the cells. This spacing allows fluid to flow more freely than just having the narrow second channel 2214 where the diaphragm 2218 is positioned.

In some instances, one or more components of the electrolyzer cell stack 2202 (e.g., gas separating plates 2211, end plates 2206, etc.) may be tapered to increase the cross-sectional area through which the electrolyte solution enters the cell. For instance, the first and/or third channels 2212, 2216 may be wider than 1 mm near the entrance for electrolyte solution (e.g., near the bottom of the electrolyzer cell stack 2202) and taper down to 1 mm near the output (e.g., near the top of the electrolyzer cell stack 2202). Such a tapering may advantageously reduce the amount of force required to circulate the electrolyte solution through the electrolyzer cell stack and, as a result, allow pumps with a smaller capacity to be employed.

Further, this design is preferable to simply increasing spacing between the positive and negative electrodes 2208, 2210 (i.e., to widen the narrow channel between the diaphragm and the electrodes) because the spacing between the positive and negative electrodes 2208, 2210 and the diaphragm 2218 influences performance of the electrolyzer cell (e.g., the impedance seen by the power supply).

4. Example Techniques for Pulsing Electrodes in an Electrolyzer System

The power supply 150 can be a direct current (DC) supply that induces a constant direct current between the anode 110 and the cathode 120. For instance, the power supply 150 can apply a sufficient voltage across the anode 110 and the cathode 120 of an electrolyzer cell to induce a constant direct current at a particular magnitude between the anode 110 and the cathode 120. However, inducing this constant direct current can result in unnecessary electrical losses at least because the impedance of the electrolyzer cell (e.g., as seen by the power supply 150) is not purely resistive.

FIG. 23 depicts representative impedance characteristics of the electrical path through an example electrolyzer cell 101 by the impedance Z_Cell. The impedance Z_Cell has some inductance and/or capacitance (i.e., the impedance Z_cell is a complex value that is a function of one or more characteristics of the applied signal) that varies with the present operating environment in the electrolyzer cell 101. For instance, the impedance Z_Cell may vary depending on the unique electrical properties of particular components used within a particular electrolyzer. In the context of an alkaline electrolyzer, the electrical path between the terminals of the power supply 150 may include one or more of the following components: (i) the anode 110, (ii) the cathode 120, (iii) a diaphragm or membrane, and/or (iv) an electrolyte solution. Each of these components can have unique electrical properties that can vary based on the operating environment. For instance, the electrical properties of an electrolyte solution can vary based on the concentration and/or temperature of the electrolyte solution.

In order to minimize the amount of power required to drive a specific amount of current through the electrolyzer (e.g., to achieve the highest electrical efficiency), the impedance Z_Cell seen by the power supply 150 is preferably minimized. The lower the value of the impedance Z_Cell, the lower the voltage that needs to be applied across the anode 110 and cathode 120 of the electrolyzer to achieve a particular current level. Given that the impedance Z_Cell is a complex value that changes with the operating environment, the value of the impedance Z_Cell is not at a theoretical minimum when a constant current is applied as described in more detail below with respect to FIGS. 24 and 25. Accordingly, the voltage (and power) required to achieve a particular current level through the electrolyzer cell 101 is unnecessarily high.

Examples herein recognize that, given the impedance Z_Cell is not purely resistive, the impedance seen by the power supply 150 varies with the characteristics of the signal applied (e.g., frequency). As a result, the impedance Z_Cell seen by the power supply 150 may be changed by pulsing the current through the electrolyzer cell 101 at different frequencies (e.g., instead of inducing a constant direct current).

FIG. 24 depicts a Nyquist plot showing example changes in the impedance Z_Cell seen by the power supply 150 as a frequency of rectangular current pulses applied to the electrolyzer cell 101 is varied from 0 Hz (i.e., DC) to 100 kHz. The data shown in FIG. 24 was projected (not captured via experimentation) for an alkaline electrolyzer with a ZIRFON diaphragm operating at 100 C and 600 mA/cm² current density. The X-axis shows the real component (e.g., real resistance) of the impedance Z_Cell while the Y-axis shows the imaginary component (e.g., reactance) of the impedance Z_Cell.

As shown in FIG. 24, the impedance Z_Cell at the DC point (i.e., when the frequency is zero) has the highest real resistance on the Nyquist plot. Accordingly, applying a constant DC current in such an operating environment would result in lower electrical efficiency than is otherwise possible (e.g., by operating at some non-zero frequency where the impedance is lower). As the frequency of the pulses increases, the real resistance shrinks while the reactance increases until ˜800 Hz before starting to shrink. Note that it is undesirable to have the power supply 150 see a substantial amount of reactance in the impedance Z_cell because such reactance impedes the changes in current flow that occur as part of pulsing current (e.g., on the rising and falling edges of a rectangular pulse). Accordingly, the theoretically preferred frequency for pulsing the electrolyzer cell 101 may be a frequency where the reactance is 0 and the real resistance is minimized. Such a frequency may also be a frequency where a magnitude of the impedance Z_Cell is also minimized. In the Nyquist plot shown in FIG. 24, such a point occurs between 8.5 kHz and 12.7 kHz where the line intersects the X-axis.

It should be noted that the relationship between the impedance values for Z_Cell and a frequency of the rectangular pulses applied changes with the present operating environment for the electrolyzer cell. For instance, the Nyquist plot of the impedance values Z_Cell as a function of frequency for an alkaline electrolyzer operating at 60 C and 400 mA/cm2 may be quite different. Accordingly, in some examples, the electrolyzer system may generate a pulsed current with a frequency that matches the particular operating environment so as to minimize the impedance Z_Cell seen by the power supply 150. The characteristics of the pulses may be selected so as to achieve a desired hydrogen production output while also minimizing the impedance Z_cell. By minimizing the impedance Z_Cell, the voltage required to induce the required/target magnitude current (and the corresponding power required) is reduced, and the electrical efficiency of the electrolyzer is improved.

FIG. 25 shows an example magnitude of the impedance Z_Cell seen by the power supply 150 when current is pulsed using a rectangular waveform at various frequencies between 0 and 2,500 Hz. The data shown in FIG. 25 was obtained using an alkaline electrolyzer operating at 80 degrees Centigrade. As shown in FIG. 25, the magnitude of the impedance Z_Cell decreases with higher frequency up to 2,500 Hz for each of the current densities tested (e.g., 200 mA/cm^2, 400 mA/cm^2, 600 mA/cm^2). The current densities shown (e.g., 200 mA/cm^2, 400 mA/cm^2, 600 mA/cm^2) are the current densities during the “on” portion of the pulsed current. The maximum frequency shown in FIG. 25 is 2,500 Hz due to limitations of the power supply 150 used in the test. In a commercial implementation, frequencies higher than 2,500 Hz may be employed. In at least some configurations, the impedance Z_Cell would continue to decrease with frequency until an inflection point after which the impedance Z_Cell would either remain unchanged or increase.

One electrochemical explanation for the reduction in impedance Z_Cell (and corresponding efficiency improvement) when a pulsed current is applied relative to a constant current (as shown in FIGS. 24 and 25) is the impact of diffusion on the electrolysis reaction. In the middle of an electrolysis reaction, there is a build up of charged ions on the surface of the electrodes that impede the reaction. By pulsing the electrodes at some non-zero frequency, the electrolysis reaction has an opportunity to relax to allow the build-up of charged ions that may otherwise impede the reaction to dissipate. Accordingly, less energy is required to drive the electrolysis reaction over a given period of time.

The controller 160 may generate control signal(s) for the power supply 150 to modify one or more characteristics of the current being induced between the anode and the cathode of a given electrolyzer cell based on one or more present operating parameter(s) of the electrolyzer system. FIG. 26A depicts a flowchart 2600 showing an example of such a process that may be implemented with the controller 150. As shown in FIG. 26A, the controller 160 can determine characteristic(s) of current output for present operating parameter(s) of the electrolyzer system (act 2610). Then, the controller 160 can generate control signal(s) for the power supply 150 based on the determined characteristic(s) of the current output (act 2620). The power supply 150 can, in turn, adjust its output based on the control signals such that the output matches the determined characteristics.

FIGS. 26B and 26C illustrate example ways in which act 2610 in FIG. 26A can be performed. Turning first to FIG. 26B, the controller 160 can identify present operating parameter(s) of the electrolyzer system (act 2630). For example, the controller 160 can monitor any process variable(s) and/or other input parameter(s) that may (directly or indirectly) impact the impedance Z_Cell seen by the power supply 150. In the context of an alkaline electrolyzer, example parameters that can impact Z_cell include, for example, the electrolyte concentration and operating temperature in addition to the age and condition of components in the electrical path including, for example, the diaphragm or membrane, the anode, and/or the cathode. The controller 160 can monitor the state of such process variable(s) by, for example, reading a state of one or more sensors (e.g., pressure sensors, temperature sensors, level sensors, etc.) coupled with the controller 160. Additionally, the controller 160 can monitor other parameters separate and apart from those that directly impact the impedance Z_Cell. For instance, the controller 160 can monitor parameters that are (directly or indirectly) impacted by the current applied to the electrolyzer cell 101, such as the hydrogen output of the electrolyzer system.

Next, the controller 160 can determine characteristic(s) of current output based on those parameter(s) (act 2640). FIG. 27 shows example characteristics of the current output that may be varied including: (i) magnitude of the pulses, (ii) duration of the pulses, (iii) shape of the pulses (e.g., rectangular pulse, triangular pulse, sawtooth pulse, raised cosine pulse, gaussian pulse, etc.), (iv) duty cycle of the pulses (e.g., ratio of on-time to off-time), and/or (v) frequency of the pulses. It should be appreciated that the current output is not limited to a sequence of pulses. In addition, the pulses do not have to return to zero current. They just need to go below a threshold current and/or voltage level to stop the electrolysis reaction. Also, in some instances, the current output may be a waveform (e.g., with some repeating pattern such as a sinusoidal wave) with a DC offset (e.g., to keep a positive current level and avoid reversing the current direction). Such a waveform can also be considered a “pulse,” as that term is used herein. FIG. 28 shows an example of such a current output formed by applying a DC shift to a sinusoidal waveform.

In another embodiment, multiple frequencies are provided in the current output. An example of such an embodiment is shown in FIG. 29 with a lower-frequency pulse that, during the “on-time,” has an embedded higher frequency pulse. Combining multiple frequencies in the current output may be beneficial for any of a variety of reasons in various circumstances. One reason to combine multiple pulse frequencies as shown in FIG. 29 is for the purpose of facilitating bubble release on the electrodes. As bubbles form on the electrodes, the surface area decreases over which the electrolysis reaction can occur. Accordingly, it may be desired to remove/reduce bubbles on the surface of the electrodes. One way to do this is pulse at a relatively-low frequency (e.g., a frequency where the impedance is still undesirably high) to allow the electrolysis reaction sufficient time to relax such that the bubbles can release from the surface of the electrodes. However, pulsing at such a low frequency may not yield the lowest impedance as seen by the power supply. Accordingly, higher-frequency pulses (e.g., at a frequency that minimizes the impedance seen by the power supply) are layered on-top of the low-frequency pulses (e.g., that facilitates bubble release), which can improve electrical efficiency as discussed herein.

In some instances, the controller 160 can employ one or more predefined mathematical relationships (e.g., in the form of a model and/or a lookup table) to determine the characteristic(s) of the current output to be employed for a particular set of operating parameter(s). For example, a manufacturer of a particular electrolyzer design may have characterized the impedance Zell for that particular electrolyzer design and constructed a lookup table using that characterization of the impedance Z_cell. In this example, the lookup table may be deployed across the entire install base of that particular electrolyzer design. The lookup table may either: (i) remain unchanged over the lifetime of an electrolyzer system embodying the particular electrolyzer design; or (ii) may be updated by each respective electrolyzer system embodying the particular electrolyzer design based on captured measurements. Alternatively, the lookup table may be constructed from scratch by each respective electrolyzer system based on captured measurements over time, such as electrochemical impedance spectroscopy (EIS) measurements. For instance, the electrolyzer system may perform parameter sweeps, record the current operating parameter(s), and record the determined current output characteristic(s) to construct a lookup table (e.g., to avoid performing a parameter sweep again in the future when the same (or similar) current operating parameter(s) arise).

Turning now to FIG. 26C, FIG. 26C illustrates another way in which act 2610 in FIG. 26A can be performed. As shown in FIG. 26C, the controller 160 can modify control signal(s) for the power supply 150 (act 2650) and measure performance change(s) based on that modification (act 2660). The controller 160 can then determine if a desired point is found (act 2670). If an optimal point is found, it can be used to generate the control signal(s) for the power supply 150 in act 2620 in FIG. 26A. However, if an optimal point is not found, the controller 160 can determine the next modification of the control signal(s) for the power supply 150 (act 2680), and the method loops back to act 2650.

So, in this embodiment, the controller 160 can employ a particular algorithm to find a minimum impedance that involves repeatedly changing one or more parameter(s) and measuring the impedance seen by the power supply 150 and/or an improvement in electrical efficiency (e.g., measure the power consumed and the hydrogen gas output) until a value for the given parameter(s) that yields the lowest impedance Z_cell and/or the highest electrical efficiency is found (for that set of operating parameters). For example, the controller 160 may cause one or more EIS measurements to be captured that may characterize the impedance of the electrolyzer cell. In some instances, as part of capturing one or more EIS measurements or as part of another measurement technique, the controller 160 can cause the power supply 150 to perform a frequency sweep where the frequency of the pulses is varied between a first value and a second value and record the impedance seen by the power supply 150. In turn, the controller 160 can identify the frequency that yielded the lowest impedance and/or highest electrical efficiency for use with that set of operating parameter(s). In another example, the controller 160 can iteratively change the value of a parameter in a particular direction and see if the performance improved or worsened. If the performance improves, the controller 160 can continue to change the parameter in the same direction to see if the performance is further improved or worsened. Otherwise, the controller 160 can switch directions and see if the performance is improved when modifying the parameter in the opposite direction.

It should be noted that a combination of the approaches in FIGS. 26B and 26C can be used. For example, the controller 160 can use a lookup table with an algorithmic approach to determine the characteristic(s) of the current output. For instance, the controller 160 can operate without a lookup table by: (i) periodically performing an algorithm to find the minimum impedance and updating the characteristic(s) of the pulses based on the output of the parameter sweep; or (ii) aperiodically perform an algorithm to find the minimum impedance when a change in one or more operating parameter(s) is detected. Alternatively, an algorithm to find the minimum impedance can be combined with a lookup table to focus the parameter sweep over a narrower range of values (e.g., to reduce the time required to perform the parameter sweep and find the minimum). For instance, the output of the lookup table for a given operation parameter may be a range of values for one or more of the pulse characteristics (e.g., instead of a particular value). In such an instance, the output of the lookup table can indicate a frequency range in which the minimum impedance Z_Cell is most likely to be located within, and the controller 160 can cause a sweep across the frequency range to find the minimum value for Z_cell for use in the particular operation parameter. A self-learning/artificial intelligence predictive algorithm can be used to perform some or all of these operations.

It should be appreciated that, in some operation parameters, the controller 160 can determine that the best set of characteristics for the current output for a particular set of operating parameter(s) is to apply a constant current and not to “pulse” at all (e.g., pulse at 0 Hz or with a 100% duty cycle). Such a situation may arise, for instance, when the desired throughput of the electrolyzer is very high (e.g., approaching a maximum output of the electrolyzer). When pulsing current at a given frequency above 0 Hz and with some duty cycle below 100%, there is a tradeoff between throughput and electrical efficiency because less power is being applied to the electrolyzer cell for a given unit of time. Further, the electrical efficiency gains of pulsing may diminish as the duty cycle is increased beyond a threshold (e.g., duty cycles above 50%) because the electrolysis reaction does not have sufficient time to relax between pulses. Accordingly, a higher throughput can be achieved by applying a constant current of a given magnitude to the electrolyzer cell than by pulsing current at some non-zero frequency at a given duty cycle (i.e., at the expense of a lower electrical efficiency). When the power supply 150 of the electrolyzer is operating at its peak current magnitude, a higher throughput can be achieved by applying a constant current to the electrolyzer cell instead of pulsing at the same current magnitude. Accordingly, the controller 160 can favor meeting the desired hydrogen production output over the electrical efficiency loss to achieve higher hydrogen production outputs.

It should be noted that the current output does not need to be entirely constructed from unipolar pulses (e.g., varying between 0 to a positive current level). In some instances, the current output may comprise one or more bipolar pulses that vary between a positive current level and a negative current level. For example, certain electrolyzers (e.g., certain PEM electrolyzers) can be operated with a reversed current direction. Operation with a reversed current for brief periods of time may, in some applications, be advantageous to, for example, generate heat. Accordingly, bipolar pulses may be employed in such situations to increase the operating temperature of the electrolyzer (e.g., without other components such as a heater).

Further, it should be noted that the controller 160 can determine a current output that is a pattern of pulses. For example, the best current output for a given operating environment may be a particular sequence of pulses with varying characteristics of the pulses within the pattern. Such pulses may be unevenly spaced (e.g., strategically allowing the electrolysis reaction to pause/relax for different lengths of time), of different magnitudes, and/or of different shapes.

FIG. 30 shows an example improvement in electrical efficiency of an alkaline electrolyzer using the pulsing techniques described above relative to using a constant-current approach. FIG. 30 was generated based on projected impedance values up to 100 kHz. In FIG. 30, each of the data points (and associated lines) represents the improvement in electrical efficiency for an alkaline electrolyzer operating at a particular temperature (in degrees C.) and using ZIRFON at a particular current density. As shown, the pulsing techniques offer an improvement in electrical efficiency between ˜4% and ˜9% depending on the particular operating state and construction of the electrolyzer.

The pulsing techniques described herein have other benefits separate and apart from the improvements to electrical efficiency. In instances where the electrolyzer is powered by a renewable energy source (e.g., in a green hydrogen production environment), the power available to the electrolyzer may vary throughout a given day because of the intermittent nature of renewable energy sources. Employing pulsing techniques may enable the electrolyzer system to operate at lower power level outputs from the renewable energy source than would otherwise be possible. Such an ability comes from the fact that the power required to generate pulses at a given current magnitude (e.g., with some duty cycle below 100%) is lower relative to applying a constant direct current at the same magnitude because the pulsed current signal has some off-time (i.e., where the power supply is not supplying a current). Accordingly, the minimum power required to pulse current into the electrolyzer cell may be lower than that of a design that employs a constant current. This allows an electrolyzer employing pulsing techniques to operate for a longer portion of the day when being powered by only renewable power sources.

Further, it should be noted that the duty cycle (e.g., the ratio between on-time and off-time) can be adjusted (e.g., within a range) in concert with changes in power availability and/or desired hydrogen output. For instance, at a minimum level of hydrogen output (and/or power availability), the electrolyzer may be pulsed at a minimum threshold duty cycle (e.g., 1% duty cycle, 5% duty cycle, 10% duty cycle, etc.). As the level of hydrogen output (and/or power availability) increases, the duty cycle of the pulses may be increased (alone or in conjunction with increases to the magnitude of the pulses). The duty cycle may be increased with the level of hydrogen output (and/or power consumption) until a transition threshold duty cycle (e.g., 50% duty cycle) where the duty cycle jumps to a predefined value and/or pulsing stops altogether (e.g., 100% duty cycle). Such a transition threshold duty cycle may be beneficial because, as mentioned above, the electrical efficiency benefits of pulsing may decay as the duty cycle is increased above a threshold (e.g., above 50%). As a result, those duty cycles where pulsing provides little (or negative) benefits may be skipped over entirely. For hydrogen output (and/or power availability) increases that occur after the transition threshold duty cycle is reached, the magnitude of the current may be increased. Accordingly, the duty cycle may be adjusted dynamically to accommodate changes in power availability and/or desired hydrogen output.

It should be appreciated that the current output may be varied in any of a variety of ways to achieve a particular duty cycle (e.g., as part of achieving a desired hydrogen output and/or power consumption level). In some examples, the off-time between pulses may be lengthened or shortened to achieve various duty cycle levels (e.g., rather than modifying the length of the on-time). For instance, the length of the on-time may be determined so as to minimize the impedance seen by the power supply (as described above) while the off-time (e.g., the gap between pulses) may be determined to achieve a particular desired hydrogen output and/or power demand. In other instances, the length of the on-time and the off-time may both be modified to achieve various duty cycle levels.

The improved dynamic range offered by pulsing (e.g., at various duty cycles) is beneficial in a wide variety of applications including, for example, in green hydrogen installations powered by renewable energy sources with variable power output. For example, the power output of a solar array may drop off near the end of the day and fall below a minimum power required for operation, such as a reversible cell potential, an equilibrium potential (e.g., the minimum voltage across the electrodes to trigger the electrolysis reaction) and/or a thermoneutral potential (e.g., the minimum voltage across the electrodes where the electrolysis reaction is exothermic) when applying a constant current. When using pulsing techniques, the power required to generate pulses at a given duty cycle below 100% (e.g., 50%) at the same current magnitude is lower relative to applying a constant direct current (e.g., because the pulsed current signal has off-time while the constant current signal does not). Accordingly, an electrolyzer system employing pulsing techniques may be able to continue operation at lower power levels from renewable energy sources relative to an electrolyzer system that solely relies on a constant direct-current approach.

FIG. 31A shows an example of the power availability of solar over the course of a day (low in the morning/night and highest at mid-day). FIG. 31B shows an example of how one can operate an electrolyzer using only a DC current (i.e., no pulsing). As shown in FIG. 31B, a sufficient available power is needed to drive the electrolysis reaction using a DC current that only starts later in the day (and ends earlier in the day). FIG. 31C shows an example of the benefits that pulsing offers in terms of dynamic range. One can start operating earlier in the day by pulsing with a low duty cycle to reduce the average power requirement per unit of time below that which is provided by the solar power source while still being above the threshold voltages/currents required for the electrolyzer. For instance, the duty cycle may be reduced by increasing the duration of the off-time (while keeping the on-time of the pulse the same) so as to increase the gap between pulses and reduce the ratio between on-time and off-time. As can be seen in FIGS. 31A-31C, the improvement in dynamic range is meaningful in solar power applications (or any other renewable/variable power source) at least because it allows one to operate for a larger time period (e.g., more of the day) than would otherwise be possible.

FIGS. 32A and 32B show examples of power availability and number of active cell stacks over time. This further improves dynamic range on the low end (e.g., minimum power required to get any hydrogen output) at least because one can reduce the number of active electrolyzer cell stacks (e.g., down to 1) to accommodate lower available input power levels.

It should be appreciated that the diagrams shown in FIG. 31A-31C and FIGS. 32A-32B are idealized diagrams and that real-world power availability (and the corresponding electrolyzer current magnitude and/or number of active cell stacks) may be substantially noisier than depicted. For instance, a cloud may pass over a solar power source and cause a dip in power output at any point throughout the day. In such an instance, the number of active cell stacks may be reduced (or otherwise adjusted) to accommodate the reduction in power output. Additionally (or alternatively), one or more characteristics of the pulsed current (e.g., magnitude and/or duty cycle) being applied to the electrolyzer may also be adjusted to accommodate the reduction in power output.

It should be appreciated that the power available (e.g., in situations where the power is varying as is common with renewable energy sources) may be another example operating parameter that may be employed to determine the characteristics of the current output. For instance, the electrolyzer cell may have a minimum current and/or voltage magnitude that is required to operate (e.g., to attain a reversible cell potential, equilibrium potential, and/or the thermoneutral potential). As the power availability declines, one or more characteristics of the current output may be modified to reduce power consumption (e.g., a duty cycle of the pulses may be reduced) in lockstep with the reduction in power available to still get some hydrogen output from the electrolyzer at increasingly low power levels.

In addition to the benefit of operating at lower power levels described above, employing pulsing techniques has a positive effect on the degradation rate of components in the electrolyzer. In particular, the pulsing techniques offer a reduction in undesirable side reactions that increase the corrosion rate experienced by various components in the electrolyzer.

It should be appreciated that the pulsing techniques described herein may be readily combined with any of the other techniques described herein. For instance, the pulsing techniques described herein may be employed to extend the dynamic range of hydrogen flow rates and/or production capacities that the electrolyzer can operate (e.g., without going below a minimum potential for the electrolyzer such as a reversible cell potential, an equilibrium potential, a thermoneutral potential, a thermally balanced potential, lower heating value potential, and/or higher heating value potential). Such an increase in the dynamic range may advantageously provide a larger range of possible operating points that the control system described herein can optimize over.

FIG. 33A depicts an example implementation of pulsing techniques over a range of hydrogen production rates from a minimum production capacity to a maximum production capacity. As shown, a production target may be set (e.g., using the control techniques described in further detail below) somewhere within the production capacity range. The production target may be associated with a set of values for one or more operating parameters (e.g., pressure(s), temperature(s), flow rate(s), voltage(s), current density value(s), current output characteristic(s) (e.g., pulsing verse not pulsing, duty cycle of pulses when pulsing, etc.), number of active electrolyzer cells and/or stacks of cells, etc.) to achieve the production target. The production capacity range includes a pulsing region between the minimum production capacity and a threshold production capacity where a pulsed current may be applied to the electrodes. In the pulsing region, the duty cycle may be increased for higher production capacities and decreased for lower production capacities. At the minimum production capacity, a duty cycle of the pulses may be at a minimum value (e.g., 1% duty cycle, 5% duty cycle, 10% duty cycle, etc.). As described in greater detail in the '906 application, the threshold production capacity may be a production capacity where the benefits of pulsing diminish (or negatively impact) performance. For instance, the duty cycle may be increased with the production target until a particular duty cycle value is reached (e.g., 50% duty cycle) after which pulsing is stopped. As shown by the no pulsing region in FIG. 33A, pulsing may be stopped (and/or the pulses may have a 100% duty cycle) for production capacities between the threshold production capacity and the maximum production capacity as shown by the no pulsing region.

It should be appreciated that the current output may be varied in any of a variety of ways within the pulsing region to achieve a particular duty cycle. In some examples, the off-time between pulses may be lengthened or shortened to achieve various duty cycle levels (e.g., rather than modifying the length of the on-time). For instance, the length of the on-time may be determined so as to minimize the impedance seen by the power supply (as described in greater detail in the '906 application) while the off-time (e.g., the gap between pulses) may be determined to achieve a particular hydrogen production target. In other instances, the length of the on-time and the off-time may both be modified to achieve various duty cycle levels. In yet other instances, a length of the off-time may be held constant while a length of the on-time may be modified. Accordingly, any of a variety of techniques may be employed to adjust the duty cycle within the pulsing region.

It should be appreciated that the dynamic range of the electrolyzer system may also be extended by independently controlling multiple electrolyzer cell stacks. For instance, the dynamic range on the low end (e.g., minimum power required to get any hydrogen output) may be improved at least because a number of active electrolyzer cell stacks may be reduced (e.g., down to 1 cell stack) to accommodate lower power levels while still operating above a minimum potential for the electrolyzer cells (e.g., a reversible cell potential, an equilibrium potential, a thermoneutral potential, a thermally balanced potential, lower heating value potential, and/or higher heating value potential). Such techniques to extend the dynamic range by independently controlling electrolyzer cell stacks may be employed alone or in combination with the pulsing techniques described above.

FIG. 33B depicts an example implementation that dynamically adjusts a number of active cell stacks over a range of hydrogen production rates from a minimum production capacity to a maximum production capacity. As shown, a production target may be set (e.g., using the control techniques described in further detail below) somewhere within the production capacity range. The production target may be associated with a set of values for one or more operating parameters to achieve the production target. The production capacity range includes a partially active region between the minimum production capacity and a threshold production capacity where a subset of the available cell stacks in the electrolyzer are active (e.g., operating). In the partially active region, the number of active cell stacks may be increased for higher production capacities and decreased for lower production capacities. At the minimum production capacity, a minimum number of cell stacks may be active (e.g., 1 cell stack). The number of active cell stacks may be increased until a threshold production capacity is reached after which all of the electrolyzer cell stacks are active (as shown by the all active region).

FIG. 34 is a graph of another example embodiment showing power availability and current magnitude over time. The improved dynamic range for an electrolyzer system further enables an electrolyzer system to better handle transient events in input power. For instance, the electrolyzer system may be powered by a solar array that may, at random times throughout the day, suddenly produce substantially less power (e.g., because of cloud cover). Such a sudden drop in power could, undesirably, cause the electrolysis cells to malfunction and mix oxygen into the hydrogen stream (and vice-versa). For instance, the voltage across the cell may fall low enough to have the polarity of the cell reverse and produce oxygen on the hydrogen side and hydrogen on the oxygen side.

During these transient events, pulsing can be activated (or the characteristics of the pulses changed) to accommodate the sudden drops in available power. This pulsing technique may be used alone or in combination with other techniques to deal with transient events to avoid getting oxygen in the hydrogen stream. For instance, the average pore size of the diaphragm may be adjustable. In such an instance, the pore size of the diaphragm may be reduced to minimize gas crossover between the sides of the electrolyzer cell. Another option would be to create a pressure gradient across the diaphragm (e.g., higher pressure on the hydrogen side than the oxygen side) to reduce the risk of oxygen crossing over into the hydrogen side.

5. Example Electrolyzer Control System

As mentioned above, hydrogen has a wide variety of applications and is used as an input for many industrial processes including, for example, ammonia production for fertilizers and oil refining. Currently, the majority of hydrogen produced in the United States is created through natural gas reforming. In natural gas reforming, air pollutants (e.g., carbon monoxide) are also undesirably released that contribute to Global Warming.

There is a trend to replace natural gas reforming (and other forms of hydrogen production that emit undesirable gasses) with green hydrogen production methods where no greenhouse gas emissions are produced. Green hydrogen production typically involves the generation of electrical energy through renewable sources (e.g., solar, wind, etc.) that is used to split water into hydrogen and oxygen through electrolysis in an electrolyzer.

FIG. 35 depicts selected systems at a facility that employs one or more electrolyzer(s) to generate an output (e.g., hydrogen gas) for a hydrogen application. As shown in FIG. 35, the facility may comprise one or more electrolyzers 3504 that convert electricity from one or more energy sources 3502 into a desirable gas (e.g., hydrogen) that may be employed in one or more hydrogen applications 3506. The energy sources 3502 may be renewable energy sources (e.g., solar panels, wind turbines, etc.), non-renewable energy sources (e.g., coal, natural gas, etc.), or any combination thereof (e.g., renewable energy sources when available and non-renewable energy sources when the renewable energy sources are not available). The particular hydrogen application(s) 3506 may vary based on the particular use case. Non-limiting examples of particular hydrogen application(s) 3506 include industrial processes that consume hydrogen, electricity generation, powering vehicles, and energy storage.

Further, each of the energy source(s) 3502, electrolyzer(s) 3504, and hydrogen application(s) 3506 may have an associated control system 3512, 3514, 3516 that controls one or more aspects of operation of the respective system. For instance, the control system 3512 for the energy source(s) 3502 may control the energy source(s) 3502 to meet current power demand (e.g., as it shifts throughout the day), and the control system 3514 for the electrolyzer(s) 3504 may control hydrogen output of the electrolyzer(s) 3504 to meet the process demand. These control systems 3512, 3514, 3516 may, in some instances, communicate with each other to coordinate operation of the respective systems and/or with other systems that are not illustrated (e.g., to obtain information from external sources).

It should be appreciated that the energy source(s) 3502, electrolyzer(s) 3504, and hydrogen application(s) 3506 do not need to be co-located at a single facility. For instance, a facility may have the electrolyzer(s) 3504 and equipment associated with a hydrogen application 3506 co-located while obtaining power from the electrical grid. As a result, the energy source(s) 3502 that make electricity for the grid may be at a separate location and connected via power lines.

As discussed above with reference to FIG. 1A, the electrolyzer system 10 can comprise various components to convert a fluid into a desirable gas (e.g., hydrogen) via electrolysis in addition to the controller 160 that controls the state of one or more controllable components in the electrolyzer system 10 based on the state of process variables (e.g., temperature(s), pressure(s), level(s), etc.) and/or one or more set points for those process variables (e.g., a target temperature(s), pressure(s), level(s), etc.). The controller 160 may determine these set points by itself in accordance with a particular control scheme or may receive these set point values from the computing device(s) 260 via the network 280. For instance, the computing device(s) 260 may determine a set point for a particular process variable (e.g., temperature) and communicate that set point to the controller 160. The controller 160 may, in turn, control one or more controllable components in the electrolyzer system 10 to change the temperature to meet the set point (e.g., the controller 160 may implement a control loop and set the set point to be the value received from the remote computing device(s) 260).

As shown in FIG. 1A, the electrolyzer system 10 takes a fluid as an input at the fluid inlet 190 that is circulated through one or more electrolyzer cell(s) 130 (e.g., in an electrolyzer cell stack) via one or more pumps 170 that are controlled by the controller 160. The type of fluid (and the state-liquid phase or vapor phase) received by the electrolyzer system 10 may depend on, for example, the type of electrolyzer (e.g., alkaline electrolyzer, proton exchange membrane (PEM) electrolyzer, solid oxide electrolyzer, etc.) and/or the application for the electrolyzer system 10. In some types of electrolyzers (e.g., an alkaline electrolyzer), the fluid inlet 190 may receive an alkaline solution (e.g., potassium hydroxide (KOH) or sodium hydroxide (NAOH)). In other types of electrolyzers (e.g., a PEM electrolyzer), the fluid inlet 190 may receive water (e.g., regular water or seawater).

As the fluid is circulated through the electrolyzer cell(s) 100, a current is induced between the anode 110 and the cathode 120 of the electrolyzer cell(s) 100 to trigger an electrolysis reaction that generates first and second output streams 186, 188. The current between the anode 110 and the cathode 120 of the electrolyzer cell 100 is induced by the power supply 150 that applies a voltage across the anode 110 and the cathode 120 of the electrolyzer cell 100. The power supply 150 behaves like a current source in that the power supply 150 applies a sufficient voltage in order to achieve a particular target current level through the anode 110 and cathode 120 of the electrolyzer cell 100.

The content of the output streams 186, 188 may vary based on the particular application and construction of the electrolyzer system 10. In water electrolysis applications, the first output stream 186 may comprise hydrogen gas while the second output stream 188 may comprise oxygen gas. In seawater electrolysis, the first output stream 186 may comprise hydrogen gas while the second output stream 188 may comprise oxygen and chlorine gasses.

It should be noted that a commercial implementation of an electrolyzer system could have additional components between the outlets of the system and any industrial process that consumes a particular desirable gas that is generated (e.g., hydrogen and/or oxygen). For instance, there could be systems to separate, purify, and/or compress the desirable gasses (e.g., oxygen/hydrogen) leaving the outlets. Further, the controller may: (1) measure one or more values other than temperature(s), pressure(s), and level(s) such as flow rate(s), electrolyte concentration(s), product gas concentration/purity/quantity/rate of production, voltage(s), and current(s); and/or (2) control other controllable components not depicted in FIG. 1A, such as a state of one or more control valves, cooling systems, and/or heating sources.

Conventional control schemes for electrolyzers focus on maximizing electrical efficiency. The electrical efficiency of an electrolyzer describes the relationship between the electrical energy consumed and the amount of hydrogen produced. A higher electrical efficiency means that less electrical energy needs to be consumed in order to create a given amount of hydrogen (and less electrical energy is lost as waste heat). One problem with such conventional control schemes is that, despite optimizing for high electrical efficiency, the electrolyzer system may still be unnecessarily expensive to operate in terms of levelized Cost of Hydrogen (LCoH₂) (i.e., the levelized cost per kg of hydrogen produced from a system (ignoring real estate cost and certain other external variables)) at least because such a scheme fails to take into account key factors that impact cost. One example of such a factor that is not taken into account by conventional control schemes is the constant fluctuations of feedstock prices. For instance, the spot price of electricity from the power grid over the course of a day can vary from a negative value per megawatt hour (MWh) (e.g., when there is an excess of electricity available due to high renewable energy generation and you are paid to consume electricity) to tens of dollars per MWh (e.g., when only non-renewable energy sources of power are available). In such an instance, a control scheme that is entirely focused on electrical efficiency (and that does not consider the changing cost of electricity) may result in an unnecessarily high LCoH₂ at least in part because of: (i) consumption of a significant amount of electricity when the prices of electricity are very high and/or (ii) underutilization of electricity (e.g., operate below a maximum hydrogen production capacity) when electricity prices are near zero (or negative). The equation in FIG. 36 shows a standard levelized cost of hydrogen formula.

In the following embodiment, a real-time model of cost (e.g., LCoH₂) is integrated into the control scheme to reduce the costs associated with operating the electrolyzer. For instance, a control system for an electrolyzer may obtain: (1) the fixed costs (e.g., capex of installing the electrolyzer), (2) real-time prices of variable costs (e.g., feedstock costs such as electricity cost, input fluid cost, heating costs, etc.), and (3) a current state of various process variable(s) (e.g., current pressure, temperature, and/or level values) and provide those values as inputs to a cost model. In turn, the cost model may generate an n-dimensional array for cost values each of which is associated with a different operating point for the electrolyzer (e.g., the cost is X if the electrolyzer is operated at Y temperature and Z pressure given the current environment). From the n-dimensional array for cost values, the control scheme may identify the optimum operating point (e.g., the point with the lowest LCoH₂) and cause the electrolyzer to operate at the identified optimum operating point. The control scheme may perform this algorithm periodically (e.g., every minute, 5 minutes, 30 minutes, hour, etc.) and/or aperiodically (e.g., when one or more variable cost(s) change by more than threshold amount).

Returning to the drawings, FIG. 37 is a diagram depicting an example implementation of a cost-based control scheme. As shown in FIG. 37, an electrolyzer system 10 comprises one or more electrolyzers 3504, each with sensor(s) 3710 and controllable component(s) 3720. The sensor(s) 3710 (e.g., pressure, level, temperature, and/or flow sensors) can be used to obtain process measurements indicative of a current state of the electrolyzer(s) 3504. FIG. 37 also shows the control system 3514 for the electrolyzer(s) 3504. In this example, the control system 3514 includes the loop control component(s) 3750 as well as a setpoint control component 3700. In one example implementation, the loop control component(s) 3750 are executed by the controller 160 described above, and the setpoint control component 3700 is implemented on the controller 160 and/or the computing device(s) 260. Also, the functions of the loop control component(s) 3750 and the setpoint control component 3700 can be implemented and distributed in any suitable way. For example, the controller 160 can do loop control and the computing device(s) 260 can do setpoint control, the controller 160 can do both, etc.)

The control system 3514 uses sensor information from the electrolyzer(s) 3504 (in combination with information from other sources) to generate control signals that control the state of the one or more controllable component(s) 3720 of the electrolyzer(s) 3504 (e.g., pumps, valves, etc.) in accordance with a determined target operating point (e.g., where the LCoH₂ is minimized). In particular, the control system 3514 employs the setpoint control component 3700 to determine the target operating point (e.g., in the form of setpoints for process variables) that is provided to the controller 160 (which is sometimes referred to herein as the loop control block), which generates the control signals so as to operate the electrolyzer(s) 3504 at the target operating point (e.g., at the setpoints for the process variables). The setpoint control component 3700 employs a cost model to generate estimated cost values that are employed by a cost optimization block to determine the target setpoints. The setpoint control component 3700 could be organized/constructed in other ways. For instance, the setpoint control component 3700 could employ a single model that combines the functions of the cost model block and cost optimization block into a single block. Similarly, the blocks could be subdivided into more granular blocks that each perform some subset of the functions of the cost model block and/or the cost optimization block.

In controlling the electrolyzer(s) 3504, the control system 3514 obtains a plurality of parameter values that are used as inputs for a cost model (e.g., a LCoH₂ model). The parameters shown in FIG. 37 include: (1) a state of process variable(s) (e.g., current temperature(s), pressure(s), voltage(s), capacity factor, feedstock consumption rate(s), number of active electrolyzer cells and/or electrolyzer cell stacks, etc.); (2) fixed cost(s) (e.g., fixed capital cost to build the electrolyzer(s), fixed operation and maintenance costs, etc.); (3) variable cost(s) (e.g., variable operation and maintenance costs, feedstock costs such as electricity costs, etc.); (4) a capital recovery factor (CRF) (e.g., generated based on the expected lifespan of the electrolyzer and an interest rate); and (5) additional parameters such as, for example, availability of current and projected resources (e.g., renewable energy such as solar and/or wind energy), current and/or projected load (e.g., hydrogen demand), and/or weather (e.g., indicative of renewable energy availability). Other such parameter values may exist.

In some instances, one or more of the input parameters may be modeled or otherwise estimated based on other parameters. For example, the operation and maintenance cost associated with operating an electrolyzer may change over time as the components in the electrolyzer degrade. Degradation tends to be somewhat electrolyzer specific, particularly in a dynamic operating environment. A bad weld in a part of the electrolyzer can adversely impact degradation. Accordingly, there can be a feedback mechanism to update any degradation model using measured data from the specific electrolyzer. One source of information regarding degradation that could be used is the I-V (current-voltage) curve since the relationship between a magnitude of current applied and the current induced changes with degradation. Accordingly, the degradation of the electrolyzer may be modeled and used as a basis to calculate the operation and maintenance costs that are, in turn, used as an input for the cost model. Such models may be implemented as any of a variety of models (e.g., regression models, neural networks, decision trees, support vector machines (SVMs), Bayesian networks, etc.) using any of a variety of techniques (e.g., machine learning techniques such as supervised learning techniques, unsupervised learning techniques, reinforcement learning techniques, or any combination thereof).

After the control system 3514 obtains the plurality of parameter values, the control system 3514 provides those values as inputs to a cost model that estimates the cost of operating the electrolyzer(s) 3504 at various operating points. The cost model may be implemented as, for example, a LCoH₂ model. FIG. 37 shows an equation for the transient LCOH₂ over a time period that starts at to and ends at t₁ that may be employed as part of the cost model block in FIG. 36. In this equation, “CRF” is Capital Recovery Factor, “Capacity Factor” is the ratio of the current hydrogen output relative to the maximum hydrogen output, “Overnight Capex” is the overnight capital expenditure to construct the electrolyzer system, “Fixed O&M” is fixed operational and maintenance cost (e.g., full time maintenance staff, periodic (e.g., annual) inspections, etc.), “Variable O&M” is variable operational and maintenance cost, and “Useful Life” refers to the useful lifespan of the electrolyzer system

It should be appreciated that the cost model is not limited to an LCoH₂ model employing the specific formula shown in FIG. 38. For instance, one or more parameters may be replaced with a constant or may be dropped from the LCoH₂ model altogether without severely decreasing the accuracy of the resulting cost estimates. For example, the fixed operating and maintenance (O&M) value may be small relative to the other costs and may be dropped altogether. Similarly, the impact of the capital recovery factor (CRF) may be small and may be dropped altogether. Accordingly, the LCoH₂ equations shown herein are just one example and other equations may be employed to generate the estimated costs of operating at various operating points.

FIG. 39 is a flow chart 3900 of an example method of an embodiment for providing cost-based control of an electrolyzer of an embodiment. In one hypothetical, the price of electricity has substantially dropped because of a glut in power on the grid. Accordingly, the price of electricity for the next time period is going to be near zero. After the method starts, the values for the inputs to the cost model in FIG. 37 are obtained (3910). This can include obtaining information regarding feedstock costs (3920) such as electricity cost and/or the state of one or more process variables (3930) such as present operating temperature, current density, pressure, hydrogen flow rate, etc. Next, using the cost model in FIG. 37 and the inputs obtained in 3910, an array of cost estimates for each of a plurality of potential operating states is generated given the current operating state (3940). For instance, “what is the levelized cost of hydrogen if the operating point of the electrolyzer is changed to X current density and Y pressure given its current state?” for a range of values of X and Y. Given that the cost of electricity (a major feedstock cost) is now low, the levelized cost of hydrogen for those potential operating states that consume a lot of electricity and generate a lot of hydrogen (e.g., states with a high current density) will have a lower levelized cost than those potential operating that consume little electricity and produce little hydrogen (e.g., states with a low current density).

Next, the target operating state from the set of potential operating states that has the lowest cost is identified (3950). For instance, a potential operating state that produces a lot of hydrogen (e.g., with a high current density) and has a low levelized cost, may be selected as the target operating state. The electrolyzer(s) 3504 are then caused to change to the target operating state (3960). In an example, this could involve instructing the electrolyzer(s) to change operating state. For instance, setpoints may be communicated to the controller 160 that, in turn, changes the state of various controllable components so as to operate the electrolyzer(s) 3504 in accordance with the target operating state. This process is repeated until the electrolyzer(s) 3504 are turned off or the operation is otherwise ceased (3970).

As discussed above, in using the cost model, the control system 3514 may generate a plurality of cost (e.g., LCoH₂) values each corresponding to a different operating point for the electrolyzer(s) 3504 that may be arranged in an n-dimensional array (e.g., in the form of one or more matrices). One simple example in the form of a matrix is shown below in FIG. 40. As shown in FIG. 40, the matrix comprises the LCoH₂ values for a plurality of pressure levels (0-n) and current density levels (0-m). For instance, the LCoH₂ value associated with operating at pressure level 2 and current density level 1 for a period of time (e.g., from t0 to t1) is shown in the location LCoH₂ [0,1]. It should be appreciated that this is a simplistic representation and the LCoH₂ values could be computed for more than two process variables. Also, these generated values can be stored in the memory, and the system can look-up/retrieve the values from the memory as part of ultimately determining the operating state at a given point in time.

In some examples, the cost model may only output cost values for combinations of parameters that are feasible for operation given the current operating environment. For instance, the current density level for a particular combination of parameters may be impossible to achieve because of limitations on the current available power. Accordingly, one or more of the LCoH₂ values shown in FIG. 40 may be empty or otherwise indicate that the combination is not possible (e.g., N/A, infinity, etc.). In other examples, the cost model may output the cost values for all combinations of parameters within a range irrespective of whether all of the combinations are actually possible. In such examples, subsequent blocks (e.g., the cost optimization block) can disregard (or otherwise not consider) those combinations that are not possible.

After the control system 3514 has generated the n-dimensional array of cost values, the control system 3514 may determine a target (e.g., optimum) operating point using the array. For instance, the control system 3514 may find the lowest LCoH₂ value in the array and identify the operating point associated with the lowest cost value as the target operating point.

In some instances, the control system 3514 may select the optimum operating point based on the cost values and one or more other values (shown as input parameters). For instance, there may be one or more constraints (e.g., minimum hydrogen production requirements, maximum hydrogen storage capacity, maximum electrical energy consumption defined by agreement or infrastructure limits, minimum amount of power required to be drawn as part of an agreement with a utility provider, etc.) that may need to be taken into account when identifying the optimum operating point. For example, an industrial process that is consuming the hydrogen generated from the electrolyzer(s) 3504 may have a minimum hydrogen supply requirement. In this example, the control system 3514 may find the lowest cost value associated with an operating point that satisfies the hydrogen supply requirement and select that operating point at the optimum operating point. Additionally (or alternatively), the amount of power available may not be unlimited (or practically unlimited in instances where the electrolyzer system 10 is coupled to the power grid). For instance, the electrolyzer(s) 3504 may only receive power from an on-site renewable power source that has variable power output over the course of a day. In such an instance, the control system 3514 may take into account the maximum amount of power that is available in identifying the optimum operating point (e.g., an operating point associated with the lowest cost value that requires an amount of power that is less than or equal to the amount of power available).

After the control system 3514 has determined the optimum operating point, the control system 3514 may control the electrolyzer system 10 in accordance with the determined optimum operating point. For example, the control system 3514 may determine one or more setpoints for various process variables that are used in the control of one or more control loops. These setpoints may be communicated to the loop control controller 160 that, in turn, controls the state of one or more controllable components of the electrolyzer(s) 3504 such that the state of the process variables is moved towards the setpoints.

The control system 3514 may perform this algorithm periodically (e.g., every minute, 5 minutes, 30 minutes, hour, etc.) and/or aperiodically (e.g., when one or more variable cost(s) change by more than threshold amount). In situations where the algorithm is performed periodically, note that the frequency at which the algorithm is repeated may (or may not) be different from the look-ahead period (e.g., t₁-t₀). For example, the algorithm may be performed every minute while the look-ahead period may be performed multiple minutes (e.g., 2 minutes, 5 minutes, etc.).

It should be appreciated that the control scheme 3514 shown in FIG. 37 may be implemented using the controller 160 from FIG. 1A alone or in combination with the computing device(s) 260. For instance, the controller 160 may perform the function of the loop block and control the state of the controllable component(s) 3720 based on associated measurements and a setpoint. The setpoint may be calculated by the computing device(s) 260 and communicated to the controller 160 over the network(s) 280. The computing device(s) 260 may calculate the setpoint by performing operations associated with the other functional blocks shown in FIG. 37 (e.g., LCoH₂ model, LCoH₂ optimization, etc.).

There are many alternatives that can be used with these embodiments. One alternative relates to future predictions for multiple time periods. In some instances, the control scheme may be constructed to optimize LCoH₂ over a longer period of time (e.g., hours, days, weeks, etc.) rather than on only a real-time basis. In such cases, the LCoH₂ model may generate estimated LCoH₂ for multiple future time periods. In turn, the LCoH₂ optimizer may employ the additional LCoH₂ values to find optimum operating points that are likely to result in lower LCoH₂ over the pertinent time frame (e.g., hours, days, weeks) even if those operating points may not have the lowest instantaneous LCoH₂ value.

FIG. 41 shows an example array of LCoH₂ values that comprises values for multiple time periods. (As with the above values, these values can be stored in memory and retrieved for use later.) As shown in FIG. 41, the 3D matrix comprises the LCoH₂ values for a plurality of pressure levels (0-n) and current density levels (0-m) for each of a plurality of time periods (0-k). For instance, the average LCoH₂ value associated with operating at pressure level 2 and current density level 1 during time period k is shown in location [0,1,k]. In some instances, the time periods shown in FIG. 41 may at least partially overlap. For instance, the time periods may all be referenced from a common starting point (e.g., present) but have different durations. For example, a first time period may be very short (e.g., instantaneous), a second time period may be 1 hour (0-1 hour), a third time period may be 6 hours (0-6 hours), and a fourth time period may be a day (0-24 hours). In other instances, the time periods shown in FIG. 41 may be non-overlapping. For instance, each successive time period may start where the previous time period ended. For example, a first time period may be the first 5 minutes (0-5 minutes), the second time period may be the next 5 minutes (5-10 minutes), and the third time period may be the next 5 minutes (10-15 minutes).

These projected LCoH₂ values for additional time periods may be employed by the cost optimization block in any of a variety of ways to determine the setpoints for various parameters. One approach would be to determine the optimal setpoint for each parameter during each time period as shown in FIG. 42. Such an approach may be suitable for parameters (or situations) where the response time of the electrolyzer system is very fast (e.g., the process variable associated with the parameter can be slewed very quickly). Here, the example parameter values are “optimal” on an instantaneous basis.

Another approach is to generate a curve that has gentle transitions between target operating points for a given parameter. Such an approach may be suitable for parameters (or situations) where the response time of the electrolyzer system is not particularly fast (e.g., the parameter takes minutes to change). Such a curve may be formed in any of a variety of ways to take into account the response time of the system. For example, the instantaneous output shown in FIG. 42 may be filtered (e.g., low-pass filtered) to smooth the output until the curve is sufficiently smooth to be inline with the real response time of the system. Such a curve may be employed as-is or may be time-shifted forward or backwards (e.g., depending on where you want the transition to take place). FIG. 43 shows an example of a filtered curve that has been slightly time-shifted to make the transitions at the end of a given time period (e.g., such that the operating point is near the target at the beginning of the subsequent time period), where the example parameter values have been smoothed.

Another approach would be to directly find the combination of parameters across each of the relevant time periods that yields a lowest aggregate cost of hydrogen for the entire period given the real-word response time of the electrolyzer (e.g., to find a combination of parameter values for each of three time periods that yields a lowest aggregate cost of hydrogen over a length of time that encompasses all three time periods). Accordingly, the determined output may, during some time periods, substantially deviate from what is “optimal” on an instantaneous basis so as to take advantage of opportunities in a future time period. For instance, the electrolyzer may operate in a fashion that is inefficient (e.g., substantially departs from what may be optimal instantaneously) in a first time period where electricity is expensive so as to be positioned to take advantage of low electricity prices in an upcoming time period where electricity is cheap (or free). In such an instance, the benefit of being properly positioned to take advantage of cheap electricity in those later time periods may more than make up for the inefficient operation in the first time period.

FIGS. 44A and 44B illustrate an example of such a scenario showing the “optimal” point for a given period when only looking at what is “optimal” on an instantaneous basis verse a curve that takes into account the response time of the electrolyzer system and produces a lower aggregate cost of hydrogen over the entire length of time that includes all three time periods. FIG. 44A shows example parameter values that are optimal on an instantaneous basis, and FIG. 44B shows example parameter values that are optimal over the entire length of time that encompasses all three time periods.

Another alternative relates to replacing cost estimation with profit estimation. The control system described herein estimates the cost associated with producing a given quantity of hydrogen with a goal of minimizing the cost to generate such hydrogen. The control system could be designed to also take into account the spot price of hydrogen and calculate the difference between the estimated cost and the spot price (e.g., to estimate profit). Such estimated profit can be maximized in the same way as cost is minimized as described above. For instance, the cost model may be replaced with a profit model that outputs the profit associated with each of a range of operating points. Similarly, the cost optimization block can be replaced with a profit optimization block that determines the best operating point that maximizes profit.

In some instances, the spot price of hydrogen could be replaced with the spot price of a product of a downstream application of hydrogen. For instance, the hydrogen output by the electrolyzer could be used to generate electricity and the spot price that is taken into account is the spot price of electricity rather than the spot price of hydrogen itself. Accordingly, the profit estimation may employ a price other than the spot price of hydrogen as a reference for the profit estimation.

Yet another alternative relates to a multi-site extension. The control systems for multiple sites (e.g., owned/operated by a common entity) may be coordinated to achieve further reductions in the cost of operation. For instance, an entity may need to fulfill an order of hydrogen (e.g., X kilograms of hydrogen) within a specified period of time (e.g., by the end of the day) from any combination of two or more nearby facilities. In such an instance, the control systems of the facilities may be coordinated to operate that facility (or facilities) with the lowest cost of generating hydrogen at higher capacities (e.g., higher than operation would be using cost alone) while operating those facilities with a higher cost of generating hydrogen at lower capacities (e.g., lower than operation would be using cost alone).

In yet another alternative, the control system described above may also include a safety scheme to shut-down the electrolyzer when one or more unsafe conditions are detected such as over-pressurization conditions, too high a concentration of hydrogen in the oxygen stream (which makes the stream highly flammable), and over temperature conditions. The safety scheme may be integrated within the control system described herein and use all of the same physical components. In such an example, the associated safety limits may be stored by the loop control block and/or the setpoint control block and immediately trigger a shutdown when the limit is reached. In other examples, the safety scheme may be implemented using separate hardware, such as separate controllable elements (e.g., values such as shutdown valves), instrumentation, and/or a separate loop control. One implementation of such a scheme employing separate safety equipment is shown in FIG. 45. As shown in FIG. 45, a safety control component(s) 3740 receives measurements from safety sensor(s) 3730 in the electrolyzer(s) 3504 can generate control signals that are provided to safety component(s) 3770 in the electrolyzer(s) 3504. The safety control component(s) 3740 may be implemented in their own separate controller, on controller 116, and/or on computing device(s) 260.

Various examples of systems, devices, and/or methods are described herein. Any embodiment, implementation, and/or feature described herein as being an “example” is not necessarily to be construed as preferred or advantageous over any other embodiment, implementation, and/or feature unless stated as such. Thus, other embodiments, implementations, and/or features may be utilized, and other changes may be made without departing from the scope of the subject matter presented herein.

6. Conclusion

Accordingly, the examples described herein are not meant to be limiting. It will be readily understood that the aspects of the present disclosure, as generally described herein, and illustrated in the figures, can be arranged, substituted, combined, separated, and designed in a wide variety of different configurations.

Further, unless the context suggests otherwise, the features illustrated in each of the figures may be used in combination with one another. Thus, the figures should be generally viewed as component aspects of one or more overall embodiments, with the understanding that not all illustrated features are necessary for each embodiment.

Additionally, any enumeration of elements, blocks, or steps in this specification or the claims is for purposes of clarity. Thus, such enumeration should not be interpreted to require or imply that these elements, blocks, or steps adhere to a particular arrangement or are carried out in a particular order.

Further, terms such as “A coupled to B” or “A is mechanically coupled to B” do not require members A and B to be directly coupled to one another. It is understood that various intermediate members may be utilized to “couple” members A and B together.

Moreover, terms such as “substantially” or “about” that may be used herein, are meant that the recited characteristic, parameter, or value need not be achieved exactly but that deviations or variations, including, for example, tolerances, measurement error, measurement accuracy limitations and other factors known to skill in the art, may occur in amounts that do not preclude the effect the characteristic was intended to provide.

It is intended that the foregoing detailed description be understood as an illustration of selected forms that the invention can take and not as a definition of the invention. It is only the following claims, including all equivalents, that are intended to define the scope of the claimed invention. Finally, it should be noted that any aspect of any of the embodiments described herein can be used alone or in combination with one another.

As discussed above, in one aspect, the present disclosure provides an alkaline electrolyzer system comprising: a power input, an electrolyzer, thermal insulation and a pump.

The power input receives power from an external power source.

The electrolyzer cell comprises a cathode coupled to the power input, an anode coupled to the power input, and a diaphragm at least partially disposed between the anode and the cathode, wherein the electrolyzer cell is configured to (i) use power from the power input to produce a first stream comprising oxygen gas and a second stream comprising hydrogen gas from an alkaline solution, and (ii) operate at a temperature in excess of 100 degrees Centigrade (C).

The thermal insulation at least partially covering the electrolyzer cell.

The pump is configured to circulate the alkaline solution through the electrolyzer cell.

In some implementations, the electrolyzer cell is further configured to operate at a temperature of up to and including about 200 degrees C.

In some implementations, the alkaline electrolyzer system further comprises a heater configured to heat the alkaline solution prior to the alkaline solution entering the electrolyzer cell.

In some implementations, the alkaline electrolyzer system further comprises a cooling system configured to cool the pump.

In some implementations, the alkaline electrolyzer further comprises a second pump configured to pressurize a conduit carrying the alkaline solution.

In some implementations, the thermal insulation or its coating is chemically resistant to the alkaline solution.

In some implementations, the alkaline electrolyzer system further comprises a heat exchanger configured to transfer heat from one or both of the first and second streams to feedwater and/or to the alkaline solution.

In some implementations, the alkaline electrolyzer system further comprises a controller configured to control operation of the pump.

In some implementations, the alkaline electrolyzer system further comprises: a first gas separator configured to separate out one or more first gasses from the first stream, wherein the one or more first gasses comprises oxygen gas; and a second gas separator configured to separate out one or more second gasses from the second stream, wherein the one or more second gasses comprises hydrogen gas.

In some implementations, the alkaline electrolyzer system further comprising: a first condenser configured to condense water vapor in the one or more first gasses; and a second condenser configured to condense water vapor in the one or more second gasses.

As described above, in another aspect, the present disclosure provides a method of operating an alkaline electrolyzer system, comprising: circulating an alkaline solution through an electrolyzer cell comprising a cathode, an anode, and a diaphragm at least partially disposed between the anode and cathode and producing a first stream comprising oxygen gas and a second stream comprising hydrogen gas at least in part by inducing a current between the cathode and the anode in the electrolyzer cell.

The method further comprises: while producing the first stream and the second stream, operating the electrolyzer at a temperature that exceeds 100 degrees Centigrade (C); separating the hydrogen gas from the second stream; and outputting the hydrogen gas.

In some implementations, the method further comprises operating the electrolyzer at a temperature of up to and including about 200 degrees C.

In some implementations, the method further comprises: causing pressure of the alkaline solution to rise to a target pressure value; and causing temperature of the alkaline solution to rise to a target temperature value.

In some implementations, the method further comprises stopping an electrolysis reaction.

In some implementations, the method further comprises removing heat from the alkaline electrolyzer system while continuing electrolysis or after the electrolysis reaction has been stopped.

As described above, in a further aspect, the present disclosure provides a diaphragm for an alkaline electrolyzer, comprising: a support structure comprising at least one of polyetheretherketone (PEEK) or polytetrafluoroethylene (PTFE), the support structure having a thickness of no more than 350 microns between a first side and a second side that is opposite the first side; and zirconium dioxide (ZrO2) disposed on each of the first side and the second side of the support structure.

In some implementations, the diaphragm comprises a maximum operating temperature over 100 degrees Centigrade (C).

In some implementations, 200 degrees C. is within the maximum operating temperature.

In some implementations, the support structure comprises a single layer.

In some implementations, the support structure comprises a plurality of layers.

As described above, in another aspect, the present disclosure provides an electrolyzer system comprising: a fluid input; an electrolyzer cell comprising a cathode and an anode; a pump coupled between the fluid input and the electrolyzer cell and configured to circulate fluid received from the fluid input through the electrolyzer cell; and a power supply coupled the electrolyzer cell and configured to induce a pulsed current between the cathode and the anode of the electrolyzer cell that varies based on a control signal.

The electrolyzer system further comprises at least one processor; and at least one non-transitory computer-readable medium comprising program instructions that are executable by the at least one processor. When executed, the program instruction cause the at least one processor to be configured to: determine at least one characteristic for the pulsed current to be induced between the cathode and the anode, the at least one characteristic for the pulsed current including a frequency of pulses; and generate the control signal for the power supply based on the determined at least one characteristic for the pulsed current.

In some implementations, at least one characteristic for the pulsed current further includes a magnitude of the pulses, a duration of the pulses, a shape of the pulses, a pattern of the pulses, and/or a duty cycle of the pulses.

In some implementations, the at least one characteristic for the pulsed current is determined from a predefined relationship that correlates a plurality of operating parameters and a respective plurality of characteristics for the pulsed current.

In some implementations, the plurality of operating parameters comprises electrolyte concentration, operating temperature, condition of the anode, condition of the cathode, condition of a diaphragm between the anode and the cathode, available power, and/or a hydrogen output of the electrolyzer system.

In some implementations, the at least one characteristic for the pulsed current is determined by repeatedly changing a value of the at least one characteristic for the pulsed current until a desired value is found.

In some implementations, the desired value provides a minimum impedance of the electrolyzer cell and/or a maximum electrical efficiency.

In some implementations, at least one of the values is obtained using a predefined relationship that correlates a plurality of operating parameters of the electrolyzer system and a respective plurality of values.

In some implementations, the current output comprises (i) a unipolar pulse that varies between zero and a positive current level, (ii) a bipolar pulse that varies between a positive current level and a negative current level, and/or (iii) a pattern of pulses with varying characteristics.

As discussed above, in a further aspect, the present disclosure provides a method of operating an electrolyzer system, comprising: dynamically determining a frequency of pulses of current that will reduce an impedance of an electrolyzer cell in the electrolyzer system; causing the pulses of current at the determined frequency to be induced between a cathode and an anode of the electrolyzer cell as solution is circulating through the electrolyzer cell; and producing a first stream comprising oxygen gas and a second stream comprising hydrogen gas at least in part due to electrolysis of the solution caused by the current induced between the cathode and the anode in the electrolyzer cell.

In some implementations, the frequency of the pulses of current is dynamically determined from a lookup table that correlates a plurality of frequencies of the pulses of current and a plurality of operating parameters of the electrolyzer system.

In some implementations, the frequency of the pulses of current is dynamically determined by determining which of a plurality of frequencies of pulses of current provides a minimum impedance of the electrolyzer cell and/or a maximum electrical efficiency.

In some implementations, at least one frequency of the plurality of frequencies of the pulses of current is identified from a lookup table that correlates a plurality of frequencies of the pulses of current and a plurality of operating parameters of the electrolyzer system.

In some implementations, the method further comprises operating the electrolyzer system at a temperature that exceeds 100 degrees Centigrade (C).

In some implementations, the method further comprises operating the electrolyzer system at a temperature up to and including about 200 degrees Centigrade (C).

As described above, in a further aspect, the present disclosure provides an electrolyzer system comprising: an electrolyzer cell comprising a cathode and an anode, wherein an impedance characteristic of an electrical path through the electrolyzer cell varies based on at least one operating parameter of the electrolyzer system; at least one processor; and at least one non-transitory computer-readable medium comprising program instructions that are executable by the at least one processor such that the at least one processor is configured to cause a pulsed current to be generated across the cathode and the anode at a frequency that reduces the impedance characteristic of the electrical path through the electrolyzer cell.

In some implementations, the frequency is determined from a predefined relationship that correlates a plurality of operating parameters of the electrolyzer system and a respective plurality of frequencies and/or by testing a plurality of frequencies to find the frequency that minimizes the impedance characteristic of the electrical path through the electrolyzer cell.

In some implementations, the predefined relationship is provided in a lookup table.

In some implementations, the predefined relationship is provided by a manufacturer of the electrolyzer system.

In some implementations, the predefined relationship is updated by at least one processor based on captured measurements over time.

In some implementations, the predefined relationship is created by at least one processor based on captured measurements over time.

As described above, in yet a further aspect, the present disclosure provides an electrolyzer system comprising: a power supply; an electrolyzer cell comprising a cathode coupled to the power supply, an anode coupled to the power supply, and a diaphragm at least partially disposed between the anode and the cathode, wherein the electrolyzer cell is configured to (i) use power from the power supply to produce a first stream comprising oxygen gas and a second stream comprising hydrogen gas from an alkaline solution, and (ii) operate at a temperature in excess of 100 degrees Centigrade (C); thermal insulation at least partially covering the electrolyzer cell; and a pump configured to circulate the alkaline solution through the electrolyzer cell.

The electrolyzer system further comprises: at least one processor; and at least one non-transitory computer-readable medium comprising program instructions that are executable by the at least one processor such that the at least one processor is configured to: determine at least one characteristic for pulsed current to be induced between the cathode and the anode, the at least one characteristic for the pulsed current including a frequency of pulses; and generate a control signal for the power supply based on the determined at least one characteristic for the pulsed current.

In some implementations, at least one characteristic for the pulsed current further includes a magnitude of the pulses, a duration of the pulses, a shape of the pulses, a pattern of the pulses, and/or a duty cycle of the pulses.

In some implementations, at least one characteristic for the pulsed current is determined from a predefined relationship that correlates a plurality of operating parameters and a respective plurality of characteristics for the pulsed current.

In some implementations, at least one characteristic for the pulsed current is determined by repeatedly changing a value of the at least one characteristic for the pulsed current until a desired value is found.

In some implementations, the electrolyzer cell is further configured to operate at a temperature of up to and including about 200 degrees C.

In some implementations, the electrolyzer system further comprises a heater configured to heat the alkaline solution prior to the alkaline solution entering the electrolyzer cell.

In some implementations, the electrolyzer system further comprises a cooling system configured to cool the pump.

In some implementations, the thermal insulation or its coating is chemically resistant to the alkaline solution.

In some implementations, the electrolyzer system further comprises a heat exchanger configured to transfer heat from one or both of the first and second streams to feedwater and/or to the alkaline solution.

In some implementations, the electrolyzer system further comprises: a first gas separator configured to separate out one or more first gasses from the first stream, wherein the one or more first gasses comprises oxygen gas; and a second gas separator configured to separate out one or more second gasses from the second stream, wherein the one or more second gasses comprises hydrogen gas.

As described above, the present disclosure additionally provides an electrolyzer comprising: an electrolyzer cell comprising a cathode and an anode, wherein the electrolyzer cell is configured to operate at a temperature in excess of 100 degrees Centigrade (C); thermal insulation at least partially covering the electrolyzer cell; and a pump configured to circulate an alkaline solution through the electrolyzer cell.

The electrolyzer further comprises at least one processor; and at least one non-transitory computer-readable medium comprising program instructions that are executable by the at least one processor such that the at least one processor is configured to: generate a cost estimate for hydrogen gas produced by the electrolyzer for each of a plurality of different operating states of the electrolyzer; select one of the plurality of different operating states based on the generated cost estimates; and control the electrolyzer; to operate according to the selected operating state.

In some implementations, the electrolyzer further comprises a power supply coupled to the electrolyzer cell and configured to induce a pulsed current between the cathode and the anode of the electrolyzer cell that varies based on a control signal.

In some implementations, the electrolyzer cell is further configured to operate at a temperature of up to and including about 200 degrees C.

In some implementations, the electrolyzer system further comprises a heater configured to heat the alkaline solution prior to the alkaline solution entering the electrolyzer cell.

In some implementations, the electrolyzer system further comprises a cooling system configured to cool the pump.

In some implementations, the thermal insulation or its coating is chemically resistant to the alkaline solution.

In some implementations, the electrolyzer system further comprises a heat exchanger configured to transfer heat from one or both of first and second streams outputted from the electrolyzer cell to feedwater and/or to the alkaline solution.

As discussed above, in another aspect, the present disclosure provides a non-transitory computer-readable medium storing program instructions that, when executed by one or more processors, cause the one or more processors to perform functions comprising: (a) calculating a cost or profit associated with producing a gas using one or more electrolyzers for each of a plurality of operating states of the one or more electrolyzers; (b) selecting one of the plurality of operating states based on a cost or profit consideration; and (c) causing the one or more electrolyzers to be configured according to the selected one of the plurality of operating states.

In some implementations, the cost or profit is calculated using a model having one or more of the following inputs: a state of process variable(s) of the one or more electrolyzers, fixed cost(s), variable cost(s), a capital recovery factor (CRF), availability of resources, load, weather, an operation cost of the one or more electrolyzers, and/or a maintenance costs of the one or more electrolyzers.

In some implementations, (a)-(c) are performed periodically.

In some implementations, (a)-(c) are performed aperiodically.

In some implementations, the cost or profit is calculated on a real time basis.

In some implementations, the cost or profit is calculated for at least one future time period.

In some implementations, the one or more electrolyzers are located at a first site, and (a)-(c) are additionally performed for an additional one or more electrolyzers located at a second site.

As discussed above, in a further aspect, the present disclosure provides a method for controlling an electrolyzer system, comprising: producing a stream comprising hydrogen gas at least in part by (i) circulating a fluid through an electrolyzer cell that comprises two electrodes; and (ii) inducing a current between the two electrodes of the electrolyzer cell; and while producing the stream: (i) identifying a cost associated with at least one feedstock of the electrolyzer system; (ii) for each of a plurality of operating states for the electrolyzer system, estimating a cost associated with the hydrogen gas based on the identified cost associated with the at least one feedstock; (iii) identifying a target operating state for the electrolyzer system based on the estimated cost associated with the hydrogen gas for each of the plurality of operating states; and (iv) controlling at least one controllable component of the electrolyzer system based on the target operating state.

In some implementations, the method further comprises inducing a pulsed current between the two electrodes of the electrolyzer cell.

In some implementations, the method further comprises detecting an unsafe condition; and shutting down the electrolyzer system in response to detecting the unsafe condition.

In some implementations, the method further comprises: generating an n-dimensional array of the estimated costs; and identifying a lowest estimated cost in the n-dimensional array; wherein the target operating state is associated with the lowest estimated cost.

In some implementations, the method further comprises generating a curve wherein transitions between target operating points in the curve for a given operating state are sufficiently smooth to be in-line with a real response time of the electrolyzer system.

In some implementations, the method further comprises identifying a combination of parameters across a plurality of time periods that yields a lowest aggregate cost of hydrogen.

As described above, in a further aspect, the present disclosure provides an alkaline electrolyzer system comprising: a power input to receive power from an external power source; an electrolyzer cell comprising a cathode coupled to the power input, an anode coupled to the power input, and a diaphragm at least partially disposed between the anode and the cathode, wherein the electrolyzer cell is configured to (i) use power from the power input to produce a first stream comprising oxygen gas and a second stream comprising hydrogen gas from an alkaline solution, and (ii) operate at a temperature in excess of 120 degrees Centigrade (C); and a gas separator configured to separate out one or more gasses from the second stream, wherein the one or more second gasses comprises the hydrogen gas.

The alkaline electrolyzer system further comprises: at least one pump configured to circulate the alkaline solution through the electrolyzer cell; at least one fluid conduit coupled between the pump and the electrolyzer cell and configured to carry the alkaline solution; and thermal insulation at least partially covering the electrolyzer cell and the at least one fluid conduit.

In some implementations, the electrolyzer cell is further configured to operate at a temperature of up to and including about 200 degrees C.

In some implementations, the alkaline electrolyzer system further comprises a heater configured to heat the alkaline solution prior to the alkaline solution entering the electrolyzer cell.

In some implementations, at least one fluid conduit is at least partially constructed from a polymeric material.

In some implementations, the alkaline electrolyzer system further comprises a second pump configured to pressurize a conduit carrying the alkaline solution.

In some implementations, the thermal insulation comprises one or more first pieces of thermal insulation at least partially covering the electrolyzer cell and one or more second pieces of thermal insulation at least partially covering the at least one fluid conduit.

In some implementations, the alkaline electrolyzer system further comprises a heat exchanger configured to transfer heat from one or both of the first and second streams to the alkaline solution.

In some implementations, the alkaline electrolyzer system further comprises an electrolyzer cell stack that comprises: a first end plate; a second end plate; and a plurality of electrolyzer cells disposed between the first end plate and the second end plate, wherein the plurality of electrolyzer cells comprises the electrolyzer cell.

In some implementations, the plurality of electrolyzer cells are connected in series.

In some implementations, the alkaline electrolyzer system further comprises a condenser configured to condense water vapor in the one or more gasses.

In some implementations, the alkaline electrolyzer system further comprises a power supply coupled to the power input and configured to induce a pulsed current between the cathode and the anode of the electrolyzer cell that varies based on a control signal.

In some implementations, the alkaline electrolyzer system further comprises: at least one processor; and at least one non-transitory computer-readable medium comprising program instructions that are executable by the at least one processor.

When the at least one processor executes the program instructions, the at least one processor is configured to: determine at least one characteristic for the pulsed current to be induced between the cathode and the anode, the at least one characteristic for the pulsed current including a frequency of pulses; and generate the control signal for the power supply based on the determined at least one characteristic for the pulsed current.

In some implementations, the alkaline electrolyzer system further comprises: at least one processor; and at least one non-transitory computer-readable medium comprising program instructions that are executable by the at least one processor.

When the at least one processor executes the program instructions, the at least one processor is configured to: generate a cost estimate for hydrogen gas produced by the alkaline electrolyzer system for each of a plurality of different operating states of the alkaline electrolyzer system; select one of the plurality of different operating states based on the generated cost estimates; and control the alkaline electrolyzer system to operate according to the selected operating state.

As described above, in another aspect, the present disclosure provides a method of operating an alkaline electrolyzer system, the method comprising: causing a temperature of an alkaline solution to rise to a target temperature that exceeds 120 degrees Centigrade (C); circulating the alkaline solution through an electrolyzer cell comprising a cathode, an anode, and a diaphragm at least partially disposed between the anode and cathode; and producing a first stream comprising oxygen gas and a second stream comprising hydrogen gas at least in part by inducing a current between the cathode and the anode in the electrolyzer cell.

The method further comprises: while producing the first stream and the second stream, operating the electrolyzer at the temperature that exceeds 120 degrees C.; separating the hydrogen gas from the second stream; and outputting the hydrogen gas.

In some implementations, the method further comprises operating the electrolyzer at a temperature of up to and including about 200 degrees C.

In some implementations, the method further comprises causing a pressure on the alkaline solution to rise to a target pressure value that exceeds 8 bar.

In some implementations, causing the pressure of the target alkaline solution alkaline solution to exceed 8 bar comprises causing the pressure on the alkaline solution to rise to about 10 bar.

In some implementations, the method further comprises transferring heat from at least one of the first stream and the second stream to the alkaline solution.

In some implementations, the method further comprises dynamically determining a frequency of pulses of current that will reduce an impedance of an electrolyzer cell in the electrolyzer system.

In some implementations, inducing the current between the cathode and the anode in the electrolyzer cell comprises inducing pulses of current at the determined frequency between the cathode and the anode of the electrolyzer cell.

In some implementations, the method further comprises: while producing the first stream and the second stream, (i) identifying a cost associated with at least one feedstock of the electrolyzer system; (ii) for each of a plurality of operating states for the electrolyzer system, estimating a cost associated with the hydrogen gas based on the identified cost associated with the at least one feedstock; (iii) identifying a target operating state for the electrolyzer system based on the estimated cost associated with the hydrogen gas for each of the plurality of operating states; and (iv) controlling at least one controllable component of the electrolyzer system based on the target operating state.

As described above, in a further aspect, the present disclosure provides a diaphragm for an alkaline electrolyzer, the diaphragm comprising: a support structure comprising at least one of polyetheretherketone (PEEK) or polytetrafluoroethylene (PTFE), the support structure having a thickness of no more than 350 microns between a first side and a second side that is opposite the first side; and a ceramic material disposed on each of the first side and the second side of the support structure, wherein the ceramic material comprises at least one of aluminum oxide (Al2O3) or zirconium dioxide (ZrO2).

In some implementations, the diaphragm comprises a maximum operating temperature over 120 degrees Centigrade (C).

In some implementations, 200 degrees C. is within the maximum operating temperature.

In some implementations, the support structure comprises a single layer.

In some implementations, the ceramic material is constructed as a ceramic fiber.

As described above, in a further aspect, the present disclosure provides an alkaline electrolyzer system comprising: a power input to receive power from an external power source; an electrolyzer cell comprising a first cathode coupled to the power input, a first anode coupled to the power input, and a first diaphragm at least partially disposed between the first anode and the first cathode, wherein the electrolyzer cell is configured to (i) use power from the power input to produce a first stream comprising oxygen gas and a second stream comprising hydrogen gas from an alkaline solution, and (ii) operate at a temperature in excess of 120 degrees Centigrade (C); and a gas separator configured to separate out one or more gasses from the second stream, wherein the one or more gasses comprises the hydrogen gas.

The alkaline electrolyzer system further comprises a pump configured to circulate the alkaline solution through the electrolyzer cell; at least one fluid conduit coupled between the pump and the electrolyzer cell and configured to carry the alkaline solution; and thermal insulation at least partially covering the electrolyzer cell and the at least one fluid conduit.

In some implementations, the electrolyzer cell is further configured to operate at a temperature of up to and including about 200 degrees C.

In some implementations, the alkaline electrolyzer system further comprises a heater configured to heat the alkaline solution prior to the alkaline solution entering the electrolyzer cell.

In some implementations, the alkaline electrolyzer system of further comprises an electrolyzer cell stack that comprises: a first end plate; a second end plate; and a plurality of electrolyzer cells disposed between the first end plate and the second end plate, wherein the plurality of electrolyzer cells comprises the electrolyzer cell.

In some implementations, the electrolyzer cell stack further comprises: a second anode; a second cathode; a second diaphragm disposed between the second anode and the second cathode; and a gas separation plate disposed between the first cathode and the second anode.

In some implementations, the electrolyzer cell stack further comprises: a first channel disposed between the first end plate and the first anode, the first channel having a first channel width; a second channel disposed between the first anode and the first cathode, the second channel having a second channel width that is smaller than the first channel width; and a third channel disposed between the first cathode and the gas separation plate, the third channel having a third channel width that is larger than the second channel width.

In some implementations, the plurality of electrolyzer cells are electrically coupled in series.

In some implementations, the electrolyzer cell stack comprises an electrical conductor that electrically connects the first cathode to the second anode, wherein the electrical conductor passes through the gas separating plate.

In some implementations, the at least one fluid conduit is at least partially constructed from a polymeric material.

In some implementations, the alkaline electrolyzer system further comprises a condenser configured to condense water vapor in the one or more gasses.

In some implementations, the thermal insulation comprises one or more first pieces of thermal insulation at least partially covering the electrolyzer cell and one or more second pieces of thermal insulation at least partially covering the at least one fluid conduit.

As described above, in another aspect, the present disclosure provides an alkaline electrolyzer system comprising: a power input to receive power from an external power source; an electrolyzer cell comprising a cathode coupled to the power input, an anode coupled to the power input, and a diaphragm at least partially disposed between the first anode and the first cathode, wherein the electrolyzer cell is configured to (i) use power from the power input to produce a first stream comprising oxygen gas and a second stream comprising hydrogen gas from an alkaline solution, and (ii) operate at a temperature in excess of 120 degrees Centigrade (C), wherein the diaphragm comprises a ceramic material and has a maximum operating temperature over 120 degrees C.; and a gas separator configured to separate out one or more gasses from the second stream, wherein the one or more gasses comprises hydrogen gas.

The alkaline electrolyzer further comprises a pump configured to circulate the alkaline solution through the electrolyzer cell; at least one fluid conduit coupled between the pump and the electrolyzer cell and configured to carry the alkaline solution; and thermal insulation at least partially covering the electrolyzer cell and the at least one fluid conduit.

In some implementations, the diaphragm comprises: a support structure at least partially constructed from a polymeric material and having a thickness of no more than 350 microns; and the ceramic material disposed on the support structure.

In some implementations, the polymeric material comprises at least one of polyetheretherketone (PEEK) or polytetrafluoroethylene (PTFE) and the ceramic material comprises at least one of aluminum oxide (Al2O3) or zirconium dioxide (ZrO2).

In some implementations, the ceramic material is constructed as a woven ceramic fiber.

As described above, in a further aspect, the present disclosure provides a method of operating an alkaline electrolyzer system, the method comprising: causing a temperature of an alkaline solution to rise to a target temperature that exceeds 120 degrees Centigrade (C); circulating the alkaline solution through an electrolyzer cell comprising a cathode, an anode, and a diaphragm at least partially disposed between the anode and cathode; and producing a first stream comprising oxygen gas and a second stream comprising hydrogen gas at least in part by inducing a current between the cathode and the anode in the electrolyzer cell.

The method further comprises while producing the first stream and the second stream, operating the electrolyzer at a temperature that exceeds 120 degrees C.; separating the hydrogen gas from the second stream; and outputting the hydrogen gas.

In some implementations, the method further comprises operating the electrolyzer at a temperature of up to and including about 200 degrees C.

In some implementations, the method further comprises causing a pressure on the alkaline solution to rise to a target pressure value that exceeds 8 bar.

In some implementations, causing the pressure of the target alkaline solution alkaline solution to exceed 8 bar comprises causing the pressure on the alkaline solution to rise to about 10 bar.

In some implementations, the method further comprises transferring heat from at least one of the first stream and the second stream to the alkaline solution.