Method for operating a fuel cell device

Abstract

The invention relates to a method (100) for operating a fuel cell device (10) comprising at least one fuel cell stack (12) and at least one fuel cell (46), wherein, in a protective mode of the fuel cell device (10), a protective voltage is applied to the fuel cell (46) which is opposite to the Nernst voltage of the anode of the fuel cell. It is proposed that the electrochemical potential of the fuel cell (46) corresponds to an electrochemical potential in normal operation of the fuel cell device (10).

Description

[0001] The invention relates to a method for operating a fuel cell device according to the preamble of the independent claim. The invention further relates to a fuel cell device configured to carry out such a method. State of the art

[0002] Depending on the design, a SOFC typically operates at temperatures between 500 °C and 900 °C. High temperature gradients within the stack lead to thermomechanical stresses, which can result in cracks in the ceramic layers or deformations of the metallic and / or non-metallic stack components. These deformations can, in turn, lead to various failure scenarios, such as leaks. The heating and cooling process of the stack is particularly critical with regard to temperature gradients and thermomechanical stresses. Therefore, a carefully controlled, often very slow, ramp-up and shutdown of the stack from operating temperature to room temperature is necessary to avoid excessively high temperature gradients and minimize the risk of failure. During heating and cooling, the anode typically needs to be purged with fuel to prevent oxidation of the nickel in the Ni-cermet anode structure.

[0003] Slowly starting up or shutting down the stack also slows the passage through the critical temperature range in which re-oxidation of the Ni-cermet anode can occur. This promotes anode oxidation, especially when no fuel is supplied and the redox potential of Ni / NiO is thereby undershot. Oxidation of the Ni in the Ni-cermet anode leads to volume expansion, cracking, spalling, and ultimately stack failure. Disclosure of the invention Advantages

[0004] The present invention describes a method for operating a fuel cell device comprising at least one fuel cell stack and at least one fuel cell. In a protective mode of the fuel cell device, a protective voltage is applied to the fuel cell, which is opposite to the Nernst voltage of the fuel cell's anode. The protective voltage is selected such that the electrochemical potential of the fuel cell corresponds to an electrochemical potential during normal operation of the fuel cell device.

[0005] This has the advantage of preventing oxidation or re-oxidation of the anode material. For example, if the anode uses nickel, particularly nickel cermet, oxidation or re-oxidation of the nickel can be prevented. Specifically, it is no longer necessary to displace or flush out oxygen from an anode path or anode compartment of the fuel cell or fuel cell stack, for example, by using fuel or steam, in order to protect the anode from oxidation.

[0006] Because the protective voltage is opposite to the Nernst voltage of the fuel cell anode, no oxygen can be incorporated on the cathode side and reach the anode via the electrolyte. This protects the anode from oxidation and re-oxidation, which would otherwise lead to fuel cell degradation. Oxidation and re-oxidation of the anode material, especially nickel or nickel in a nickel-cermet anode, leads to volume expansion, cracking, spalling, delamination of the cell layers or the cell assembly, and ultimately to the failure of the fuel cell and / or the fuel cell stack.

[0007] It is essential that the protective voltage applied to the fuel cell is designed so that the electrochemical potential of the fuel cell corresponds to the electrochemical potential during normal operation of the fuel cell device. In this way, the potential can be maintained during normal operation of the fuel cell device even without fuel on the anode side or in the anode compartment.

[0008] A fuel cell device shall be understood in particular to be a device which forms, in particular, a functional component, especially a structural and / or functional component, of a fuel cell system or the entire fuel cell system.

[0009] In this context, a fuel cell system shall be understood to mean in particular a system for the stationary and / or mobile generation of energy, especially electrical and / or thermal energy, using at least one fuel cell unit.

[0010] In this context, a fuel cell stack is understood to be a unit comprising at least one fuel cell, preferably a plurality of fuel cells. A fuel cell is specifically designed to convert at least the chemical reaction energy of at least one fuel gas, particularly hydrogen, and at least one oxidizing agent, particularly oxygen, into electrical energy. The fuel cell can advantageously be configured as a solid oxide fuel cell (SOFC).

[0011] A fuel cell is an electrochemical cell comprising an anode and a cathode, as well as an electrolyte located between the anode and cathode. Typically, a fuel cell is planar or plate-shaped, and the fuel cells can advantageously be stacked on top of each other in a fuel cell stack. Advantageously, the anode and cathode are each designed as an anode layer and a cathode layer, respectively, and the electrolyte is advantageously designed as an electrolyte layer between the anode and cathode layers. The anode of solid oxide fuel cells can advantageously consist of nickel, in particular a nickel cermet.

[0012] In the context of a fuel cell, the anode is the fuel electrode and the cathode is the air electrode. The fuel cell, or fuel cell stack, is designed such that a fuel, particularly hydrogen, can be supplied to the anode and air can be supplied to the cathode. Typically, each fuel cell is associated with a planar component, which is arranged in the fuel cell stack between two immediately adjacent fuel cells and separates the opposite anode side of one fuel cell from the cathode side of a second fuel cell. This planar component is also called an interconnector.

[0013] The interconnector is designed, for example, by a structure of raised features extending from its surface, such that two fluid chambers are formed between the interconnector and each of the two adjacent fuel cells. These chambers are designated for either fuel or air, and are intended for supplying air to the fuel cell at the cathode facing the interconnector and fuel to the fuel cell at the anode facing the interconnector. The fluid chamber formed between the interconnector and the fuel cell at the cathode facing the interconnector is also referred to as the cathode chamber. The fluid chamber formed between the interconnector and the fuel cell at the anode facing the interconnector is also referred to as the anode chamber. Advantageously, the anode chamber and the cathode chamber are fluidically separated from each other.

[0014] An anode path is defined in particular as a fluid path used to supply the anode with fuel or to remove the anode exhaust gases. The anode path includes, in particular, a fuel supply line, an anode exhaust line, and functional components for regulating the fuel flow and / or the anode exhaust gas flow, as well as functional components for the thermal, mechanical, and / or chemical treatment or conversion of the fuel and / or anode exhaust gas, such as heat exchangers, compressors, fans, valves, reformers, afterburners, filters, and / or catalysts. It is particularly conceivable that the anode path includes an anode recirculation circuit, or simply recirculation circuit.

[0015] In this context, a recirculation circuit or anode recirculation circuit is understood to mean, in particular, a fluid connection or fluid path designed to supply fuel- or hydrogen- and water-containing anode exhaust gas from the fuel cell unit to a mixing point – the recirculation inlet point – where the anode exhaust gas is mixed with the fuel or hydrogen, which is at least substantially pure. Specifically, the recirculation circuit is designed to supply unreacted fuel or hydrogen back to the anode side of the fuel cell unit. Specifically, the mixture of the anode exhaust gas and the fuel or hydrogen is intended to be supplied to the anode side of the fuel cell unit.

[0016] In this context, a reformer or reformer unit is understood to be a chemical-technical unit for the processing of at least one hydrocarbon-containing fuel, in particular by steam reforming and / or partial oxidation and / or autothermal reforming, especially for the production of at least one fuel gas, particularly hydrogen, and / or for breaking down higher-chain alkenes. The reformer typically incorporates a catalyst or at least a catalytic layer for the reforming process. Possible catalyst materials include precious metals, such as platinum.

[0017] The fuel cell device includes a control unit configured to execute the method according to the invention. In particular, the control unit includes a processing unit and a memory and is equipped with appropriate interfaces to receive information, especially measurement data, and to send control commands to functional components of the control unit. The control unit is optionally configured to receive external signals, for example, from an input device for inputting control commands by an operator or from a higher-level control unit—for example, a higher-level overall system consisting of several fuel cell devices or a cloud. The control unit is configured to process received measurement signals and / or control commands and, if necessary, to convert them into control commands.The control unit is specifically designed to supply the fuel cell stack with electricity, in particular to apply a voltage to the stack, and also to further utilize any electrical power generated by the electrochemical reaction during normal operation. For this purpose, the control unit can, for example, include power electronics electrically connected to the fuel cell stack or control a power electronics component electrically connected to the fuel cell stack accordingly.

[0018] In the context of this invention, "normal operation" can be understood as the planned operation of the fuel cell system or fuel cell device under typical stress. In particular, the fuel cell system is also capable of achieving its maximum possible efficiency during normal operation. The operating temperature of the fuel cell of the fuel cell system is typically between approximately 600 °C and 800 °C. During normal operation, hydrogen or synthesis gas and air supplied to the fuel cell stack are at least partially converted to water, generating an electric current.

[0019] A protective mode is defined as a special operating mode of the fuel cell stack in which a protective voltage is applied. It is advantageous to activate the protective mode when the fuel cell or fuel cell stack is not supplied with fuel, for example, during startup, shutdown, and / or in the event of a fuel cell system malfunction that interrupts the fuel supply to the fuel cell stack. For instance, the protective mode could be initiated when the control unit receives a fault signal from a fuel supply component—such as a fuel pump—that indicates an interruption of the fuel supply to the fuel cell stack.It is also conceivable that the protective mode is activated if the fuel supply to the fuel cell stack needs to be interrupted, for example, to protect the fuel cell stack. This may be necessary, for instance, if a fuel conditioning component or a fuel filter fails, or if the fuel has an excessively high concentration of a pollutant that could be particularly harmful to the fuel cell stack – for example, excessive sulfur content.

[0020] Advantageous further developments of the method are possible due to the features listed in the dependent claims.

[0021] A further improvement is possible if the protective voltage is chosen so that it generates a potential sufficient for the conversion of free oxygen on the anode side, especially in the anode compartment of the fuel cell.

[0022] If the protective voltage is selected to generate a potential sufficient to convert the free oxygen present on the anode side, particularly in the anode compartment of the fuel cell, the oxygen ions are transported from the anode side—i.e., the fuel gas side—through the electrolyte to the cathode side—i.e., the air side. The current generated in this way is also known as the pump current. According to Faraday's law, the pump current is proportional to the oxygen concentration or partial pressure in the anode compartment. Therefore, if air or oxygen is still present on the anode side or in the anode compartment, applying the protective voltage transports the oxygen to the cathode side or cathode compartment, leaving only nitrogen or N₂ on the anode side or in the anode compartment. This protects the anode from oxidation or reoxidation, which would otherwise lead to the degradation of the fuel cell.

[0023] In normal operation, however, the oxygen ions are transported in the opposite direction, from the cathode side through the electrolyte to the anode side, where they recombine with the hydrogen supplied there to form water.

[0024] The pump current can be increased within a specific voltage range by raising the protection voltage, thereby increasing the speed or rate of oxygen transport. A higher protection voltage allows oxygen to be transported out of the anode compartment more quickly. It is conceivable that different protection voltages are applied to the fuel cell during protection operation, depending on the specific protection scenario. For example, a lower protection voltage may suffice during a controlled start-up phase, in which the fuel cell stack is heated slowly, compared to a protection phase triggered by a sudden fuel supply failure, which necessitates faster oxygen transport and thus a higher protection voltage.

[0025] Advantageously, when determining the protective voltage, the cell resistance, its variation in different fuel cells, and its changes due to aging are also taken into account. It is conceivable that the control unit includes a function or model that selects the protective voltage in such a way that the desired pumping current is achieved, considering the actual cell resistance.

[0026] It is further advantageous if the fuel cell has a nickel anode and the protective voltage is at least the redox potential of nickel oxidation to nickel oxide. This prevents nickel oxidation at the anode-electrolyte interface. Oxidation at the anode-electrolyte interface can also lead to delamination of the fuel cell's layered structure.

[0027] A further improvement is possible if the protection voltage is at least between 675 mV and 800 mV, preferably at least between 700 mV and 775 mV, and particularly preferably at least between 725 mV and 750 mV. These parameter values ​​have proven to be a good lower limit for the protection voltage to overcome the redox potential of oxidation of the anode material under typical operating conditions of solid oxide fuel cells.

[0028] A further improvement is possible if the protection voltage is selected to achieve a limiting current corresponding to the IU characteristic curve of the fuel cell. The IU characteristic curve refers to the functional relationship between the protection voltage and the pump current under the current operating conditions of the fuel cell – in particular, temperature and oxygen concentration or partial pressure on the anode side. This has the advantage that the protection voltage is sufficiently high to transport the oxygen out of the anode or anode compartment very quickly.

[0029] Typically, the pump current initially increases with increasing protection voltage and, from a protection voltage of 0 V, exhibits a linear or nearly linear increase. This increase flattens out with increasing protection voltage and typically converges to a limiting current. In variants with this improvement, the pump current generated by the protection voltage is the control parameter. For this purpose, the pump current is measured, and the protection voltage is increased until the pump current reaches the limiting current.The pump current can be the oxygen ion current from the anode towards the cathode and / or a resulting electron current between adjacent fuel cells - for example, an electron current via an interconnector between two adjacent fuel cells - from the cathode of a first fuel cell towards an anode of a second fuel cell, wherein the cathode of the first fuel cell and the anode of the second fuel cell are facing each other and are advantageously fluidically separated and electrically connected by an interconnector arranged between the two fuel cells.

[0030] The term "reaching a limiting current" means, in particular, that the pumping current generated by the protective voltage reaches the limiting current. The term "the pumping current reaching the limiting current" means, in particular, that the magnitude of the pumping current is at least 90% of the magnitude of the limiting current, preferably at least 95%, and most preferably at least 99%.

[0031] The limiting current is advantageously determined by applying and increasing a protective voltage to the fuel cell, measuring the resulting pumping current generated by the fuel cell under this protective voltage, and identifying the limiting current when the pumping current reaches a plateau. This enables a particularly reliable and precise determination of the limiting current, which depends on the current operating conditions of the fuel cell. This method has the advantage that the determination and evaluation of operating parameters that influence, or could influence, the limiting current is unnecessary.

[0032] The term "plateau" for the pump current refers specifically to the pump current reaching its limit. This can be detected, for example, if the pump current remains constant despite an increase in the protective voltage, or if it remains constant within the limits of the measurement accuracy. It is also conceivable that the plateau is reached when the pump current increases only slowly with an increase in the protective voltage. For instance, a minimum rate of increase for the pump current, dependent on the protective voltage, could be defined—that is, a value with the unit A / V, which, when multiplied by the positive change or delta of the protective voltage, yields the minimum value by which the pump current must increase—and the plateau is detected when this minimum rate of increase is not reached.However, other methods known from control engineering are also conceivable, for example determining the derivative of a curve representing the IU characteristic curve and falling below a minimum slope of this curve to detect the plateau, optionally in combination with the evaluation of a second derivative and the like.

[0033] It is also conceivable that reaching the plateau is detected by determining a limit value of the pump current from a fit of the measured IU characteristic curve to a model of the IU characteristic curve - for example by a polynomial function and / or power functions - for example by convergence analysis or limit value analysis - and other methods known from mathematics and control engineering - and that reaching the plateau is detected when the pump current reaches the determined limit value.

[0034] A further improvement is possible if an oxygen partial pressure is determined on the anode side, particularly in the anode compartment, and the limiting current is calculated as a function of this oxygen partial pressure. In this variant, characteristic curves are advantageously stored in a control unit, from which a value for the limiting current can be derived as a function of the measured oxygen concentration or oxygen partial pressure on the anode side. This has the advantage that the limiting current can be derived very easily directly from a measurement. In particular, no analysis of the IU characteristic curve is necessary. It is also conceivable that a function is stored in the control unit which, given the oxygen partial pressure as input, provides the appropriate value for the limiting current.In variations of this variant, it is conceivable that the limiting current is determined depending on one or more further operating parameters, for example depending on a determined stack temperature.

[0035] Advantageously, the protective operation of the fuel cell device is used in operating phases when the stack temperature is between a first temperature and a second temperature, where the first temperature is lower than the second temperature.

[0036] Preferably, the method according to the invention is a start-up operation or part of a start-up operation and / or a shutdown operation or part of a shutdown operation.

[0037] The risk of anode material oxidation in the fuel cell stack typically arises above a critical temperature if air or oxygen is present in the anode compartment. This leads to degradation of the fuel cell stack and consequently to performance losses. Therefore, when the stack temperature exceeds the critical temperature, no air or oxygen should be present in the anode compartment or at the anode of the fuel cell. The critical temperature depends on the properties of the fuel cell stack and, depending on the cell chemistry and stack design, is approximately between 430°C and 490°C, typically between 450°C and 470°C.

[0038] Therefore, the initial temperature is advantageously set below the upper critical temperature. This protects the fuel cells before the critical temperature is reached, or before the fuel cell stack has cooled sufficiently below the critical temperature.

[0039] It is further advantageous if the first temperature is between 380°C and 430°C, preferably between 390°C and 420°C, and particularly preferably between 400°C and 410°C. These temperature ranges enable particularly efficient and safe operation, especially when no protective voltage is applied or no protective operation is used below the first temperature. Therefore, the highest possible first temperature saves energy, as it prevents the protective voltage from being applied too early (during startup) or for too long (during shutdown). On the other hand, to maintain stack health, the activation of the protective voltage should be reliably below the upper critical temperature. The specified temperature ranges have proven to be parameters that enable high safety with high efficiency.

[0040] Advantageously, the second temperature lies between 480°C and 530°C, preferably between 490°C and 520°C, and most preferably between 500°C and 510°C. These temperature ranges enable particularly economical and safe operation, especially when no protective voltage is applied or protective operation is used above the second temperature. Typically, during normal operation or from a sufficient stack temperature, the reforming in the reformer, particularly with a recirculation circuit, and the synthesis gas and steam thus generated, ensure the anode is protected from oxidation. In this case, the additional application of the protective voltage or the use of protective operation is no longer necessary.

[0041] The values ​​for the first temperature and / or second temperature can be advantageously stored on a control unit.

[0042] The term "stack temperature" refers specifically to a temperature characteristic of the operation of the fuel cell stack. Typically, the stack temperature can be measured as the temperature at the inlet of the air flowing into the fuel cell stack or the medium flowing into the cathode compartment of the fuel cell stack, or as the temperature at the inlet of the fuel flowing into the fuel cell stack or the medium flowing into the anode compartment of the fuel cell stack.

[0043] Furthermore, it is possible to measure the stack temperature as the temperature at the outlet of the air flowing out of the fuel cell stack or of the medium flowing out of the cathode compartment of the fuel cell stack, or as the temperature at the outlet of the fuel flowing out of the fuel cell stack or of the medium flowing out of the anode compartment of the fuel cell stack.

[0044] Within the scope of the present invention, it is advantageous if the stack temperature is measured as the temperature of the medium flowing from the cathode compartment of the fuel cell stack. The numerical values ​​for specific temperatures refer, within the scope of the present invention, to embodiments in which the stack temperature is measured as the temperature of the medium flowing from the cathode compartment of the fuel cell stack.

[0045] The invention further relates to a fuel cell device comprising at least one fuel cell stack and at least one fuel cell, as well as a control unit which is configured to carry out a method according to the present invention.

[0046] A major advantage of the method according to the invention, besides stack protection, is the fact that, unlike prior art stack protection measures such as protection by purging the stack with steam or hydrogen, no additional components are required in the fuel cell device. In particular, the use of the method according to the invention eliminates the need for components for the supply, transport, and mixing of hydrogen or steam, such as hydrogen tanks, water tanks, corresponding pumps and metering devices, mixing points, vaporization devices, corresponding lines, and so on.This allows for cost savings in the manufacture of the fuel cell device, the reduced complexity enables a simpler and more efficient design, and ultimately increases durability and reliability while reducing maintenance requirements.

[0047] The fuel cell device can be improved with an oxygen sensor, in particular a lambda probe, on the anode path. In this way, an advantageous variant of the method according to the invention can be implemented, in which, during protection mode, the limiting current is determined as a function of the oxygen concentration or the oxygen partial pressure on the anode side, particularly in the anode compartment, to set the value of the protection voltage. Advantageously, the oxygen sensor is arranged on the anode compartment. However, it is also conceivable that the oxygen sensor is arranged in the anode path near the anode compartment, and that the oxygen concentration or the oxygen partial pressure in the anode compartment is derived from the measurement of the oxygen concentration or the oxygen partial pressure in the anode path.For example, it is conceivable that the oxygen sensor is located at the end of the fuel supply line leading into the fuel cell stack, or at the end of the anode exhaust line leading out of the fuel cell stack. This enables fast and reliable regulation and control of the fuel cell device, and allows for the provision of a particularly safe – especially fail-safe – and durable fuel cell system. Drawings

[0048] The drawings illustrate exemplary embodiments of the method for operating a fuel cell device and of the fuel cell device itself, which are explained in more detail in the following description. They show Fig. 1 a schematic circuit diagram of an exemplary embodiment of a fuel cell device, Fig. 2 a schematic representation of a fuel cell in the fuel cell stack, Fig. 3 a schematic flowchart of a procedure for operating the fuel cell device and Fig. 4 different IU characteristic curves. Description

[0049] In Fig. Figure 1 shows a schematic circuit diagram of an embodiment of a fuel cell device 10. The fuel cell device 10 comprises two fuel cell stacks 12, which have a plurality of fuel cells 46, in this case solid oxide fuel cells (SOFC), and a plurality of processor units 14.

[0050] In the context of this invention, a processor unit 14 shall be understood to be, in particular, a unit or component of the fuel cell device 10 that is not a fuel cell and / or a fuel cell stack 12. In the present case, the processor units 14 are units for the chemical and / or thermal pre- and / or post-treatment of at least one medium to be converted and / or converted in the fuel cell stack 12, such as, for example, a fuel gas, air and / or exhaust gas.

[0051] One of the processor units 14 is a heat exchanger 18 arranged in an air supply line 16 for heating the air L supplied to one of the fuel cell stacks 12. In this case, the air L, for example in normal operation, is supplied to a cathode compartment 20 of each of the fuel cell stacks 12, while reformed fuel RB is supplied to an anode compartment 22 of each. In the fuel cell stacks 12, the reformed fuel is electrochemically converted to generate electricity and heat.

[0052] The reformed fuel RB is produced by supplying fuel B, in this case natural gas, to the fuel cell device 10 via a fuel supply line 24 from a fuel supply 25 or fuel source, which is reformed in a further processor unit 14, in this case a reformer 26.

[0053] Furthermore, the fuel cell stacks 12 are connected on the exhaust side to another processor unit 14, in this case an afterburner 28. Exhaust gas from the fuel cell stacks 12 is supplied to the afterburner 28, in this case via a cathode exhaust duct 30 (cathode exhaust KA) and an anode exhaust duct 32 (anode exhaust AA). The cathode exhaust contains predominantly unburned air L, while the anode exhaust AA contains, among other things, unreacted fuel B. Furthermore, the anode exhaust AA typically contains water vapor and carbon dioxide. By means of the afterburner 28, the anode exhaust AA, or rather the unreacted fuel B contained therein, is combusted with the addition of the cathode exhaust KA, or rather the air L contained therein, thereby generating additional heat.

[0054] The hot exhaust gas A produced during combustion in the afterburner 28 is discharged from the afterburner 28 via an exhaust gas duct 34 and a further processor unit 14, in this case via a heat exchanger 36. The heat exchanger 36 is fluidically connected to the reformer 26, so that heat is transferred from the hot exhaust gas A to the fuel B supplied to the reformer 26. Accordingly, the heat from the hot exhaust gas A can be used for reforming the supplied fuel B in the reformer 26.

[0055] Downstream of the heat exchanger 36, another processor unit 14, in this case the heat exchanger 18, is located in the exhaust gas duct 34, so that the remaining heat of the hot exhaust gas A can be transferred to the supplied air L in the air supply line 16. Similarly, the remaining heat of the hot exhaust gas A can be used to preheat the supplied air L in the air supply line 16. Downstream of the heat exchanger 18, the exhaust gases A are discharged, for example, via a chimney from the fuel cell device 10.

[0056] Furthermore, the fuel cell device 10 has a return line 38 by means of which anode exhaust gas AA can be partially diverted from the anode exhaust line 32 and fed to the fuel supply line 24. The return line 38, together with the fuel supply line 24, forms an anode recirculation circuit 40 by means of which anode exhaust gas AA can be returned to the anode 22 of the fuel cell stack 12, so that any unreacted fuel B in the anode exhaust gas AA can subsequently be converted, thereby further increasing the efficiency of the fuel cell device 10. The anode recirculation circuit 40 includes a compressor 42. The compressor 42 is fluidically connected to the return line 38, so that the recirculation rate of the anode exhaust gas AA can be regulated via the compressor 42. The compressor 42 is a recirculation blower.The compressor 42 is also fluidically connected to the fuel supply 25 via a fuel supply line 24 and in particular also enables the control of the supply of fuel B in the fuel supply line 24 to the reformer 26 or to the fuel cell stacks 12.

[0057] In the anode recirculation circuit 40, downstream of the compressor 42, a heat exchanger 44 is arranged. This heat exchanger transfers the heat from the anode exhaust gas AA flowing into the anode recirculation circuit 40 via the return line 38 to the fuel B or anode exhaust gas AA flowing from the anode recirculation circuit 40 to the reformer 26 via the fuel supply line 24. Downstream of the heat exchanger 44, in the direction of flow of the fuel B or anode exhaust gas AA, is the heat exchanger 14. This heat exchanger transfers the heat from the hot exhaust gases A flowing from the afterburner 28 via the exhaust gas duct 34 to the fuel B or anode exhaust gases AA flowing to the reformer via the fuel supply line 24.

[0058] Furthermore, the fuel cell device 10 has a starter burner 54 for heating the fuel cell device 10 during start-up and for controlled cooling during shutdown. The starter burner 54 combusts fuel B – for example, natural gas – and air L to produce hot exhaust gases A. The exhaust gases A are routed from the starter burner 54 via a further exhaust gas duct 34 to the afterburner 28. In this way, the afterburner 28 can be brought up to operating temperature during start-up. From the afterburner 28, the hot exhaust gases A flow to the heat exchanger 36 and to the heat exchanger 18.

[0059] During start-up and shutdown, air L is transported via air supply line 16 to the cathode compartments 20 of the fuel cell stacks 12, whereby the air L is heated via the heat exchanger 18 by the exhaust gases A of the start-up burner 54. Furthermore, during start-up and shutdown, air L can be transported via compressor 42, fuel supply line 24, and reformer 26 to the anode compartments 22 of the fuel cell stacks 12. In the off state, the anode recirculation circuit 40 and the anode compartment 22 are at least partially, preferably substantially, filled with air L. Residues of the anode exhaust gas AA, including residues of unreacted fuel B, may also be present. This gas mixture is transported via compressor 42 to the anode compartments 22, where it is heated via heat exchanger 36 by the exhaust gases A of the start-up burner 54.

[0060] Fig. Figure 2 shows a detail from a sectional view of a fuel cell stack 12, depicting a partial view of a section of two adjacent fuel cells 46 stacked directly above one another. In the figure above, a first fuel cell 46a is arranged, and a second fuel cell 46b is arranged below it. An interconnector plate 48 is arranged between the two fuel cells 46.

[0061] The fuel cells 46 each have a metal support 50. An anode layer 52 is arranged on the metal support. This layer is, for example, a nickel cermet. The anode layer 52 is completely covered by an electrolyte layer 54. A cathode layer 56 is arranged on the electrolyte layer 54. The metal supports 50 have a plurality of perforations or holes, which are designed to conduct a fuel B to the anode layer 52.

[0062] The interconnector plate 48 has protrusions extending towards either the first fuel cell 46a or the second fuel cell 46b. The interconnector plate 48 makes contact with the first fuel cell 46a on the anode side via its protrusions, thereby contacting the metal support 50. The interconnector plate 48 also makes contact with the second fuel cell 46b on its cathode side, directly contacting the cathode layer 56. The structuring of the interconnector plate 48 with its protrusions thus creates two fluidically separated fluid spaces.

[0063] A first fluid chamber exists between the first fuel cell 46a and the interconnector plate 48 and is located on the anode side of the first fuel cell 46a; this first fluid chamber is part of the anode chamber 22 of the fuel cell stack 12, or the anode chamber 22 of the first fuel cell 46a. A second fluid chamber exists between the second fuel cell 46b and the interconnector plate 48 and is located on the cathode side of the second fuel cell 46b; this second fluid chamber is part of the cathode chamber 20 of the fuel cell stack 12, or the cathode chamber 20 of the second fuel cell 46b.

[0064] Fig. Figure 3 illustrates a method 100 for operating the fuel cell device 10. The sequence of a start-up operation, which includes a protective operation, is explained as an example. In this process, the fuel cell stacks 12 are heated via the start burner 54, the heat exchanger 36, and the heat exchanger 18. For this purpose, warm air is transported on the cathode side via the air supply line 16 and through the heat exchanger 18 into the cathode chambers 20.

[0065] In a first step S1, the recirculation blower 42 is started. This moves the fluids in the anode path, in particular through the recirculation circuit 40 itself, through the fuel supply line 24, through the reformer 26, through the anode chambers 22, and through the anode exhaust duct 32. The aim is to incorporate oxygen L from the air into the reformer 26 or the catalyst material of the reformer 26. The incorporation of oxygen into the reformer 26 is made possible by the rising stack temperature. The fuel supply 25 is closed, so that no fresh fuel B is introduced into the anode path, in particular the anode recirculation circuit 40 and the anode chambers 22.

[0066] During the process, the stack temperature is measured by temperature sensors (not shown) located directly on the respective cathode exhaust ducts 30, adjacent to the fuel cell stacks 12 and cathode chambers 20. These temperature sensors transmit the stack temperature to a control unit (not shown), which executes the process 100. The control unit is connected to the necessary controllable pumps, valves, and other actuators required to start, enable, stop, or suppress the fluid flows necessary in the respective steps. Furthermore, the control unit regulates the fuel cell stacks 12, their electrical power, the voltage applied to them, and the current flow, in particular a protective voltage applied to the fuel cell stacks 12.

[0067] As soon as the stack temperature reaches the first temperature T1, which is 400°C in this example, the protection operation is started in step S2 and a protection voltage is applied to the fuel cell stacks 12 and the fuel cells 46. This protection voltage is opposite to the Nernst voltage of the anode of the fuel cells 46. The protection voltage is continuously increased until the pump current generated by the protection voltage reaches a limit current 104 of an IU characteristic curve 102. The control unit determines and evaluates the pump current. The limit current 104 is determined by detecting a plateau in the pump current as the protection voltage is increased. The limit current 104 is in Fig. Figure 4 illustrates this. The protective voltage transports oxygen trapped in the anode material to the cathode as oxygen ions. In this way, the anode path is further cleared of oxygen.

[0068] As soon as the stack temperature reaches the second temperature T2, which is 500°C in this example, the protection operation is terminated and the protective voltage is switched off in step S3. Additionally, the fuel supply 25 is opened, so that the fuel supply line 24 now carries fresh fuel B. With the addition of fuel B, in this case natural gas, the fatty gas components of the natural gas release the oxygen that was incorporated into the catalyst material of the reformer 26 in step S1. The fatty gas components are reacted by the oxygen, producing water vapor and CO2, which are returned to the reformer 26 via the recirculation circuit 40. The steam reforming that now begins in the reformer produces synthesis gas with a sufficient hydrogen concentration for current-driven heating operation.

[0069] In a subsequent step S4, the fuel cell stacks 12 are controlled by the control unit so that they can deliver electrical power; in particular, reformed fuel RB or hydrogen can be electrochemically converted. During this phase, stack protection is achieved via the water or water vapor D contained in the recirculated anode exhaust gases AA. Additional protection by the protective voltage is no longer necessary. During this phase, the stack temperature reaches the operating temperature.

[0070] During shutdown, the stack temperature is cooled down from the operating temperature to room temperature. The steps of procedure 100 S1 to S4 essentially run in reverse order compared to the start-up phase.

[0071] Fig.Figure 4 illustrates the limiting current 104 using IU characteristic curves 102 and the functional relationship between the protection voltage and the pump current. The protection voltage is plotted on the abscissa 106, and the pump current on the ordinate 108. Three different IU characteristic curves 102 are shown, each for a different oxygen concentration or partial pressure on the anode side.

[0072] The first IU characteristic curve 102a shows the relationship at a very low oxygen concentration on the anode side of 1 vol.%. At the first voltage U1 of, for example, 0.05 V for the protection voltage, the pump current is an initial current I1 of 0.14 A. When the protection voltage is increased, the pump current rises only slightly and at the second voltage U2 of 0.35 V already reaches the first limiting current 104a of 0.23 A, which remains largely constant up to a voltage of 1.00 V.

[0073] The second IU characteristic curve 102b shows the relationship at an oxygen concentration on the anode side of 10 vol.%. At the first voltage U1, the pump current is, for example, the second current I2 of 0.41 A. With an increase in the protection voltage, the pump current rises almost linearly with a slope of approximately 5.5 A / V. From a third voltage U3 of, for example, 0.24 V and a third current I3 of 1.45 A, the rate of increase of the pump current drops significantly, and the pump current slowly converges to the second limiting current 104b of approximately 1.75 A, which it reaches at approximately 1.00 V.

[0074] The third IU characteristic curve, 102c, shows the relationship at an oxygen concentration of 25 vol.% on the anode side. At the first voltage, U1, the pump current is, for example, the second current, I2. With an increase in the protection voltage, the pump current rises almost linearly and almost exactly as in the second IU characteristic curve, 102b. From the third voltage, U3, the pump current, unlike in the second IU characteristic curve, 102b, does not saturate but continues to rise quasi-linearly, for example, even with an increased slope of approximately 8.0 A / V. Only when the fourth voltage, U4, of, for example, 0.62 V, is reached does the rate of increase of the pump current drop sharply, and the pump current slowly converges to the third limiting current, 104c, of approximately 4.50 A, which it reaches at approximately 1.00 V.

[0075] As can be clearly seen from the three IU characteristic curves 102, the limiting current 104 depends strongly on the oxygen concentration or the oxygen partial pressure. Likewise, the value of the protection voltage at which the pump current approaches the limiting current 104 is strongly dependent on the oxygen concentration or the oxygen partial pressure.

Claims

[1] Method (100) for operating a fuel cell device (10) comprising at least one fuel cell stack (12) and at least one fuel cell (46), wherein in a protective mode of the fuel cell device (10) a protective voltage is applied to the fuel cell (46) which is opposite to the Nernst voltage of the anode of the fuel cell, characterized by , that the protective voltage is selected such that the electrochemical potential of the fuel cell (46) corresponds to an electrochemical potential in normal operation of the fuel cell device (10). [2] Method (100) according to claim 1, wherein the protective voltage is selected such that it generates a potential which is sufficient for the conversion of free oxygen on the anode side, in particular in the anode space (22) of the fuel cell (46). [3] Method (100) according to any one of the preceding claims, characterized by, that the fuel cell (46) has a nickel anode and the protective voltage is at least the redox potential of the oxidation of nickel to nickel oxide. [4] Method (100) according to any one of the preceding claims, characterized by that the protective voltage is at least between 675 mV and 800 mV, preferably at least between 700 mV and 775 mV, particularly preferably at least between 725 mV and 750 mV. [5] Method (100) according to any one of the preceding claims, characterized by , that the protection voltage is selected such that a limiting current (104) of an IU characteristic curve (102) of the fuel cell (46) is reached. [6] Method (100) according to claim 5, characterized by, that the limiting current (104) is determined by applying and increasing a protective voltage to the fuel cell (46), recording a pumping current resulting from the protective voltage through the fuel cell (46), and detecting the limiting current (104) when the pumping current reaches a plateau. [7] Method (100) according to claim 5 or 6, characterized by , that an oxygen partial pressure is determined on the anode side, in particular in the anode space (22) and the limiting current (104) is determined as a function of the oxygen partial pressure. [8] Method (100) according to any one of the preceding claims, characterized by , that the protective operation of the fuel cell device (10) is applied in operating phases when a stack temperature is between a first temperature (T1) and a second temperature (T2), wherein the first temperature (T1) is less than the second temperature (T2). [9] Fuel cell device (10) comprising at least one fuel cell stack (12) comprising at least one fuel cell (46) and a control unit configured to perform a method (100) according to any of the preceding claims. [10] Fuel cell device (10) according to claim 9 with an oxygen sensor, in particular a lambda probe on the anode path (24, 32, 40).

Images