JP2002520778A - Sensor cells for electrochemical fuel cell stacks - Google Patents

Abstract

(57) [Summary] An electrochemical fuel cell stack has a plurality of fuel cells. At least one of the fuel cells is a sensor cell. The sensor cell has at least one structural difference that differs from the plurality of residual fuel cells. The structural differences include, for example, a reduced sensor cell electrochemically active area, a reduced electrocatalyst loading, a modified anode or cathode flow area, a different electrocatalyst composition, or a modified coolant flow area configuration. included. The sensor cell operates under substantially the same conditions as the remaining cells in the stack. However, in response to changes in specific stack operating conditions, an electrical or thermal response, preferably a voltage change, is induced in the sensor cell, which does not occur simultaneously with the remaining fuel cells. Thus, the sensor battery can detect an undesired condition and use its response to initiate a corrective action. More than one sensor cell specified for different types of conditions may be used in the stack. In the absence of undesired conditions, the sensor cell can serve as a power generating fuel cell.

Description

DETAILED DESCRIPTION OF THE INVENTION

TECHNICAL FIELD [0001] The present invention relates to electrochemical fuel cells, and in particular, incorporates one or more specific sensor fuel cells into a fuel cell stack and other components in the stack. It relates to detecting a condition in question before the battery is adversely affected by the condition in question.

BACKGROUND OF THE INVENTION Electrochemical fuel cells convert fuel and oxidants into electricity and reaction products. Solid polymer electrochemical fuel cells generally use a membrane electrode assembly (MEA),
It consists of a porous electrically conductive sheet material, for example, an ion exchange membrane or solid polymer electrolyte interposed between two electrodes consisting of a layer of carbon fiber paper or carbon woven fabric. . The MEA has a catalyst layer at each membrane / electrode interface, typically in the form of finely divided platinum, to cause the desired electrochemical reaction. In operation, the electrodes are electrically connected, providing a circuit for conducting electrons between the electrodes through an external circuit.

At the anode, a fuel stream travels through a porous anode substrate and is oxidized at an anode electrocatalyst layer. At the cathode, the oxidant stream travels through the porous cathode substrate and is reduced at the cathode electrode catalyst layer to form a reaction product.

In fuel cells using hydrogen as fuel and oxygen-containing air (or substantially pure oxygen) as oxidant, the catalytic reaction at the anode produces hydrogen cations (protons) from the supplied fuel. The ion exchange membrane facilitates the transfer of protons from the anode to the cathode. In addition to producing protons, the membrane separates the hydrogen-containing fuel stream from the oxygen-containing oxidant stream. In the cathode electrode catalyst layer, oxygen reacts with protons passing through the membrane to form water as a reaction product. The reaction between the anode and the cathode in a hydrogen / oxygen fuel cell is represented by the following formula: Anode reaction: H ₂ → 2H ⁺ + 2e ⁻ Cathode reaction: 1 / 2O ₂ + 2H ⁺ + 2e ⁻ → H ₂ O

[0005] In a typical fuel cell, the MEA is located between two electrically conductive fluid flow area plates or separator plates. The fluid flow area plate has at least one flow passage formed in at least one of its flat major surfaces. The flow channels allow fuel and oxidant to flow toward or across the respective electrodes, ie, the anode on the fuel side and the cathode on the oxidant side. The fluid flow area plate acts as a current collector, providing support for the electrodes, providing grooves for fuel and oxidant to go to the anode and cathode surfaces, respectively, and reaction products such as water formed during operation of the cell. To provide grooves for removal. The separator plate has no flow passages formed on its surface, but is used in combination with an adjacent material layer that provides a passage for fuel and oxidant to the anode and cathode electrocatalysts, respectively.

[0006] Two or more fuel cells can be electrically connected together in series to increase the overall output power of the assembly. In a tandem arrangement, one side of a given fluid flow area or separator plate may serve as the anode plate of one battery, and the other side of the fluid flow area or separator plate may serve as the cathode plate of an adjacent battery. it can.
Such a multi-layer fuel cell array is referred to as a fuel cell stack and usually holds a series of fuel cell assemblies together in an assembled state with connecting rods and end plates sandwiched between a pair of end plates. The stack includes a fluid fuel stream (eg, substantially pure hydrogen, reformed methanol or reformed natural gas, or a methanol-containing stream in a direct methanol fuel cell) and a fluid oxidant stream (eg, substantially pure oxygen, oxygen). It typically has inlets and manifolds for passing the contained air or oxygen in a carrier gas such as nitrogen) to the individual fuel cell reactant flow passages. The stack generally has an inlet and a manifold for directing a flow of fluid coolant, typically water, to internal passages in the stack to absorb heat generated by the operating fuel cell. The stack generally has exhaust manifolds and outlets for discharging spent reaction streams and reaction products, such as water, as well as exhaust manifolds and outlets for the coolant stream exiting the stack. In a power plant, various fuel, oxidant, and coolant conduits move their fluid streams into and out of the fuel cell stack.

[0007] The performance of a fuel cell stack is typically monitored by detecting the voltage of individual cells or groups of cells in the stack. A typical stack typically has 30 to 200 cells. Voltage detection for individual cells or groups of cells is expensive, and detects and determines voltage conditions that are outside of a predetermined voltage range and takes corrective action, or correct operating conditions (i.e., desired or desired). Stopping the stack until conditions within the preferred range are restored requires complex data collection equipment and control algorithms. A typical method of monitoring fuel cell performance using voltage sensing is described in U.S. Pat. No. 5,170,124. U.S. Pat. No. 5,170,124 describes an apparatus and method for measuring and comparing a voltage of a group of cells in a fuel cell stack to a reference voltage. If the measured voltage differs from the reference voltage by more than a predetermined amount, an alarm signal or process control action is initiated, and a shutdown procedure can be performed or a corrective action can be initiated. Although this voltage sensing solution determines the presence of an abnormal condition, the solution is incorrect with respect to the cause and / or nature of the problem that causes the abnormal condition.

DISCLOSURE OF THE INVENTION In an electrochemical fuel cell stack having a plurality of fuel cells, each fuel cell
An anode having an anode electrode catalyst, a cathode having a cathode electrode catalyst, and an ion exchange membrane interposed between the anode and the cathode are provided. At least one of the plurality of fuel cells is a sensor cell having at least one structural difference different from the plurality of remaining fuel cells. During operation of the stack, the structural differences cause an electrical and thermal response to the sensor cell, which does not occur simultaneously with the remaining fuel cells under substantially the same operating conditions.

[0009] Accordingly, the sensor cells preferably operate under substantially the same conditions as the remaining non-sensor cells. However, in response to changes in certain conditions, electrical and / or
A thermal response (preferably a voltage change) is triggered in the sensor cell, but not simultaneously in the remaining fuel cells. Sensor cells react differently to undesired conditions before they adversely affect other batteries in the stack and can therefore be used to detect those conditions. Different electrical and / or thermal responses of the sensor cell to specific operating conditions can provide diagnostic information or provide an alarm signal or initiate a specific correction procedure to restore the stack to a normal operating condition. Or can be used to stop the stack if it cannot be restored to normal. More than one type of sensor cell specific for different types of conditions can be used in the fuel cell stack.

During operation of the fuel cell stack, the anode has a fuel flow directed thereto, typically through an anode flow area, and the cathode has an oxidation directed there, typically through a cathode flow area. Each of the fuel stream and the oxidant stream having an agent flow and directed to the fuel cells of the stack has an inlet pressure and an associated stoichiometry. Each of the plurality of fuel cells generates a current per unit area of the cathode, producing water at the cathode and producing heat at a nominal operating temperature. Said structural differences are due to the fact that in the sensor cell, under substantially the same operating conditions of reactant supply, e.g., cell inlet reactant pressure and stoichiometry, electrical And / or preferably cause a thermal response. The plurality of fuel cells are connected by a manifold such that reactants are supplied in a parallel supply form. This arrangement generally allows the cells to easily operate under substantially the same conditions with respect to the reactant feed.

[0011] Sensor cells incorporated into the stack can also serve as useful power generating cells. Thus, during operation to generate power for the stack, the sensor battery (s) and the remaining battery are connected to provide power. A variable electrical load can be applied through the fuel cell stack with the sensor cell (s).

[0012] In the above aspect, the change that is induced in the sensor cell but not simultaneously in the remaining fuel cells in response to a change in specific stack operating conditions is preferably a voltage. The sensor cell is designed so that its voltage drops before or even more quickly than the voltage of other cells in the stack in response to undesirable operating conditions.

Other responses that occur in the sensor cell in response to changes in specific stack operating conditions, but not simultaneously in the remaining fuel cells, include current per unit area of cathode of the sensor cell (current density). Changes in the voltage distribution within the components of the sensor cell; changes in the electrical resistance of the sensor cell; and / or changes in the temperature of the sensor cell. The voltage (or other characteristic) of the sensor cell may be substantially the same as or different from the remaining cells under normal operating conditions, but what is important for the functioning of the sensor cell is the condition The difference in their response to change.

Various structural differences can be incorporated therein to provide increased sensitivity to the sensor cell for particular stack operating conditions. For example, in a preferred embodiment, at least one of the anode and cathode of the sensor cell may have an electrochemically active area smaller than the electrochemically active area of the remaining fuel cell. Reactants that go to at least one of the anode and cathode electrocatalysts in the sensor cell may, for example, block some of the reactant flow passages or channels in the porous electrode or flow area plate compared to the rest of the fuel cell. Alternatively, it may be inhibited by removing or masking certain areas of the cell. The addition amount of the electrode catalyst of at least one of the anode and the cathode of the sensor cell may be smaller than the addition amount of the electrode catalyst of the remaining fuel cell. In general, the reduction in the electrochemically active area and / or the reduction in the amount of electrocatalyst added
Increases sensor cell performance sensitivity to various undesirable fuel cell operating conditions. However, such solutions alone may not be sufficient if the sensor cell must distinguish between different conditions that affect performance.

In another preferred embodiment, the configuration of the reactant flow region of the sensor cell is different from that of the remaining fuel cell. For example, the reactant flow zone configuration of the sensor cell may cause a greater pressure drop than occurs in the remaining fuel cell reactant flow zone configuration. The sensor battery is
The remaining fuel cell may include an anode or cathode electrocatalyst different in composition from the corresponding anode or cathode electrocatalyst. In a stack where each of the fuel cells has a coolant flow that is routed along a coolant flow area while conducting heat with each of the fuel cells, the coolant flow area configuration of the sensor cell is such that the remaining fuel cell coolant You may make it different from a distribution area form.

Structural differences include one or more specific stack operating conditions, such as fuel stoichiometry, oxidizing agent stoichiometry, carbon monoxide poisoning of the anode electrocatalyst, fuel flow. Air bleed level, fuel cell operating temperature, methanol poisoning, electrode overflow (
The sensor cells are selected to be more sensitive to changes in flooding, dehydration and / or mass transfer problems. In a direct methanol fuel cell stack, it is useful to incorporate a sensor cell where the structural differences make the sensor cell more sensitive to methanol concentration. In some embodiments, the condition is exacerbated therein so that the sensor cell responds more quickly, while in other embodiments, the sensor cell is simply made more sensitive to the condition.

Thus, in a preferred embodiment, the sensor cell is operated in an undesired fuel cell operating state (compared to the other cells in the stack), for example:-low reactant stoichiometry; -Low or excessive fuel cell operating temperature;-carbon monoxide poisoning or other types of poisoning of the electrocatalyst;-as by accumulation of inerts such as nitrogen in the reaction stream. More sensitive to at least one particular reaction dilution;

A sensor cell is affected by one or more of the above or other undesirable conditions before other cells. They detect the need for corrective action and preferably do not cause degradation or damage to the overall stack by inducing it.
It can be used to prevent such undesirable conditions. Early detection of undesirable or potentially harmful operating conditions is especially important in large stacks with many individual fuel cells. This is because the demands for stack reliability and longer stack life are increasing. Incorporating one or more sensor cells also generally reduces the cost of the stack control system, simplifies it, and also increases the stability and reliability of the stack.

A method of operating an electrochemical fuel cell system comprising a fuel cell stack and a sensor cell as described in any of the above aspects comprises: at least one of the sensor cell electrical and thermal operating parameters; Monitoring at least one of the same parameters of the battery of the sensor cell and comparing the monitored operating parameter of the sensor battery with that of the at least one remaining cell, if the compared parameter indicates an undesirable operating condition. Generating an output signal if it differs by more than the predetermined threshold amount shown.

The preferred parameter to monitor is the voltage or temperature of the battery, but, as noted above, many other parameters can be used. The parameter measured and compared may be, for example, the actually measured voltage value or the rate of voltage change occurring in the sensor cell versus at least one other cell in response to changes in operating conditions.

The output signal may include a number of signals, including the generation of an alarm signal to alert an operator of the device to the presence of an undesirable operating condition, the initiation of data recording, or the triggering of a corrective action to restore the desired operating conditions of the fuel cell system. Can be obtained by connecting to the control system. Monitoring of the selected parameter (s) can be performed, for example, continuously or periodically during operation of the stack.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 illustrates a typical fuel cell 10. The fuel cell 10 includes a membrane electrode assembly 12 interposed between an anode flow area plate 14 and a cathode flow area plate 16.
Having. The membrane electrode assembly 12 has two electrodes, an anode 21 and a cathode 2.
The ion exchange membrane 20 is interposed between the two. In a conventional fuel cell, the anode 21 and cathode 22 are made of a porous electrically conductive sheet material 23 and 24, respectively.
, For example, a substrate of carbon fiber paper or carbon woven fabric. Each substrate has a thin layer of electrocatalyst 25 and 26, respectively, disposed on one surface thereof at the interface with the membrane 20 to electrochemically activate each electrode.

Further, as shown in FIG. 1, the anode flow area plate 14 has at least one fuel flow groove 14 a dug, roughened, or molded on the surface facing the anode 21. Similarly, the cathode separator plate 16 has its cathode 22
It has at least one oxidizing agent flow groove 16a dug, roughened, or molded on the surface facing the surface. When the cooperating surfaces of the electrodes 21 and 22 are assembled together,
Grooves 14a and 16a form reactant flow area passages for fuel and oxidant, respectively. The flow area plate is electrically conductive.

Referring now to FIG. 2, the fuel cell stack 100 has a plurality of fuel cell assemblies,
The series is shown as 111 in FIG. . Each of the fuel cell assemblies has a membrane electrode assembly 112 interposed between a pair of fluid flow area plates 114,116. The fuel cell stack 100 also has a first end plate 130 and a second end plate 140.

The plate 130 has fluid inlets 132, 134, 136 for introducing fluid fuel, oxidant, and coolant streams, respectively, into the stack. Plates 140 provide fluid outlets 142, 1, 1, respectively, for discharging fluid fuel, oxidant, and coolant streams from the stack.
44 and 146. The fluid outlets are in flow communication with corresponding fluid inlets through passages in the stack 100.

The fuel cell assembly has a series of openings formed therein, which cooperate with corresponding openings in adjacent assemblies in the fluid manifolds 152, 1, 1.
54, 156, 162, 164, and 166 are formed. The fuel flow entering the stack through the fuel inlet 132 passes through the manifold 152 to the individual fuel flow area plates. After passing through the fuel flow area plate grooves, the fuel flow is collected at the manifold 162 and exits the stack through the fuel outlet 142. Similarly, oxidant streams entering the stack through oxidant inlet 134 pass through manifolds 154 to individual oxidant flow area plates. After passing through the oxidant flow area plate grooves, the oxidant stream collects at manifold 164 and exits the stack through oxidant outlet 144. Fluid coolant (typically water) introduced through coolant inlet 136 passes through manifold 156 to a coolant plate assembly (not shown) in stack 100. The coolant stream is collected at manifold 166 and exits the stack through coolant outlet 146.

The connecting rods 170 extend between the end plates 130 and 140, and in the assembled state of the stack 100, tightening nuts 172 disposed at both ends of each connecting rod, and the tightening nuts 172 and the end plates 130, Disc spring 17 inserted in between
4. Compress and fix.

FIG. 3 is a schematic diagram of a fuel cell power generation device 200 having a fuel cell stack 210. The fuel cell stack 210 includes-and + bus plates 212, 21.
4 to which an external circuit having a variable load 216 can be electrically connected by closing switch 218. This device has a fuel (hydrogen) circuit,
It has an oxidizer (air) circuit and a cooling water circuit. Reactant and coolant streams circulate in the apparatus through various conduits schematically illustrated in FIG.

A hydrogen supply 220 is connected to the stack 210, and the supply pressure is controlled by a pressure regulator 221. Water in the hydrogen stream exiting stack 210 is knocked out (k
nockout) accumulates on drum 222, which can be exhausted by opening valve 223. Unreacted hydrogen is recirculated to stack 210 by pump 224 in recirculation path 225. An air supply 230 is connected to the stack 210, the pressure of which is controlled by a pressure regulator 231. Water in the air stream exiting stack 210 accumulates in reservoir 232, which can be drained by opening valve 233, and the air stream is exhausted from the device through valve 234.

Through the cooling water path 240, water exits the reservoir 232 by the pump 241 and circulates through the stack 210. The temperature of the water is regulated by the heat exchanger 242. A purge system 250 is used to purge the hydrogen and oxidant passages in the fuel cell stack 210 with a low humidity non-reactive gas. The flow of gas from the purge gas supply 260 to the hydrogen and air inlet conduits 261, 262 is controlled by valves 263, 264 and three-way valves 267, 266. The nitrogen pressure is controlled by a pressure regulator 265.

In one embodiment, the sensor cell has a higher performance sensitivity for low reactant stoichiometry than the remaining cells in the stack. "Stoichiometry" is the ratio of the amount of reactants supplied to the fuel cell stack to the amount of reactants actually consumed in the fuel cell stack (unconsumed reactants exit the fuel cell stack). . 1.35 hydrogen (H ₂
) Stoichiometry is 135 per 100 parts actually consumed in the fuel cell stack
Means that part of the hydrogen was supplied to the fuel cell stack. In the context of reactant stoichiometry, it is generally important that insufficient reactant stoichiometry or reactant starvation occur in the fuel cells of the stack. Reactant starvation will cause the voltage of the battery to drop, and the battery may undergo battery voltage reversal if proper actions are not taken. Effects such as cell voltage reversal and concomitant material degradation and overheating can result in damage to fuel cell components, particularly the membrane electrode assembly. In the case of fuel starvation, for example, a large positive potential is reached at the anode, which results in a negative total cell voltage, in some cases oxidation of the fuel cell components and associated permanent damage to the fuel cell. It may be. Early detection of an undesirable low reactant stoichiometric state can be achieved by incorporating at least one sensor cell in the stack that causes low voltage or overheating prior to other fuel cells in the stack. The sensor cell (s) can be coupled to the stack operation control system and initiate a corrective action on the remaining cells to avoid the deleterious effects of insufficient reactant stoichiometry.

The stoichiometric sensitivity in the sensor cell can be determined, for example, by modifying the pressure drop in the flow channel of the sensor cell to, for example, reduce the length and open area of the groove in a stack that is not a sensor cell. This can be achieved by varying the depth of the grooves while maintaining the same size as those of other fuel cells. Such sensitivity can also be achieved, for example, by modifying the groove length, width, surface friction coefficient, and / or number. In addition, the stoichiometric sensitivity depends on the sensor cell (s)
Can be achieved by modifying the mass transport properties of the membrane electrode assembly (MEA) of the present invention relative to other cells in the stack. Some specific MEA modifications include: adding an ionomer coating to a porous electrode substrate; increasing the ionomer content of the electrocatalytic layer in the electrode; by changing the hydrophobicity of the cathode Changes in water treatability; and, if one is used, increasing the thickness of the sublayer and / or decreasing its porosity (the sublayer is located between the electrocatalyst layer and the underlying substrate). Layer within the electrode). With any of these technologies, the MEA can be
Or it can be applied in a stepwise or continuously changing manner.

A sensor cell in which the effect of low stoichiometry is intentionally exacerbated,
For example, it is preferably designed to be more tolerant or robust to battery voltage reversal. For example, for sensor cells that are sensitive to fuel stoichiometry, the anode catalyst may be platinum black, which tends to be more tolerant of cell voltage reversal than the carbon-supported platinum catalyst used in other cells. it can.

When achieving sensitivity by modifying the voltage drop according to the Darcy equation (see US Pat. No. 5,260,143), the inlet and outlet of the gas flow in the fuel cell fluid flow area. Pressure drop between fluid density, coefficient of friction, length of flow passage,
And increases as fluid velocity increases. Conversely, the pressure drop between the gas flow inlet and outlet decreases as the diameter of the flow passage (groove) increases. Increasing the sensor cell pressure drop relative to the remaining cells in the stack at a given flow rate will reduce the flow of reactants to the sensor cell according to Darcy's equation. In this regard, the reactant stoichiometry is lower for the sensor cells compared to the remaining cells in the stack, thereby increasing the performance sensitivity to reactant stoichiometry over the remaining cells. become. The modification (one or more) of the flow zone grooves to provide the desired performance sensitivity to reactant stoichiometry is dependent on the range of conditions that the fuel cell experiences under normal operating conditions (eg, such as current density and pressure). ) Must be included.

In large stacks with many individual fuel cells, for example, the effect of pressure in the manifold due to the effect of Bernoulli, or the effect of liquid water, results in different reactant flow distributions. In this regard, in a large fuel cell stack for a rugged vehicle such as a bus, the cell furthest from the internal wetting device (i.e., the cell furthest from the inlet of the moistened fuel stream) may be inverted due to lack of fuel. Some tendency has been observed. Therefore, in such large stacks, it may be desirable to incorporate several sensor cells at different locations in the stack (eg, at both ends and in the middle, etc.) to account for stack related changes in the battery environment.

[0036] In order to initiate a suitable system response, it is also desirable to distinguish between reduced oxidant stack performance and low fuel stoichiometry. This means that by using more than one sensor cell, for example, one sensor cell designed to be more sensitive to oxidant stoichiometry and more sensitive to fuel stoichiometry This can be achieved by using another sensor cell designed as follows. This can best be achieved if the cell voltage is monitored by separating the low oxidant stoichiometric sensor cell from the low fuel stoichiometric sensor cell. More complex, but with a single sensor cell, based on the characteristic voltage response or the perimeter closest to the cell flow area pressure drop or outlet of each flow area being monitored
er) Based on the voltage, it is also possible to distinguish between low oxidant stoichiometry and low fuel stoichiometry. (The voltage at a given electrode or at all points in the flow area is not constant in the loaded fuel cell. Voltage differences can be easily detected, especially along the reactant flow path in the flow area. The magnitude of these voltage differences in the fuel cell is
It can be indicative of abnormal conditions such as low reactant stoichiometry. Thus, a suitable sensor cell can be made responsive to a voltage drop or voltage distribution within one component of the sensor cell).

The solution described above in connection with low stoichiometry sensitive sensor cells can be applied to other hazardous fuel cell operating conditions. These conditions include the following: overflow, which is the shape of the MEA or the flow area, for example by a sensor cell similar to a low stoichiometric cell, or to be more sensitive to overflow; Can be detected in advance by altering: dehydration, for example by a sensor cell with increased reactant flow (with increased stoichiometry) to amplify the effects of dehydration, or Can be detected in advance by using different membranes or electrodes that are more sensitive to dehydration. An undesired operating temperature, which can be detected in advance, for example by means of a sensor cell with a modified coolant flow area plate which magnifies the temperature change in the sensor cell; reactant dilution (for example nitrogen accumulation) ), Which can be detected in advance, for example, by means of the low-stoichiometric sensor cells described above and / or MEAs which are more sensitive to mass transport phenomena; It can be detected beforehand by a sensor cell with a more sensitive catalyst and / or a lower effective catalyst loading.

In a preferred embodiment in the latter case, the sensor cell has an increased sensitivity to carbon monoxide poisoning as compared to the remaining cells in the stack. The use of such a sensor cell is advantageous in fuel cell systems incorporating a reformer when carbon monoxide impurities are typically present in the reformate stream. A sensor cell that is sensitive to carbon monoxide can be obtained by using less or more carbon monoxide-sensitive anode electrode catalyst in the sensor cell (s) than the other fuel cells in the stack. Can be. The catalyst in the sensor cell should be robust with respect to carbon monoxide poisoning (the sensor cell will not be permanently damaged), but the performance of the sensor cell will be more sensitive to the effects of carbon monoxide It is desirable that

In such an apparatus, an air bleed may be used to reduce the adverse effects of carbon monoxide present in the reformate stream. (Air bleed refers to the manner in which small amounts of oxygen, typically air, are introduced into a fuel stream and reacted with and removed from carbon monoxide adsorbed on the surface of the anode electrocatalyst). Insufficient air bleed levels result in carbon monoxide poisoning and can be detected by carbon monoxide-sensitive sensor cells (which can then cause increased air bleed levels. ). However, excessively high air bleed levels can result in excessive heat generation at the anode, which can cause degradation. Therefore, it is desirable in such devices to use sensor cells that are sensitive to poisoning and temperature and to use these responses to control air bleed levels.

Several examples are provided below illustrating certain features of a sensor cell suitable for use in a gas-supplied solid polymer electrolyte fuel cell. They should not be considered as limiting. For example, the sensor cell is suitable for use in a liquid feed solid polymer electrolyte fuel cell (eg, as a methanol concentration sensor in a direct methanol fuel cell).

The performance of low oxidizer stoichiometry and low fuel stoichiometry sensor cells has been demonstrated in an eight cell stack. 4A and 4B show oppositely facing surfaces 322 of a bipolar cathode flow area plate 310 for a low oxidizer (air) stoichiometric sensor cell.
332 are shown. The surface 332 in FIG. 4A has an oxidant inlet 326 and an oxidant outlet 328.
Six meandering flow grooves or depressions 324 are formed therein for sending an oxidant stream (preferably air) between the two. Flow channel 3 in cathode plate of sensor battery
The depth of 24 is less than the depth of the corresponding grooves in the cathode plates of each of the remaining seven fuel cells of the stack. A decrease in the flow channel 324 in the sensor cell cathode plate causes a pressure drop in the cathode flow region, which, as discussed above with respect to the Darcy equation, causes the sensor cell to have a low oxidant stoichiometry. Make it more sensitive. Surface 322 contacts the cooperating flat major surface of the membrane electrode assembly (not shown in FIG. 4A).

The surface 332 of FIG. 4B allows the coolant fluid flow (preferably liquid water) to
Four meandering channels or recesses 334 for sending from 36 to coolant outlet 338
Are formed therein.

As further shown in FIGS. 4A and 4B, the cathode flow area plate 310 has a plurality of manifold openings 342, 344, 346, 348, 350, 352 formed therein, each of which has a surface. It extends between 322 and 332. The manifold opening 342 directs the incoming wet oxidant stream toward the plurality of fuel cells that make up the stack. Manifold opening 344 receives the spent oxidant stream and directs it to the oxidant outlet of the fuel cell stack. The manifold opening 346 directs the incoming wet fuel stream toward the plurality of fuel cells that make up the stack. Manifold opening 348 receives the spent fuel stream and directs it toward the fuel outlet of the fuel cell stack. The manifold opening 350 directs the incoming coolant stream toward the coolant jacket associated with the fuel cells that make up the stack. Manifold opening 352 receives the spent coolant stream and directs it to the coolant outlet of the fuel cell stack. Thus, the cells in the stack are connected by a manifold in parallel for each fluid from the common supply manifold header to the common exhaust manifold header.

FIGS. 5A and 5B show oppositely facing surfaces 422, 432 of a bipolar anode flow area plate 410 for a low fuel (H ₂ ) stoichiometric sensor cell. Surface 42
2 is flat (no depression) and contacts the cooperating surface 332 shown in FIG. 4B to form a coolant jacket. The surface 432 in FIG. 5B has one meandering flow groove or depression 434 formed therein for delivering the fuel flow (preferably hydrogen) between the fuel inlet 436 and the fuel outlet 438. Anode plate 4 for sensor battery
The depth of the flow grooves 434 in 10 is less than the depth of the corresponding grooves in the anode plates of each of the remaining seven fuel cells of the stack. The reduction of the flow channels 434 in the sensor cell anode plate causes a pressure drop in the anode flow area, which makes the sensor cell more sensitive to low fuel (hydrogen) stoichiometry. Surface 43
2 is in contact with the other flat major surface of the membrane electrode assembly (not shown in FIG. 5B) and also on the opposite side by surface 322 shown in FIG. 4A.

As further shown in FIGS. 5A and 5B, the anode flow area plate 410 has a plurality of manifold openings 442, 444, 446, 448, 450, 452 formed therein, each of which has a surface 422. 432. Cathode plate 310
And the anode plate 410, the manifold openings 442, 444, 446,
448, 450, 452 have openings 342, 344, 346, 348, 350, 35
Alongside the two, they form a manifold for the incoming fuel, oxidant, and coolant streams and their exhaust streams.

FIG. 6 a shows three different current densities: 100 A / ft in plot 6A. ^Two (ASF), 500 ASF for plot 6B, 1000 AS for plot 6C
Composite of cell voltage as a function of oxidizer (air) stoichiometry of fuel cell at F
It is a plot figure. FIG. 6b shows 2.0H_TwoStoichiometric, 30/30 psig air
/ H_Two, Battery temperature = 75 ° C., active area = 0.25 ft^Two, Dow experimental membrane
Mark XUS 13204.10), polytetrafluoroethylene at the cathode
6%, 10/2 pass H_Two8 different fuel cells operated in / airflow area
Composite polarization probe of cell voltage as a function of current density for different air stoichiometry
FIG. Figure 6b shows 6 (plot 6D), 4 (plot 6E), 3 (pro
6F), 2.5 (Plot 6G), 2 (Plot 6H), 1.75 (Plot
G1), 1.5 (Plot 6J), and 1.25 (Plot 6K) air chemistry
Figure 3 shows a stoichiometric polarization plot.

FIGS. 6 a and 6 b show that sensitivity to stoichiometry increases with increasing current density. Thus, decreasing the active area of the sensor cell makes it more sensitive to stoichiometric changes. Reducing the active area of both the anode and cathode results in increased sensitivity to stoichiometric changes for both reactants. However, the reduction of the active area of only one of the electrodes still causes the active area of the other electrode to be substantially separated from the former electrode if much of the latter electrode area is physically far away from the former. If there is no passage, it is effectively reduced. Thus, in order to make the sensor cell sensitive to only one reactant stoichiometry, the active area of one electrode should be replaced by the active area of the electrode with the reduced area of all electrodes. Decrease in a way that is still physically close. This is achieved by using an electrode that has a similar geometric size to the other electrode, but has a small distribution of small inactive areas (eg, a series of closely spaced bands on the electrode surface). By depositing an electrocatalyst). Alternatively, the same effect can be achieved similarly by masking the reactant supply side of the electrode in the MEA with a gas impermeable strip pattern. Alternatively, for example, in embodiments having multiple channel flow areas, every other groove in the flow area could be blocked. Such methods may also take advantage of the variation in current density and reactant concentration that exists over the active electrode area. Therefore, it may also be desirable to use a non-uniform mask pattern or catalyst deposition.

FIG. 7 a shows a plot of cell voltage as a function of fuel stoichiometry for the average of the sensor cells and the remaining cells in the same stack. The sensor cell is structurally similar to the other cells in the stack, except that the sensor cell is about 40%
Of the anode active area of the MEA was achieved by masking the fuel supply side of the MEA with a series of gas impermeable zones. (The distribution of the bands is non-linear and the bands are wider and closer apart near the fuel inlet). As shown in FIG. 7a, the sensor cell was significantly more sensitive to reduced fuel stoichiometry than the average cell. The sensitivity of the sensor cell to reduced oxidant stoichiometry was not significantly different from that of the average cell.

FIG. 7 b shows a plot of cell voltage as a function of oxidant stoichiometry for the average of the sensor cells and the remaining cells in the same stack. The sensor cell is structurally similar to the other cells in the stack, except that the sensor cell mimics a reduction in cathode active area of about 20%, which means that the channel in the multiple linear channel oxidant flow area Was achieved by blocking As shown in FIG. 7b, the sensor cells were significantly more sensitive to reduced oxidant stoichiometry than the average cell. The sensitivity of the sensor cell to reduced fuel stoichiometry was not significantly different from that of the average cell. Blocking the flow channel, whether the reactant is a fuel or an oxidant, creates a sensor cell that is sensitive to a single reactant stoichiometry over a wide range of current densities. Has proven to be effective. The masking method was effective when the reactant was a fuel, but was not as effective over the same range of current densities when the reactant was an oxidant.

FIG. 8 shows the average of the sensor cell (shown as cell 8) and the other seven cells of the stack at 100 A / ft ² (AS) at three different oxidant flow pressures.
FIG. 4 is a composite plot of battery voltage as a function of oxidant (air) stoichiometry in F). Plot 8A shows cell voltage as a function of air stoichiometry for an average of seven cells at 100 ASF and 10 psig. Plot 8B is 1
4 shows the cell voltage at 00 ASF and 10 psig as a function of air stoichiometry for a sensor cell 8 having the cathode plate illustrated in FIG. 4A. Plot 8C
Shows cell voltage as a function of air stoichiometry for an average of seven cells at 100 ASF and 30 psig. Plot 8D shows 100 ASF and 30 ps
5 shows cell voltage as a function of air stoichiometry for a sensor cell in ig.
Plot 8E shows cell voltage as a function of air stoichiometry for an average of seven cells at 100 ASF and 65 psig. Plot 8F shows 100 ASF
And cell voltage as a function of air stoichiometry for a sensor cell at 65 and 65 psig. FIG. 8 shows that the cell voltage of the sensor cell drops dramatically compared to the average corresponding voltage drop of the other seven fuel cells in the stack in response to the reduced air stoichiometry. Show. FIG. 8 also shows that the sensor cell voltage drop is induced by higher air stoichiometry at higher pressure levels. By changing the flow area design, the air stoichiometry that induces the voltage drop can be adjusted based on the expected range of operating conditions.

FIG. 9 shows the average of the sensor cell (cell 8) and the other seven cells in the stack as a function of oxidant (air) stoichiometry at 30 psig at three different current densities. It is a composite plot of battery voltage. Plot 9A is 100 AS
Shown is the cell voltage as a function of air stoichiometry for the average of seven cells at F. Plot 9B shows cell voltage as a function of air stoichiometry at 100 ASF for a sensor cell having the cathode plate illustrated in FIG. 4A. Plot 9
C shows cell voltage as a function of air stoichiometry for an average of seven cells at 1000 ASF. Plot 9D shows cell voltage as a function of air stoichiometry for a sensor cell at 1000 ASF. Plot 9E is 600 AS
Shown is the cell voltage as a function of air stoichiometry for the average of seven cells at F. Plot 9F shows cell voltage as a function of air stoichiometry for a sensor cell at 600 ASF. FIG. 9 shows that the cell voltage of the sensor cell drops dramatically compared to the average corresponding voltage drop of the remaining seven fuel cells in the stack in response to the reduced air stoichiometry. Show. FIG. 9 also shows that the sensor cell voltage drop is induced by higher air stoichiometry at higher current densities. Again, the air stoichiometry that induces the voltage drop can be adjusted based on the expected range of operating conditions.

FIG. 10 shows 100 ASF, three different fuel (H ₂ / N ₂ ) flow pressures for the sensor cell (cell 8), the adjacent cell (cell 7), and the average of the other seven of the stacks. FIG. 4 is a composite plot of battery voltage as a function of H ₂ stoichiometry.
Plot 10A, for an average of seven cells in 10 psig, indicating the battery voltage as a function of H ₂ stoichiometry. Plot 10B, for a sensor cell having an anode plate illustrated in FIGS. 5A and 5B at 10 psig, indicating the battery voltage as a function of H ₂ stoichiometry. Plot 10C, for an average of seven cells in 30 psig, indicating the battery voltage as a function of H ₂ stoichiometry. Plot 10D, for cell 7 adjacent in 100ASF and 30 psig, indicating the battery voltage as a function of H ₂ stoichiometry. Plot 10E, for sensor cells in 30 psig, indicating the battery voltage as a function of H ₂ stoichiometry. The plot 10F is
The average of seven of the battery at 65 psig, indicating the battery voltage as a function of H ₂ stoichiometry. Plot 10G shows the H ₂ for the sensor cell at 65 psig.
Show battery voltage as a function of stoichiometry. FIG. 10 shows that the cell voltage of the sensor cell drops dramatically compared to the average corresponding voltage drop of the other seven fuel cells in the stack in response to the reduced H ₂ stoichiometry. Is shown.

FIG. 11 shows the average of the eight cells in the stack, the sensor cell (here shown as cell 7), the adjacent terminal cell (shown as cell 8), and three different current densities at 30 psig. FIG. 2 is a composite plot of battery voltage as a function of H ₂ stoichiometry of FIG. Fuel cell data includes 18% N ₂ and 82% H ₂ . Plot 11A shows cell voltage as a function of air stoichiometry for an average of eight cells at 100 ASF. Plot 11B shows cell voltage as a function of air stoichiometry for sensor cell 7 at 100 ASF. Plot 11
C is the adjacent terminal cell 8 with the anode plate illustrated in FIG. 5B at 100 ASF.
4 shows cell voltage as a function of air stoichiometry for. Plot 11D shows 6
The cell voltage is shown as a function of air stoichiometry for an average of eight cells at 00 ASF. Plot 11E shows cell voltage as a function of air stoichiometry for sensor cell 7 at 600 ASF. Plot 11F shows cell voltage as a function of air stoichiometry for adjacent end cell 8 at 600 ASF. Plot 11G shows cell voltage as a function of air stoichiometry for an average of eight cells at 1000 ASF. Plot 11H shows cell voltage as a function of air stoichiometry for sensor cell 7 at 1000 ASF. Plot 11 I
Shows the cell voltage as a function of air stoichiometry for the adjacent terminal cell 8 at 1000 ASF. FIG. 11 shows that the cell voltage of the sensor cell dropped dramatically compared to the average corresponding voltage drop of the adjacent cell 7 and the eight fuel cells in the stack in response to the reduced H ₂ stoichiometry. To do so. FIG. 11 also shows that the sensor cell voltage drop is induced by lower H ₂ stoichiometry at higher current densities.

FIG. 12 is a composite plot of cell voltage as a function of H ₂ stoichiometry for each of the eight cells in the stack including sensor cell 8 at 600 ASF and 30 psig. Plots 12A and 12B show the voltage drop curves for two experiments sensors battery 8 in the case of slowly decreasing the flow rate of H _2. Plots 12C and 12D show two experiments of sensor cell 8 when the H ₂ flow was reduced much faster than in plots 12A and 12B (the elapsed time between data points was the same for all plots). 2 shows a voltage drop curve for the above. FIG.
This indicates that a voltage drop of the sensor battery 8 occurs regardless of the speed at which the H ₂ flow rate is reduced.

The sensor cell method of the present invention can provide one or more of the following advantages over conventional fuel cell performance monitoring methods: (1) Sensor cells are specific to the fuel cell stack that occurs in the device. It can be designed to detect undesirable operating conditions. (2) Sensor cells can detect undesirable operating conditions prior to other cells in the fuel cell stack, triggering corrective action, thereby resulting in overall stack cell damage and / or power or fuel cells. Can be used to prevent performance degradation. (3) The sensor battery can reduce the period and complexity of battery monitoring and simplify the control system. (4) The sensor cell can serve as a productive power generating fuel cell in the absence of undesirable operating conditions that can cause an early drop in output power. (5) Sensor cells can be used to provide improved stack safety, longevity and reliability by inducing corrective action and protecting the remaining cells of the stack from harmful conditions. (6) Sensor cells can be designed to be more tolerant or robust to the consequences of undesirable operating conditions being detected, such as, for example, cell reversal and its deleterious effects. .

While specific elements, aspects and applications of the present invention have been shown and described, it will be understood that the invention is not limited thereto. This is because those skilled in the art can make modifications, particularly with reference to the above teachings. It is therefore intended by the following claims to cover such modifications as incorporating those features falling within the spirit and scope of the invention.

[Brief description of the drawings]

FIG. 1 is a fragmentary side view of a typical solid polymer electrochemical fuel cell having a membrane electrode assembly interposed between two fluid flow area plates.

FIG. 2 is a partially cutaway perspective view of an example of an electrochemical fuel cell stack.

FIG. 3 is a system diagram of a fuel cell power generation device.

FIG. 4A illustrates an oppositely facing surface of a bipolar cathode flow area plate of an example of a low oxidizer (air) stoichiometric sensor cell.

FIG. 4B illustrates an oppositely facing surface of a bipolar cathode flow area plate of an example of a low oxidizer (air) stoichiometric sensor cell.

FIG. 5A illustrates an oppositely facing surface of a bipolar anode flow area plate of an example of a low fuel (H ₂ ) stoichiometric sensor cell.

FIG. 5B illustrates an oppositely facing surface of a bipolar anode flow area plate of an example of a low fuel (H ₂ ) stoichiometric sensor cell.

FIG. 6a shows three different current densities: 100 A / ft ² (ASF), 500 ASF
FIG. 3 is a composite plot of cell voltage as a function of oxidizer (air) stoichiometry of a fuel cell at, and 1000 ASF.

Figure 6b 2.0H ₂ stoichiometry, 30/30 psig air / H _2, cell temperature = 75 ° C., the active region = 0.25ft ^2, Dow experimental membrane (trademark symbol XUS 13,204.10)
Polytetrafluoroethylene 6% in the cathode composite 10/2 pass H ₂ / air stream region, in the battery voltage shown on eight different air stoichiometry of engineered fuel cell as a function of current density It is a polarization plot figure.

FIG. 7a shows the average of the cells in a fuel cell stack and the sensor cells in the same stack.
FIG. 3 is a plot of cell voltage as a function of fuel stoichiometry. The sensor cell was masked to resemble an approximately 40% reduction in the active area of the anode.

FIG. 7b for cell average in a fuel cell stack and sensor cells in the same stack
FIG. 4 is a plot of battery voltage as a function of oxidant stoichiometry. Certain grooves in the oxidant flow area in the sensor cell were blocked to mimic an approximately 20% reduction in the active area of the cathode.

FIG. 8: Oxidant (air) stoichiometry at 100 A / ft ² (ASF), at three different oxidant flow pressures, on average for the sensor cell (cell 8) and the other seven cells in the stack. FIG. 4 is a composite plot of battery voltage as a function.

FIG. 9: Composite plot of cell voltage as a function of oxidant (air) stoichiometry at 30 psig, at three different current densities, averaged over the sensor cell (cell 8) and the other seven cells in the stack. FIG.

FIG. 10 at 100 ASF, at three different fuel (H ₂ / N ₂ ) flow pressures, the average of seven cells that are not sensor cells (cell 8), adjacent cell (cell 7), and sensor cells in the stack, FIG. 4 is a composite plot of battery voltage as a function of H ₂ stoichiometry.

FIG. 11: 30 psig, at three different current densities, the average of seven cells that are not sensor cells (cell 7), adjacent terminal cell (cell 8), and sensor cells in the stack,
FIG. 4 is a composite plot of battery voltage as a function of H ₂ stoichiometry.

FIG. 12 is a composite plot of cell voltage as a function of H ₂ stoichiometry for each of the eight cells in the stack, including sensor cell 8, at 600 ASF and 30 psig.

[Procedure amendment]

[Submission date] February 20, 2001 (2001.2.20)

[Procedure amendment 1]

[Document name to be amended] Drawing

[Correction target item name] Fig. 6a

[Correction method] Change

[Correction contents]

FIG. 6a

──────────────────────────────────────────────────の Continued on the front page (72) Inventors Knights, Shana, De Canada British Columbia, Vancouver, East Tense Avenue 55, Number 103 Street 4372 F term (reference) 5H026 AA06 CC10 5H027 AA06 KK21 KK31 KK46 KK51 KK54 KK56

Claims (39)

[Claims] 1. An improved electrochemical fuel cell stack having a plurality of fuel cells, wherein each of the fuel cells has an anode having an anode electrocatalyst; a cathode having a cathode electrocatalyst; An ion exchange membrane interposed between the fuel cell stack and at least one of the plurality of remaining fuel cells, wherein at least one of the plurality of fuel cells is different from the plurality of remaining fuel cells. A sensor cell having differences; and furthermore, during operation of the stack, the structural differences do not occur simultaneously in the remaining fuel cells under substantially the same operating conditions. Causing at least one of the responses to the sensor cell; 2. During operation of the stack, each of the anodes has a fuel flow directed thereto along an anode flow region, and each of the cathodes has an oxidant flow directed therethrough along a cathode flow region. Wherein each of said fuel stream and said oxidant stream comprises:
Having an inlet pressure and associated stoichiometry, each of the plurality of fuel cells generates a current per unit area of the cathode, generating water at the cathode and generating heat at a nominal operating temperature, the structural difference. Wherein the sensor cell has at least one of an electrical and thermal response that does not occur simultaneously in the remaining fuel cell under substantially the same operating conditions with respect to the inlet pressure and stoichiometry of the fuel and oxidant streams. The fuel cell stack according to claim 1, wherein 3. The fuel cell stack according to claim 1, wherein during operation of the stack, the sensor cells and the remaining cells provide power. 4. The fuel cell stack according to claim 1, wherein during operation of the stack, a variable electrical load is applied through the fuel cell stack having the sensor cells. 5. The fuel cell stack according to claim 1, wherein a plurality of fuel cells are connected by a manifold so that reactants are supplied in a parallel supply form. 6. The fuel cell stack according to claim 1, wherein the at least one response is a voltage change occurring in the sensor cell. 7. The fuel cell stack according to claim 1, wherein at least one response is a change in current per unit area of the cathode occurring in the sensor cell. 8. The fuel cell stack according to claim 1, wherein the at least one response is a change in a voltage distribution occurring within a component of the sensor cell. 9. The fuel cell stack according to claim 1, wherein the at least one response is a change in electrical resistance that occurs in the sensor cell. 10. The fuel cell stack according to claim 1, wherein the at least one response is a temperature change occurring in the sensor cell. 11. The fuel according to claim 1, wherein the structural difference is that at least one of the anode and cathode of the sensor cell has an electrochemically active area smaller than the electrochemically active area of the remaining fuel cell. Battery stack. 12. The fuel cell stack according to claim 1, wherein reactant access to at least one of the anode and cathode electrocatalysts of the sensor cell is inhibited as compared to the remaining fuel cells. 13. The fuel cell stack according to claim 1, wherein the structural difference is that at least one of the anode and the cathode of the sensor cell has a smaller amount of electrocatalyst than the remaining fuel cell. . 14. The fuel cell stack according to claim 1, wherein the structural difference is that the sensor cell has an anode flow area configuration that is different from the anode flow area configuration of the remaining fuel cells. 15. The fuel cell stack according to claim 14, wherein the anode flow region configuration of the sensor cell causes a voltage drop that is greater than the voltage drop that occurs in the remaining fuel cell anode flow region configuration. 16. The fuel cell stack according to claim 1, wherein the structural difference is that the sensor cell has a cathode flow area configuration that is different from the cathode flow area configuration of the remaining fuel cells. 17. The fuel of claim 16, wherein the cathode flow area configuration of the sensor cell causes a greater voltage drop along the flow area than the corresponding voltage drop that occurs in the remaining fuel cell cathode flow area configuration. Battery stack. 18. The fuel cell stack according to claim 1, wherein the anode catalyst of the sensor cell has a different composition from the anode catalyst of the remaining fuel cells. 19. The fuel cell stack according to claim 1, wherein the cathode catalyst of the sensor cell has a different composition from the cathode catalyst of the remaining fuel cells. 20. Each of the fuel cells has a coolant flow that flows along a coolant flow area while conducting heat with each of the fuel cells, and the structural difference is that of the remaining fuel cells. The fuel cell stack according to claim 1, wherein the fuel cell stack has a sensor cell coolant flow area configuration different from the coolant flow area configuration. 21. The fuel cell stack according to claim 1, wherein structural differences make the sensor cell more sensitive to fuel stoichiometry. 22. The fuel cell stack of claim 1, wherein structural differences make the sensor cell more sensitive to oxidant stoichiometry. 23. The fuel cell stack according to claim 1, wherein the structural differences make the sensor cell more sensitive to carbon monoxide poisoning. 24. The fuel cell stack according to claim 1, wherein the structural differences make the sensor cell more sensitive to air bleed levels in the fuel stream. 25. The fuel cell stack according to claim 1, wherein the structural differences make the sensor cell more sensitive to operating temperature. 26. The fuel cell stack according to claim 1, wherein the structural differences make the sensor cell more susceptible to methanol poisoning. 27. The fuel cell stack according to claim 1, wherein the fuel cell stack is a direct methanol fuel cell stack, and the structural differences make the sensor cells more sensitive to methanol concentration. 28. The improved fuel cell stack according to claim 1, wherein structural differences make the sensor cell more susceptible to overflow. 29. The fuel cell stack according to claim 1, wherein the structural differences make the sensor cell more susceptible to dehydration. 30. The fuel cell stack according to claim 1, wherein the structural differences make the sensor cell more sensitive to mass transport phenomena. 31. A method of operating an electrochemical fuel cell device having a fuel cell stack comprising a plurality of solid polymer fuel cells, wherein at least one of the plurality of fuel cells comprises: Is a sensor cell having at least one different structural difference; during operation of the stack, the structural differences do not occur simultaneously in the remaining fuel cells under substantially the same operating conditions Causing at least one of an electrical and thermal response in the sensor cell; the method of operation comprising: at least one of the electrical and thermal operating parameters of the sensor cell; Comparing the monitored operating parameter of the sensor battery with that of the at least one remaining battery, if the comparing Generating an output signal if the determined parameter differs significantly from a predetermined threshold amount and indicates an undesirable operating condition. 32. The method of claim 31, wherein the monitored parameter is a battery voltage. 33. The method according to claim 31, wherein the parameter to be monitored is temperature. 34. The method of claim 31, wherein the parameter monitored is the rate of change of the battery voltage. 35. The method of claim 31, wherein the output signal issues an alarm signal alerting an operator of the device. 36. The output signal records a monitored parameter.
The method of claim 1. 37. The method of claim 31, wherein the output signal initiates a corrective action that restores the device to a desired operating condition. 38. The method of claim 31, wherein monitoring is performed continuously during operation of the stack. 39. The method of claim 31, wherein monitoring is performed periodically during operation of the stack.