The present invention relates to a fuel cell stack including a heating system which comprises electrical heating elements for a rapid heating-up of the fuel cell stack. In particular, the present invention relates to fuel cell stack wherein there is included a resistance wire in the fuel cells of the fuel cell stack and/or resistive elements in bipolar plates of the fuel cells of the fuel cell stack.

A fuel cell apparatus delivers an electric current (and heat) through an electro chemical process wherein hydrogen or a gas with a high content of hydrogen, also called reformate gas, is used as fuel for the fuel cell stack. A fuel cell apparatus, including a fuel cell stack, may therefore be considered an electric generator.

Provided that the hydrogen gas is completely clean this process may be carried out at a temperature of about 80-90 ⁰ C in the fuel cell stack with a relatively high efficiency and without much degrading or contamination of the catalysts used in the membrane. For clean hydrogen, PEM fuel cell systems are widely used. However, problems related to accessibility of clean hydrogen gas, including transport and storage, limits a wider use of this technology.

There has been developed several different fuel cell systems (HTPEM, SOFC, etc.) some of which are characterized by that they operate at higher temperatures, i.e. in the range of 150-950 ⁰ C and have a higher tolerance for impurities in the hydrogen gas. This is an attractive feature since

- the production/manufacturing of the hydrogen gas is simplified, and hence the price of hydrogen gas is reduced, different ways of manufacturing hydrogen rich gas or reformate gas is made possible, - the robustness and lifespan of the fuel cell stack is increased, which may contribute to a considerably more widespread use of fuel cell systems for production of electric power, in stationary as well as mobile applications.

Depending on the particular application of a fuel cell system, different demands and circumstances necessitates particular designs of the whole fuel cell apparatus, which is therefore an important factor for what kind of applications that the fuel cell technology may find its operative and economic use. When using "unclean" hydrogen gas or reformate gas, a correct and evenly distributed operating temperature is required throughout the fuel cell stack. In addition, if quick start-up of the fuel cell system (i.e. the fuel cell stack) is required or frequent start-stop of the fuel cell stack takes place, integrated technical solutions are needed. These are particularly requirements that arise in connection with mobile applications, such as for example use of fuel cells in vehicles. The need for temperature control of the fuel cell stack is known, and different technical solutions have been proposed and developed. Basically, these solutions control the temperature by circulating air or a liquid medium such as water, oil etc., on the outer surfaces and/or through ducts in the fuel cell stack. Heating elements in the form of electric resistance elements on or around the fuel cell stack has also been used. The cross sectional design of the fuel cell stack may also be varied in different ways, i.e. square, rectangular etc., in order to achieve optimal and even heating of the surface and the core or inner parts of the fuel cell stack. Liquid medium is considerably more efficient than air, but make strict demands to sealing, connections etc. Ducts through the stack give rise to different technical challenges and costs, like for example if the ducts are arranged in the longitudinal direction of fuel cell stack through the elements of fuel cell stack (the layers of the fuel cells).

In existing solutions the fuel cell stack is normally heated through the outer surfaces and through ducts in order to achieve evenly distributed heating and a relatively fast heating-up. These technical solutions, however, requires pumps, fans, electrical heating elements, reservoir tanks for liquid medium, connections, control system, etc. which all taken together, can become fairly extensive affecting the size, complexity and cost of the overall system.

Since clean hydrogen gas (i.e. with substantially no impurities) is not easily available due to problems related to production, transport and storage, the use of "unclean" hydrogen gas and reformat gas may be the solution for a more widespread use of fuel cell technology. Contrary to "clean" hydrogen, the production of "unclean" hydrogen gas and reformate gas can be carried out with considerably higher flexibility through new technology and by the use of easily available fuel such as gas, oil or other liquid hydrocarbons.

It would be desirable if such an electric generator could be run "at demand" and also on a vehicle. To increase the process temperature within the fuel cell stack quickly and to maintain a stable temperature throughout the whole fuel cell stack during the time of operation is a requirement if such a development of the fuel cell technology is to take place. Medium-temperature fuel cell systems, i.e. fuel cell systems with working temperatures above 125 ⁰ C (for example HTPEM — High Temperature Polymer Electrolyte Membrane), will have further advantages when using "clean" hydrogen including increased efficiency, reduced need for water etc. Furthermore, high temperature PEM is currently the most suitable type of fuel cell which utilises reformed fuels since they offer the advantage of increased tolerance towards the carbon monoxide content in fuels generated when the fuel is reformed. When operating a HTPEM fuel cell stack with pure hydrogen the elevation of the temperature of the fuel cell stack from the dew point to the operating temperature of about 180 ⁰ C can be achieved autothermally by producing electric and thermal energy through the fuel cell reaction. The fuel cell stack temperature must be brought up above the dew point of the feed reactants in order to avoid condensation inside the stack and especially the MEA (Membrane Electrode Assembly). This dew point temperature is usually far below the operational temperature of HTPEM fuel cell stacks. Thus, the HTPEM fuel cell stack must be preheated before it can be operated.

The preheating can be done by means of external heating measures applied to the stack surface area or indirectly heating the cooling fluid (air or other another gas or a liquid coolant). When a HTPEM fuel cell stack is fed with reformat gas containing a significant CO-level inhibiting the fuel cell reaction at temperature levels below the nominal operational temperature, the preheating temperature of the stack has to be close to the operational temperature. The required preheating temperature correlates to the fractional amount of carbon monoxide contained in the reformat gas which is fed into the fuel cell stack which means that the fuel cell stack can not be heated autothermally and the required amount of preheating energy increases accordingly.

A HTPEM fuel cell stack is commonly preheated by using electrical energy. This can be done by mounting a heating mat or similar devices on the surface of the outer geometry of the fuel cell stack assembly. This method is, however, encumbered with limitations in that the available surface area on the fuel cell stack assembly is limited. Therefore a high surface temperature would be required to induce as much heat energy as is needed to achieve an acceptable preheating time. The upper temperature limitation for the MEA is typically 180 ⁰ C which means that the heat flux towards the core of the fuel cell stack assembly is limited by using this preheating method. The preheating can be improved when the cooling media of the stack is heated externally and then passed through the stack assembly because the heating power can be increased by high flow rates. However, the upper temperature limit of the temperature of the cooling fluid is still about 180 ⁰ C and heat losses must be accepted when heating up the coolant media externally of the fuel cell stack assembly. In addition, heat transfer from the (heated) cooling fluid to the stack materials to be heated is limited due to poor conductivity and heat transfer coefficient. For air and some other gases the heat transfer is also limited by the heat capacity of the fluid. Energy is also needed to circulate the (heated) cooling fluid, particularly through the internal ducts in the fuel cell stack for cooling fluid. In principle, the major disadvantage of preheating is that the entire fuel cell stack assembly has to be preheated close to the operational temperature of about 180 ⁰ C. The heat capacity and the conductivity of the materials used in the fuel cell stack assembly define the constraints with respect to an achievable time for preheating the fuel cell stack assembly. The bipolar plates, making up more than 80% of the weight of the fuel cell stack, is determining for preheating time. The bipolar plates are commonly made of compound of graphite and polymer with a typical heat capacity between 0,75-0,85 J/gK and heat conductivity around 20 W/mK. It is also known to use bipolar plates made of metal, but problems related to corrosion of the metallic materials in a phosphor acid environment at high temperatures and electrochemical potential make graphite based materials for the bipolar plates the most preferred option.

In the German patent DE 197 57 318 Cl there is disclosed a method of heating a fuel cell stack wherein a layer (Sperrschicht) is included in the MEA which is made of a material which will only let protons (H ⁺ -ions) through and at the same time has sufficient resistivity to produce heat energy when an electric current is passed through it. To provide an additional proton conducting layer in the MEA is, however not an optimal solution since the additional layer introduces an increased risk of problems related to operational reliability of the fuel cell stack. It is also necessary to somehow avoid the electric current being passed through the additional layer from going astray within the fuel cell. How this may be achieved is not discussed in the German patent publication. Furthermore, the same material must be able to produce sufficient energy when a current is passed through it and at the same time only let protons diffuse through it. The selection of such materials is obviously limited.

In the American patent US 7,201,981 B2 there is disclosed a system for heating a fuel cell stack where an electric heater is provided at the ends of the fuel cell stack such that the two fuel cells at the two ends of the fuel cell stack are warmed up. When the two fuel cells at the end of the fuel cell stack has been sufficiently heated, the two fuel cells will be started. The heat produced when the two fuel cell are operated is used to heat the two fuel cells adjacent to the two fuel cells at the end of the fuel cell stack. When these two fuel cells are sufficiently warm, they are also started, and the heat produced is used to heat up the next adjacent fuel cell. The process continues until all the fuel cells in the fuel cell stack has been heated at which point normal operation of the fuel cell commences. In one embodiment there is also disclosed a heating element provided between two fuel cells in the middle of the fuel cell stack such that the heating process is started from both ends of the fuel cell stack and in both directions from the middle of the fuel cell stack. Making this system work involves a complicated system for successively switching on the fuel cells as they reach a desired temperature and independent loads for the fuel cells as they are switched on and start to operate.

The objective has therefore been to develop a simple and efficient way to heat up a fuel cell stack and to provide an even distribution of heat energy within the fuel cell stack, particularly in the start-up phase of a fuel cell system when the fuel cell stack needs to be heated to a temperature where contamination of the fuel cell stack by the impurities contained in the hydrogen (as for example in a reformate gas) is not a problem for the operation of the fuel cell stack.

This is solved by the claimed invention as defined in the attached independent claims. Further embodiments of the inventions are defined in the dependent claims. There is provided a MEA which comprises at least the following layers:

- a PEM,

- an anode layer and a cathode layer which are arranged on either side of the PEM, a GDL (gas diffusion layer) which is arranged next to the anode layer and on the opposite side of the anode to the PEM, and a GDL which is arranged next to the cathode layer and on the opposite sides of the cathode to the PEM, wherein the MEA is provided with at least one resistance wire for warming up of the MEA.

A fuel cell basically comprises a membrane, which can be a polymer electrolyte membrane (PEM), in the middle of the fuel cell. On each side of the membrane there is provided an electrode (anode and cathode) and outside the electrodes there is provided a gas diffusion layer (GDL). Furthermore, the anode and the cathode are normally provided with a catalyst. These layers (membrane, electrodes and gas diffusion layers) are arranged between two bipolar plates, normally separated from the bipolar plates with a sealing gasket. When a number of such fuel cells are arranged next to each other in a longitudinal direction (i.e. normal to the surface of the layers that each fuel cell is made up of) a fuel cell stack is formed.

The resistance wire may be embedded in one of the layers of the MEA, in some of the layers of the MEA or all of the layers of the MEA, where the layers, as mentioned above, at least comprises a PEM, anode, cathode and GDLs.

In order to avoid disturbance of the diffusion of protons in the fuel cell, the resistance wire or resistance wires are preferably arranged outside of any proton conducting parts of the layers of the MEA. That can for example be achieved by providing the resistance wire in an outer zone close to the rim or edge of the layers of the MEA. Alternatively, the resistance wire may be embedded in a foil which is either arranged between a pair of adjacent layers of the MEA, or attached to one of the layers of the MEA.

In a preferred embodiment, the foil is made of proton conducting material where the proton conducting areas of the foil is covered with the electrode (anode or cathode), attached to the proton conducting foil, while the resistance wire is arranged around the rim zone of the foil through which there is substantially no conduction of protons taking place. In an embodiment of the MEA there is provided two or more foils in the MEA. In this embodiment one or more of the foils may be attached to a surface of their respective layers of the MEA, for example to the anode and the cathode.

The foil may be attached or joined to the layers of the MEA with glue, printing, welding or any other suitable means for attaching foils to the layers.

There is further provided a fuel cell stack comprising a plurality of fuel cells where each fuel cell comprises bipolar plates and a MEA (membrane electrode assembly), as described above, arranged between the bipolar plates. A plurality of the fuel cells are provided with at least one resistance wire which is arranged between the bipolar plates in the individual fuel cells such that the fuel cells are heated by the resistance wires when an electric current is passed through the resistance wires.

Each fuel cell may be provided with more than one resistance wire depending on the need for supplied heat energy. Often, the two end sections of the fuel cell stack requires more heat energy to be heated than the mid section of the fuel cell stack. The fuel cells in end sections may therefore be provided with a larger number of resistance wires than fuel cells in the mid section of the fuel cell stack.

As explained above, the MEA comprises at least the following layers:

- a PEM (proton exchange member),

- an anode layer and a cathode layer which are arranged on either side of the PEM, a GDL (gas diffusion layer) which is arranged next to the anode layer and on the opposite side of the anode to the PEM, and a GDL which is arranged next to the cathode layer and on the opposite sides of the cathode to the PEM, wherein the at least one resistance wire is included in the MEA.

In an embodiment of the fuel cell stack, the resistance wire may be embedded in one of the layers of the MEA, some of the layers of the MEA or all of the layers of the MEA. Alternatively, the at least one resistance wire may be arranged between two adjacent layers of the MEA as the MEA is assembled, or attached to an outer surface of one of the layers of the MEA which may be done when the respective layers are manufactured.

The resistance wire may have a substantially circular cross section. Alternatively, the resistance wire may be substantially flat, i.e. the cross section of the resistance wire is rectangular with a width which is greater than, preferably substantially greater than, the height. The resistance wire is preferably covered with an electrically insulating material if the resistance wire is attached to the outer surface of one of the layers of the MEA. In an embodiment of the fuel cell stack, a resistance wire is provided between two or more different pairs of adjacent layers of the MEA, which will provide a faster heating of the fuel cell stack. All or some of the fuel cells in the fuel cell stack may be provided with two or more resistance wires in this way. It would also be possible to provide two foils with a resistance wire attached both sides of any layer of the MEA.

In an embodiment of the fuel cell stack, the at least one resistance wire is arranged between the MEA and the bipolar plates, either both of the bipolar plates or one of the bipolar plates. The resistance wire or wires provided between the MEA and the bipolar plates may be the only resistance wire or wires in a fuel cell, or they may be provided in addition to any resistance wires within the MEA.

In an embodiment of the MEA, the at least one resistance wire is embedded in a foil. The foil may be either arranged between a pair of adjacent layers of the MEA, or attached to an outer surface of one of the layers of the MEA. The foil may cover substantially the whole area of the surface of the layers of the MEA. Alternatively, the foil may have cut-outs for the proton conductive areas of the layers of the MEA.

In an embodiment, the foil is made of proton conducting material where the proton conducting areas of the foil is covered with the electrode (anode or cathode), attached to the proton conducting foil, while the resistance wire is arranged around the rim zone of the foil.

In an embodiment of the fuel cell stack there is provided two or more foils in the MEA of the fuel cells. In this embodiment one or more of the foils may be attached to a surface of their respective layers of the MEA, for example to the anode and the cathode.

The foil may be attached or joined to the layers of the MEA of the fuel cells with glue, printing, welding or any other suitable means for attaching the foil to the layers.

In an embodiment of the fuel cell stack, the at least one resistance wire is embedded in a foil, and such a foil is arranged between the MEA and either both bipolar plates or one of the bipolar plates.

In a further embodiment of the fuel cell stack, the resistance wire is embedded in one of or both of the gas diffusion layers. Alternatively, the resistance wire is embedded in a separate, proton conducting foil, as explained above, which is attached to the outer surface, or surfaces, of the GDLs.

It would also be possible to attach separate resistance wires, preferably provided with an electrically insulating material, to the outer surface of the anode layer, cathode layer or the GDLs, preferably to parts of the outer surface of the layers where the conduction of protons are not affected, for example the outer rim zone of the layers.

The fuel cells are normally provided with sealing gaskets which are arranged between the MEA and the bipolar plates. In an embodiment of the fuel cell stack, a resistance wire is embedded in one or both sealing gaskets.

As already mentioned, the resistance wire or resistance wires, whether embedded in a foil or not, are preferably arranged outside of any proton conducting zones or parts of the layers of the MEA such that the conduction of protons are unaffected by presence of the resistance wire. The resistance wires may therefore be provided in the rim zones and between proton conducting zones if the layers of the MEA are provided with two or more proton conducting zones separated by non-conducting zones.

There is also provided a fuel cell stack comprising a plurality of fuel cells, each fuel cell comprising a MEA (membrane electrode assembly), as explained above, and a bipolar plate arranged on both sides of the MEA. Furthermore, a plurality of the fuel cells are provided with at least one resistive element which is embedded in one of the bipolar plates or both bipolar plates such that the bipolar plates, and hence the fuel cells, are heated by the at least one resistive element when an electric current is passed through the at least one resistive element.

In an embodiment of the fuel cell, the resistive element comprises a foil with at least one embedded resistance wire. Alternatively, the resistive element comprises a foil which is made of a resistive material, a resistance wire or any other element providing resistance to the flow of an electrical current. Preferably, the at least one resistive element is electrically insulated from the bipolar plate.

In an embodiment of the fuel cell stack, the electric currents being passed through the resistance wires of the MEA and/or the resistive elements of the bipolar plates are individually adjustable such that the amounts of heat produced in individual fuels cells are independently adjustable. In this way, the production of heat can be controlled and regulated such that there is a higher production of heat energy in fuel cells towards the two end sections of the fuel cell stack than in the fuel cells towards the mid-section. This is useful in cases where one or both of the end sections require a larger amount of heat energy than the mid section to be heated to a temperature where the fuel cell stack can be operated.

An alternative to regulating every single fuel cell separately would be to divide the fuel cell stack into for example two end sections and a mid section, wherein the fuel cells in the same section receives equal electric currents. This approach will simplify the equipment needed to provide and regulate the electric current, but will also provide less flexibility in regulating the temperature of fuel cell stack. Dividing the fuel cell stack into more than three sections could improve the temperature control of the fuel cell stack such that a substantially even temperature along the fuel cell stack is achieved. To control the temperature of the fuel cell stack, the fuel cell stack may be provided with a number of temperature probes which communicates with a control unit. Based on the measured temperatures in the fuel cell stack, the control unit can regulate the electric currents being passed through the electric resistance wires in the MEA and/or the resistive elements in the bipolar plates. During normal operation of the fuel cell stack, i.e. when the fuel cell stack has reached its operating temperature, the control unit preferably also controls and regulates the cooling fluid flowing through the fuel cell stack.

There is also provided a use of an electric resistance wire for heating of a fuel cell comprising a layered MEA arranged between two bipolar plates, the layers including a PEM, an anode, a cathode and GDLs, wherein the electric wire is embedded in or attached to at least one of the layers of the MEA.

There is also provided a use of an resistive element for heating of a fuel cell comprising a MEA arranged between two bipolar plates, the resistive wire being embedded in at least one of the bipolar plates.

It should be noted that throughout this application the term "resistance wire" should also be taken to comprise a strip of a resistive material which is attached to the foil, or any other electrically conductive means forming a conductive path which can be arranged in a zone or zones of the layers of the MEA through which there is substantially no conduction of protons taking place.

A description of preferred embodiments of the present invention will now be described with reference to the drawings, where

Figure 1 discloses a fuel cell stack with a plurality of fuel cells where each fuel cell is provided with resistive heating means. The electric current supplied to the heating means in each cell can be individually controlled and regulated. Figure 2 shows a schematic view of a fuel cell stack with four fuel cells.

Figure 3 shows an exploded view of a MEA with sealing gaskets on each side.

Figure 4 shows schematic view of a proton conducting foil with an embedded resistance wire in the rim zone of the foil. It is also shown an electrode, i.e. anode or cathode, attached to the outer surface of the proton conducting foil. Figure 5 shows a further example of a foil comprising two proton conducting zones and a resistance wire arranged in non-conductive zones of the foil. The areas within the dashed lines indicate the proton conductive areas.

Figure 6 shows a further example of a foil comprising a proton conducting zone and a resistance wire arranged in a loop in the rim zone of the foil. The area within the dashed line indicate the proton conductive area.

Figure 7 shows a cross sectional view of an embodiment of a resistance wire, with a substantially circular cross section, which is covered with an electrically insulating material. Figure 8 shows a cross sectional view of a further embodiment of a resistance wire, with a substantially rectangular cross section, which is covered with an electrically insulating material.

Figure 9 shows a cross sectional view of a resistance wire, with a substantially circular cross section, embedded in a strip of an electrically insulating material. Figure 10 shows a cross sectional view of a resistance wire, with a substantially rectangular cross section, embedded in a strip of an electrically insulating material.

Figure 11 shows a front view of a section through a bipolar plate with an embedded resistive element.

Figure 12 shows a side view of a section through the bipolar plate shown on figure 11.

Figure 1 shows a schematic embodiment of the fuel cell apparatus 10 including a fuel cell stack 12 with a first end section 14, a mid section 15 and a second end section 16. The fuel cell stack 12 is formed by a number of fuel cells 13.

The fuel cell apparatus may be provided with one or more fluid lines 35 through which a cooling fluid may be passed from a cooling fluid supply 34. The fluid line 28 is arranged in fluid communication with ducts (not shown in the figures) in the fuel cell stack 12 such that heat energy produced when the fuel cell apparatus 10 is operating, can be removed. The cooling fluid may for example be air or water. In case the cooling fluid is air, the cooling fluid supply 34 is normally the surrounding atmosphere. Furthermore, the fluid line or lines 35 are preferably provided with valve means 36 such that the flow of cooling fluid can be controlled and regulated in order to help keeping the temperature in the fuel cell stack 12 within a desired operating temperature interval. The cooling fluid is discharged through at least one cooling fluid discharge line 38 as indicated on figure 1. Hydrogen is supplied to the fuel cell stack from a hydrogen supply 24 through a fluid line 25. The fluid line 25 is preferably provided with valve means 26 such that the flow of hydrogen through the fluid line 25, and subsequently to the fuel cell stack 12, can be controlled and regulated. The hydrogen supply 24 may comprise a tank or a similar device which can store hydrogen safely. Alternatively, the hydrogen supply 24 may comprise a steam reformer or any other type of apparatus which can produce the required hydrogen on site.

The oxidant is supplied from an oxygen supply 28 which is connected to the fuel cell stack through a fluid line 29. Normally, the oxygen is supplied through air, in which case the oxygen supply 28 is simply the surrounding atmosphere. The fluid line 29 is preferably provided with valve means 30 such that the flow of oxygen or air through the fluid line 29 and consequently to the fuel cell stack 12, can be controlled and regulated. The process in each fuel cell 13 of the fuel cell stack 12 involves hydrogen combining with oxygen to form steam or liquid water. The steam and/or the water is discharged, together with remaining parts of the air, through a fluid discharge line 32 as indicated on figure 1. On figure 1 there is also indicated that the fuel cell stack 12 is connected to a power consumer 40. The power consumer can for example be an electro motor of a vehicle or other devices including an electro motor. The power consumer can also be a battery or accumulator, which is charged by the fuel cell apparatus 10 when it is operating. Other types of power consumers 40 can also be conceived. A plurality of the fuel cells 13 is provided with at least one resistive element, like a resistance wire, to produce heat energy. This is explained in more detail below. The required electric current is supplied to the resistive elements through electric conductors 45 which are connected to a power source 42 and to the fuel cells 13 which are provided with one or more resistive elements. When the electric power source 42 is switched on, the heat energy is immediately produced in the resistive elements and the fuel cells are heated from within, and the temperature is raised to the desired operating temperature in a short period of time. As can be seen on figure 1, the fuel cell apparatus 10 is provided with a number of regulators 47 which can be used to regulate the size of the electric current that flows through the fuel cells. On figure 1 the fuel cell apparatus is provided with a regulator 47 for each fuel cell 13 such that the electric current, and hence the production of heat energy in each fuel cell, can be regulated independently. If a more basic system for regulating the electric current flowing through the resistive elements in the fuel cells 13 is desired, an electric circuit with a number of regulators 47 corresponding to the number of sections could be provided such that each regulator 47 regulates the electric current flowing through the fuel cells 13 in one section of the fuel cell stack 12. All fuel cells 13 within the same section, as for example a first end section 14, a mid section 15 and a second end section 16 as indicated in figure 1, then receives equal electric currents (provided the resistance of the resistive elements is equal in all fuel cells), and the same amount of heat energy is produced in each fuel cell 13 within the same section, but may be regulated such that the sections 14, 15, 16 are heated differently. The power supply 42 may be a battery or an external supply of electrical energy like the mains.

The fuel cell stack 12 may also be provided with at least one, but preferably three or more, temperature probes 18. If there are provided three temperature probes 18, as indicated on figure 1 , there is preferably provided one temperature probe 18 in each end section 14, 16 of the fuel cell stack 12, and one temperature probe 18 in the mid section 15 of the fuel cell stack 12. The temperatures recorded by the temperature probes 18 are transmitted as signals 19, 20, 21 to a control unit 70 through signal wires or by means of wireless communication means like Bluetooth or any other suitable wireless communication means. The control unit 70 processes the signals 19, 20, 21 from the temperature probes 18, and based on the desired temperature profile of the fuel cell stack 12, sends control signals as required to one or more of the regulators 47 (only one is indicated with a reference number on figure 1) and/or one or more of the valve means 26, 30, 36, and possibly the power supply 42.

Figure 2 shows a schematic drawing of a fuel cell stack 12. The fuel cell stack 12 is made up of a plurality of fuel cells 13 where each fuel cell 13 comprises a MEA 50 (membrane electrolyte assembly) which is sandwiched between two bipolar plates 51. A fuel cell stack 12 is formed when a plurality of fuel cells 13 are connected in a longitudinal direction as shown on figure 2. The bipolar plates 51 of adjacent fuel cells 13 are arranged next to each other such that there is electrical connection between the bipolar plates 51. The bipolar plates 51 are made of an electrically conductive material. When the fuel cell stack 12 is operating, a potential difference is created between the bipolar plate 51 on the anode side and the bipolar plate 51 on the cathode side. If the two bipolar plates 51 of a fuel cell 13 are interconnected with an external electric wire, an electric current would flow through the wire.

It should be noted that the size of the MEA is in reality much smaller than indicated in figure 2. When all the layers of the MEA are joined together, the MEA is still just a thin film. Although the fuel cell stack 13 shown in figure 2 comprises only two fuel cells 13, it should be obvious that any number of fuel cells 13 may be used to form the fuel cell stack 12.

Figure 3 shows the shows the structure of the MEA 50 together with sealing gaskets 53 which separates the MEA from the bipolar plates 51. The MEA 50 comprises a proton exchange membrane (PEM) 56, which normally comprises a proton- conducting polymer membrane, arranged between the anode 55 on one side and the cathode 57 on the other side. The anode layer 55 is provided with a platinum catalyst which causes hydrogen molecules to split into protons which diffuses through the PEM 56. On the other side of the anode layer 55 there is provided a GDL 54 (gas diffusion layer). On the opposite side of the cathode 57 there is also provided a GDL 58. As mentioned, the PEM 55, the anode layer 55, the cathode layer 57 and the GDLs 54, 58 normally form a thin film.

The plate 22 on the anode side is formed with ducts (not shown on the figures) for the supply of hydrogen gas. The hydrogen gas diffuses through the gas diffusion layer 54 to the anode 55 where a platinum catalyst causes the hydrogen molecules to split into H ⁺ -ions (protons) which pass through the PEM 56. The bipolar plate 51 on the cathode side is provided with ducts (not shown on the figures) through which an oxidant, normally air is fed. The protons which pass through the membrane 56 to the cathode 57 combine with oxygen in the air, which has diffused through the gas diffusion layer 58 to the cathode 57, to form water or steam, depending on the working conditions in the fuel cells 13.

Figures 4-6 shows different ways to configure a resistance wire or a strip of a resistive material embedded in a proton conducting foil arranged in the MEA. In figure 4 there is shown a proton conducting foil 60. A resistance wire 62 (or a strip of a resistive material) has been attached to the non-conductive rim zone 61 of the foil 60. Each end of the resistance wire is connected to electric wires 63 which are connected to the electric power source 42 (see fig. 1) through electric wires 45. To the proton conducting zone 64 the anode layer 55 or the cathode layer 57 may be attached.

In figure 5 there are indicated two separate proton conductive zones 64. The resistance wire 62 may therefore be arranged in the non-conductive zone between the proton conducting zones 64 in a substantially U-shaped loop. In figure 6 the configuration of the resistance wire 62 is similar to the configuration shown in figure 4. The resistance wire 62 is, however configured in a loop such that there are two parallel resistance wires 62 in the non-conductive rim zone 61 around substantially the whole circumference of the foil 60.

An alternative to arranging the electric conductors 45 on the outside of the fuel cell stack 12, as shown in figure 1, would be to configure the bipolar plates and the MEAs with one or more through-going holes in the longitudinal direction of the fuel cell 12 in the rim zone 61 such that electric wires, connected to the resistance wires/elements 62, 68 in the fuel cells 13, can be passed through the fuel cell stack

12 in the longitudinal direction. The electric wires/elements 62, 68 can then be connected to the electric conductors 45, and power source 42, at one or both ends of the fuel cell stack 12.

A more elegant way to connect the resistance wires/elements 62 within the fuel cells

13 would be to arrange an electric conductor through each bipolar plate 51 in the longitudinal direction of the fuel cell stack 12, where the electric conductor is electrically insulated from the bipolar plates 51 and with corresponding connections on each side of the bipolar plates 51 such that the electric conductor of adjacent bipolar plates and the resistance wires 62 are automatically connected when the bipolar plates are placed next to each other with the MEA arranged between them. The resistive elements 68 (see figure 11-12) would preferably be connected to the through-going electric conductor within the bipolar plates 51 during the manufacturing process of the bipolar plates 51.

Figures 7-10 shows different configurations of a resistance wire where the resistance wire is not embedded in a proton conductive foil. The resistance wire 62 is shown as having a substantially circular cross section in figure 7 and a rectangular cross section in figure 8. Obviously, many other shapes of the cross section would be possible. If the height of the rectangular cross section, as shown in figure 8, is made sufficiently small, the resistance wire 62 would become like a strip of a resistive material as mentioned above. The resistance wires 62 shown in figure 7 and 8 are covered with an insulating material and may be attached to one or more layers of the MEA, preferably in zones of the MEA through which there is substantially no conduction of protons taking place.

In figures 9 and 10 the two resistance wires 62, disclosed in figures 7 and 8, are sandwiched between two layers of an electrically insulating material 65 such that a strip 66 with an embedded resistance wire 62 is formed. One or more such strips 66 can be attached to one or more layers of the MEA 50, preferably in zones of the MEA through which there is substantially no conduction of protons taking place.

In figure 11 and 12 there is shown a bipolar plate 51 in which there is embedded a resistive element 68. The resistive element 68 is connected to wires 63 which in turn are connected to the electric power source 42 via the electric conductors 45. The resistive element may be a foil with an embedded resistance wire as shown in figures 4-6. As there is no conduction of protons through the bipolar plates 51, the resistance wire or wires 62 may be configured in any pattern in the foil, and alternatively the foil may be made of a resistive material covered with an electrically insulating material. Another alternative would be to embed a strip or strips 66, as shown in figures 9-10, in any desired pattern within the bipolar plates 51.