Bipolar plate for a fuel cell, fuel cell comprising such bipolar plates and method for producing such a fuel cell

Abstract

The invention relates to a bipolar plate which has two plates, anode and cathode, which are provided with channels for circulating a reactive fluid. The bottom walls of the channels form identical parallel corrugations (26, 36) which are regularly distributed along a distribution axis (Y26, Y36) and are spaced apart in pairs by a pitch, denoted p. Each of the corrugations defines a mean axis (X26, X36) which is perpendicular to the distribution axis. Each corrugation (26, 36) is delimited by two opposite and parallel longitudinal edges (26A, 26B, 36A, 36B) which are spaced apart by a corrugation width, denoted I1 or l2, and define a neutral fibre (26C, 36C) midway between the two longitudinal edges. Each corrugation (26, 36) includes first and second top portions (26.1, 36.1, 26.2, 36.2) which are arranged on the same first or second side of the mean axis of the corrugation and aligned in a direction parallel to the mean axis. The corrugations have the same offset D along the distribution axis (Y26, Y36). The offset is such that D = p + x where min(l1, l2) / 2 < x < (I1 + l2) / 2.

Description

[0001] Bipolar plate for a fuel cell, as well as fuel cell comprising such bipolar plates and method of manufacturing such a fuel cell

[0002] The present invention relates to a bipolar plate of a fuel cell. It also relates to a fuel cell comprising such bipolar plates, as well as to a method of manufacturing such a fuel cell.

[0003] A fuel cell is a device for generating electricity by electrochemical reaction between a fuel, typically hydrogen, and an oxidizer, typically oxygen contained in the air. We are interested here in fuel cells comprising several cells, each of which constitutes an electrochemical generator and which are stacked against each other in a stacking direction. Schematically, each cell comprises an anode plate and a cathode plate, which extend perpendicular to the stacking axis and between which an electrolyte is interposed. The invention is particularly applicable in the case where the electrolyte is in the form of a proton exchange membrane (PEM).In such a case, the membrane is covered or in contact, on its face facing the anode plate, with an anodic catalytic layer and, on its face facing the cathodic plate, with a cathodic catalytic layer. The assembly of the membrane and the anodic and cathodic catalytic layers forms, generally with gas diffusion layers arranged on each side of the membrane, a membrane-electrode assembly, generally designated by its acronym "AME". Generally, the catalytic, cathodic and / or anodic layer is deposited, for example by coating, printing, etc., either on the corresponding face of the membrane or on the face of the corresponding gas diffusion layer which is facing the membrane.For each pair of two immediately adjacent cells, the anode plate of one of the two cells of the pair and the cathode plate of the other of the two cells of the pair are applied directly against each other along the stacking axis and form a bipolar plate in its operational state. In some embodiments, in the operational state of the bipolar plate, the anode plate and the cathode plate are fixed to each other, for example by welding, brazing or gluing. In other embodiments, in the operational state of the bipolar plate, the anode plate and the cathode plate are tightened against each other, in particular by the compression tightening imposed on the stack of cells of the battery.Each of the two plates, respectively anode and cathode, of each bipolar plate delimits, on its face facing away from the other plate of the bipolar plate, channels for the circulation of a reactive fluid, respectively hydrogen and oxygen typically, which are covered respectively by the membrane-electrode assemblies belonging respectively to the two cells sharing the bipolar plate. Between the two plates of each bipolar plate in the operational state of the latter, free passages are delimited for the circulation of a cooling fluid, such as glycolated water.

[0004] The invention relates more specifically to bipolar plates whose channels in which the reactive fluids, typically hydrogen and oxygen, circulate do not extend lengthwise in a rectilinear manner but in an undulating manner, the channel undulations of each of the two plates of each bipolar plate being in phase opposition with the channel undulations of the other plate of the bipolar plate, as for example in US 7,781,122.The respective bottom walls of the channels of each of the two plates of each bipolar plate are provided to be generally flat so as to, at the assembly interface between the two plates, rest in substantially flat contact against the bottom walls of the other of the two plates in the operational state of the bipolar plate: due to the undulation of the channels and their phase opposition, the contact between the two plates of each bipolar plate is made between a multitude of contact zones, each of which is not very extensive and which is distributed at the assembly interface between the two plates. In principle, the overall pattern formed by these contact zones at the interface between the two plates of each bipolar plate is predictable since it is a function of the geometric characteristics of the undulated bottom walls of the channels of each of the two plates and of the relative positioning of these two plates in the operational state of the bipolar plate.This overall pattern makes it possible to control both the mechanical stability of the assembly of the two plates and the electrical connection between them: we therefore generally seek to maximize the cumulative extent of the contact zones while ensuring the distribution of the latter at the contact interface. However, in practice, the relative positioning of the two plates in the operational state of the bipolar plate very often suffers from a tolerance much greater than the manufacturing tolerance of the corrugated channels of each plate, in particular because the stacking of the plates induces a positioning uncertainty which is intrinsic to the assembly operations of the fuel cell.It follows that, in the operational state of each bipolar plate, the overall pattern actually formed by the contact zones at the interface between the two plates of the bipolar plate ultimately does not correspond to the overall pattern that was desired but, on the contrary, risks inducing degraded performance for the mechanical stability of the assembly of the two plates and / or for the electrical connection between them. For its part, US 2021 / 0143451 discloses a bipolar plate comprising an anode plate and a cathode plate, applied directly against each other along a stacking axis. Between two stacked bipolar plates is interposed a membrane-electrode assembly.US 2021 / 0143451 provides, for each of its bipolar plates, that (i) the cathode plate delimits, on its surface facing away from the anode plate, in other words on its face facing the membrane-electrode assembly, channels forming a flow field for a gas containing oxygen, these channels being delimited by undulating reliefs of the cathode plate, and (ii) the anode plate delimits, on its face facing away from the cathode plate, in other words on its face facing the membrane-electrode assembly, channels forming a flow field for a gas containing hydrogen, these channels being delimited by undulating reliefs of the anode plate.The undulating reliefs of the cathode plate include cathode peaks that are pressed directly against the membrane-electrode assembly, more precisely against a cathode face thereof, while the undulating reliefs of the anode plate include anode peaks that are pressed directly against the membrane-electrode assembly, more precisely against an anode face thereof. The cathode peaks are drawn in elevation in Figure 4 of US 2021 / 0143451 while the anode peaks are drawn in elevation in Figure 3 of US 2021 / 0143451. In Figure 5 of US 2021 / 0143451, the superposition of the cathode peaks and the anodic peaks is drawn, highlighting the high surface pressure zones where the membrane-electrode assembly is locally pinched between respective parts, aligned with each other along the stacking axis, of the cathode peaks and the anodic peaks.These high surface pressure zones are shown alone in Figure 7 of US 2021 / 0143451. Thus, US 2021 / 0143451 details the relative arrangement of the tops, respectively cathode and anodic, of the cathode and anodic plates of each of its bipolar plates, but does not detail the relative arrangement of the bottom walls of the channels delimited respectively by the cathode plate and the anodic plate, that is to say the walls of these channels, applied directly in contact with each other.

[0005] The aim of the present invention is to propose a bipolar plate with corrugated channels in phase opposition, the mechanical and electrical performances of which of the contact interface between the two plates of the bipolar plate are controlled despite the uncertainty of the assembly positioning between the two plates.

[0006] For this purpose, the invention relates to a bipolar plate for a fuel cell, as defined in claim 1. The operational state of the bipolar plate is that found when the cell is assembled and operational.

[0007] One of the ideas underlying the invention is to seek to maximize the mechanical and electrical performance of the contact interface between the two plates of the bipolar plate, while taking into account a usual positioning tolerance between the two plates in the operational state of the bipolar plate. To do this, the invention provides that the corrugations, identical and parallel to each other, of the anode plate and the corrugations, identical and parallel to each other, of the cathode plate are in phase opposition and have the same pitch, while being able to have a corrugation width which is either identical, or which is different between the two plates.According to the invention, the offset of the corrugations of the anode plate and that of the corrugations of the cathode plate, which are the same, are provided to be equal to the sum between the aforementioned pitch and a predetermined value which is between, inclusive, half of the smaller of the two corrugation widths and, inclusive, half of the sum of the two corrugation widths.In a nominal configuration of the operational state of the bipolar plate, in which the mean axes of the corrugations of the anode plate are respectively aligned, along the stacking direction of the two plates, with the mean axes of the corrugations of the cathode plate, the contact areas between the two plates at their assembly interface form an overall pattern which exhibits good mechanical and electrical performance: in particular, as described in more detail below, each top portion of each corrugation of one of the two plates is applied against one of the top portions of the corrugations of the other of the two plates, forming a double-diamond contact area whose two diamonds are connected by a strip whose minimum width is between zero and half of the larger of two corrugation widths.One of the advantages of the invention is to retain these double diamond contact zones when the two plates of the bipolar plate are not, in the operational state, exactly in the nominal configuration but in a non-nominal configuration where the mean axes of the corrugations of the anode plate are offset relative to the mean axes of the corrugations of the cathode plate, this offset being due to the usual positioning tolerance of the two plates during their assembly and being typically of the same order of magnitude as half the sum of the two corrugation widths.Thanks to the invention, whatever the positioning difference that is actually present between the two plates in the operational state of the bipolar plate, within the limits of the usual positioning tolerance, the offset prescribed by the invention makes it possible to guarantee contact zones between the two plates, which are both sufficiently extensive in total to ensure a good electrical connection between the two plates, and sufficiently distributed at the interface between the two plates to ensure good mechanical stability of the assembly of the two plates, and, in doing so, good performance for a fuel cell comprising a stack of bipolar plates according to the invention. Furthermore, the control of the mechanical and electrical performance of the bipolar plate according to the invention is advantageously optimized by playing on the geometric characteristics of the corrugations of the two plates, as described in more detail below.

[0008] Thus, additional advantageous characteristics of the bipolar plate according to the invention are specified in claims 2 to 8.

[0009] The invention also relates to a fuel cell, as defined in claim 9.

[0010] The invention also relates to a method for manufacturing a fuel cell as defined above, this method being as defined in claim 10.

[0011] The manufacturing method according to the invention takes advantage of the fact that the mechanical and electrical performance of the fuel cell is guaranteed regardless of the actual positioning difference between the two plates of each bipolar plate during assembly of the latter, within the limits of the usual positioning tolerance.

[0012] The invention will be better understood by reading the following description, given solely by way of example and with reference to the drawings in which:

[0013] - figure 1 is an exploded perspective view of a fuel cell according to the invention;

[0014] - figure 2 is a partial schematic section along plane II of figure 1;

[0015] - figure 3 is a diagram illustrating channel undulations of one of the two plates of a bipolar plate, according to the invention, belonging to the fuel cell of figures 1 and 2;

[0016] - figure 4 is a view similar to figure 3, illustrating the superposition of the respective undulations of the two plates of the bipolar plate;

[0017] - Figure 5 is a diagram of contact zones between the two plates of the bipolar plate, resulting from the superposition of the undulations illustrated in Figure 4;

[0018] - figure 6 is a view similar to figure 4, illustrating a dimensioning variant for the respective corrugations of the two plates of the bipolar plate according to the invention;

[0019] - Figure 7 is a view similar to Figure 5 but applied to the superposition of the corrugations illustrated in Figure 6; - Figure 8 is a view similar to Figure 4, illustrating another dimensioning variant for the respective corrugations of the two plates of the bipolar plate according to the invention; and

[0020] - Figure 9 is a view similar to Figure 5 but applied to the superposition of the undulations illustrated in Figure 8.

[0021] Figure 1 shows a fuel cell 1. The fuel cell 1 is for example intended to be installed in a vehicle and to produce, from reactive fluids such as hydrogen and oxygen, electricity supplying an electric motor ensuring the propulsion of the vehicle.

[0022] The fuel cell 1 comprises a stack of cells 2, in which the cells 2 are arranged against each other along a stacking axis Z. The stack of cells 2 is held between two end plates, which are not shown in FIG. 1 and which in particular make it possible to keep the stack of cells 2 compressed, i.e. clamped, along the stacking axis Z and to supply this stack with both a reactive fuel fluid and a reactive oxidant fluid.According to a preferred example which will be relied upon subsequently, the fuel reactive fluid consists of hydrogen in gaseous form and the oxidant reactive fluid consists of oxygen in gaseous form, it being noted that, in practice, the fuel cell 1 may not be supplied with pure oxygen, but with air, or even with a gaseous mixture comprising oxygen, for example a mixture of oxygen and nitrogen, a mixture which may be in a ratio between oxygen and nitrogen which is similar to that of air or in a different ratio. Here, the terms “oxygen” and “dioxygen” are undifferentiated; likewise, the terms “hydrogen” and “dihydrogen” are undifferentiated. For the sake of simplicity, the terms “oxygen” and “dioxygen” will also cover air or any gaseous mixture containing oxygen in a significant proportion.

[0023] As shown in Figures 1 and 2, each cell 2 comprises a membrane-electrode assembly 10, a detail of a section of which is integrated in Figure 1.

[0024] Each cell 2 also comprises two polar plates which are arranged on either side, along the stacking axis Z, of the membrane-electrode assembly 10, namely an anode plate 20 and a cathode plate 30. For each cell 2, the anode plate 20 makes it possible to supply the cell 2 with hydrogen while the cathode plate 30 makes it possible to supply this cell with oxygen. For this purpose, and as clearly visible in FIG. 2, the anode plate 20 delimits, on its face facing the membrane-electrode assembly 10, channels 21 for the circulation of hydrogen while the cathode plate 30 delimits, on its face facing the membrane-electrode assembly 10, channels 31 for the circulation of oxygen.For each pair of two cells 2 immediately adjacent to each other, the anode plate 20 of one of the two cells and the cathode plate 30 of the other of the two cells are arranged directly against each other along the stacking axis Z and form, in the operational state of the fuel cell 1, a bipolar plate 3. Thus, in the operational state of the fuel cell 1, the two cells 2 of each pair of cells immediately adjacent to each other share the same bipolar plate 3 which separates from each other the membrane-electrode assembly 10 of one of these two cells 2 and the membrane-electrode assembly 10 of the other of these two cells 2.In the operational state of each bipolar plate 3, the channels 21 are delimited by the face of the anode plate 20 of the bipolar plate 3, facing away from the cathode plate 30 of the bipolar plate 3, while the channels 31 are delimited by the face of the cathode plate 30 of the bipolar plate 3, facing away from the anode plate 20 of this bipolar plate 3. Also in the operational state of each bipolar plate 3, the anode 20 and cathode 30 plates of the bipolar plate 3 form between them free passages 4 for the circulation of a cooling fluid, separated from the channels 21 and 31. The cooling fluid does not play any direct role in the electrochemical reactions occurring in cells 2, but the circulation of this cooling fluid makes it possible to control the temperature of cells 2, therefore the temperature of the electrochemical reaction, therefore the operating conditions of the electrochemical reaction.

[0025] In practice, each of the plates 20 and 30 is a shaped sheet, in particular stamped, made of metal. However, the invention can also be implemented with other embodiments of the plate, it being possible for it to be made of other materials, in particular for example graphite, and / or it being possible for it to be shaped other than by stamping, for example by machining, by chemical attack, by electroerosion, or by additive manufacturing.

[0026] Before describing the channels 21 and 31 in more detail, structural and functional characteristics of the membrane-electrode assembly 10 of each cell 2 are described in more detail below, it being emphasized that the specificities of these characteristics are not limiting. Thus, in the embodiment considered in the figures, the membrane-electrode assembly 10 of each cell 2 comprises two gas diffusion layers 11 and 12 which are arranged on either side, along the stacking axis Z, of a proton exchange membrane 13 whose two opposite faces are respectively covered by, or in contact with, an anodic catalytic layer 14 and a cathodic catalytic layer 15.In the assembled state of the fuel cell 1, each cell 2 thus comprises, successively along the stacking axis Z, the anode plate 20, the gas diffusion layer 11, the anode catalytic layer 14, the proton exchange membrane 13, the cathode catalytic layer 15, the gas diffusion layer 12, and the cathode plate 30.

[0027] For each cell 2, the gas diffusion layer 11 covers the channels 21 and allows the transport of hydrogen from the channels 21 of the anode plate 20 to the anode catalytic layer 14. Similarly, the gas diffusion layer 12 covers the channels 31 and allows the transport of oxygen from the channels 31 of the cathode plate 30 to the cathode catalytic layer 15. In practice, the gas diffusion layers 11 and 12 are formed of a porous material such as a carbon fiber fabric or porous carbon paper.

[0028] For each cell 2, the proton exchange membrane 13 allows hydrogen ions, or protons, to be conducted from the anodic catalytic layer 14 to the cathodic catalytic layer 15, while preventing the flow of electrons. The proton exchange membrane 13 is for example made of a perfluorinated sulfurized polymer material, such as the material known under the trade name “Nation”. The proton exchange membrane 13 also allows the flow of water in liquid form, for the purpose of removing the water produced at the cathodic catalytic layer 15 from the cell 2.

[0029] Within each cell 2 when the fuel cell 1 is in operation, an oxidation reaction occurs at the anodic catalytic layer 14 and consists of catalytically splitting the hydrogen supplied by the gas diffusion layer 11 into protons and electrons. The protons thus produced pass through the proton exchange membrane 13 until they reach the cathodic catalytic layer 15, while the electrons thus produced are captured by the anodic plate 20 and then conducted to the cathodic plate 30 of the bipolar plate 3 to which the anodic plate 20 of the cell 2 in question belongs.At the same time, a reduction reaction occurs at the cathode catalytic layer 15 and consists of reacting the oxygen supplied by the gas diffusion layer 12 with the protons passing through the proton exchange membrane 13, as well as with electrons supplied by the cathode plate 30 of the cell 2 considered, to form water molecules.

[0030] In practice, the anodic 14 and cathodic 15 catalytic layers are made up of porous structures, which allow the transport of reactive fluids, i.e. here hydrogen and oxygen, inside these catalytic layers, and which are formed from three different materials, namely:

[0031] - a material for transporting protons, for example the same material as the proton exchange membrane 13, such as the “Nation” material mentioned above,

[0032] - a material for transporting electrons, for example carbon, and - a material for catalyzing the electrochemical oxidation and reduction reactions described above, for example platinum.

[0033] In practice, the membrane-electrode assembly 10 of each cell 2, which is interposed between the anode 20 and cathode 30 plates of the cell 2 in question, also comprises a support frame 40. Advantageously, the support frame 40 is formed of two films superimposed and glued to each other 16, for example PET or PEN films themselves in the form of a frame for which to delimit an internal window whose contour is slightly smaller than the contour of the proton exchange membrane 13, the contour of the latter being trapped, along the contour of the internal window, between the two films of the support frame 40.

[0034] Also before describing the channels 21 and 31 in more detail, it will be noted that, as schematically illustrated in FIG. 1, the fuel cell 1 comprises a hydrogen inlet 50 supplying hydrogen to the anode plates 20 and a hydrogen outlet 51 discharging the hydrogen from the anode plates 20. The fuel cell 1 also comprises an oxygen inlet 52 supplying oxygen to the cathode plates 30 and an oxygen outlet 53 discharging the oxygen from the cathode plates 30. The fuel cell 1 also comprises a cooling fluid inlet 54 supplying cooling fluid to the bipolar plates 3 and a cooling fluid outlet 55 discharging the cooling fluid from the bipolar plates 3.In the embodiment illustrated in the figures, the inlets 50, 52 and 54, as well as the outlets 51, 53 and 55 are formed by openings provided in the anode plates 20, the cathode plates 30 and the support frames 40. In practice, these inlets and outlets are connected to corresponding openings, which are provided in one of the two aforementioned end plates and which are themselves connected to hydrogen, oxygen and cooling fluid supply circuits. In a variant not shown, these inlets and outlets are formed by conduits provided around the stack of cells 2, in an ad hoc housing of the fuel cell 1.

[0035] As shown schematically in Figure 1, each cathode plate 30 comprises three adjacent zones which successively connect the oxygen inlet 52 to the oxygen outlet 53, namely a homogenization zone 30A, an active zone 30B which is connected to the oxygen inlet 52 by the homogenization zone 30A, and a homogenization zone 30C connecting the active zone 30B to the oxygen outlet 53. The channels 31 of the cathode plate 30 extend in length over the entire active zone 30B, each connecting the homogenization zone 30A to the homogenization zone 30C. The active zone 30B has substantially the same dimensions as the window delimited by the support frame 40 of the membrane-electrode assembly 10 of the cell 2 to which the cathode plate 30 in question belongs, and it is generally expected that the gas diffusion layer 12 completely covers at least the active zone 30B.The homogenization zones 30A and 30C, which connect the oxygen inlet 52 and the oxygen outlet 53 to the active zone 30B respectively, make it possible to distribute the oxygen over the entire transverse extent of the active zone 30B, that is to say in all of the channels 31.

[0036] Each anode plate 20 comprises an active zone and two homogenization zones, which are respectively similar to the active zone 30B and to the homogenization zones 30A and 30C of each cathode plate 30: the two homogenization zones of each anode plate connect to the active zone of this anode plate respectively the hydrogen inlet 50 and the hydrogen outlet 51 and make it possible to distribute the hydrogen over the entire transverse extent of this active zone, that is to say in all of the channels 21 extending in length over this entire active zone, from one to the other of the two homogenization zones.

[0037] In the embodiment considered in the figures, the bipolar plates 3 and the support frames 40 have a rectangular peripheral contour. In addition, the hydrogen inlet 50 and the hydrogen outlet 51 are located diagonally to each other while the oxygen inlet 53 and the oxygen outlet 53 are also located diagonally to each other, which makes it possible to obtain a more homogeneous distribution of the reactive fluids on the respective active areas of the anode 20 and cathode 30 plates.

[0038] We will now examine in more detail the structural and dimensional characteristics of the channels 21 of the anode plates 20 and the channels 31 of the cathode plates 30, based on Figures 2 to 9. It will be noted that these characteristics differ between the embodiment illustrated in Figures 4 and 5, the embodiment illustrated in Figures 6 and 7, and the embodiment of Figures 8 and 9, as explained below.

[0039] As shown schematically in Figure 2, the channels 21 of each anode plate 20 are, over their entire length, separated two by two by tooth walls 22 belonging to the anode plate 20. In the assembled state of the fuel cell 1 to which the anode plate 20 belongs, these tooth walls 22 are entirely applied, along the stacking axis Z, directly against the membrane-electrode assembly 10 of the cell 2 concerned, more precisely against the gas diffusion layer 11 of this membrane-electrode assembly 10.Similarly, the channels 31 of each cathode plate 30 are, over their entire length, separated two by two by tooth walls 32 belonging to the cathode plate 30: in the assembled state of the cell 2 to which the cathode plate 30 belongs, the tooth walls 32 are entirely applied, along the stacking axis Z, directly against the membrane-electrode assembly 10 belonging to the cell 2 concerned, more precisely to the gas diffusion layer 12 of this membrane-electrode assembly 10. In practice, the tooth walls 22 and 32 are each substantially flat, within manufacturing tolerances.

[0040] Also as shown schematically in Figure 2, each of the channels 21 is delimited, over its entire length, by both a bottom wall 23, a side wall 24 which connects the bottom wall 23 to one of the two tooth walls 22 adjacent to the channel 21 in question, and a side wall 25 which connects the bottom wall 23 to the other of the two aforementioned tooth walls 22. The bottom wall 23 and the side walls 24 and 25 belong to the anode plate 20. Similarly, each of the channels 31 is delimited, over its entire length, by both a bottom wall 33, a side wall 34 and a side wall 35, all of which belong to the cathode wall 30, the side wall 24 connecting the bottom wall 23 to one of the two tooth walls 32 adjacent to the channel 31 in question while the side wall 25 connects the bottom wall 23 to the other of these two tooth walls 32.In the operational state of each bipolar plate 3, the bottom walls 23 of the anode plate 20 belonging to the bipolar plate 3 considered are directly in contact, along the stacking axis Z, against the bottom walls 33 of the cathode plate 30 belonging to the bipolar plate 3 considered, forming, between the anode 20 and cathode 30 plates, contact zones C3. In practice, for this purpose, the bottom walls 23 and 33 are substantially planar, so that, at the level of the contact zones C3, these bottom walls 33 and 23 are in substantially planar contact. For each bipolar plate 3, the contact zones C3 are arranged in a geometric plane TT which is perpendicular to the stacking axis Z. The free passages 4, which are formed between the anodic 20 and cathodic 30 plates of each bipolar plate 3 for the circulation of the cooling fluid, extend within the extent of the geometric plane TT, outside the contact zones C3.

[0041] As illustrated very schematically for the channels 31 in Figure 1, the channels 21 and 31 do not extend lengthwise in a straight line, but in an undulating manner, at least over a major part of their length. Indeed, it can be provided that the channels, in particular at their connecting end with the homogenization zones, comprise segments, in particular end segments, whose geometry is different from the geometry that they have in a main segment, in particular a central segment which extends between these possible end segments. The geometry of such main segments of the channels is considered below.More precisely, for each of the bipolar plates 3, the bottom walls 23 and the bottom walls 33 form, in orthogonal projection on the geometric plane TT, respective corrugations 26, respectively 36: in FIG. 3, the plane of which corresponds to the geometric plane TT, two of the corrugations 26 are shown alone, while in FIG. 4, the plane of which also corresponds to the geometric plane TT, both several of the corrugations 26 and several of the corrugations 36 are shown in the operational state of the corresponding bipolar plate 3. It is understood that, in FIG. 4, the regions of the geometric plane TT, at which the corrugations 26 of the anode plate 20 and the corrugations 36 of the cathode plate 30 overlap, correspond to the contact zones C3. In Figure 5, whose plane also corresponds to the geometric plane TT, only the contact zones C3 are drawn and shaded in order to visualize them better.As detailed below, we will distinguish, among the C3 contact zones, on the one hand, summit contact zones C3.1, and on the other hand, axial contact zones C3.X.

[0042] As clearly visible in Figures 3 and 4, for each bipolar plate 3, the corrugations 26 of a given bipolar plate 3 are identical to each other and parallel to each other, being distributed in the geometric plane TT in a regular manner along a distribution axis Y26. These corrugations 26 are thus spaced two by two by a constant pitch, which is noted p in Figures 2, 3, 4, 6 and 8. Similarly, for each bipolar plate 3, the corrugations 36 of a bipolar plate 3 are identical to each other and parallel to each other, being distributed in the geometric plane TT in a regular manner along a distribution axis Y36 which, in the operational state of the bipolar plate 3, is, apart from positioning tolerances, coincident with the distribution axis Y26. The corrugations 36 are thus spaced two by two at a constant pitch, which is equal to the pitch p of the corrugations 26 and which is therefore also noted p in figures 2, 4, 6 and 8.

[0043] As clearly visible in Figures 4, 6 and 8, for each bipolar plate 3, the undulations 26 of the anode plate 20 of a given bipolar plate 3 are, in the operational state of the bipolar plate 3, in phase opposition with the undulations 36 of the cathode plate of said given bipolar plate 3.

[0044] As shown in Figures 3, 4, 6 and 8, for each bipolar plate 3, each of the corrugations 26 of the anode plate 20 defines a mean axis X26, which is perpendicular to the distribution axis Y26 and along which the corresponding corrugation 26 extends in length while alternating on either side of this mean axis X26. The respective mean axes X26 of the corrugations 26 all belong to the geometric plane TT and are parallel to each other. Similarly, for each bipolar plate 3, each of the corrugations 36 of the cathode plate 30 defines a mean axis X36, which is perpendicular to the distribution axis Y36 and along which the corresponding corrugation 36 extends in length while alternating on either side of the mean axis X36. The respective mean axes X36 all belong to the geometric plane TT and are parallel to each other.In the operational state of each bipolar plate 3, the mean axes X26 and the mean axes X36 are parallel to each other: more precisely, in a nominal configuration of the operational state of the bipolar plate 3 concerned, the mean axes X26 are respectively merged with the mean axes X36, as illustrated in Figures 4, 6 and 8. Conversely, in a non-nominal configuration of this operational state, resulting from a usual relative positioning tolerance which is intrinsic to the manufacture of each bipolar plate 3 by stacking the anode 20 and cathode 30 plates of the latter, the mean axes X26 are not merged with the mean axes X36, but are offset from the latter along the distribution axes Y26 and Y36, the corresponding offset being called 5 in the following.The value of this offset 5 is in any case strictly less than half of the step p which is common on the one hand to the undulations 26 and on the other hand to the undulations 36, preferably strictly less than a quarter of this step p.

[0045] In all cases, it will therefore be possible to define two corrugations 26 and 36, one belonging to the anode plate 20 and the other belonging to the cathode plate 30, as being two corresponding corrugations of the bipolar plate 3, when the mean axis X26 of one coincides with the mean axis X36 of the other, or when their mean axes X26, X36 are offset from each other by an offset less than half the pitch p which is common on the one hand to the corrugations 26 of the anode plate 20 and on the other hand to the corrugations 36 of the cathode plate 30, preferably strictly less than a quarter of this pitch p.

[0046] Since the corrugations 26 of the anode plate 20 of a given bipolar plate 3 are, in the operational state of the bipolar plate 3, in phase opposition with the corrugations 36 of the cathode plate 30 of said given bipolar plate 3, each corrugation 26 of the anode plate 20 and the corresponding corrugation 36 of the cathode plate 30 necessarily intersect at their mean axis to form, among the contact zones 3, the axial contact zones C3.X.

[0047] As illustrated in Figures 3 and 4, for each bipolar plate 3, each of the corrugations 26 of the anode plate 20 is delimited, over its entire length, by two longitudinal edges 26A and 26B, which are opposite each other along the distribution axis Y26 and which, here, each undulate on either side of the mean axis X26 of the corrugation 26 concerned. These longitudinal edges 26A and 26B are parallel to each other, being spaced apart from each other by a width of the corresponding corrugation 26, denoted 11 in FIGS. 2, 3, 4, 6 and 8. Insofar as the corrugations 26 are identical to each other, the width 11 is the same for all the corrugations 26 and, for each of these corrugations 26, is constant over the entire length of the corrugation.It is thus possible to define, for each of the corrugations 26 of the anode plate 20 of each bipolar plate 3, a neutral fiber 26C as being the corrugated line which extends both parallel and midway between the two longitudinal edges 26A and 26B of the corrugation 26 concerned.

[0048] In the same way, and as illustrated in Figure 4, for each bipolar plate 3, each of the corrugations 36 of each cathode plate 30 is delimited, over its entire length, by two longitudinal edges 36A and 36B, which are opposite each other along the distribution axis Y36 and which, here, each undulate on either side of the mean axis X36 of the corrugation 36 concerned. These longitudinal edges 36A and 36B are parallel to each other, being spaced apart from each other by a width of the corresponding corrugation 36, denoted I2 in FIGS. 2, 3, 4, 6 and 8. Insofar as the corrugations 36 are identical to each other, the width I2 is the same for all the corrugations 36 and, for each of these corrugations 36, is constant over the entire length of the corrugation.It is thus possible to define, for each of the undulations 36 the cathode plate 30 of each bipolar plate 3, a neutral fiber 36C as being the undulated line which extends both parallel and midway between the two longitudinal edges 36A and 36B of the undulation 36 concerned.

[0049] It will be noted that the width 11 and the width I2 are equal to each other both in the embodiment illustrated in Figures 4 and 5 and in the embodiment illustrated in Figures 6 and 7. On the other hand, in the embodiment illustrated in Figures 8 and 9, the width 11 is different from the width I2: here, the width 11 is strictly greater than the width I2; however, in a variant not shown, the width 11 can be provided to be strictly less than the width I2.

[0050] As illustrated in Figures 3, 4, 6 and 8, each of the corrugations 26 of each anode plate 20 includes, as longitudinal portions furthest from the mean axis X26 of the corrugation 26 concerned along the distribution axis Y26, both first top portions 26.1, which are arranged on the same first side of the mean axis X26, and second top portions 26.2, which are arranged on the same second side, opposite the aforementioned first side, of the mean axis X26. The top portions 26.1 of each corrugation 26 are aligned with each other in a direction parallel to the mean axis X26 concerned. Similarly, the top portions 26.2 are aligned with each other in a direction parallel to the mean axis X26 concerned.Similarly, and as illustrated in Figures 4, 6 and 8, each of the corrugations 36 of each cathode plate 30 includes, as longitudinal portions furthest from the mean axis X36 of the corrugation 36 concerned along the distribution axis Y36, both first top portions 36.1, which are arranged on the same first side of the mean axis X36, and second top portions 36.2, which are arranged on the same second side, opposite the aforementioned first side, of the mean axis X36. The top portions 36.1 of each corrugation 36 are aligned with each other in a direction parallel to the mean axis X36 concerned. Likewise, the top portions 36.2 are aligned with each other in a direction parallel to the mean axis X36 concerned.

[0051] According to a preferred embodiment, which is implemented in the embodiments illustrated in the figures and which have, among other advantages, to simplify the design and manufacture of the anode 20 and cathode 30 plates of each bipolar plate 3, each of the undulations 26, respectively 36, of each anode 20, respectively cathode 30 plate, is constituted by the repetition of a undulation pattern M26, respectively M36. More precisely, the undulation pattern M26, which is here identical for the different undulations 26 of each anode plate 20, includes only one of the first top portions 26.1 and only one of the second top portions 26.2, and is repeated several times successively along the mean axis X26 of each undulation 26 to constitute the entirety of the latter.In the same way, the undulation pattern M36, which is here identical for the different undulations 36 of each cathode plate, includes only one of the first top parts 36.1 and only one of the second top parts 36.2, and is repeated several times successively along the mean axis X36 of each undulation 36 to constitute, in certain examples, the entirety of the latter.

[0052] In order in particular to further reinforce the simplification of the design and manufacture of the anode 20 and cathode 30 plates of each bipolar plate 3, the corrugation pattern M26, respectively M36, is preferably made up of two sections which follow one another along the mean axis X26, respectively X36, and which are arranged on either side of this mean axis, namely a first section M26.1, respectively M36.1, which includes the first top portion 26.1, respectively 36.1, of the corrugation pattern M26, respectively M36, and a second section M26.2, respectively M36.2, which includes the second top portion 26.2, respectively 36.2, of the corrugation pattern M26, respectively M36. The first section M26.1, respectively M36.1, thus includes, in addition to the first summit part 26.1, respectively 36.1, of the undulation pattern M26, respectively M36, two longitudinal parts 26.3 and 26.4 of the corresponding undulation 26, respectively 36.3 and 36.4 of the corresponding corrugation 36, which each connect the first top portion 26.1, respectively 36.1, to the mean axis X26, respectively X36, of the corresponding corrugation. Similarly, the second section M26.2, respectively M36.2 includes, in addition to the second top portion 26.2, respectively 36.2, of the corrugation pattern M26, respectively M36, two longitudinal portions 26.5 and 26.6 of the corresponding corrugation 26, respectively 36.5 and 36.6 of the corresponding corrugation 36, which each connect the second top portion 26.2, respectively 36.2, to the mean axis X26, respectively X36, of the corresponding corrugation. The longitudinal parts 26.4 and 26.5, respectively 36.4 and 36.5, connect directly to each other, at the level of the mean axis X26, respectively X36. The first section M26.1, respectively M36.1, has an axis of symmetry YM26.1, respectively YM36.1 , which extends perpendicular to the mean axis X26 of the corresponding corrugation 26, respectively X36 of the corresponding corrugation 36, and which passes through the first top portion 26.1 of the corrugation pattern M26, respectively 36.1 of the corrugation pattern M36: this axis of symmetry YM26.1 , respectively YM36.1 , constitutes an axis of symmetry for the first top portion 26.1 , respectively 36.1 , and the longitudinal portions 26.3 and 26.4, respectively 36.3 and 36.4, are symmetrical to each other with respect to this axis of symmetry YM26.1 , respectively YM36.1. Furthermore, the second section M26.2, respectively M36.2, is symmetrical to the first section M26.1, respectively M36.1, with respect to a point OM26, respectively OM36, which is formed at the intersection between the corresponding mean axis X26, respectively X36, and the corresponding neutral fiber 26C, respectively 36C: thus, the longitudinal part 26.5, respectively 36.5, is symmetrical to the longitudinal part 26.4, respectively 36.4, with respect to this point of symmetry OM26, respectively OM36; the second summit part 26.2, respectively 36.2, is symmetrical to the summit part 26.1, respectively 36.1, with respect to this point of symmetry OM26, respectively OM36; and the longitudinal part 26.6, respectively 36.6, is symmetrical to the longitudinal part 26.3, respectively 36.3, with respect to this point of symmetry OM26, respectively OM36.

[0053] In all cases, each corrugation 26 of each anode plate 20 has an offset, denoted D in figures 3, 4, 6 and 8, this offset D corresponding to the spacing, along the distribution axis Y26 and measured at the level of the neutral fiber 26C of the corrugation 26 concerned, between the first top parts 26.1 and the second top parts 26.2 of the corrugation 26 concerned. To the extent that the corrugations 26 of the anode plate 20 are identical to each other, this offset D is the same for all the corrugations 26. In the same way, each corrugation 36 of each cathode plate 20 has an offset corresponding to the spacing, along the distribution axis Y36 and measured at the level of the neutral fiber 36C of the corrugation 36 concerned, between the first top parts 36.1 and the second top parts 36.2 of the corrugation 36 concerned: this offset of the corrugations 36 of the cathode plate 30, which is the same for all the corrugations 36 since the latter are identical to each other, is the same as the offset D of the corrugations 26 of the anode plate 20. and is therefore also noted D in figures 4, 6 and 8. This offset D of the corrugations 26 and 36 of each bipolar plate 3 is dimensioned so as to respect the following relationship:.

[0054] D = p + x with min(l1 , l2) / 2 < x < (11 + l2) / 2.

[0055] In other words, the offset D is equal to the sum of the pitch p and a predetermined value x which is between, inclusive, half of the smallest of the widths 11 and I2 and, inclusive, half of the sum of the widths 11 and I2.

[0056] In the embodiment of Figures 4 and 5, where the widths 11 and 12 are equal, the offset D is chosen to be equal to the sum of the pitch and half of the width 11 or 12. In other words, D = p + 11 = p + 12.

[0057] In the embodiment of Figures 6 and 7, where the widths 11 and 12 are also equal to each other, the offset D is chosen to be equal to the sum of the pitch p and the width 11 or 12. In other words, D = p + 11 / 2 = p + 12 / 2.

[0058] In the embodiment of figures 8 and 9, where one of the widths, here the width 11, is strictly greater than the other width, here the width I2, with in addition the larger width of the two, here therefore the width 11, strictly greater than the smaller width of the two, here the width I2, the offset D is chosen equal to the sum between the pitch p and half of the smaller width I2. In other words, D = p + I2 / 2.

[0059] In all cases, the above relationship makes it possible, in the nominal configuration of the operational state of each bipolar plate 3, and for all the corrugations 26 of the anode plate 20 with the exception of the end corrugations 26 along the distribution axis Y26, to bring the top portions 26.1 of each corrugation 26 of the anode plate 20 into contact with respectively the top portions 36.2 of an adjacent corrugation 36 of the cathode plate 30, and to bring the top portions 26.2 of the same corrugation 26 of the anode plate 20 into contact with respectively the top portions 36.1 of another adjacent corrugation 36 of the cathode plate 30, forming between them top contact zones C3.1 among the contact zones C3.For a given corrugation 26 of the anode plate 20, the two adjacent corrugations 36 of the cathode plate 30 are the two corrugations which are immediately adjacent to the corresponding corrugation 36 of the cathode plate 30 as defined above, namely having the mean axis X36 coincident with the mean axis X26 of the given corrugation 26 or having an offset strictly less than half the corrugation pitch P-.

[0060] Conversely, the above relationship makes it possible, in the nominal configuration of the operational state of each bipolar plate 3, and for all the corrugations 36 of the cathode plate 30 with the exception of the end corrugations 36 along the distribution axis Y36, to bring the top portions 36.1 of each corrugation 36 of the cathode plate 30 into contact with respectively the top portions 26.2 of an adjacent corrugation 26 of the anode plate 20, and to bring the top portions 36.2 of the same corrugation 36 of the cathode plate 30 into contact with respectively the top portions 26.1 of another adjacent corrugation 26 of the anode plate 20, forming between them top contact zones C3.1 among the contact zones C3.For a given corrugation 36 of the cathode plate 30, the two adjacent corrugations 26 of the anode plate 20 are the two corrugations 26 which are immediately adjacent to the corresponding corrugation 26 of the anode plate 20 as defined above, namely having the mean axis X26 coincident with the mean axis X36 of the given corrugation 36 or having an offset strictly less than half the corrugation pitch p.

[0061] More precisely, it is noted that each given corrugation 26 of the anode plate 20, with the exception of the end corrugations 26 along the distribution axis Y26, is thus in contact with exactly 3 distinct corrugations 36 of the cathode plate 30, namely: with the corresponding corrugation 36 of the cathode plate 30, to form the axial contact zones C3.X for this given corrugation 26;

[0062] - with a first corrugation 36 adjacent to the cathode plate 30, at its top parts 26.1, to form one half of the top contact zones C3.1 for this given corrugation 36;

[0063] - and with a second adjacent corrugation 36 of the cathode plate 30, at its top portions 26.2, to form the other half of the top contact zones C3.1 for this given corrugation 26.

[0064] Conversely, each given corrugation 36 of the anode plate 30, with the exception of the end corrugations 36 along the distribution axis Y36, is thus in contact with exactly 3 distinct corrugations 26 of the anode plate 20, namely: with the corresponding corrugation 26 of the anode plate 20, to form the axial contact zones C3.X for this given corrugation 36;

[0065] - with a first corrugation 26 adjacent to the anode plate 30, at its top parts 36.1, to form one half of the top contact zones C3.1 for this given corrugation 36;

[0066] - and with a second adjacent corrugation 36 of the cathode plate 30, distinct from the first, at its top parts 36.2, to form the other half of the top contact zones C3.1 for this given corrugation 36.

[0067] Particularly advantageously, the above relationship allows that, for each corrugation 26 of the anode plate 20 and each corrugation 36 of the cathode plate 30, there is necessarily a summit contact zone C3.1 between two immediately successive axial contact zones C3.X. Thus, with the exception of the end corrugations 26, 36 along the distribution axis Y26, Y36, there is no summit zone which is not in contact with a corrugation of the opposite plate.

[0068] As clearly visible in Figures 5, 7 and 9, each of the summit contact zones C3.1 has a double diamond shape, including two diamonds and a strip which connects the two aforementioned diamonds and whose minimum width is between zero, as in the embodiment of Figures 4 and 5, and half of the larger of the two widths 11 and 12, as in the embodiment of Figures 6 and 7, as well as in the embodiment of Figures 8 and 9. It will be noted that the term "diamond" used here is not understood in its strict geometric sense, but as an evocation of shape. In the case where the operational state of one of the bipolar plates 3 is not in the nominal configuration, the offset 5 mentioned above does not call into question the obtaining of the summit contact zones C3.1, even if the effective contour of each of these summit contact zones C3.1.1 differs somewhat from the double diamond shape described above for the nominal configuration, since this offset 5 is of the same order of magnitude as the widths 11 and 12, typically being equal to the min value (11, 12) of the smallest of the width 11 and the width 12. In this respect, by way of non-limiting example, the widths 11 and 12 are typically between 0.1 mm and 1 mm, preferably between 0.1 and 0.5 mm, while the usual positioning tolerance between the anode 20 and cathode 30 plates of each bipolar plate 3 induces a value of approximately 0.15 mm for the offset 5.

[0069] Thus, and more generally, by applying the above relationship, the obtaining of the contact zones C3, in particular the summit contact zones C3.1, is guaranteed and in this way the mechanical and electrical performances at the contact interface between the anodic 20 and cathodic 30 plates of each bipolar plate 3 are controlled, whatever the actual positioning of these two plates during their assembly within the limits of the usual positioning tolerance.

[0070] In practice, to control the value of the offset D, it is possible to play on various geometric characteristics of the corrugations 26 and 36 of each bipolar plate 3. According to a possibility that is both practical, efficient and economical, which is implemented here, the top parts 26.1 and 26.2 of each corrugation 26, respectively 36.1 and 36.2 of each corrugation 36, are each curved, preferably following a circular profile: in this way, it is possible to play on the radius of curvature of these top parts 26.1 and 26.2, respectively 36.1 and 36.2, it being noted that the top parts 26.1, respectively 36.1, have a concavity which is opposite to that of the top parts 26.2, respectively 36.2. That being said, other possibilities are conceivable: as examples not shown, the summit parts 26.1 and 26.2, respectively 36.1 and

[0071] 36.2, are each curved but following a sinusoidal profile, or these summit parts 26.1 and 26.2, respectively 36.1 and 36.2, are each rectilinear, following a flat profile.

[0072] Taking into account the explanations given so far, it is understood that the fuel cell 1 is preferably manufactured by first attaching one of the anode plates 20 and one of the cathode plates 30 on either side, along the stacking axis Z, of each of the membrane-electrode assemblies 10, so as to form the different cells 2, then by assembling together the anode 20 and cathode 30 plates of each bipolar plate 3. That being said, other possibilities are conceivable for manufacturing the fuel cell 1. Generally speaking, the anode 20 and cathode 30 plates of each bipolar plate 3 can be assembled to each other in different ways, including by brazing, by welding, by gluing, or by simple assembly by compression of the stack of cells.

[0073] Finally, various arrangements and variants of the bipolar plate 3 and the fuel cell 1 described so far are conceivable. For example, the different variants mentioned at different points in the description above can be combined with each other, at least partially.

Claims

CLAIMS 1. Bipolar plate (3) for a fuel cell (1), comprising two plates, respectively anode (20) and cathode (30), which, in the operational state of the bipolar plate (3), are applied against each other along a stacking axis (Z), in which each of the two plates (20, 30) delimits, on its face facing away from the other plate, channels (21, 31) for the circulation of a reactive fluid, which each have a bottom wall (23, 33) in contact with a bottom wall (33, 23) of a single channel (31, 21) of the other of the two plates, in which the bottom walls (23, 33) of the channels (21, 31) of each of the two plates (20, 30) form, in orthogonal projection on the same geometric plane (rr) perpendicular to the axis stacking (Z), respective corrugations (26, 36) which, for each of the two plates (20, 30): - are identical and parallel to each other, being distributed regularly along a distribution axis (Y26, Y36) and spaced two by two by a pitch, noted p, the pitch of the corrugations (26) associated with the anode plate (20) being equal to the pitch of the corrugations (36) associated with the cathode plate (30), and - in the operational state of the bipolar plate (3), are in phase opposition with the undulations associated with the other of the two plates, in which each of the undulations (26, 36) associated with each of the two plates (20, 30) defines a mean axis (X26, X36): - which is perpendicular to the distribution axis (Y26, Y36), - according to which the undulation extends in length while alternating on either side of the mean axis, and - which, in a nominal configuration of the operational state of the bipolar plate (3), is coincident with one of the mean axes of the corrugations associated with the other of the two plates, in which each of the corrugations (26, 36) associated with each of the two plates (20, 30) is delimited by two longitudinal edges (26A, 26B, 36A, 36B) which: - are opposite each other along the distribution axis (Y26, Y36), - are parallel to each other, being spaced apart from each other by a corrugation width, the corrugation width of the corrugations (26) associated with the anode plate (20) being noted 11 while the width of the corrugations (36) associated with the cathode plate (30) is noted I2, and - define a neutral fiber (26C, 36C) of the corrugation (26, 36), which extends both parallel and midway between the two longitudinal edges, in which each of the corrugations (26, 36) associated with each of the two plates (20, 30) includes, as longitudinal parts furthest from the mean axis (X26, X36) along the distribution axis (Y26, Y36), first and second top parts, the first top parts (26.1, 36.1) being both arranged on the same first side of the mean axis of the corrugation and aligned with each other in a direction parallel to the mean axis, while the second top parts (26.2, 36.2) are both arranged on the same second side, opposite the first side, of the mean axis of the corrugation and aligned with each other in a direction parallel to the mean axis, and in which the corrugations (26) associated with the anode plate (20) and the corrugations (36) associated with the cathode plate (30) have the same offset, noted D, which corresponds to the spacing, along the distribution axis (Y26, Y36) and measured at the level of the neutral fiber (26C, 36C), between the first summit parts (26.1, 36.1) and the second summit parts (26.2, 36.2) of each corrugation, the offset being such that D = p + x with min(l1, l2) / 2 < x < (11 + l2) / 2.

2. Bipolar plate according to claim 1, in which the corrugation width 11 of the corrugations (26) associated with the anode plate (20) is equal to the corrugation width I2 of the corrugations (36) associated with the cathode plate (30).

3. Bipolar plate according to claim 1, in which the corrugation width 11 of the corrugations (26) associated with the anode plate (20) is strictly greater than the corrugation width I2 of the corrugations (36) associated with the cathode plate (30).

4. Bipolar plate according to claim 1, in which the corrugation width 11 of the corrugations (26) associated with the anode plate (20) is strictly less than the corrugation width I2 of the corrugations (36) associated with the cathode plate (30).

5. A bipolar plate according to any preceding claim, wherein each of the corrugations (26, 36) associated with each of the two plates (20, 30) comprises a corrugation pattern (M26, M36) which: - is identical between the undulations of each of the two plates, - includes only one of the first summit parts (26.1, 36.1) and only one of the second summit parts (26.2, 36.2) of the undulation, and - is repeated several times successively along the mean axis (X26, Y26) of the ripple to constitute the ripple.

6. Bipolar plate according to claim 5, in which the undulation pattern (M26, M36) of each of the undulations (26, 36) associated with each of the two plates (20, 30) consists of a first section (M26.1, M36.1) and a second section (M26.2, M36.2), which follow one another along the mean axis (X26, X36) of the undulation and which are arranged on either side of the mean axis of the undulation, respectively including the first top portion and the second top portion of the undulation pattern, in which the first section (M26.1, M36.1) has an axis of symmetry (YM26.1), which extends perpendicular to the mean axis (X26, X36) of the undulation (26, 36) and which passes through the first top portion (26.1, 36.1) of the undulation pattern (M26, M36), and in which the second section (M26.2, M36.2) is symmetrical to the first section (M26.1, M36.1) with respect to a point (OM26) formed at the intersection between the mean axis (X26, X36) and the neutral fiber (26C, 36C) of the undulation (26, 36).

7. A bipolar plate according to any preceding claim, wherein the first and second top portions (26.1, 26.2, 36.1, 36.2) of each of the corrugations (26, 36) associated with each of the two plates (20, 30) are each curved.

8. Bipolar plate according to claim 7, in which each of the first and second top portions (26.1, 26.2, 36.1, 36.2) of each of the corrugations (26, 36) associated with each of the two plates (20, 30) follows a circular profile.

9. Fuel cell (1), comprising bipolar plates (3), which each conform to any one of the preceding claims and which are stacked against each other in the stacking direction (Z), with the interposition of a membrane-electrode assembly (10) between two successive bipolar plates.

10. Method for manufacturing a fuel cell (1) according to claim 9, in which, before assembling together the anode (20) and cathode (30) plates of each of the bipolar plates (3), each of the membrane-electrode assemblies (10) is assembled with one of the anode plates (20) and one of the cathode plates (30), attached on either side, along the stacking axis (Z), of the membrane-electrode assembly concerned.

Images