CZ309461B6 - Fuel cell distribution board - Google Patents

Abstract

The invention is a distribution board in fuel cells as sources of electrical energy into which gaseous fuel and air as an oxidizer are supplied — and the generated electric current is output. One of the advantages of these cells is that the by-products of the reaction do not pollute the air because it is pure water. This water, in the form of droplets held by surface tension, blocks the cavities for the flow of gases in existing designs of distribution boards. This impairs the conversion efficiency and is best removed. According to the invention, this is solved by specially shaping the cavities in the distribution plates, which induce self-excited periodic oscillations of the flowing gas in the entrance part of the cavities. For part of the oscillation period, the short-term increased flow exerts an intense hydrodynamic force on the water droplets and drags them with it towards the outlet opening from the plate. This will suppress the previous problems caused by flooding of part of the cavities with water. Flow oscillations also increase the efficiency of the ongoing conversion process by mixing.

Description

Fuel cell distribution board

Field of technology

The subject of the invention are distribution boards in fuel cells. These cells are known as potentially advantageous sources of electrical energy in a wide variety of technical fields. Both fuel and oxidizer are fed into them, usually both in the gas phase - and then the generated electric current is taken out of them. These cells - and the systems assembled from them, i.e. cell batteries - are especially considered for the near future as an advantageous alternative for the electric drive of motor vehicles. Compared to chemical cells, they have a small weight and do not require long charging times - and in principle they do not generate any harmful emissions, as the resulting product of the ongoing reactions is clean water.

Current state of the art

The individual fuel cells and the batteries formed from them are assembled from plates that are not stacked on top of each other, and the distribution plates according to the invention are an important part of them. In the cells, processes of electric current and/or voltage generation take place. Other important plates in the cell are mainly the pair of plate electrodes, the anode and the cathode. Between them is a polymer membrane, also shaped like a plate of the same overall shape. The electricity generation depends on the special properties of this membrane. The electrodes are placed each on its own side of this membrane and are each in contact with the distribution board on its own side. The purpose of this distribution board is to distribute the reactant through the small hole through which the reactant is supplied, which is usually hydrogen on one side and, similarly, air on the opposite side, as much as possible over the entire surface of the electrode. A system of channels usually created by selecting material on one side of the board is used for this distribution. It is precisely these distribution boards that are the subject of the invention.

The plates of both electrodes - anodes and cathodes - in fuel cells are usually porous and thin and therefore very delicate from a mechanical point of view. Also, the membrane between them is most often made of a soft polymeric material and therefore does not have any great mechanical rigidity and strength in current designs. In order to strengthen the bundle of plates, it is therefore customary to provide it with the necessary mechanical rigidity by adding a solid supporting end plate at both ends of the bundle - so that the cell can be easily handled, for example during assembly and similar situations.

Quite often, in the design and manufacture of the cell, the outer distribution plate of the battery is combined with the adjacent end plate. Both thus form a single unit. The end plate gives this combination mechanical rigidity, and at the same time, the system of channels made in it covers the surface of the adjacent electrode plate, just like distribution plates. Another relatively common combination is the connection of two adjacent distribution boards into one mechanical unit. The resulting board is then referred to as a bipolar distribution board. In both cases of such combinations, the function of the distribution channels does not change in any way and thus no functional advantage is achieved, except that a greater spatial compactness of the battery of cells can be achieved.

Feedstock hydrogen, which is most often thought of as fuel for vehicle fuel cells, is produced at some cost. It does not occur alone in nature. Therefore, although not large, but - especially with the expected mass production - non-negligible funds are spent on its production, both investment and operational. It is therefore desirable that the hydrogen in the fuel cell is used as best as possible, so that its consumption is as small as possible. To which the highest possible efficiency of the processes in the article is required. Even a small improvement in efficiency can be significant given the large production volumes expected overall for vehicle operation. The requirement for good efficiency also results from the very large specific volume of hydrogen that it has in its normal state. This characteristic of it brings difficulties with its storage, for example in a vehicle or in other facilities where it is to be temporarily preserved for a certain period of time for later consumption. The bigger

- 1 CZ 309461 B6 is the efficiency of its use in the fuel cell, the smaller the hydrogen stock stored in the vehicle (or indeed elsewhere) can be. Of course, its volume can be reduced by compression for more convenient storage, but this brings other problems. Solving this issue of functional efficiency is one aspect of the invention.

One of the practical problems of existing fuel cell arrangements is the blocking of flow cavities for gases in the distribution plates with water droplets. This water is present in the liquid phase as a product of the reaction taking place at the cathode, and it also reaches other cavities in the form of water droplets. However, these water droplets block the available cavities for hydrogen, which impairs the flow of hydrogen and thus reduces the overall efficiency of the cells. Ways to suppress this blocking effect are mentioned in the literature. Although a number of them are known, they do not provide a complete solution. The most considered is the periodic opening of the cavity in the distribution boards to the atmosphere. The gaseous fuel then flowing out at high speed drags water droplets with it and carries them away. However, this removal of gaseous fuel by discharge is unfavorable from the point of view of the cell's energy balance. The flow must be quite intense, and this causes a substantial loss of hydrogen. In addition to other undesirable circumstances, such as complications with the mechanical drive of a necessary opening and closing valve in the outlet to the atmosphere, potential problems such as the possibility of fire also increase with the release of hydrogen.

The essence of the invention

The previously mentioned shortcomings are suppressed by the distribution board of the fuel cells with created cavities for fluid flow, one end of which is connected to the supply opening and open on one side to the surface of the adjacent electrode with cavities for fluid flow, one end of which is connected to the supply opening for fuel fluid and after open on one side, the essence of which is that the cavities have a shape in plan with branching into two branches, which are the first channel and the second channel, where each of these two branches leads separately to one of the two outlet openings, with the place of this branching into two branches is a narrowing of the cross-section of the cavity created by the mutual approximation of the inlet walls located opposite each other which reach up to the inlet opening, with the first inlet wall located on one side of the cavity and the second inlet wall located on the other opposite side of the cavity and behind this point of narrowing in the direction of flow is between the oppositely located by the retaining walls - both the first retaining wall and the second retaining wall - on the contrary, in the flow direction of the expansion of the cross-section, where at the end of this expansion there is a separator created between the entrances to the first channel and the second channel.

It may therefore be expedient that the distribution plate according to the invention between the ends of the two inlet walls and the ends of the two retaining walls in the flow direction has on each side one of the two recesses formed in the wall of the distribution plate.

According to the invention, it can also be expedient for the distribution board to have a protrusion on each side of the recesses of both walls of the distribution board, between the outer wall of which and the inner wall of the recess there is an empty space.

It may also be expedient for the distribution board to be arranged according to the invention with both inlet walls being formed on opposite sides of the part of the distribution board forming one continuous projection, the supply opening being formed in this continuous projection and on the outside of this continuous projection being between and the other parts of the distribution board an empty gap.

In this arrangement according to the invention, the problem of the presence of water in the cavities for hydrogen, reducing the efficiency of the fuel cell, is solved by the special shaping of the cavities in the distribution plates, where self-excited periodic oscillations are generated. In one of the two channels, for the duration of half the oscillation period, a more intense flow towards the outlet opening is created - while simultaneously in the other of the two channels, a flow that is less intense is created for a short time - or even a flow in the reverse direction occurs there. In the following half of the period, these intensity changes on the left and right alternate with each other. In the short term, the increased flow has an intense effect

-2CZ 309461 B6 hydrodynamic force on individual drops of water and drags them with it to the outlet opening. This will suppress the problems caused by the flooding of the cavities in the plate and in the electrode with water, moreover, the induced flow oscillations will allow the mixing of the flowing gas and the suppression of boundary layers in the liquid to increase the efficiency of the ongoing conversion process.

Clarification of drawings

A total of eleven alternative examples of the implementation of a fuel cell distribution board according to the invention are shown in the attached twenty-five figures, and examples of stacking of distribution boards to form a fuel cell and a battery of fuel cells are also shown.

The individual described examples of the realization of the distribution board differ from each other in two respects. Either they have different designs of cavities in which self-excited oscillations occur caused by aerodynamic instability in the supplied fluid (Fig. 2, 10, 12, 14, 20, 21), or they have different designs of channels (Fig. 2, 10, 18), in which hydrogen and air flow and thus flow along the surface of the electrodes. In a series of separate drawings (fig. 4, 5, 6, 7, 11, 13, 15, 16, 17) some details of the implementation of distribution boards are depicted in an enlarged version.

The stacking of two circuit boards to form a single fuel cell is shown in cross-section in Fig. 1. The details of stacking multiple circuit boards to form a multi-cell battery are shown in cross-section in Figs. 3, 8, 9. The possibilities of storing adjacent circuit boards in a fuel cell are shown in Fig. 22, 23, 24, 25. The solution of the passages for supply and discharge of gaseous reactants and products in the distribution board is shown in Fig. 19.

Examples of implementation of the invention

Example 1

In this example, shown mainly in figures 1 and 2 but with some enlarged details also in figures 4, 5, 6 and 7, the distribution boards according to the invention are parts of a fuel cell designed to supply both hydrogen as fuel and oxygen from the atmosphere as an oxidizer. In this case, the fuel cell is intended for applications requiring an electrical supply with a very low voltage, at the level of around one volt. With such a low required voltage, there is no need to assemble a battery from several cells - as is the case in most other applications - as a single cell is sufficient for such a low voltage. This simplicity is its advantage for an easier initial explanation of the meaning of the invention. All other implementation examples have a certain mutual similarity and differ only in the execution of some details.

The basis on which the idea of this fuel cell is built is the selective permeability of the membrane 60. Through the catalyst, hydrogen atoms are broken down into positively charged protons and free electrons. Membrane 60 only allows protons but not electrons. Flat electrodes, anode 50 and cathode 70, are located on both sides of the membrane 60, between which, due to the accumulation of electrons on the anode 50, a usable voltage difference is created, which can be taken at the output terminals 400. Both electrodes are porous, looped and mechanically delicate. Membrane 60 also has a relatively low mechanical resistance. Therefore, both electrodes, the anode 50 and the cathode 70, are supported on the outside by solid plates, a distribution plate 80 on the side of the anode 50 and a second distribution plate 180 on the opposite side, i.e. the cathode side 70. Both these distribution plates 80 and 180 are of solid material , so that the nuts or screw heads holding the whole bundle of all the plates together can reliably rest on them. To feed both fuel and oxidizer and then distribute both gases over the surface of the electrodes 50, 70, the distribution plates 80 and 180 have gas-flowed channels on their inner side - for example, created by removing material. In an imaginary cross-section of this fuel cell in Fig. 1, the first channel 31 is visible on the side of the anode 50 - whereas on the side of the cathode 70, the first opposite channel 91 is visible. Into the first channel 31, an inlet is introduced from the left

-3 CZ 309461 B6 of hydrogen 200. which according to the invention passes through the first oscillator 1000 in which the flow oscillates. On the opposite, lower side, an air supply 300 is introduced into the first opposite channel 91, which similarly passes through the second oscillator 1001. On the right side of the cell, in Fig. 1, there are outputs of the products 600 of ongoing reactions.

Fig. 2 shows the first distribution board 80 according to the invention, rectangular in shape, together with a combination of possible arrangements of self-excited oscillators. This picture, Fig. 2, has the character of a general view of this first distribution board 80 and, in particular, of a view of the cavities through which the supplied gaseous fuel flows from the supply opening 20 to the branch. An enlarged detail of this inlet part is shown in Fig. 4. From this point of flow bifurcation, two parallel channels are created for the flow of gaseous fuel, the first channel 31 and the second channel 32. In this case, both of these channels are arranged in the form of repeating meanders. The purpose of this meandering arrangement is to cover as much of the surface area of the anode 50 as possible. The first channel 31 leads from the meanders to the first outlet opening 41, while the end of the second channel 32 is similarly introduced into the second outlet opening 42.

Hydrogen is supplied to the supply opening 20 through the hydrogen supply 200 in a direction perpendicular to the plane of the first distribution board 80. Behind this supply opening 20, the hydrogen flow path leads between two side walls, the first inlet wall 21 and the second inlet wall 22. These walls (in particular according to Fig. 4) they bring each other closer to each other in the flow direction. Behind the place of greatest flow narrowing, the cross-section of the channel widens again between the side walls located opposite each other, the first retaining wall 27 and the second retaining wall 28. The arrangement of these two walls, the first retaining wall 27 and the second retaining wall 28, shows a section by the plane E - E marked in Fig. 4. An image of this planar section is in Fig. 7. It can be seen there that the first retaining wall 27 is formed on the inner surface of the first protrusion 25 and symmetrically to this, the second retaining wall 28 is formed on the inner surface of the second protrusion 26. The first the protrusion 25 is, according to Fig. 6, located inside the first recess 23, which is in Fig. 5. Symmetrically to this, the second protrusion 26 is located inside the second recess 24. The hatched area drawn in Fig. 5 is located higher perpendicular to the drawing than the surface selections that are drawn as empty in the image. On the contrary, again in Fig. 6, the hatched area of the surface is lower against the drawing line than the area drawn as empty.

It should be noted in Fig. 4 that there is a narrow gap between the first inlet wall 21 and the first retaining wall 27 to allow the flow of fluid. Behind the upper end of the largest expansion of the flowed cross-section between the first retaining wall 27 and the second retaining wall 28, given by the mutual opening of these retaining walls 27 and 28, a separator 29 stands in the path of direct flow. On both sides of the separator 29, there are entrances to a pair of channels, the first channel 31 and of the second channel 32, which then pass from there through their meandering sections. On the basis of the experiments carried out, it is found that the separator 29 has a groove-like recess on its inlet face. This facilitates tilting of the fluid stream 121 and bypassing the separator 29 into the channels 31 and 32.

On both sides of the path for hydrogen in the first distribution board 80, there are retaining walls on both sides. It is specifically the first retaining wall 27 and the second retaining wall 28. Throughout their extent, both the first recess 23 and the second recess 24 have a constant depth, as can be discerned in the section along the plane E - E shown in Fig. 4. When shaping the first recess 23 when manufacturing the first distribution board 80, a first protrusion 25 is left inside the first recess 23 so that a gap of approximately constant width is left between the outer wall of the first recess 23 and the outer wall of the first protrusion 25. Also, when forming the second recess 24 on the opposite side, a second protrusion 26 is left such that a gap is also left between the outer wall of the second recess 24 and the outer wall of the second protrusion 26, which also has a roughly constant width. The only significant reduction in this width is at the inlet opening 20 at the end of the first inlet wall 21 and similarly also at the end of the second inlet wall 22.

The function of the described first embodiment is shown in Fig. 4 by means of drawn arrows as found in experimental investigation. The hydrogen supplied through the inlet 20 flows through the constriction between the first inlet wall 21 and the second inlet wall 22 as shown by the arrow between them. By this narrowing, its speed is increased and with this considerably high speed

-4CZ 309461 B6 comes as a fluid stream 121 into the space between the first retaining wall 27 and the second retaining wall 28. The mutual expansion angle of these two retaining walls 27 and 28 is so large that the fluid stream 121 cannot be held simultaneously by both retaining walls 27 and 28. In the situation indicated by the arrows in Fig. 4, the fluid stream 121 has adhered to the first retaining wall 27. This adhesion is primarily dependent on the pressure conditions given by the surrounding fluid being entrained into the fluid stream 121. Since the liquid stream 121 is very close to the first retaining wall 27 due to its inclination, the gas carried away by entrainment in the stream cannot be replaced by another gas from the surroundings. In this narrow space between the stream 121 and the first retaining wall 27 there is thus a negative pressure. On the opposite side, between the fluid stream 121 and the second retaining wall 28, there is not such a large negative pressure, because gas can come there as external suction 124 from the second channel 32. It is precisely this pressure difference between the two sides of the fluid stream 121 that holds the fluid stream 121 in its deflected position. By this inclination, the gas is led from the end of the first retaining wall 27 as a channel flow 123 into the first channel 31. There, however, the gas must flow into the left half of the meander, which causes a certain resistance to the flow due to the friction of the gas against the walls. Part of the supplied fluid stream 121 is therefore diverted at the end of the first retaining wall 27 and passes as a return flow 127 through the gap between the outer wall of the first recess 23 and the outer wall of the first protrusion 25. Two circumstances lead to this. It is, on the one hand, a certain inability to overcome the frictional resistance of the first channel 31. The second circumstance is again that the fluid stream 121 sucks fluid from its surroundings and induces a negative pressure in its surroundings, which is a phenomenon generally known from jet ejectors. It is applied as an internal suction 128 and a negative pressure in the gap between the first recess 23 and the projection 25. In the second channel 32, the already mentioned suction effect causes an external suction 124, which causes a return flow into the space between the first retaining wall 21 and the second retaining wall 22, from where the fluid is torn into the fluid stream 121. In the meander of the second channel 32, a reverse flow is thus induced, whereas in the first channel 31, the flow directed to the first outlet opening 41 prevails. Only a smaller part is diverted as a return flow 127, which thus in the gap between the outer wall of the first recess 23 and the outer wall of the first protrusion 25 causes an energetic pulsating flow impinging on the fluid stream 121 from the side.

This energetic flow coming from the side causes the negative pressure, which the fluid stream 121 has so far held at the first retaining wall 27, to be canceled. Furthermore, the flow momentum of the discharge impinging on the side from the side is applied to the fluid stream 121. Both cause the fluid stream 121 to be guided so far into the first channel 31 flips to the other side. It begins to be held at the second retaining wall 28. From there it is then led into the second channel 32. Because the fluid stream 121 is very close to the second retaining wall 28 due to its overturning, the entrained gas cannot now be replaced there by gas from the surroundings. In this narrow space between the overturned fluid stream 121 and the second retaining wall 28, a negative pressure will now arise. Due to the geometric symmetry of the internal shape of the cavities, the tasks are now reversed. In the meander of the second channel 32, the direction of flow is thus reversed and a forward flow is induced into the second outlet opening 41, whereas in the first channel 31, reverse flow prevails. It is now possible to take the individual details of the above-mentioned flow step by step as they reverse after the fluid flow 121 is reversed. However, the result of the changed conditions in this way will be to produce a force effect on the overturned fluid stream 121, which in turn will return it to the original initial state, as drawn in Fig. 4. This is a state with outflow into the first channel 31 and reverse flow in the second channel 32.

Alternating flow directions in this way have two desirable consequences. The first of them is the intense force effect of the pulsations on any randomly present water droplets, which, in the case of fuel cells, reach in particular by back diffusion into the cavities in the first distribution board 80 and cause a loss of efficiency of the fuel cell there. The flow pulse of oscillating hydrogen drags a drop of water with it and carries it to the respective outlet opening 41, 42. This restores the efficiency of electricity generation in the given cell. The second beneficial consequence of induced changes in flow direction is intensive diffusion into the anode 50. This increase in transport is a known effect, which can otherwise be achieved - but much more demanding - by mechanical mixing or similar agitation of the liquid in the cavities. The mechanical design of the oscillator is possible, but from an economic point of view, it is a disadvantageous alternative for the fuel cell compared to the self-excited oscillations according to the invention.

-5CZ 309461 B6

Example 2

In this second example, which relates to the illustration in Fig. 3, 8 and 9, slightly modified but basically similar distribution boards according to the invention are part of a bundle of boards forming a battery of fuel cells. The imaginary section of the bundle of distribution boards is guided according to the broken line A - B C - D indicated in Fig. 2 and thus the image of the cross section in Fig. 3 is obtained. . In most practical applications, higher electrical voltages are required. These are most easily achieved by lining up several fuel cells one after the other in a series. In Fig. 3, there is a battery of cells in which there are three fuel cells in series, physically connected together in a battery forming one bundle of plates.

In addition to the above-described plates of a single fuel cell such as the membrane 60 and the electrodes 50, 70, there are two other types of plates in the bundle in this case. First of all, there are the end plates, namely, in Fig. 3, the first end plate 10 at the top and the second end plate 100 at the bottom. the necessary electrical insulation 155 to prevent mutual short-circuiting both through the bolts 54 and through the tubular components in which the supply holes 20 are. In the end plates 10. 100, outlets and fluid supplies are also attached.

The actual functional boards in the battery are basically the same as the example in Fig. 1. However, a significant difference can be seen in Fig. 3 below the cathode 70, as there is a very different bipolar terminal board 190. Unlike other distribution boards 80 has a bipolar end plate 190 in itself two systems of formed cavities. One is on the upper side of this bipolar end plate 190, and the first opposite channel 91 can be seen from it in the cross-section in Fig. 3. The second system of formed cavities is then on the lower side of the bipolar distribution plate 190. It is visible from it in the cross-section in Fig. 3 first channel 31. The advantage of this arrangement in this long example of implementation, i.e. with bipolar plates, is the greater compactness of the resulting battery of cells and the smaller volume that this battery then occupies. A comparison of the two embodiments of the invention is made possible by the partial cross-sections in Fig. 8 and Fig. 9, where in Fig. 9 there is a bundle of four fuel cells clamped in the end plates 10, 100. There are only one-sided distribution plates 80. In contrast, in Fig. 8 are bipolar distribution boards 190 and at both ends of the bundle are combination boards that take on the role of both end boards 10. 100 and distribution boards 8CX They are in Fig. 8 the top combination board 800 and the bottom combination board 801.

Example 3

This example of implementation concerns both the depiction of the distribution board 80 in Fig. 10 and, on the other hand, a detail of the cavities at the gas inlet to the channels 31, 32 of the distribution board 80 in which oscillations are generated in Fig. 11. The flow inside the cavity is explained there by means of drawn long curved arrows of the instantaneous flow direction. It is an example of an embodiment arranged in such a way that it allows to achieve a very high frequency of self-excited oscillations. This leads to a very intensive mixing of the flowing gas and thus to an increase in the efficiency of the fuel cell with distribution plates 80 according to the invention. The basic principle of generating oscillations is quite close to the principle explained in the first embodiment and differs only in the details enabling high frequency.

The frequency of the oscillations is basically determined by the length of the path that the flowing gas must travel before a force in the opposite direction begins to act on it in the long half of the oscillation cycle. Therefore, if the desired high frequency is to occur, it is necessary to shorten the paths in the corresponding design. There are two of these cuts in Fig. 10 and 11. The first of them is a different coverage of the surface area of the electrodes 50. 70 channels for gas flow. This coverage, as in the first embodiment described above, is also divided into two symmetrical halves, namely the first channel 31 on the left in Fig. 10 and the second channel 32, which is on the right in this figure. A short path and good coverage at the same time are achieved by dividing the flow into a large number of parallel channels, namely thirteen first parallel channels 35 on the left side and identical thirteen second parallel channels 36 on the opposite right side. Both are very short compared to the meanders in Fig. 2. At the exit

-6CZ 309461 B6 of these systems of parallel channels 35. 36 is on the left the first collecting branch 33 leading to the first outlet opening 41. On the right is the second collecting branch 34 leading to the second outlet opening 42. The second of the two aforementioned shortenings of the path is indicated in the lower part Fig. 10 and at the same time enlarged in Fig. 11. Instead of the rather long gas-flowed gap between the recesses 23, 24 and the protrusions 25, 26 in Fig. 4, there are now only short recesses 23, 24 with inserted protrusions 25, 26 in these figures a gap through which the gas can quickly pass in a very short time to the inlet wall 21, 22 as a return flow 127 with internal suction 128.

Example 4

This example of the cavity of the distribution board 80 is shown in Fig. 12, and the detail of its cavities in which the desired flow oscillations are induced is drawn in an enlarged form in the next Fig. 13. For the flow of gas in the shown cavities of the distribution board 80, two channels are also used here, the first channel 31 and second channel 32 with a common inlet opening 20 but with different outlet openings - the first outlet opening 41 for the first channel 31 and the second outlet opening 42 for the second channel 32. Both channels 31, 32 are in turn arranged in meanders to cover as much as possible the surface area of the electrodes 50, 70 with which the cavities of the distribution boards are covered. For the generation of oscillations in the flowing gas, that part of the cavities is used, beyond which the initially common flow branches into the first channel 31 and the second channel 32 in the flow direction. This branching is in the lower part of Fig. 10. From the inlet opening 20, the flow cavity is directed of gas in the direction of Fig. 10 upwards between the first inlet wall 21 and the second inlet wall 22. These walls lean towards each other in the flow direction. It causes the velocity of the gas here to increase and its pressure to decrease. There is a narrow gap above the upper end of both inlet walls 21 and 22, and above it are the beginnings of another pair of walls, the first retaining wall 27 and the second retaining wall 28. Both of these retaining walls 27 and 28 move away from each other in the flow direction ._The opening angle that they hold together is so large that this pair of walls does not function as a diffuser. The flowing gas breaks away from one of the two retaining walls and is held only by the opposite remaining wall. Behind the upper end of both retaining walls 27 and 28, in the path of the vertical flow, there is a separator 29, also here with a groove-like recess on its front. A substantial difference from the above-described arrangement of recesses and protrusions in this fourth embodiment is that there is a double-sided recess 39 reaching to the left and right of the supply opening 20 and inside it there is a continuous projection 30. The supply opening 20 is made here in this continuous projection 30 When making a double-sided recess, the procedure is such that a gap is created between its outer concave vertical wall and the convex outer vertical wall of the continuous protrusion 30, for example as in Fig. 10, a gap with the same width everywhere - with the exception of the narrowing of the width at the entrance walls, i.e. at the first entrance wall 21 and the second entrance wall 22.

The long and generally curved arrows in Fig. 13 indicate the nature of the flow precisely in the situation where the gas supplied to the supply opening 20 cannot flow through the adjacent parts of the cavities generating oscillations in a straight upward direction because there is a separator 29 in its path. It must therefore deviate from the straight direction and in the situation shown in Fig. 13, it passes into the first channel 31. Inside the cavities, this fluid stream 121, after exiting from the space between the first and second inlet walls 21 and 22, is inclined to the left in Fig. 13, where it flows around the separator 29 on the left side. the meanders of the first channel 31 flows into the first outlet opening 41. From the second channel 32, the suction effect of this fluid stream 121, on the other hand, leads to a backward-directed external suction 124. Thus, the fluid is sucked into the fluid stream 121 and is led away with it together with this main flow into the first channel 31 At the first retaining wall 27, the influence of the suction effect is also manifested in the fluid stream 121 as a local negative pressure. The gas in the form of an overturning flow 129 is sucked in through a narrow gap between the first inlet wall 21 and the first retaining wall 27. In contrast, in the gap on the opposite side, at the beginning of the second retaining wall 28, the pressure is greater. This creates a return flow 127 passing from this higher pressure to the overturning flow 129. It passes through the gap between the edge of the continuous projection 30 and the inner wall of the double-sided recess 39. This flow in the gap does not have a constant size, but its intensity gradually increases with time. It reaches a very high kinetic energy in the narrow gap between the first entrance wall 21 and the first retaining wall 27. There it encounters the fluid stream 121, which is very sensitive to the action of transverse forces in these places. Under this effect, the fluid stream 121 first straightens, but in a straightened manner

-7 CZ 309461 B6 direction cannot be maintained - already because it hits the recessed groove on the face of the separator 29. Its tipping continues until it ends up adhering to the second retaining wall 28. However, this completely changes the asymmetric conditions in the otherwise geometrically symmetrical cavities. This manifests itself in the fact that there is suction into the first channel 31 from the first outlet opening 41. On the contrary, the fluid stream 121 increased by this suction then flows out through the second channel 32. However, the pressure ratios across the fluid stream 121 are also reversed, where the pressure is higher on the right and lower pressure on the left. This, of course, also reverses the flow through the gap between the edge of the continuous protrusion 30 and the inner wall of the double-sided recess 39. The intensity of this reversed flow gradually increases and causes the fluid stream 121 to flip back to the original position already described above. In this way, self-excited oscillations will occur, during which the flow directions in the channels 31, 32 alternate with both of the above-mentioned desirable phenomena, i.e. an intense force action on the adhered water droplets and their removal - and an improvement in the efficiency of the cell by increasing the intensity of diffusion transport.

Example 5

This other example of the design of the distribution plate with oscillating gas mixing in its cavities is shown in the following figures 14, 15, 16 and 17. There is, first of all, in figure 14 a general view of the end plate of the fuel cell, where it is particularly interesting in the lower parts of the image integrally forming the channel made generator of periodic oscillations. The next pictures show sub-pictures of the flow.

The majority of the cavities on the surface of the distribution board in Fig. 14, roughly in the upper two-thirds of its surface, have exactly the same system of cavities created by excavating as the previous embodiment in Fig. 10. The difference - and that is relatively small - is in the lower third, where a system of specially shaped input cavities acting as a self-excited oscillation generator. These are also oscillations with a relatively high frequency. The gas is supplied through the supply opening 20, from where it emerges in the vertical direction in Fig. 14 as a fluid stream 121. On both sides of this path of the fluid stream 121, there are arc-shaped recesses 23, 24. Unlike Fig. 10, when making the recesses 23, 24, they are not left here protrusions 25, 26 with retaining walls 27, 28. However, the observed flows showed that the events in the oscillation generator are practically the same as if the protrusions 25, 26 were there. The observed flows are indicated by the long curved arrows in Figs. 15, 16 and 17. At the far left of Fig. 15 is the situation where the gas (hydrogen or air) supplied through the inlet 20 by flow through the narrowed cross-section creates a fluid stream 121. This cannot flow in with its initial direction oriented straight up in Fig. 15, because there is a divider 29 against it. It must therefore lean to pass into one of the channels, for example the first channel 31 as it is in Fig. 15. The entrance to the channel is at sharp-edged on both sides. Sharp edges from the fluid stream 121 cut off the marginal portions and return them. A smaller part is thus separated to the right and, relative to the groove on the face of the separator 29, creates a small standing recirculation vortex 125. In contrast, the part cut off to the left from the fluid stream 121 forms an intense return flow 127 flowing along the inner side of the arc-shaped first recess 23. Suction into the fluid stream 121 causes negative pressure in the cavity. This leads the second channel 32 to temporarily reverse the flow direction as the external suction 124. This only lasts until the return flow 127 arrives along the inner side of the arcuate first recess 23 to the constricted point. There, the fluid stream 121 is very sensitive to external stimuli that can change its flow direction. As shown in Fig. 16, the return flow 127 diverts the fluid stream 121 to the right and this leads it to the entrance of the second channel 32. This finally leads to the situation indicated in Fig. 17, which is actually a mirror image of the phase indicated in Fig. 15. The path for the return flow 127 is then defined on the outside by the wall of the second recess 24 and on the inside by the standing vortex movement 225. The sharp edges of the inflow into the second channel 32 in turn cut off the marginal parts from the fluid stream 121 and the second phase of the oscillation cycle occurs.

Example 6

This example represents a more complete demonstration of the arrangement of the plates - in particular, of course, the distribution plates 80 - in a small fuel cell. The example is drawn in a total of four images, Fig. 18, 19, 20 and 21. At the same time, the first of them, Fig. 18, consists of two partial views with an explanatory detail

-8CZ 309461 B6 of this arrangement. In the upper of the two partial illustrations, Fig. 18 shows the contours of cavities created in the first distribution board 80 by removing material. In these cavities, self-excited oscillations are generated especially at their beginning, at the supply opening 20. In the subsequent channel branched into two parts, the first channel 31 and the second channel 32, electrical energy is then generated. The contours of the cavities in Fig. 18 here fill the overall surface structure, which can be described as a square placed diagonally on one of its edges. Cavities in which self-excited oscillations of the hydrogen flow are generated by aerodynamic instability are in the lower part of this imaginary circumscribed square. This lower part of the picture shows a planar imaginary section marked as G - G. The area of this section is shown in Fig. 18 by the lower of the two sub-pictures. The central part of the first distribution board 80 is visible there. The cavities in which the action takes place are created by one-sided recessing to the same depth everywhere. The cavities are closed during the assembly of the fuel cell by sealing them against leakage from the side on which these cavities are open (because that is where the plate material was taken - in section G - G. in Fig. 18, it is the upper side of the first distribution plate 80) of hydrogen by adding additional plates, namely anode 50, membrane 60 and cathode 70. To which a second end plate 100 is attached from the outside. All these five plates have six circular holes in them. All of them also have the above-mentioned external shape, drawn in the following Fig. 19. This shape can again be described as a square with rounded corners placed diagonally on one of its corners (here, of course, the corners are rounded). The outer contours of the end plates 10. 100 have the same external square shape if they are used to provide the necessary mechanical rigidity to the entire battery of cells. An example of the corresponding end plate 10 is in Fig. 19. It has four holes for screws 55 in the rounded corners of the square and the basic trio of inlets with outlets, namely the inlet opening 20 and the corresponding first outlet opening 41 - and opposite it the second outlet opening 42. Fig. 19 is yet another such trio used for air, if hydrogen is distributed in the first distribution board and vice versa. If a low frequency of generated self-excited oscillations is required, a cavity formation with as long as possible a propagation path of backward flowing pressure pulses is required. Figures 20 and 21 show how it is possible to increase the length of the return flow loops in the square design of Figure 18.

Example 7

In this last description of the arrangement of the distribution board with an oscillating inlet flow of hydrogen and air, the findings concerning the mutual relationship of the channels 31, 32 in the adjacent planes of the bundle of plates are summarized. In order for the desired catalytic decomposition into protons and electrons to occur, it is necessary that the starting gases of this reaction flowing in the channels 31, 32 of the parallel plates can be brought into close proximity to each other. A commonly known solution is shown in an enlarged local detail in Fig. 22. It consists in the fact that the directions of the first channel 31 and the first opposing channel 91 cross in mutually perpendicular directions when viewed in the direction of the arrow 160. It means that there are small places in the system of plates where hydrogen and air are separated from each other only by the membrane 60 and the electrodes 50, 70. The small size of these places is due to the condition of small mechanical stress on the membrane 60 and the electrodes 50, 70.

The arrangement then corresponds to the example in Fig. 23. On the left is an example from Fig. 12 described above, through which the simple first opposite channel 91 leading from the opposite supply opening 90 to the first opposite outlet 93 is drawn on the left side of the picture. This is again a view in the direction of arrow 160. when otherwise mutually invisible (separately located in different planes parallel to the drawing plane of the image) appear simultaneously. Since the requirement is economy of working with hydrogen - while air is available in any quantity - the economy of the improved generation of oscillations in the distribution board in Fig. 23 is carried out only in the hydrogen board.

Two other examples of small fuel cell implementations are in the last pictures, Figs. 24 and 25. There are two views in the direction of arrow 160 of the overlapping channels 31 and 32 of the square distribution plate of Fig. 18. The advantage of the square configuration is the two axes of symmetry of the square , differing by an angle a. Under this angle, the flow directions in the channels 31, 32 can be rotated to each other for hydrogen and opposite channels for air, such as the first opposite channel 91 in Fig. 24 when viewed in the direction of arrow 160. In the example in Fig. 24 it is a mutual rotation by an angle a = 90° in the usual arrangement and by an angle a = 180° with opposite flow directions. In this second example, a greater use of the distribution board area is achieved, but the membrane is slightly larger

-9CZ 309461 B6 mechanically stressed by the pressure difference. The connecting channel 83 leading to the common outlet 40 and the opposite connecting channel 84 leading to the opposite connecting outlet 47 are indicated in Fig. 24 only symbolically. In fact, common outlets can be made through short vertical openings such as those in Fig. 3.

Claims (4)

PATENT CLAIMS 1. Distribution board of a fuel cell with cavities created for fluid flow, one end of which is connected to the inlet opening (20) and open from one side to the surface of the adjacent electrode (50, 70), characterized by the fact that the cavities have a bifurcated shape branches, which are the first channel (31) and the second channel (32), where each of these two branches leads to one of the two outlet openings (41, 42), and the place of this bifurcation into two branches is the narrowing of the cross-section of the cavity formed by mutual approach against between the inlet walls (21, 22) of the distribution board, the first inlet wall (21) located on one side of the cavity and the second inlet wall (22) on the other opposite side, and behind this point of narrowing in the flow direction, there is a first by the retaining wall (27) and the second retaining wall (28), on the other hand, the extension of the cross-section, and at the end of this extension, between the entrances to the two branches, there is a separator (29). 2. The plate according to claim 1, characterized in that between the ends of the two retaining walls (27, 28) and the beginnings of the two entrance walls (21, 22) there is one of two arc-shaped recesses (23, 24) on each side. 3. A plate according to claim 2, characterized in that inside the recess (23, 24) there is a projection (25, 26) on each side, between the outer wall of which and the inner wall of the recess (23, 24) there is a continuous gap. 4. A plate according to claim 3, characterized in that both projections (25, 26) on opposite sides of the plate wall are connected to one another into one continuous projection (30), while the inlet opening (20) is formed in this continuous projection (30 ).