A method for bonding fuel cell components
By using electromagnetic radiation cured aqueous cationic epoxy resin adhesives in the visible or ultraviolet range, poor adhesion and material waste in the bonding of PEM fuel cell components and frames are solved, and high-strength bonding and material utilization are achieved. Improved, suitable for industrial production.
Patent Information
- Application Number
- CN202080067797.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-09-30
- Filing Date
- 2020-06-08
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2040-06-08
AI Technical Summary
In the prior art, the combination of PEM fuel cell components and frames has problems such as poor adhesive adhesion, high positioning accuracy, waste of materials and complex processes, resulting in high leakage rate of fuel cell and disproportionately using materials.
The aqueous cationic epoxy resin adhesive cured by electromagnetic radiation in the visible or ultraviolet range is activated by electromagnetic radiation and heated by electromagnetic radiation to reduce its viscosity to achieve good permeability and high strength bonding of the porous surface.
The full surface coverage of the catalyst coating and gas diffusion layer is achieved, which reduces leakage rates, improves bond strength and material utilization, simplifies the process flow, and is suitable for industrial production.
Smart Images

Figure CN114450828B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method for bonding components of a fuel cell to a frame and / or to each other and to the use of such a method. Background Art
[0002] It is known in the prior art to combine the components of a PEM fuel cell with a frame and / or with each other by means of an ultraviolet-curable adhesive. For example, US2018 / 0159160A1 discloses a frame that is completely transparent to ultraviolet light so that the joint can be cured through the frame. Here, the frame is bonded to a membrane having a catalytic coating, and a gas diffusion layer is also integrated into the composite. The disadvantage of this structure is that, on the one hand, it requires a frame that is mostly transparent to ultraviolet radiation, which is not the case with the commonly used polyethylene naphthalate (PEN) frame. DE 102015117077A1 also describes the combination of components of a PEM fuel cell in a so-called membrane electrode assembly (MEA). In this case, an ultraviolet-curable adhesive is also used, and here and in a similar document JP 2016-201183A1, the adhesive is first activated so that the corresponding components are subsequently connected to each other, and the final curing of the adhesive is not achieved before the components are placed relative to each other, and the adhesive wets the components to which the adhesive was not initially applied accordingly. In this case, the components are bonded to each other, for example the gas diffusion layer is bonded to the membrane, and also to the frame, for example the membrane is bonded to the frame and one of the gas diffusion layers is bonded to the frame.
[0003] A feature of this method is that the UV-curable adhesive is first activated by a small amount of radiation. The curing of the adhesive then begins at this point in time. Activation without curing the adhesive would result in a slow hardening of the adhesive, which would then continue to harden more and more.
[0004] During this process, the adhesive becomes increasingly viscous, with the result that all surfaces to be bonded with the adhesive become difficult to wet.
[0005] The combination of a catalyst-coated membrane (CCM) (catalyst coated membrane) with a frame is also crucial for the sealing of the fuel cell. Typically, when combining a full-surface coated membrane, only the surface of the catalyst is bonded to the surface of the frame. Due to the poor adhesion of the catalyst to the actual membrane or its ionomer, it is now very common to ensure that hydrogen and oxygen mix and therefore cause leakage of defective fuel cells, which may cause the entire fuel cell stack to fail due to the corresponding high leakage rate. Therefore, in practice, it is common and conventional to use the material of the catalyst-coated membrane only in the required area. There is thus no catalyst at the edge of the membrane bonded to the frame. Therefore, they can be well combined with the frame accordingly, so that the material of the membrane is tightly adhered to the frame, and there is no catalyst between them.
[0006] However, this construction has serious disadvantages: The membrane, which is only partially coated with the catalyst, must be positioned very precisely. In addition, the membrane material in the uncoated areas is usually designed to be relatively large in size to ensure a strong bond. The excess material is then cut off and discarded, so that a disproportionate amount of material is used in production.
[0007] Furthermore, coating only part of the membrane is very complicated, requiring ink catalysts and specialized equipment that are compatible with direct membrane coating. This is another drawback.
[0008] It is therefore desirable to use a membrane that is completely coated with catalyst, so that the entire material, for example unrolled on a roll, can also be used for the construction of a fuel cell. Furthermore, the positioning accuracy required for a membrane that is completely coated with catalyst is much lower than in the case of a membrane that is only partially coated with catalyst.
[0009] US 9,831,504 B2 has tried this. In order to minimize the above-mentioned disadvantages of poor adhesion, an adhesive is used which comprises a heat-curable component on the one hand and a UV-curable component on the other hand. As described in the literature on UV-curable components, once the adhesive has been placed in place, it remains there and no longer flows. Subsequently, the frame and the film are joined, wherein the previously positioned, precisely placed adhesive is finally cured by thermal effects, wherein the heat is introduced, for example, by UV irradiation of a frame made of polyethylene naphthalate (PEN), which itself is impermeable or only very slightly permeable to the UV radiation of the usual 365 nm wavelength required for the curing of the adhesive, so that the heat is released mainly by UV radiation.
[0010] However, the inventors of the present invention have not been able to determine in experiments using this method a significantly better binding performance than in the prior art mentioned at the outset.
[0011] Reference may also be made to US 2008 / 0118802 A1 for other prior art. Summary of the invention
[0012] The object of the present invention is to provide an improved method for bonding components of a PEM fuel cell to a frame and / or to each other. This object is achieved by the following method.
[0013] A method for bonding components of a PEM fuel cell to a frame and / or to each other, for which purpose an adhesive that can be cured by electromagnetic radiation in the visible or ultraviolet range is applied to the frame and / or at least one of the components, wherein the adhesive is activated by electromagnetic radiation and the adhesive is heated after the frame and / or components are brought into contact; or the frame and / or components are brought into contact and the adhesive is exposed to electromagnetic radiation for activation and heating; wherein the heating reduces the viscosity of the adhesive before the adhesive is finally cured; characterized in that the adhesive is a water-containing cationic epoxy resin, wherein the water content of the cationic epoxy resin is 100-500 ppm.
[0014] The inventors have found in their experiments that adhesives that can be cured by electromagnetic radiation, such as preferably photocurable or UV-curable adhesives, applied to these components and / or frames are ideally suitable for this bonding. Now, the inventors' decisive understanding is that after the activation of the adhesive, the viscosity of the adhesive is reduced by heat input, that is, the adhesive has a lower viscosity. On the one hand, this allows to offset the initial curing phase, and on the other hand, due to the very low viscosity characteristics of the adhesive thus obtained, a good penetration of the particularly porous or especially microporous surfaces on these components is achieved. Here, due to its greatly reduced viscosity caused by heating, the adhesive that penetrates the porous surface is also absorbed by the smallest pores, thereby allowing complete or at least almost complete penetration of the porous surface. Therefore, due to the contact of the adhesive with the substrate of the component, that is, with the ionomer in the case of the membrane, a much higher bonding strength is achieved in the bonding.
[0015] The method according to the invention provides two different ways to achieve this great advantage. In the first case, the adhesive is activated by light or preferably by UV light before the contact is made between the component and the frame and / or between the components. Thereafter, heat is applied. In the other case, the component and the frame and / or between the components are brought into contact. Thereafter, the adhesive is activated and heated in one exposure by UV light or preferably by light.
[0016] In both cases, aqueous cationic epoxy resins are used to achieve very good penetration of porous or microporous surfaces or layers of the components, which involves corresponding advantages in terms of quality and service life of the respective fuel cell or its membrane electrode assembly.
[0017] As the inventors have discovered, cationic epoxy resins allow activation by ultraviolet light as well as blue visible light. Therefore, it is ideal for both cases of the two methods (pre-activation and one-time exposure) according to the present invention. Adhesives with a water content between 100 ppm and up to 1.5% by weight (but not limited thereto) result in delayed curing of the adhesive. Compared to the faster start of curing of anhydrous cationic epoxy resins, this delayed curing after activation with light or ultraviolet light is ideal for use in the present invention. In addition, a small amount of water allows faster curing after the start of delayed curing compared to anhydrous cationic epoxy resins. This is also advantageous for the method of the present invention because the rapid curing after the start of delayed curing reduces the flow of adhesive into unwanted areas.
[0018] According to one embodiment of the present invention, the water content is about 100 to 500 ppm, preferably 200 to 450 ppm. In addition, the viscosity of the cationic epoxy resin used at 75°C is less than 100 mPas. This is generally further reduced with increasing heat. At a temperature of about 120-170°C, which is a preferred temperature range for bonding the parts together according to another embodiment of the present invention, the viscosity of the cationic epoxy resin can be less than 10 mPas.
[0019] Another advantageous embodiment of the invention is characterized in that the component of the PEM fuel cell has a microporous surface at least on the surface to be bonded to another component and / or the frame. In particular, those components having such a microporous surface layer, such as a catalyst on a CCM or a microporous layer (MPL) on a gas diffusion layer (GDL), benefit from the bonding method of the invention, which allows bonding of the base material below the microporous surface layer at least almost directly, without the need to remove the microporous surface layer or to omit the application of the microporous surface layer in the bonding area.
[0020] In a preferred further improvement of the method of the invention, heat is generated in the component by applying electromagnetic radiation, preferably in the visible or ultraviolet range. By using this heating implemented by electromagnetic radiation, in particular visible light or ultraviolet light, heat can be applied very directly to the exact area where heat is needed, such as a microporous surface layer. Heat is usually generated at the layer with the lowest transparency, which is usually the microporous catalyst coating of the membrane or the MPL of the gas diffusion layer. Generally, these can be approximately equivalent to the properties of a black body that absorbs all light and converts it into heat. Therefore, heat input is generated by radiation from the frame side or the back side of the component, mainly in the area where the adhesive will be liquefied to a very low viscosity, so that heat input can occur via capillary forces passing through the holes of the very strong heating catalyst, which is also heated very quickly due to its low thermal mass.
[0021] Heating is achieved by two different methods or a combination thereof. The first method is to generate heat in the microporous surface layer itself. The microporous surface layer, i.e. the catalyst layer of a CCM or the MPL of a GDL, is usually very dark in color. It acts as a kind of black body in which heat can be generated by exposure to light or UV light. This is ideal since this is the surface or layer that will be penetrated by the adhesive while its viscosity is reduced. In addition, the heating of the adhesive is limited to the radiation area. Therefore, the risk of the adhesive flowing into unwanted areas is reduced. The second method is to generate heat in and / or through the frame. Typically, the frame is transparent. But even if it is transparent, for example made of PEN, it will transmit little or even no UV radiation with a wavelength of, for example, 365nm. However, since this radiation is partly converted into heat in the PEN frame and partly due to the wavelength shift caused, there will still be a corresponding heat input. The wavelength-shifted radiation can then pass through the PEN frame and cause a black body effect in the microporous surface layer. This means that the first and second methods are applied here in a mixed manner. In order to use radiation in the visible range, for example at 430nm, the PEN frame is transparent. Therefore, the first method is applied alone.
[0022] Prior to final curing, the pre-activated adhesive can penetrate into the pores of the microporous surface layer. This structure also has the advantage that the heat is introduced in a very targeted manner and almost only in the region of the catalyst or gas diffusion layer and the adhesive. As a result, the adhesive can be prevented from penetrating into adjacent components, in particular in the subsequent structure of the PEM fuel cell, the electrochemically active regions of the catalyst-coated membrane and / or the gas diffusion layer, respectively, which are arranged adjacent thereto.
[0023] The use of electromagnetic radiation, preferably in the visible or ultraviolet range, to heat the adhesive has further advantages. The use of light or ultraviolet light allows a high rate of temperature rise. Heating of the area to be heated can usually be completed in less than 5 seconds. This allows penetration of the microporous surface layer and a very fast and efficient bonding process. Therefore, the method according to this embodiment is ideal for efficient industrial production.
[0024] According to an advantageous further improvement of the present invention, the adhesive does not contain a heat cross-linking component, and is therefore significantly different from the method in the above-mentioned prior art. Unlike this prior art, the inventor does not intend to keep the adhesive in place. Instead, an activated (activated once via ultraviolet radiation, such as a short pulse of ultraviolet radiation) but still liquid adhesive is applied between the membrane or other parts coated with the catalyst and the frame, and heated. Unlike the prior art, crosslinking is not immediately induced by heating the adhesive. On the contrary, the adhesive becomes much thinner by heating, so its viscosity is reduced. The adhesive is already basically located where it should be at this time. The adhesive will flow and pass through the microporous catalyst surface or layer on the membrane or the MPL of the gas diffusion layer by capillary force, wherein further flow in the plane is limited by rapid solidification. The catalyst layer is porous only for very low viscosity liquids, so very thin adhesives with very low viscosity can pass through these holes by means of capillary force. As a result, experiments studying cross-sectional areas under electron microscopes and corresponding experiments of pulling apart the combined multiple layers have obtained much higher bonding strengths.
[0025] The inventors assume that the adhesive, which has become very thin due to the heating, penetrates the entire thickness of the catalyst layer in the area where the adhesive is applied, so that, despite the presence of the catalytic coating, a bond is ultimately achieved between the material of the frame on the one hand and the ionomer (i.e. the actual membrane material of the membrane coated with the catalyst) on the other hand. This is confirmed by the corresponding cross-sectional images studied by electron microscopy. The same applies to the gas diffusion layer. Due to its porosity, the layer is also deeply penetrated into the material by the adhesive, which has become very low in viscosity, so that this bond achieves an ideal adhesion, for example also to the frame or the membrane coated with the catalyst, whose coating material is also penetrated.
[0026] An advantageous development of the method according to the invention further provides that the components and / or the frame are held together or pressed against each other by a holding-down device and that the electromagnetic radiation is applied through at least one window in the holding-down device which is transparent to radiation. Thus, such a holding-down device or a pressing tool with a suitable, not too high pressure can be used to position the components and / or the frame relative to each other. In this way, a window which is transparent to radiation can be provided at the corresponding location so that the radiation can pass through the window, for example to the component, in particular to the gas diffusion layer, and thus the required heat can be input before the radiation-activated adhesive finally hardens.
[0027] As described above, the heat input can occur by radiation, which ultimately ensures that the component itself or the catalyst coating of the component is heated in a very point-like and targeted manner. However, in a very advantageous embodiment of the method, at least one area of the clamping device adjacent to the transparent window of the clamping device can be limited to be cooled, in particular actively cooled. In this way, the heat input can be limited to the part corresponding to the transparent window in a very targeted manner, so that the membrane and / or gas diffusion layer and the frame as the components to be bonded are not subjected to unnecessary thermal stress outside the gluing point. The cooling or active cooling also helps the heat to dissipate in the edge area and not return to the bonding area, wherein the cooling can be achieved, for example, passively by cooling fins, or actively by cooling fluid, which, for example, moves along the cooling fins or passes through cooling ducts in these areas. In this way, the cycle time for forming the bond can be reduced in the end, thereby achieving economic advantages.
[0028] According to an improved embodiment of the invention, in the case of bonding the frame to a component (e.g. a membrane coated with a catalyst), electromagnetic radiation at least for heating is applied from the frame side through the frame. This allows very easy handling and enables the method to be used in an ideal manner for industrial production lines without the need for complex turning over of the unsprayed bonding material or a second radiation source.
[0029] In the embodiment of the invention described as a single exposure method, a further improvement defines that in the case of contacting the frame and / or the component and heating the adhesive by electromagnetic radiation, the activation and heating are accomplished by electromagnetic radiation in the same single exposure of the adhesive to electromagnetic radiation. This can preferably be accomplished using visible blue light in the range of 430 nm, but is not limited thereto.
[0030] As already indicated, a very advantageous application of the method according to the invention is defined by the bonding of full-surface coated membranes to a frame. By using the method according to the invention, a membrane completely coated with catalyst as one of the components can be firmly and reliably bonded to a frame, which is otherwise only possible with membranes without a catalyst coating in the bonding area, which would however lead to the above-described disadvantages.
[0031] An alternative but equally suitable application of the method according to the invention is to implement the method in order to bond at least one gas diffusion layer as one component to a catalyst-coated membrane and / or a frame as another component. The gas diffusion layer (for an MEA, usually two such gas diffusion layers are included, one on the anode side and one on the cathode side) is microporous at least on the side facing the membrane in order to distribute the arriving gas as evenly as possible to the usable area of the membrane. The method described is therefore particularly suitable for bonding such a gas diffusion layer, in particular directly to the respective side of the membrane and, if required, to the frame.
[0032] A particularly advantageous combination of this application is to bond at least one, but preferably two gas diffusion layers to a pre-bonded and cured composite of a catalyst full-surface coated membrane and a frame. This particularly advantageous application of the method according to the invention defines that in a first method step, the catalyst-coated membrane (CCM) is correspondingly bonded to a frame surrounding the membrane. For this purpose, an adhesive can preferably be applied to the edge region of the catalyst-coated membrane and the adhesive is activated accordingly. Then, a frame is placed on it and irradiation is continued through the frame. Depending on the wavelength used, the radiation heats the PEN membrane and thus the adhesive, and / or directly heats the catalyst layer and thus the adhesive with the above-mentioned effect, that is, the adhesive becomes very low in viscosity and completely penetrates the catalytic coating, so that the catalytic coating of the membrane can be omitted around the edge region. Subsequently, this composite of the catalyst full-surface coated membrane and the frame is correspondingly cured and can be obtained as a so-called semi-finished product. The composite is bonded using one, preferably two gas diffusion layers. One diffusion layer, for example the diffusion layer on the anode side, is located on the catalyst-coated membrane in a frame and is provided with an adhesive in the edge region, which stays between the gas diffusion layer and the membrane, while the other gas diffusion layer is superimposed on the entire membrane and preferably protrudes minimally beyond the membrane. In the edge region, it is also coated with an adhesive. In this case, the adhesive is activated again by radiation. Then, these parts are correspondingly joined and held in position relative to each other using a pressing device or a pressing device, and radiation is preferably irradiated through the transparent window of the pressing device in order to heat the adhesive accordingly and achieve the desired bonding of the composite to the two gas diffusion layers, so that as a whole, an inherently tight and stably bonded structure is generated for the membrane electrode assembly. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Further advantageous embodiments result from this exemplary embodiment, which is described in more detail below with reference to the drawings.
[0034] Shown are:
[0035] Figure 1 A first step according to a first method of the present invention;
[0036] Figure 2 A graph of the UV radiation energy from a UV source as a function of wavelength;
[0037] Figure 3 A second step according to the first method of the present invention;
[0038] Figure 4 The third step of the first method according to the present invention;
[0039] Figure 5 The fourth step of the first method according to the present invention;
[0040] Figure 6 according to Figure 2 Energy diagram of the frame penetrating radiation of the radiation source;
[0041] Figure 7 Schematic representation of an electron micrograph of a cross section through a structure having a bond according to the prior art;
[0042] Figure 8 Schematic representation of an electron micrograph of a cross section through a structure having a bond according to the invention;
[0043] Fig. 9 a step according to a second method of the invention;
[0044] Fig.10 Schematic diagram of applying UV radiation and activating the adhesive on two gas diffusion layers;
[0045] Fig.11 Schematic diagram of components provided with activated adhesive prior to bonding;
[0046] Fig.12 Similar to Fig.11 , with additional hold-down means for positioning the components during the joining process;
[0047] Fig.13 Schematic diagram of the combined structure of the gas diffusion layer, the catalyst-coated membrane and the frame; and
[0048] Fig.14 Schematic diagram illustrating the effect of water content on cure of cationic epoxy resins. DETAILED DESCRIPTION
[0049] The present invention uses a membrane coated with a catalyst on the entire surface to produce a composite of a frame and a membrane (CCM) coated with a catalyst for a PEM fuel cell, wherein no portion is provided for bonding, in which the membrane is not coated with a catalyst or the catalyst catalytic coating has been removed. The membrane coated with the catalyst is bonded to the frame to generate a composite. In addition, the composite is subsequently bonded to two gas diffusion layers to form the core of a membrane electrode assembly (MEA). However, the method can also be used for other bonding tasks, or is only a part of the described bonding.
[0050] exist Figure 1 , a part of the frame is shown and is denoted by reference numeral 1. An adhesive 2 which can be activated by UV radiation is attached to the frame 1, for example by screen printing or inkjet printing in the desired areas. The applied adhesive 2 is then briefly irradiated with UV radiation via a UV source 3 in order to activate it. The adhesive 2 remains liquid in this case, so that the activation energy and / or the length introduced via the UV source 3 must be adjusted accordingly. Figure 2 In the diagram, the spectral irradiance E is expressed in mW / cm 2 / nm is plotted on the y-axis of the graph, while the wavelength is plotted in nm on the x-axis. Under normal conditions, Figure 1 In the illustrated structure, the UV source 3 emits UV radiation directly onto the adhesive 2. Its peak is at 365 nm. As already mentioned, the UV curing adhesive 2 without heat-curing components is therefore only activated, but remains liquid in the area where it is applied. Instead of UV radiation, for example, in the case of the above-mentioned wavelength and the above-mentioned parameters, it is also possible to irradiate at a wavelength of, for example, about 430 nm, which means in the visible light range. Everything said and described below can be used similarly for radiation of this wavelength.
[0051] exist Figure 3 In the illustration of , it can be seen how the catalyst-coated membrane 4 is now placed in the exact position on the frame 1 and the adhesive 2. Figure 4 In the illustration of FIG. 1 , it can be seen that the previous structure of the frame 1, the adhesive 2 and the catalyst-coated membrane 4 is reversed. The frame 1 is made of a transparent material, preferably polyethylene naphthalate (PEN). PEN has the property that ultraviolet radiation cannot penetrate this material or only penetrates it to a small extent, so when using Figure 5 In the case where the UV source 3 is shown to re-irradiate the structure, only a very small amount of UV light is introduced into the adhesive 2, which is not the main reason for curing the adhesive because the adhesive has been Figure 1 The process shown is active. However, the UV radiation is converted into heat in the PEN frame 1, which causes the adhesive 2 in direct contact therewith to heat up.
[0052] Of course, the process described can also be implemented in a different way than described and illustrated here. For example, the adhesive 2 can also be applied directly to the membrane 4 coated with the catalyst on the entire surface. Accordingly, the adhesive 2 can be activated from above by means of UV radiation by means of a UV source 3. Subsequently, the frame 1 can be positioned on the applied activated adhesive 3 without having to turn over or move the structure. In this case, the catalytic coating of the membrane 4 can be heated by means of UV radiation through the frame 1 via the same UV source 3. This structure is also simpler in terms of operation.
[0053] Figure 6 The diagram is similar to Figure 2 Graph showing how much energy and how much wavelength penetrates a frame 1 made of PEN. It is noteworthy that almost no UV radiation occurs below 375 nm. The peak of the UV radiation is at about 385 nm, which is higher than the peak of the wavelength according to Figure 2 The situation is much lower for radiation. In the region above 400 nm, part of the light also appears as visible light in the violet and blue region. This means that for a wavelength of 365 nm, PEN causes a wavelength shift, whereby a small part of the UV radiation and visible light penetrates the frame 1 made of PEN, while most of the UV energy is converted into heat to heat the transparent adhesive 2. For a wavelength of 430 nm, most of the UV radiation passes through the frame 1 and also through the essentially transparent adhesive 2 to the catalytic coating of the film 4. There, it encounters the catalytic coating, which is relatively dark and ultimately acts with the properties of a black body. It therefore absorbs the light energy and converts most of it into heat. Due to the relatively low thermal mass, there is a relatively strong heating of the adhesive. The adhesive becomes thinner and thus obtains a very low viscosity, since its viscosity decreases with increasing heat. For the very thin adhesive 2, which has not yet been cured at this point, due to the Figure 1 The curing by UV radiation in the adhesive 2 has been activated but not yet completed, so that the adhesive 2 can very easily enter the pores of the catalytic layer on the membrane 4 due to the thinning caused by heating. Therefore, it basically penetrates the porous catalytic layer and directly bonds with the ionomer of the catalyst-coated membrane 4, thus producing a very deep and stable connection, although the catalytic coating 5, 8 is not removed in the bonding area.
[0054] In a series of successful experiments, the inventors used LOCTITE EA 3060LC as an adhesive. This is a UV-curable cationic epoxy resin. Its water content specification is 300+ / -150PPM. The actual data collected by the inventors for more than 200 units confirmed that the water content ranged from 200-450PPM, with an average of 369PPM. In the clean room where they conducted the experiments, the relative humidity was 42+ / -2%. The temperature was controlled at 21+ / -1.5℃. They expected that the humidity of the room would slightly affect the moisture content of the adhesive when it was exposed to the room for a period of time.
[0055] The viscosity specification of LOCTITE EA 3060LC is 10-14 Pas at room temperature, which is confirmed by the applicant's measurements. The inventors found that it reached a viscosity of less than 100 mPas at 75°C. They expect that at the temperatures used in the bonding process (about 150+ / -20°C), the viscosity can be less than 10 mPas, close to the viscosity of water (1 mPas).
[0056] This was confirmed by experiments carried out by the inventors in cross sections studied by electron microscopy. Figure 7 In the illustration of , such an image is shown purely as an example. A cross section with a thickness of about 30 μm from bottom to top shows the catalytic coating on the cathode side, here denoted by 5, in the area to the right of the adhesive 2. Then there is the ionomer of the membrane 4, denoted by 6, and a reinforcement layer, denoted by 7, which is surrounded by a second layer of ionomer 6. On this further ionomer layer 6, a catalytic coating with an anode catalyst, denoted by 8, is shown, the catalytic coating 5 being the cathode catalyst. The catalytic coatings 5, 8 form a microporous surface or layer on the ionomer 6. This is followed by a layer of adhesive 2. Tear tests for the bond according to the prior art have now been produced. Figure 7 The image shown, i.e., the bonding is produced in such a way that the adhesive 2 is bonded to the surface catalytic coating, here the catalytic coating 8 on the anode side, at the surface with a limited degree of adhesive penetration, so that when tearing, cohesive failure occurs inside the catalyst layer. This corresponds to the problem according to the prior art.
[0057] exist Figure 8 In the illustration of , the same image is shown again for the type of combination described here. Figure 8 In the figure, on the left, the structure is shown again, wherein, according to the examination by electron microscopy, it has been recognized that the catalyst has undergone some changes in this area and is clearly completely penetrated by the binder 2. Due to the limited display options in the line drawing, this illustration has been omitted here. However, the decisive Figure 8The tear test results are shown in the right area. It will be appreciated that, unlike the prior art, the ionomer 6 splits along the reinforcement layer 7 and remains firmly bonded to the catalytic coating 8 and the adhesive 2 on the anode side. The strength of the bond surface is therefore significantly higher than the strength of the actual membrane 4, so that a particularly strong and compact structure can be achieved by the method according to the invention.
[0058] Alternative methods may also be used, preferably using visible light. Fig. 9 As shown, it is actually Figure 5 Almost identically, the adhesive 2 can be applied to the frame 1 and / or the film 4. The previously described step of activating the adhesive 2 by first exposure to electromagnetic radiation is obvious. Alternatively, activation and heating are performed by the same exposure to electromagnetic radiation, preferably blue visible light with a wavelength of about 430 nm. The inventors have found that light of about this wavelength enables activation of the adhesive and thus delays the curing time of the adhesive, as well as heating before the final curing of the adhesive, which is a cationic epoxy resin with a small amount of water. This single exposure or irradiation allows the bonding method to be performed very quickly.
[0059] This composite of frame 1 and membrane 4 described in the previous figures can now be kept as a kind of semi-finished product. The next step can provide that the two gas diffusion layers (GDL) 9 are coated with adhesive 2 in their edge regions, for example, in turn by a deposition method such as screen printing or inkjet printing, as was previously done for frame 1 or membrane 4. This is Fig.10 The example of two gas diffusion layers 9 is explained in more detail in the illustration of FIG. As an exemplary and non-limiting design and process, one gas diffusion layer 9 is coated with a relatively large portion in the edge area and the other gas diffusion layer 9 is coated with a correspondingly smaller portion. In this case, the two gas diffusion layers are the gas diffusion layer 9 for the cathode of the fuel cell on the left and the gas diffusion layer for the anode of the fuel cell on the right. Each gas diffusion layer 9 is provided with a microporous layer MPL 10 on one side, which side will then face the membrane 4. The adhesive 2 is applied to the MPL 10 of the respective gas diffusion layer 9, and as shown Fig. 9 As indicated by the arrows in the illustration and the UV source or light source 3 , the adhesive 2 is briefly irradiated with electromagnetic radiation to activate the adhesive 2 .
[0060] Fig.11The illustration shows a structure prepared for bonding two gas diffusion layers 9 to a composite of a frame 1 and a membrane 4, which has already been cured. The frame 1 with the membrane 4 is introduced between the two gas diffusion layers 9 with the corresponding adhesive 2. The gas diffusion layer 9 for the anode of the fuel cell (shown here in the upper part) is placed above the membrane 4 in the frame 1 and is coated with the adhesive in the edge area. The gas diffusion layer 9 of the cathode shown below the composite of the frame 1 and the membrane 4 has a correspondingly wider adhesive coating, so that it is also bonded to the membrane 4 at the same location as the other gas diffusion layer is bonded to the membrane 4, while ensuring that the area of the membrane 4 that is already bonded to the frame 1 is also bonded to the gas diffusion layer. In addition, the width of the applied adhesive 2 is so large that a bond between the shown gas diffusion layer 9 of the cathode and the frame 1 is also achieved in the edge area. The components shown are adapted to each other in shape but still spaced apart from each other, and are now positioned relative to each other using a clamping device 100 or a pressing tool and are held against each other under slight pressure. This is correspondingly shown in Fig.12 . The pressing device here comprises a lower part 11 and an upper part 12, which are pressed against each other under slight pressure in order to position the frame 1 as well as the membrane 4 and the two gas diffusion layers 9 as the components to be bonded, against each other in a reliable manner. The activated, UV-curable adhesive is heated accordingly by means of a UV radiation source 3, for example but not limited to through the two transparent windows 13 shown in the two parts 11, 12 of the pressing device 100. In this case, the transparent window 13 is preferably transparent, for example made of glass, quartz or the like. That is, the UV radiation penetrates the transparent window 13 and thus heats the corresponding gas diffusion layer 9 by the black body effect in the area in which the transparent window 13 is formed, that is, in particular in the area where the adhesive 2 is also applied. Finally, the adhesive 2 is heated in the manner and method already described so that the adhesive 2 becomes low-viscosity and, on the other hand, penetrates through the microporous layer 10 of the corresponding gas diffusion layer 9 and, on the other hand, penetrates through the catalytic coating 5, 8 of the membrane 4.
[0061] At the same time, the adhesive 2 adheres to the anode side of the structure on the frame 1 side, and the GDL or its MPL 10 adheres to the frame 1, which is laterally penetrated by the adhesive 2 near the membrane 4, thus forming an overall very tight structure, which again in Fig.13. Here, the area designated by I represents the connection between the catalyst-coated membrane 4 and the frame 1, and the catalytic coating 8 on the anode side of the membrane 4 is penetrated by the adhesive 2. The two areas designated by II show the bond between the gas diffusion layer 9 or its MPL 10 and the membrane 4. Here, the adhesive 2 penetrates into the corresponding MPL 10 and the corresponding catalytic coating 5, 8. The area designated by III shows the bond between a gas diffusion layer or its MPL 10 assigned to the lower part of the fuel cell cathode and the frame 1. The side of the frame 1 in the area of the other gas diffusion layer 9 is also wetted by the adhesive in the area designated by IV, so that a sealing of the structure also occurs here. Thus, in Fig.13 In the illustration of FIG. 1 , the active area of the membrane 4 and the gas diffusion layer 9 which can be used for active reactions of the substances in the fuel cell starts on the right and close to the binder 2 .
[0062] In order to protect the electrochemically active region on the one hand and to keep the heat input for liquefying the adhesive 2 in a targeted manner within a desired range, cooling regions may be provided in the upper part 12 and / or the lower part 11 of the pressing device 100. Fig.12 In the illustration of , two cooling regions are respectively arranged on the right side near the transparent window 13, both regions are designated by 14. In addition, other regions more or less directly adjacent to the window 13 can also be cooled. These cooling regions 14 ensure that heat is accumulated in a targeted manner only in the region of the adhesive 2, or is introduced into said region, and that the subsequent electrochemically active region is not damaged by this heat input. In addition, the cooling in the region 14 allows a very targeted heat input and a very targeted time for which the heat input can be stopped, without allowing the heat accumulated in the region of the gas diffusion layer 9 adjacent to the transparent window 13 to return to the region of the adhesive 2. The third advantageous effect leads to a reduction in the temperature in some parts of the gas diffusion layer and the catalyst-coated membrane 4 adjacent to the cooling region 14. In other words, the viscosity of the adhesive 2 in these regions increases, so that the preferred active cooling of the region 14 also affects the spatial extent of the adhesive 2 in an advantageous manner, or it can be limited.
[0063] Overall, this construction allows an ideal combination of the key elements of the membrane electrode assembly, in particular the frame 1 and the membrane 4 and the gas diffusion layer 9 , which are collectively referred to as “components”.
[0064] As mentioned above, the binder 2 may be a cationic epoxy resin. Preferably, the cationic epoxy resin contains a small amount of water. Fig.14The graph shows the effect of water in cationic epoxy resins. The horizontal axis is time. The vertical axis shows the epoxide conversion. The solid line represents a completely water-free cationic epoxy resin. The dashed and dotted lines show the curing of the epoxy resin when it has a certain water content, that is, between about 100ppm and up to 1.5wt.%. The water content increases with the changed line form, as shown in the explanation box of the graph. Compared with the water-free cationic epoxy resin, this small amount of water has the effect of delaying the curing after the initial activation. After this delay, the curing of the cationic epoxy resin with water will be faster than the curing of the water-free cationic epoxy resin. Therefore, a very small amount of water is sufficient. When the epoxy resin is used in an environment with high air humidity, it can be introduced into the epoxy resin.
[0065] This delay in curing time t is ideal for the method of the present invention because the time delay allows for heating and helps the thin adhesive 2 to penetrate the microporous surface well. In addition, the faster curing afterwards ensures that the curing is completed before the thin adhesive 2 can flow into areas where the adhesive is not needed.
Claims
1. A method for bonding components (4, 9) of a PEM fuel cell to a frame (1) and / or to each other, for which purpose an adhesive (2) which can be cured by electromagnetic radiation in the visible or ultraviolet range is applied to the frame (1) and / or to at least one of the components (4, 9), in, contacting the frame (1) and / or the components (4, 9) and exposing the adhesive (2) to electromagnetic radiation for activation and heating; wherein the heating reduces the viscosity of the adhesive (2) before the adhesive (2) is finally cured; It is characterized in that The binder (2) is a water-containing cationic epoxy resin, wherein the water content of the cationic epoxy resin is 100-500 ppm, and the viscosity of the cationic epoxy resin at 75° C. is less than 100 mPas; applying electromagnetic radiation at least for heating from the frame (1) side through the frame (1) while the frame (1) is bonded to the component (4, 9); and In case the frame (1) and / or the components (4, 9) are brought into contact and the adhesive (2) is heated by electromagnetic radiation, activation and heating by electromagnetic radiation are accomplished in the same single exposure of the adhesive to electromagnetic radiation.
2. The method according to claim 1, It is characterized in that The water content of cationic epoxy resin is 200-450ppm.
3. The method according to claim 1 or 2, It is characterized in that The heating is performed to a temperature of 100-200°C.
4. The method according to claim 1 or 2, It is characterized in that A component (4, 9) of a PEM fuel cell has a microporous surface layer (5, 8, 10) at least on the surface that is bonded to other components (4, 9) and / or a frame.
5. The method according to claim 1 or 2, It is characterized in that By applying electromagnetic radiation, heat is generated in the components (4, 9).
6. The method according to claim 1 or 2, It is characterized in that The adhesive (2) does not contain any thermally cross-linking component.
7. The method according to claim 1 or 2, It is characterized in that The components (4, 9) and / or the frame (1) are held together by means of a hold-down device, wherein the irradiation with electromagnetic radiation takes place through at least one radiation-transmissive window (13) in the hold-down device.
8. The method according to claim 7, It is characterized in that At least one area of the pressing device adjacent to the transparent window (13) is cooled.
9. Use of the method according to any one of claims 1 to 8 for bonding a membrane (4) coated with a catalyst on the entire surface as one of the components to a frame (1).
10. Use of the method according to any one of claims 1 to 8 for bonding at least one gas diffusion layer (9) as a component to a catalyst-coated membrane (4) as a further component and / or to a frame (1).
11. The use according to claim 9 or 10, It is characterized in that At least one gas diffusion layer (9) is bonded to a pre-bonded and cured composite of a membrane (4) coated with a catalyst on its entire surface and a frame (1).
