Method for producing a seal on a layer or layer of an electrochemical cell

Abstract

The invention relates to a method for producing a seal (1) on a layer or layer (2) of an electrochemical cell by means of stencil printing, wherein a stencil (3) with at least one recess (5) spanned by webs (4) is used, which is filled with a sealant and / or adhesive (6) in a doctor blade process, so that after the sealant and / or adhesive (6) is released from the stencil (3), the webs (4) leave impressions in the sealant and / or adhesive (6). According to the invention, the production of the seal (1) includes a quality inspection comprising the steps: a) measuring an average profile height (h) over a length (l) of the seal (1), b) processing the measured values ​​by means of Fourier transform (FFT) to determine amplitudes (a1, a2, a3, a4, ...an) in an amplitude spectrum that are characteristic for the seal (1), c) comparing the amplitudes (a1, a2, a3, a4, ...an) with a previously determined target amplitude and / or a mean amplitude (aAVG) for the detection of deviations that suggest a printing error.

Description

The invention relates to a method for producing a seal on a layer or ply of an electrochemical cell, in particular a fuel, electrolysis or battery cell. The layer or ply can be, for example, a mono- or bipolar plate, a separator plate and / or a membrane electrode arrangement.Prior ArtElectrochemical cells, such as fuel cells, electrolysis cells or battery cells, have a multilayer or multilayer structure. This construction requires intermediate seals to separate media supplied during operation of a cell. To form the seals, a sealing and / or adhesive is applied to a layer or layer of the cell. In particular, dispensing or printing methods are used here.Mass production of electrochemical cells requires a high processing speed, i.e. short cycle times, and a high process stability. This requirement is met in particular by stencil printing. The printing process during stencil printing is divided in principle into the following three sub-processes: 1. application of the sealing and / or adhesive to the stencil, which was previously placed on the layer or ply to be printed, 2. filling of the stencil cutout or stencil cutouts with the sealing and / or adhesive in a doctoring process, 3. release of the sealing and / or adhesive from the stencil cutout or the stencil cutouts by separating the stencil from the layer or ply.The stencil thickness determines the layer height of the applied structure, wherein typical layer heights are in the range from 100 μm to 300 μm. For seals of electrochemical cells, however, layer heights of up to 1500 μm are required in order to achieve tolerance compensation and to be as free as possible in the design of the structures to be applied. This is because-depending on the respective sealing concept-the structures to be applied can be large and / or complex.In order to achieve the layer heights required for seals of electrochemical cells during stencil printing, thicker stencils must accordingly be used. However, this increases the risk that air bubbles are incorporated during the doctoring process for filling the stencil cutout or stencil cutouts with the applied sealing and / or adhesive, which air bubbles subsequently lead to defects. This is because the stencil thickness also increases the volume of the stencil cutout or stencil cutouts to be filled with the sealing and / or adhesive.However, not only when filling the stencil cutout or the stencil cutouts with the sealing and / or adhesive, bubbling may occur, but also when releasing the sealing and / or adhesive. This applies in particular if, for the realization of closed contours, templates are used which have webs above the template recesses. This is because then a part of the sealing and / or adhesive can remain adhering to the underside of the webs and form threads. These are stretched or stretched to the length during the separating process, generally during the lowering of the layer or ply to be printed, until they tear or become detached from the stencil. Air bubbles can then form when the threads collapse, as illustrated by way of example in FIG. 1.The properties of the webs of a stencil have a direct influence on the printing result, in particular on the corrugation / roughness and thus on the topology of the resulting seal. By using a template with webs, a seal with a regularly corrugated surface is produced.The present invention is concerned with the object of optimizing the process of producing a seal by means of stencil printing in order to avoid printing errors and to achieve a higher dimensional accuracy.To achieve the object, the method having the features of claim 1 is proposed. Advantageous further developments of the invention can be found in the dependent claims.Disclosure of the InventionA method for producing a seal on a layer or ply of an electrochemical cell by means of stencil printing is proposed. In the method, a stencil with at least one recess spanned by webs is used, which is filled with a sealing and / or adhesive in a doctoring process, so that after the sealing and / or adhesive has been released from the stencil, the webs leave impressions in the sealing and / or adhesive. According to the invention, the production of the seal includes a quality test, comprising the steps: a) measuring a mean profile height h over a length l of the seal, b) processing the measured values by means of Fourier transformation (FFT) in order to determine amplitudes a 1, a 2, a 3, a 4,... a n in an amplitude spectrum, which are characteristic of the seal, c) comparing the amplitudes a 1, a 2, a 3, a 4,... a n having a previously determined setpoint amplitude and / or an average amplitude a AVG for detecting deviations which indicate a pressure error.Since pressure errors, such as air bubbles, flaws and the like, lead to deviating profile heights, pressure errors can be detected with the aid of the quality test and measures can be taken in a targeted manner, on the other hand. The measures can include an adjustment of the printing parameters, for example the doctor blade speed and / or the separation speed, and / or an adjustment of the rheology, for example the viscosity, the surface tension and / or the thixotropy, of the printing material. With the aid of the measures, the production process can be stabilized, so that printing errors are less caused. The seals printed according to the proposed method accordingly have a higher dimensional accuracy. At the same time, the number of shots can be decreased.In the context of the quality test, the topology of the surface of the printed seal is assessed. This represents an individual characteristic variable, since the webs of the stencil leave impressions in the sealing and / or adhesive. A seal produced according to the method accordingly has a specific, in particular wavy, topology which can be measured and evaluated. However, the assessment does not use the measured mean profile heights h as a basis, but rather the amplitudes a 1, a 2, a 3, a 4,... a n. determined from this by means of FFT. This procedure has the advantage that process instabilities or printing errors are more easily recognizable.The detection of deviations in the determined amplitude spectrum corresponds to an indirect detection of a pressure error. Moreover, this method enables the identification of specific recess regions of the template, in which the probability of bubble formation is particularly high. By identifying such recess regions, targeted measures for reducing bubble formation can be taken more quickly.According to a preferred embodiment of the invention, in step a) the mean profile height h is measured with the aid of a profiler or a laser scanner. The laser scanner is particularly suitable for the series production of seals. The mean profile height of the seal can be measured over its entire length or over a length section which is particularly critical and / or representative.In step b), the measured mean profile heights are then processed by means of FFT and the corresponding amplitude spectrum is determined with the amplitudes a 1, a 2, a 3, a 4,... a n which are characteristic of the seal.In step c), the amplitudes a 1, a 2, a 3, a 4,... a n can then be compared with a previously determined setpoint amplitude and / or an average amplitude a AVG in order to detect a deviation which indicates a pressure error. The desired amplitude can be determined beforehand under optimized process conditions (desired conditions). For this purpose, the mean profile heights are likewise first measured and processed by means of FFT. If the mean amplitude a AVG is to be used as a reference value in step c), this can be determined according to the following equation:When producing the seal by means of stencil printing, the sealing and / or adhesive is preferably applied to the layer or ply in a layer thickness of 300 to 1500 μm, further preferably in a layer thickness of 500 to 1500 μm. Layer heights in the specified ranges permit tolerance compensation and provide a greater freedom of design. Since the thickness of the stencil increases with the height of the layer, the risk of air bubble formation and other printing errors is increased at the same time. With the aid of the proposed method, pressure errors can be detected and the process parameters can be adjusted accordingly.Furthermore, the sealing and / or adhesive is preferably applied by means of stencil printing to a mono- or bipolar plate, a separator plate and / or a membrane electrode arrangement of the electrochemical cell. The seal produced from the sealing and / or adhesive then serves for the media separation within the electrochemical cell and / or seals it off from the outside.The invention and its advantages are explained in more detail below with reference to the attached drawings. These show: FIGS. 1a) to e) are longitudinal sections through a layer of an electrochemical cell during printing by means of a stencil printing process, FIG. 2 is a plan view of a template, FIG. 3 is an enlarged detail of FIG. 2, FIG. 4 is a cross-section through the template of FIG. 2, FIG. 5 shows a perspective illustration of a seal produced by a method according to the invention, FIG. 6 a) shows a diagram for illustrating the mean profile height h of a seal produced by a method according to the invention over a length l, b) of an enlarged section of FIG. 6 a), FIG. 7 a) shows a diagram for illustrating the mean profile height h of a seal produced according to a method according to the invention over a length l, b) of a corresponding amplitude spectrum determined by means of FFT, FIG. 8 a) shows a diagram for illustrating the mean profile height h of a seal produced according to a method according to the invention over a length l, b) of a corresponding amplitude spectrum determined by means of FFT, and FIG. 9 a) shows a diagram for illustrating the mean profile height h of a seal produced according to a method according to the invention over a length l, b) of a corresponding amplitude spectrum determined by means of FFT.DETAILED DESCRIPTION OF THE DRAWINGSFIG. 1 shows, by way of example, the process of producing a seal 1 by means of stencil printing. For this purpose, a stencil 3 having a recess 5 spanned by webs 4 is placed on a layer or ply 2 to be printed. The recess 5 is then filled with a sealing and / or adhesive 6 in a doctoring process (see FIG. 1 a )). To release the sealing and / or adhesive 6, the stencil 3 is separated from the layer or ply 2. For this purpose, the layer or layer 2 is moved downward away from the template 3 (see FIG. 1 b )). During this movement, a part of the sealing and / or adhesive 6 remains adhering under the webs 4, so that threads 7 form (see FIG. 1 c )). These tear and collapse (see FIG. 1 d )), so that in the worst case an air bubble 8 is formed (see FIG. 1 e )). The air bubble 8 leads to a pressure error which impairs the dimensional stability of the seal 1.Whether an air bubble 8 is formed depends in particular on the stencil geometry, the rheology of the printing material and further printing parameters, such as the doctor blade speed and / or separation speed. A stencil geometry for stencil printing of seals 1 is illustrated by way of example in FIGS. 2 to 4.The top view of FIG. 2 shows a recess 5 of a template 3 which is spanned by a plurality of webs 4. As can be seen from FIG. 3, the webs 4 have a web width b S and are arranged S to one another at a regular distance a. The web width b S and the distance a S result in a web-to-axle distance a A. As can be seen from FIG. 4, the webs 4 have a web length l S which corresponds to the width b A of the recess 5. The height h A of the recess 5 is reduced in the region of the webs 4 by the web height h S. The webs 4 accordingly lead to impressions remaining in the sealing and / or adhesive 6 after the release of the sealing and / or adhesive 6 from the stencil 3, such that the finished seal 1 has a wavy surface structure or topology.A seal 1 with a corresponding topology is illustrated by way of example in FIG. 5. The topology can be measured, for example with the aid of a profiler or a laser scanner. Ideally, an average profile height h is measured along a length l. The measured values then result in a height profile over the length l, as shown by way of example in FIG. 6 a). From this, an average amplitude a AVG can be determined. The distance between two wave peaks corresponds to the web axial distance a A.As illustrated by way of example in the following figures, an amplitude spectrum corresponding to the height profile can be created in each case with the aid of the Fourier transformation (FFT).FIG. 7 a) shows a height profile of a first seal 1 and FIG. 7 b) shows the corresponding amplitude spectrum. This has three characteristic amplitudes a 1, a 2, a 3. The amplitude a 1 always appears at the frequency f=1 / a A. The frequencies at which the other amplitudes a 2, a 3 appear also have a relationship with the web center distance a A. The amplitude a 2 appears at f=2*(1 / a a) and the amplitude a 3 at f=3*(1 / a a).The same applies analogously to the height profile of a second seal 1 shown in FIG. 8 a ) and the corresponding amplitude spectrum shown in FIG. 8 b ), only the four characteristic amplitudes a 1, a 2, a 3, a 4 appear here. The amplitude a 4 appears at a frequency f=4*(1 / a A).The amplitudes a 1, a 2, a 3 and optionally a 4 contain characteristic or individual information about the measured profile. The amplitudes can thus be used as characteristic values for the quality test. For example, the amplitudes a 1, a 2, a 3 and optionally a 4 can be compared with setpoint values or reference amplitudes which have been determined beforehand under optimized process conditions. Alternatively or additionally, an average amplitude a AVG can be determined from the amplitudes a 1, a 2, a 3 and optionally a 4 and used as a reference value. The reference value a A for the seal 1 of FIG. 8 can be calculated, for example, according to the following formula:The amplitude a 4 is not taken into account in the calculation, since it does not appear in all wave profiles or its magnitude is so low (<5 μm) that it can be neglected for the calculation of a AVG.If the comparison of the amplitudes a 1, a 2, a 3 and, if appropriate, a 4 yields a deviation from the respective reference value, this indicates a pressure error.FIGS. 9 a) and 9 b) show, by way of example, the height profile and the corresponding amplitude spectrum of a third seal 1. Local profile elevations h 1, h 2 lead to the amplitude a 1 appearing higher in the amplitude spectrum. In addition, further amplitudes a X around a 1 can be seen. These appear due to the blistering, which results in an irregular surface structure. The amplitudes a X appear as "noise" in the amplitude spectrum. The further amplitudes a 2, a 3, a 4 are greatly attenuated by pressure errors, so that the amplitude reduction can likewise be used to identify a seal 1 which has not reached the required quality standard.

Claims

Method for producing a seal (1) on a layer or ply (2) of an electrochemical cell by means of stencil printing, wherein a stencil (3) having at least one recess (5) spanned by webs (4) is used, which recess is filled with a sealing and / or adhesive (6) in a doctoring process, such that after the sealing and / or adhesive (6) has been released from the stencil (3), the webs (4) leave impressions in the sealing and / or adhesive (6), characterized in that the production of the seal (1) comprises a quality test, comprising the steps: a) measuring an average profile height (h) over a length (l) of the seal (1), b) processing the measured values by means of Fourier transformation (FFT) in order to determine amplitudes (a 1, a 2, a 3, a 4,... a n) in an amplitude spectrum which are characteristic of the seal (1), c) Comparison of the amplitudes (a 1, a 2, a 3, a 4,... a n) with a previously determined desired amplitude and / or an average amplitude (a AVG) for detecting deviations which indicate a pressure error.Method according to claim 1, characterised in that in step a) the mean profile height (h) is measured with the aid of a profiler or a laser scanner.Method according to Claim 1 or 2, characterized in that, in step c), the mean amplitude (a AVG) is determined by the following equation: a A V G = 2 ( a 1 + a 2 + a 3 + a 4 + ⋯ a n 2 ) Method according to one of the preceding claims, characterized in that the sealing and / or adhesive (6) is applied to the layer or layer (1) by means of stencil printing in a layer thickness of 300 to 1500 μm, preferably of 500 to 1500 μm.Method according to one of the preceding claims, characterized in that the sealing and / or adhesive (6) is applied by means of stencil printing to a mono- or bipolar plate, a separator plate and / or a membrane electrode arrangement of the electrochemical cell.

Images