Method for producing a stack of solid oxide cells

Abstract

Method for producing a stack of solid oxide cells (SOxC), namely a stack of solid oxide fuel cells (SOFC) or of solid oxide electrolyzer cells (SOEC), in which at least two SOxC are arranged one above the other and joined together to form the stack using a bonding agent, such as a glass solder, in a multi-phase joining and conditioning process, during which the arrangement consisting of the at least two SOxC is exposed to at least different temperatures, characterized in that the joining and conditioning process, avoiding a rigid temporal sequence, takes place at least partially as an impedance-controlled process.by monitoring the electrical impedance profile during this process and controlling the sequence of the joining and conditioning phases, which differ at least in their temperature profiles, based on measurements obtained from repeated electrical impedance measurements.

Description

[0001] The invention relates to the production of stacks of solid oxide cells (SOxC), namely stacks of solid oxide fuel cells (SOFC) or solid oxide electrolyzer cells (SOEC). It relates to a process in which several such SOxC cells are arranged one above the other and joined together to form a corresponding stack by applying temperature and, optionally, force, using a glass solder.

[0002] Solid oxide fuel cells (SOFCs) and solid oxide electrolyzer cells (SOECs) with ceramic cells are high-temperature variants of fuel cells and electrolyzer cells, respectively. They operate at temperatures between 600 and 1000°C and are characterized by high electrical efficiencies. Generally, solid oxide fuel cells and solid oxide electrolyzer cells are referred to collectively as solid oxide cells, sometimes also as SOxC.

[0003] The assembly of multiple SOxC cells into a stack occurs in a multi-phase joining and conditioning process. During this process, the cells are not only physically (i.e., geometrically) joined, at least under the influence of temperature and, depending on the type, possibly also under the influence of force. Rather, material changes also occur in the cells themselves, such as in the ceramic components they contain, as well as in the materials used as bonding agents between the cells, such as glass solders of varying properties.Furthermore, a sintering process takes place, in which particles of the main ceramic component of the cells fill previously existing cavities when heated to just below the melting temperature. This results in a reduction in the volume (shrinkage) of the cells and the stack formed from them, while largely maintaining the shape of the cells. Additionally, the binder is removed from the green ceramic body of the cells in the form of vapor during a process called thermal debinding. This can be achieved, for example, by heating at ambient pressure in an oxidative or non-oxidative atmosphere or in a vacuum. Debinding requires a precise start at the required temperature, control of the chemical reactions within the product, and high temperature homogeneity.

[0004] According to the process name, the cells, their components, and the component stacks also undergo conditioning in the joining and conditioning process, both with regard to the individual cell and the overall arrangement. This conditioning process reduces mechanical stresses arising during the joining process, allows individual components of the arrangement to crystallize, and reduces reducible components of the stack through chemical reactions.

[0005] Depending on the type of cells assembled into the stack, as well as their number and the stack's geometry (dimensions), different temperature profiles and, if applicable, force profiles are applied to the stack during the individual phases of the joining and conditioning process, which typically lasts several hours. For the different types of stacks used with regard to the cells employed, manufacturers utilize temporal temperature regimes and, if applicable, force regimes that result from experiments and experience gained during production. Where reference is made above, below, and in the patent claims to a force profile or force profiles, these do not refer to profiles in the sense of a spatial force distribution within the stack, but rather to temporal profiles of the force or forces acting upon the stack.

[0006] Based on the aforementioned empirical data, the individual phases of the joining and conditioning process, as well as the process as a whole, are time-controlled according to the state of the art. The time intervals for each phase are selected to ensure that the objective of that phase—for example, melting a glass solder used to join the cells—is reliably achieved, or that certain reduction processes are completed.

[0007] After completion of the assembly and conditioning process, the quality of the overall assembly, or stack, is checked, among other things, by an impedance measurement, specifically by measuring the complex AC resistance at different frequencies. For solid oxide fuel cells (SOFCs) in particular, this impedance measurement provides information about the electrical properties of the resulting stack. The impedance measurement is performed using a suitable measuring device via electrochemical impedance spectroscopy (EIS).

[0008] The use of impedance spectroscopy for quality assessment is also mentioned in connection with a process for manufacturing electrochemical solid oxide devices, namely stacks of solid oxide cells, described in DE 10 2006 056 986 A1. According to this process, after a first stage of the otherwise fixed-time assembly process for stack production, referred to as a pre-sealing process, in which initially only a portion of the connections to be formed between the elements of the stack, and thus a partially completed stack referred to as a pseudo-stack, is produced, an intermediate inspection is carried out. Following this quality control, for example using impedance spectroscopy, and provided that none of the cells of the pseudo-stack exhibit defects, or at least all cells are functional, the creation of all further connections required to form the finished stack then proceeds.

[0009] The previously described timing control of the joining and conditioning process has the disadvantage that, with regard to reliably achieving the objectives in each phase of the process, the overall process often takes too long. This means that the best possible manufacturing efficiency is not achieved.

[0010] The object of the invention is to avoid this disadvantage in order to increase the overall efficiency of manufacturing corresponding SOxC stacks. Naturally, it should still be ensured that the objectives pursued in the individual phases of the joining and conditioning process are reliably achieved. A suitable method for this purpose is to be specified.

[0011] A method that solves the problem is characterized by claim 1. Specific, or advantageous, embodiments of the method are given by the dependent claims.

[0012] The proposed method for manufacturing a stack of solid oxide cells (SOxC), i.e., stacks of solid oxide fuel cells (SOFC) or stacks of solid oxide electrolyzer cells (SOEC), also assumes that at least two of the aforementioned SOxC cells are arranged one above the other and joined together in a joining and conditioning process using a bonding agent, such as glass solder. In this respect, the proposed method relates in particular to the control of the multi-phase joining and conditioning process, during which the arrangement consisting of the at least two SOxC cells to be joined is exposed to different temperatures and, if necessary, different forces, usually with the addition and removal of chemical media. According to the invention, this joining and conditioning process proceeds, at least partially, as an impedance-controlled process.The sequence of phases of the joining and conditioning process, which differ from each other at least by different temperature profiles and possibly also by force profiles, is controlled depending on measured values ​​obtained from repeated measurements of the electrical impedance.

[0013] The core idea is therefore to use impedance measurement not only to assess the quality of the SOxC stacks resulting from the joining and conditioning process, but also to monitor the impedance and its behavior during the process itself, thus moving away from a rigid time-based regime, at least for parts of this process. Surprisingly, it has been shown that impedance measurements can be used not only to make quality statements about the finished product, but also to gain insights into the progress of the joining and conditioning process and to use these insights for targeted process control.The statement made above and in the characterization of the method in claim 1, according to which the joining and conditioning process is at least partially impedance-controlled, means that not the entire process necessarily has to be impedance-controlled, but that in individual cases and depending on the type and / or nature of the manufactured product, a fixed time sequence can also be adhered to with regard to individual phases.

[0014] Repeated measurement of the electrical impedance during the joining and conditioning process is preferably carried out using electrochemical impedance spectroscopy (EIS) with an impedance spectrometer. In this process, a multitude of impedance measurements are recorded at different frequencies of a measurement signal applied to the arrangement or parts thereof.

[0015] In practical implementation, the process can be designed such that, first, a number of stacks of the same type are experimentally produced, with repeated measurements of the electrical impedance during the joining and conditioning process. The quality of each individual stack is then assessed. One of the impedance profiles recorded during the experimental production of stacks that meet a defined quality standard (quality parameters) is then used to control the joining and conditioning process in the serial production of stacks of this type.

[0016] The latter can be achieved, for example, by dynamically controlling the joining and conditioning process in series production, at least in phases, with regard to the temperature profile and, if necessary, the temporal profile of the forces introduced into the stack, based on the impedance profile recorded during the experimental production of high-quality stacks. This process involves repeated impedance measurements. Alternatively or cumulatively, the temporal sequence of the phases of the joining and conditioning process can also be controlled in series production, based on the findings from the experimental production of the stacks, with repeated impedance measurements.

[0017] In any case, it should be noted that the way in which the recorded impedance profile is used to influence the joining and conditioning process in individual cases during series production can vary considerably and depends heavily on a multitude of different factors. These include, for example, the specific type of SOxC assembled into a stack, their number in the stack, and the geometry of the stack manufactured according to the process. Furthermore, the components used in the production process, and especially in the joining and conditioning process, such as the type of fastener, certainly have a significant influence on the impedance profile during manufacturing and thus on the control regime ultimately implemented based on the impedance.

[0018] In this respect, the respective manufacturing process and its design must ultimately be aligned with the manufacturer's experience and the insights gained from the stacks they have produced. This also applies, among other things, to the method used to measure the impedance values ​​of a given stack. For example, the design of the process flow can be based on the continuously measured impedance of the entire stack, or rather, of all the SOxC components assembled into the stack.

[0019] Alternatively or cumulatively, it is also conceivable to measure the impedance for one or more individual SOxCs of the stack and / or for one or more groups of SOxCs to be combined into the stack. The respective, repeatedly performed measurements can be carried out sequentially, but preferably also simultaneously, particularly when measuring multiple individual cells or groups of cells (with possible simultaneous measurement of the entire stack).

[0020] The following section will discuss some aspects of the invention again using an exemplary embodiment and the accompanying drawings. The drawings show in detail: Fig. 1: the part of a stack of solid oxide cells SOxC, Fig. 2: an exemplary course of the impedance measured during the joining and conditioning process on a stack of SOxC, Fig. 3: an exemplary temperature profile during a joining and conditioning process for a stack of SOxC.

[0021] The Fig. Figure 1 shows a section of a stack of solid oxide cells, specifically a stack of solid oxide fuel cells (SOFCs) or solid oxide electrolyzer cells (SOECs), depicted as a (conceptual) breakout from such a stack. The illustration of the section shows three SOFCs arranged geometrically parallel to each other, or one above the other. The individual SOFCs consist, among other things, of a ceramic material, interconnector plates, and conductive layers, and are connected to each other, for example, using a glass solder, which simultaneously seals the entire assembly (sealing glass).

[0022] For this purpose, the arrangement shown in part is subjected to a joining and conditioning process that typically lasts several hours. In this joining and conditioning process, which can be divided into several (not necessarily rigid) phases, the arrangement consisting of the parallel SOxC components and the bonding agent is subjected to at least time-varying temperatures – and, depending on the specific stack type, possibly also to different forces.

[0023] In the Fig. Figure 3 illustrates an example of a possible temperature profile over time for such a joining and conditioning process, which, according to this example, can be divided into essentially four phases. It should be noted here that the temperature profile shown in the Fig. 3 shown course (as well as the one in the Fig. The figures shown (2) are not representative. Rather, the course of this process varies considerably depending on the type of cells connected to form the stack, the cell and stack geometry, the number of cells, and the materials used for the cells and their connection. Accordingly, the process may be divided into significantly more phases. With regard to the Fig. 3 is merely intended to clarify once again that the temperature regime shown there as an example and the sequence of the individual phases (four phases in the example) follow fixed time specifications according to the state of the art, which are usually based on the experience of the respective manufacturer.

[0024] According to the example in the Fig. The four phases of the joining and conditioning process, as illustrated in Figure 3, can be structured as follows, for example. First, the glass solder paste is dried at a low temperature while maintaining the force acting on the stack. During this process, some solvents evaporate from the container. This step usually lasts several hours (Phase I).

[0025] The temperature is then increased to over 850°C (Phase II), at least according to the example shown. The heating rate is, for example, between 1 K / min and 3 K / min. During the heating process, the glass solder melts. As the glass solder melts, the stack shrinks slightly. It is necessary to maintain a constant force despite this shrinkage.

[0026] After reaching the joining temperature, the cell is held at a constant temperature for an extended period (Phase III). The final temperature from Phase II may, if necessary and deviating from the illustration, be slightly higher than the joining temperature from Phase III. Depending on the target configuration of the stack to be manufactured, the temperature profile curve is modified accordingly to achieve specific properties.

[0027] During the extended period of Phase III, the glass solder largely crystallizes into a glass-ceramic. The bond between the glass solder and the SOxC cell is improved by the reduction of stresses. For successful and uniform crystallization of the glass solder, it is essential that the various cells are heated evenly, which is one reason for the long duration of this phase. Irregularities can lead to quality defects and internal stresses within the cells, potentially reducing their lifespan.

[0028] Corresponding to the Fig. 3 shows the Fig. Figure 2, which also serves as a non-representative example illustrating how the impedance, repeatedly measured for the entire stack, for a single SOxC, or for a group of such cells, can develop during the assembly and conditioning process. Regarding the depicted temporal profile of the impedance, it should be explicitly noted again that this can vary considerably depending on the overall configuration. As a general rule, it can only be assumed that the impedance will gradually decrease as the process progresses.

[0029] Based on empirical data obtained from a series of tests for a specific type of stack, the joining and conditioning process can be impedance-controlled. This means that, deviating from rigid time phases—such as those exemplified here in the... Fig.As 2 has been reiterated, the entire process is designed depending on the impedance profile. This means that, for example, individual phases shown here as examples can be shortened based on the impedance development, whereby, for instance, the process in phase II can continue with an exponential temperature increase instead of a linear one. The modifications to the joining and conditioning process mentioned above are, of course, only examples.

Images