Multi-Level Mathematical Modeling of Solid Oxide Fuel Cells

Abstract

Solid oxide fuel cells (SOFCs) are power units operating at 600–1000°C which produce electricity through the electrochemical conversion of the chemical energy of a fuel, thus providing a clean and efficient alternative for energy production in the near future. Although nowadays the first commercialization of residential and portable power units has just started, the optimization of this technology for bigger power system installations remains challenging. Indeed, SOFCs are complex systems, showing nonlinear interactions and strong coupling among phenomena occurring at different length scales. Therefore, modeling tools and simulation techniques offer a valid contribution to gain a deep understanding of the elementary processes in order to support the research in this field. In this thesis, an integrated microstructural–electrochemical modeling framework for SOFCs is presented. At the microscale, the model numerically reconstructs the microstructure of the electrodes, which are random porous composite media wherein the electrochemical reactions occur. The effective properties of the electrodes are evaluated in the reconstructed microstructures and used, as input parameters, in physically–based electrochemical models, consisting of mass and charge balances written in continuum approach, which describe the transport and reaction phenomena at the mesoscale within the cell. Therefore, the strong coupling between microstructural characteristics and electrochemical processes can be conveniently taken into account by the integrated model. The presented modeling framework represents a tool to fulfill a from–powder–to– power approach: it is able to reproduce and predict the SOFC macroscopic response, such as the current–voltage relationship, knowing only the powder characteristics and the operating conditions, which are the same measurable and controllable parameters available in reality. As a consequence, empirical, fitted or adjustable parameters are not required, feature which makes the model fully predictive and widely applicable in a broad range of conditions and fuel cell configurations as an interpretative tool of experimental data and as a design tool to optimize the system performance. In this thesis several particle–based microstructural models are presented, covering all the main morphological features and electrode architectures adopted in SOFC technology (e.g., conventional composite electrodes, infiltrated electrodes, arbitrary particle shapes and agglomerates). Electrochemical models for the description of electrodes (button cell configuration) and cells (within a stack) are discussed. These models are applied for both conventional conducting materials, typically adopted in SOFC technology, and for innovative fuel cell configurations. Each model is satisfactorily validated at the corresponding scale, then the whole framework is tested for the microstructural–electrochemical simulation of short stacks, showing an excellent agreement with experimental data. The application of the integrated model confirms that there is a strong coupling between electrode microstructure and cell electrochemical behavior: only by taking into account this interaction a model can provide quantitative information and sound predictions. Concluding, this thesis provides an approach to fill the gap between the microstructural and the electrochemical modeling, offering a predictive tool which does not rely on empirical and adjustable parameters, capable to reproduce the macroscopic electrochemical behavior of an SOFC unit starting from the powder characteristics.