| As a clean, highly efficient, fuel flexible power generating device, solid oxide fuel cell (SOFC) is an important part of the new energy technologies. As the experiments are of high cost, time consuming and difficult to comprehensively explore the effects of various material parameters, structural designs and the working conditions on the SOFC performance, theory and modeling are considered as important tools for accelerating the development of SOFC technologies.This dissertation focuses on developing a multi-scale and multi-physics coupled model for the theoretical study and parametric optimizations of SOFCs. The multi-scale mainly includes the development of percolation micro models for composite electrodes, multi-physical coupled macro models for cell processes and the combination of the micro and macro model for the simulations of the cell unit. The multi-physics coupling here means that the model comprehensively considers the coupled behavior of the detail electrochemical reaction process, the conductions of electronic and ionic currents, the transport of reacting gas species and the mixed heat transfer through the solid and gas parts.In chapter one, the development history of fuel cells and their classifications are briefly introduced. The basic working principle, components and stack structures of SOFCs are then described. The expressions for the Nernst potential at open circuit and the local Nernst potential are deduced in detail based on underlying thermodynamics theory. Finally, a brief literature overview about the SOFC theory and simulation on three different length scales is given.In chapter two, a detailed review about the present micro scale electrode models are presented. The advantages and disadvantages of Bouvard and Suzuki's percolation micro model based on the randomly packing of spheres are discussed. Then, a novel percolation micro model, based upon significant earlier literatures, is developed to predict effective electrode properties from microstructure parameters. This model is applicable to both the binary and multi-component mixtures of particles. The model predicts effective ionic and electronic conductivities, three-phase boundary lengths, and hydraulic pore radii. The effective properties depend upon primary physical characteristics, including average particle-radii, volumetric packing densities, particle contact angles, and porosity. All results are presented in nondimensional form, which provides considerable generality in their practical application.In chapter three, percolation micro-model is extended to predict the effective properties in composite electrode formed by polydisperse electronic and ionic conductors (i.e., normal standard distribution for each phase), such as the percolated triple-phase-boundary (TPB) lengths, hydraulic pore radius, intra- and inter-particles conductivities and so on. And the independent microstructure parameters include:the mean particle size of electrode-and electrolyte-materials, volume fraction of electrode-material, the relevant standard deviations of each phase and porosity. The validity of this model is verified by comparing the calculated results based on the percolation micro-model equations and the results based on random packing reconstruction reported by Kenney et al.. Finally, the specific natures of anode and cathode are considered into the percolation micro-model for discussing what kind of composite electrodes with actual particle size distributions, would maximize the performance of anode and cathode, respectively. The results shows that:the composite electrode with small mean particle size of electrode- and electrolyte-materials and relatively narrow particle size distribution would be helpful for enhancing the performance of composite cathode (LSM and YSZ); And a higher SOFC composite anode performance can be obtained by using a composite electrode with larger electrolyte-particle and relatively wide particle size distribution.In chapter four, the processes of electrochemical reactions, electronic and ionic conductions within an SOFC are clearly described through a proposed equivalent circuit model. With this model, the local electronic and ionic electric potential profiles in the electrode- and electrolyte-phases can be described based on local gas compositions and electrochemical kinetic analysis. Correspondingly, the cell-level scale macro-model and the percolation micro-model, described in chapter two, are combined to systematically examine the effects of various parameters on the performance of a composite cathode inter-layer. The examined parameters include the thickness, effective electronic and ionic conductivities, exchange current density, operating temperature, output current density, electrode- and electrolyte-particle radii, composition and porosity of the cathode inter-layer. The comprehensive study shows conclusively that a cathode inter-layer thickness in a range of 10-20μm is optimal for all practical material choices and microstructure designs.Chapter five can be considered as a extending of chapter four. As most of the SOFC activation overpotential is caused within cathode inter-layer, understanding the effects of each parameter on the distribution of cathode electrochemically active zone (CEAZ) is essential to enhance the SOFC performance. Although CEAZ can be exactly simulated through the numerical model, such as the cell lever model described in chapter 4, these kinds of model often rely on complicated numerical procedures and are difficult for the experimentalists to use with different material parameters and working conditions. In this chapter, analytical equations are deduced for predicting the upper limited thickness of CEAZ as a function of operational conditions and electrode properties such as, the output current density, effective ionic conductivity, percolated triple phase boundary length and the exchange transfer current per unit TPB length. And the analytical expressions are validated by numerical models without the effective circuit approximation.In chapter six, a summary for this thesis are presented.
|
PDF Full Text Request