(490g) Predictive Modeling of Effective Conductivity in Infiltrated Solid-Oxide Fuel Cell Anodes

The Solid Oxide Fuel Cell (SOFC) is an electrochemical device that converts chemical energy into electrical energy by directly and efficiently oxidizing hydrocarbon fuels. The SOFC anode, which is where the fuel oxidation reaction occurs, must have the following four properties to be functional: 1) porosity for mass transport, 2) catalytic activity for the oxidation reaction, 3) oxygen ion conductivity to provide an oxidizing agent to the fuel, and 4) electronic conductivity to transport the electrons released during the oxidation reaction. Recent work has shown that anode fabrication and modification by infiltration of an electronic conductor and oxidation catalyst into an oxygen ion conducting porous scaffold can result in outstanding electrochemical performance. In addition, infiltration allows one to separate the calcination temperature of the various active components, requires less electronic conducting material to form an interconnected conductive pathway, relieves dimensional instabilities within the fuel cell, and allows the porosity and pore morphology to be easily controlled. While the infiltration method provides great flexibility in composite composition and structure, the influence of the controllable parameters are not well understood. Furthermore, fabrication and testing of anode composites representing the full range and number of working parameters would be time consuming. Thus, a model to guide experiments is desired. 

In this talk, we will highlight our recent work on using a model to predict properties of the fuel cell anode as a function of the important processing variables.  The model is based upon the underlying physics of the process including the mixing of the anode materials, evaporation of the solvent, burning off of the pore formers, sintering of the remaining YSZ backbone and infiltrating with conducting particles.  The model then is used to predict the effective conductivity of the fuel cell anode as a function of conductor particle fraction.  These results are favorably compared to existing experimental results and demonstrate effective prediction with no adjustable parameters both in regard to the conductor fraction where conduction first takes place and the quantitative effective conductivity at realistic loadings.  We will also highlight other important characteristics such as the conductor particle coordination number.  Beyond the validation of the model, we also will demonstrate the models capability of predicting the effective conductivity as a function of important model parameters such as pore former and YSZ particle size as well as pre-infiltrate porosity.  Finally, we will discuss the opportunities to advance this type of modeling to incorporate other properties of the anode such as thermal expansion to provide design optimization.