(631e) Validation of a Phenomenonological Steady-State Model for Solid Oxide Fuel Cell (Sofc)

Several models have been proposed to address the operating characteristics of tubular Solid Oxide Fuel Cells (SOFC). Most models consider: mass/species and momentum conservation in the gas flow channels, diffusion of the reactants and the products to the reaction sites, energy conservation, and temperature dependent ohmic resistance. In this talk, we will present our modeling work and the validation of our model on a commercial SOFC. We will focus on the various physical phenomena that occur in a cell and point out their importance vis a vis a predictive model for SOFC under certain operating conditions.

The data that supports the modeling was collected from a counter-flow anode-supported tubular SOFC over a wide range of temperature (700-850 oC) and flow (21-51 ml/Min). The complexity of the model was gradually increased by observing the deviation of the simulated results compared to the experimental values. It is observed that if the momentum conservation equations are not taken into consideration, there may be substantial error in the calculation of the concentration profiles in the gas flow channels, especially in the cathode gas flow channel due to the non-uniform velocity profile in the channels. This error is reflected in the simulated I-V characteristics of the system. It was also observed that diffusion through the electrodes play a very significant role in all the operating conditions. Modeling of oxygen ion conduction through the electrolyte becomes necessary to match the experimental results in a wide temperature range. Ionic conduction through the YSZ electrolyte was modeled by Nernst-Planck equation coupled with Poisson equation which yields the concentration profile of oxygen ions from the cathode Triple Phase Boundary (TPB) to the anode reaction sites. Thus a full cell model of SOFC was developed considering all the significant physical phenomena that play a critical role in determining the resultant power-density of the system. The predictive capability of the model will be demonstrated in the feasible operating range of the cell. Special solution strategies that were used to solve the inherently stiff nonlinear differential equations will also be discussed.