Semester

Summer

Date of Graduation

2006

Document Type

Dissertation

Degree Type

PhD

College

Statler College of Engineering and Mineral Resources

Department

Mechanical and Aerospace Engineering

Committee Chair

Ismail Celik.

Abstract

Fuel cells are one of the promising eco-friendly and efficient electricity generators for future energy infrastructure. Rigorous research has been underway for over a decade to develop fuel cell technology as a viable alternative to the conventional energy sources. Numerical modeling has played a prominent role in such research endeavors. Detailed multi-dimensional models reveal important information regarding the performance of a fuel cell but they are computationally intensive. Relatively simple zero- and one-dimensional models on the other hand average out the details that could be critical. The topic of this dissertation is a new strategy for modeling fuel cells which is not as complex as the multi-dimensional models but at the same time retains important details of three dimensional distributions inside the important components of a Solid Oxide Fuel Cell (SOFC). The objective is to develop a new reduced order model for transient three dimensional modeling of SOFCs. The reduction in complexity is achieved by using one-dimensional models for the gas channels and three dimensional modeling for solid and porous regions. This approach circumvents the problem of solving three-dimensional Navier Stokes equations inside the channels but still resolves the details inside the more important components, electrodes and electrolyte. Another unique feature of the new approach is the electrochemistry model which calculates the electric potential jump across the anode/electrolyte and cathode/electrolyte interfaces separately. The electrochemistry model is tested separately, validated and then incorporated into the SOFC model. The computer code for the model is developed on the foundation of the Navier Stokes solver, DREAM, developed by Dr. Ismail Celik and his co-workers and hence it is named DREAM SOFC. The new model has the advantage of faster run time for transient simulations compared to a complex three dimensional model while resolving almost as many details. This makes the new model more suitable for modeling multi-cell SOFC stacks consisting of as many as 50 cells. The computer code is first verified using the numerical results from literature and also a multi-dimensional fuel cell model FLUENT SOFC. Following the validation, parametric studies were performed to study the effect of parameters such as electrolyte thickness, convective heat transfer coefficient etc. which yielded interesting results. Numerical uncertainty in the results was found out to be small by means of Richardson extrapolation using computations on two grids. The temperature dependence of electrical conductivity of the SOFC materials was found to be making the current distribution more uniform in the co-flow configuration and more non-uniform in counter-flow configuration. It was shown that while thinner electrolytes give better power output, they produce highly non-uniform current distribution inside the SOFC. The start-up transients of a co-flow SOFC were simulated and it was observed that it takes about 30 min for the cell to reach steady state.

Share

COinS