| The objective of this dissertation work is to numerically simulate three-dimensional aspects of flow in proton-exchange-membrane (PEM) fuel cells. As of today, PEM fuel cells cannot be satisfactorily run for a sustainable period. This research provides internal information on flow, gas composition, heat transfer, and water management that will help to overcome the current operational problems. A computational fluid dynamics model is used to predict current density distribution in two dimensions on the membrane for both straight channel and full-cell fuel cells. The flow domain includes two main flow channels, a membrane, and two diffusion layers. The electrochemical reactions, exothermic effects, phase change of water, diffusion through the thin membrane, etc. are essential elements of fuel cell simulation. These different aspects of fuel cell are incorporated mainly by developing new source terms in transport equations. In this work, the effects of diffusion layer, membrane thickness, operating cell voltage, inlet humidity, and temperature distribution on PEM fuel cell performance are studied and predicted results are compared with the available experimental data. Predictions clearly show the conditions where insufficient water lowers the membrane conductivity and yields low currents as well as the conditions where excess water leads to flooding of the electrode and low currents due to a decreased reaction area. Predictions also indicate that water evaporation by temperature rise inside fuel cell dehydrates the membrane and drops its performance. This work is the first to clarify characteristic three-dimensional flow, species transport, and current density distribution inside fuel cell and these numerical methods can easily be incorporated in commercial flow solvers to decrease the design tune of fuel cells. |
