Two-Dimensional PEM Fuel Cells Modeling using COMSOL Multiphysics

A fuel cell has been considered as an efficient and clean alternative power source for automobile industry since the energy crisis forced people to find a substitution for fossil fuels. A proton exchange membrane fuel cell, also known as polymer electrolyte membrane fuel cell (PEMFC), becomes a prime candidate for applications in vehicles because of its following features: the PEMFC operates at a relative low temperature (less than 90°C); the PEMFC can start quickly; the PEMFC has a higher current density due to thin membrane electrodes assembly (MEA) compared to other types of fuel cells; and there are no corrosive fluid hazards since there is no liquid electrolyte present in the PEMFC. Nevertheless, the wide application of a PEMFC is limited due to high capital cost, fuel availability, and durability etc. The difficulty of maintaining suitable thermal management and water management also affects the fuel cell performance significantly. For example, too much water produced on the cathode side will fill the pores of gas diffusion layers (GDLs), and therefore block the diffusion of reactants to reach the catalyst layer. Too little water on the anode side will dry up the membrane so that protons cannot migrate through it. Both cases result in a decrease in the cell output power. Experimental research and numerical simulation have been used in fuel cell design in order to improve the performance of fuel cells. Experimental data is useful to validate the models. The computational models are efficient in predicting the cell performance under a variety of design parameters. Fuel cell models can be classified into 1D, 2D and 3D according to dimensions. The accuracy of 1D model (Springer et al.,1991), (Gurau et al., 2000) is sacrificed due to some assumptions made in order to simplify the problem to 1D. A 3 D model simulates the reactant gas flow in the directions along the flow channel and perpendicular to the flow channel simultaneously, which results in more accurate results but requires longer computational time and larger computing capacity facility (Haralldsson & Wipke et al., 2004) (Berning et al., 2002). A 2D fuel cell model (Siegel et al., 2003) (Biyikoglu, 2005) (Hwang, 2006) combines the benefits of 1D and 3D models and gains its popularity in PEM fuel cell modeling due to its higher computational efficiency compared to 3D models and better simulation accuracy compared to 1D models. A two-dimensional mathematical model of a PEM fuel cell can be conducted in two different modes: parallel or perpendicular to the gas flow direction in the gas channel