Optimum design of polymer electrolyte membrane fuel cell with graded porosity gas diffusion layer

• Published in: International journal of hydrogen energy Vol. 41; no. 20; pp. 8412 - 8426
• Main Authors: Zhang, Y., Verma, A., Pitchumani, R.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.06.2016
• Subjects: Channels; Current density; Current density distribution; Fuel cell reliability; Fuel cells; Gas diffusion; Graded porosity electrode; Local current; Mathematical models; Membrane durability; Optimization; Porosity; Proton exchange membrane (PEM) fuel cell; Current density distribution; Membrane durability; Graded porosity electrode; Fuel cell reliability; Proton exchange membrane (PEM) fuel cell; Optimization
• ISSN: 0360-3199; 1879-3487
• DOI: 10.1016/j.ijhydene.2016.02.077

Variation in the local current density along a polymer electrolyte membrane (PEM) fuel cell often causes sharp temperature and stress gradients that affect the membrane durability and service life of fuel cells. Towards minimizing the variation in local current density, and potentially improving fuel cell reliability, this paper explores the use of functionally graded porosity in the gas diffusion electrode layers along the flow direction. A computational model for fuel cell is used together with a numerical optimization method to determine the optimum porosity distribution along the length of the channel, with the objective of maximizing power density while limiting the current density variation. The optimum porosity distribution and the corresponding local current density distribution are compared and discussed in detail for different operating conditions. Experimental studies are conducted to measure the local current density distribution in a fuel cell with a segmented current collector and to evaluate the effects of graded porosity distribution in the gas diffusion layer. It is shown that use of an optimally graded porosity distribution improves the uniformity in current density along the length of the channel by up to a factor of 10, while maximizing the power density.