Date of Award


Document Type

Open Access Dissertation


Chemical Engineering

First Advisor

John W. Van Zee


An analysis of liquid water transport and removal in Proton Exchange Membrane Fuel Cells (PEMFCs) as affected by different membranes and the geometry and surface roughness of bipolar plates on is presented. Four topics are considered. First, the channel dimension and shape of various flow fields have been shown to affect the cell performance and the uniformity in the distributions of current. Typical variations in the channel width, height, and undercut that may occur with manufactured metal plates are studied. These sample-to-sample variations and distributions are studied and compared with laboratory-scale graphite plates. The goal of the work is to provide fundamental information that can be used to develop tolerance and design principles for manufacturing metal bipolar plates. Secondly, the effect of roughness was studied experimentally to characterize liquid water droplet movement that may result from significant liquid droplet accumulation on the surface of the flow channel on either side of the membrane. Liquid water droplet movements were analyzed by considering the change of the contact angle as a function of flow velocity. Also, various stainless steel surfaces having different surface roughness were used to determine the relationships between flow rate and the contact angles. The pressures drop and channel characteristics are presented through dimensionless analysis and with a force balance equation. The result shows that the surface roughness has a great impact on pressure drop and liquid droplet removal. A unique relationship between surface roughness and onset of droplet movement has been discovered that may describe the relationship between surface properties and liquid droplet movement on any surface in the PEMFC.

For the third aspect, a flexible low-cost technique for determining the current distribution was developed and used to understand the transport of water across a PEMFC for various membrane and cell geometries. This aspect built on the knowledge that non-uniform current distributions in PEMFCs result in local over-heating, accelerated ageing, and lower power output than expected. Liquid water transport is also known to qualitatively correlate with these distributions, especially when a fuel cell experiences water flooding. Present-day methods to measure these current distributions may significantly affect the flow path, break up diffusion media, and are usually very expensive. In this dissertation, a cost-effective method of mapping the current distribution in a cell was developed which overcomes many of the above limitations. A current distribution board was designed to add minimal internal resistance as well as minimize the disruption of the flow pattern when used in a cell.

These current distribution boards were used to study the forth aspect of this dissertation: the quantitative correlations between ex-situ measurements of water diffusion coefficients and electro-osmotic drag for different membrane materials, in-situ measurements of water transport, and numerical predictions of the current and water distributions as verified by water balances. The ex-situ measurements were shown to provide the parameters for the 3-D PEMFC mathematical model. The improved knowledge of this model proves to provide a better understanding the water management of the cell.

In addition, different membrane materials were used to study the effect of water transport properties on overall fuel cell performance. In this aspect, the alternative material (hydrocarbon type membrane) was studied and compared with standard membrane material (perfluorinated sulfonated copolymer, Nafion®). Current distribution behaviors of two different membranes were studied in the different operating condition of fuel cell such as humidity of inlet gas to understand the effect of water transport properties from different membrane material. Water balances experiment was also used to analyze water transport for these membranes.