Figure 2 shows the basic design of a typical PEM fuel cell.  At the anode, the hydrogen molecules disassociate into H+ ions which become hydrated with a water molecule (H₃0⁺) and migrate through the membrane to the cathode, and electrons, which are removed to power an electrical device. 

At the cathode, oxygen molecules react with the returning electrons and the migrating  H₃0⁺ ions to form water.  A theoretical electric potential of 1.2 volts is generated from these two reactions. This potential difference drives the current through the external load making the fuel cell a source of power.

In the cell, the membrane acts as both separator and electrolyte.  The conductivity of this membrane is directly proportional to its hydration state.  During operation, the effect of proton migration causes water molecules to move from the anode to the cathode resulting in membrane dehydration on the anode side and flooding on the cathode side (additional water is produced at the cathode by the reaction).  Therefore, water needs to be supplied to the anode and removed from the cathode  

1.  DEVELOPMENT OF MEA'S

Recent development of the PEM fuel cells has led to many studies of platinum (Pt) supported on gas-diffusion electrodes. High catalytic activity requires manufacturing a highly-dispersed-Pt-loaded carbon electrode.  The target for Pt loading is around 0.1-0.2 g-Pt/kW.  At this level the PEM system will be cost comparative to the internal combustion engine.  Figure 3 shows the typical and ideal Pt distributions at the interface between the membrane and the carbon electrode.

A new method to deposit a thin platinum layer on a carbon substrate of the electrode is being developed by our research group. The process consists of electrodeless deposition of platinum onto a carbon substrate by in-situ reduction of an ionic platinum solution with ethanol.  Figure 4 shows a scanning electron micrograph of a carbon substrate completely covered with the platinum deposit. MEAs using the proposed platinum deposition method are currently being developed in our laboratory.

  2.  WATER AND HEAT MANAGEMENT  

During operation, water molecules are carried from the anode side to the cathode side of the membrane by electro-osmosis (proton migration), and if this transport rate of water is higher than that by back-diffusion of water, the membrane will eventually become dehydrated and too resistive to conduct high current.   Consequently, to prevent membrane dehydration a sufficient amount of water must be added to the anode stream.  Figure 5 shows humidification designs that can be used for PEM fuel cells. Proper thermal management, such as heat removal is also needed to achieve optimal water.

  3.  DEVELOPMENT OF GAS DISTRIBUTOR  

The effects of the flow distributor on the performance of the PEM fuel cells are very important. The gas distributor needs to be optimized to provide adequate gas transport, water supply to the anode, and water removal from the cathode.  Figure 6 and Figure7 show the conventional and proposed (Interdigitated) gas distributor designs for PEM fuel cells. The performance improvements using the interdigitated flow distributors are shown in Figure 8 and Figure 9.  

Problems with the Conventional Design  

Gas transport limitation by diffusion: The reaction rate is limited by the diffusion rates of hydrogen and water through the anode and oxygen through the cathode from the channels to the catalyst sites.

Water flooding in the cathode: Assuming that oxygen (air) coming into the cell is not saturated, the difference between the partial pressures of water vapor inside the electrode and in the flow channels causes water that is generated at the catalyst sites to be transported out of the porous electrode.  However, as the gas transverses the channels, it becomes saturated at some point. Beyond this point there is no longer a driving force to remove water from the inner layers of the electrode.  As a result the cathode becomes partially flooded and under utilized.  

Advantages of the Interdigitated Design  

Changes the transport mechanism: The reactant gases are forced to flow into the porous electrodes in order to exit. This design, in effect, has converted the transport of the reactant/product gases to/from the catalyst layers from a diffusion mechanism to a convection mechanism with a greatly reduced gas diffusion layer over the catalyst sites.

Liquid water removal: The shear force of this gas flow helps remove most of the liquid water that is entrapped in the inner layers of the electrodes, thereby significantly reducing the electrode flooding problems.  

4.  INCREASING INTERFACIAL AREA  

To achieve optimal performance in a fuel cell, the interfacial surface area between the electrode and the membrane must be maximized.  Surface roughening of the electrode layer has been shown to be an effective method of improving a cell’s performance.  However, more structured, organized and controlled methods of increasing interfacial reaction surface area are needed.  Our research group is currently investigating the gas plasma etching process as a potential technique of providing a quantified increase in the interfacial area.

  ACKNOWLEDGEMENTS  

This research is supported by the University of Kansas Graduate Research Funds, K*STAR/NSF-EPSCoR program, NSF, KU-ERC, Los Alamos National Laboratory, and General Motors.  

STUDENTS INVOLVED IN THIS RESEARCH

 

Graduate:

Haitao Huang, M.S, ‘96

David Wood, M.S., ‘97

Jung Yi, Ph.D., ‘98

Ganesh Venkatasubramanian, M.S., present

Wensheng He, Ph.D., present

Dilip Natarajan, Ph.D., present

Mack Knobbe, M.S., present

Undergraduate:

Jeff Heidrick, ‘95

Jeff Doris, '95

Sara Sawyer, ‘95 

Jason Swink, ‘95

Robert Babst, ‘96

Francis Orzulak, 96

Tat Yan, ‘96

Stephen Weller, ‘97

Jason Voogt, 97

Matt Byrne, ‘97

Cheok Har, ‘97

Darin Bowman, ‘97

Chris Polonchek, ‘97

Carrie Nelson, ‘97

Mack Knobbe, ‘98-99

Adam Tobia, ‘98-99

James Stork, ‘98

Michael Zalvis, ‘99

Minh Vu Nguyen, present