The Path a Proton Takes Through a Fuel Cell Membrane

Many experts believe that fuel cells may someday serve as revolutionary clean energy conversion devices for transportation and other portable power applications. Because they generate electricity by converting chemical hydrogen and oxygen into water, fuel cells generate energy much more efficiently than combustion devices, and with near-zero pollutant emissions.

Although NASA’s space program began using fuel cells decades ago to power probes, satellites and space capsules; there are still a number of design issues that engineers need to sort out before these devices can ubiquitously power civilization on Earth.

One challenge is to develop a relatively inexpensive and robust polymer membrane that effectively conducts protons. In a step toward achieving that goal, the Pacific Northwest National Laboratory’s (PNNL) Ram Devanathan and Michel Dupis are running computer simulations to understand how protons move through different polymer membranes.

When the duo ran their simulations on supercomputers at the National Energy Research Scientific Computing Center (NERSC) and the Environmental Molecular Sciences Laboratory (EMSL), they found that contrary to conventional wisdom protons do not take the path of least resistance when traveling through membranes. In fact, protons will often “mingle” with the sulfur and oxygen clusters along the way. The surprising results were published in a recent issue of Physical Chemistry Chemical Physics (PCCP).

"Whether you are using membranes for separation or water purification, you are interested in selective transport through the membrane," says Dupuis. "Specifically, we are interested in how changing the design would affect proton transport through the membrane."

What’s in a Membrane?

Generally speaking, most fuel cells consist of an anode (negative-side), a cathode (positive-side) and a proton exchange membrane (PEM) that separates the two-sides. PEMs are semi-permeable membranes designed to conduct only protons and block everything else. Because PEMs must be hydrated to transfer protons, researchers need to find materials that will continue to effectively conduct protons in a variety of environments like those with sub-zero temperatures, low humidity and high operating temperatures. It is also imperative that the membrane is robust enough to withstand the harsh, acidic environment inside of the fuel cell or battery.

Using supercomputers at NERSC and EMSL, the team simulated thousands of molecules of the short-side chain perfluorosulfonic acid membrane. They wanted to determine what would happen if they changed different aspects of the molecules used to construct the membrane.  The team ran simulations where they changed different aspects of the molecule, including the length of the side chain that dangles the sulfonate group into the water wire.

Conventional wisdom said that making the side chains shorter would make the central water channel larger and give the protons more room to move, allowing them to go faster. Surprisingly, the team found that the chain length did not matter. When the chain is long it folds back and doesn't block the channel. When the chain is short, it just stays out of the way.

“Conventional wisdom was that the protons would zip along the center of the water channel, where there were no sulfonate groups to slow them down. What we found was that the proton doesn't go through the highway, but rather goes on sightseeing trips mainly along the sulfonate groups," said Devanathan, a materials scientist at PNNL who led the study. “Because this membrane is essentially the conductor that generates electricity, we want protons to go fast through the material, but don’t want water and other molecules to go along with the proton.”

The team’s results compared well with experimental data collected on the perfluorosulfonic acid membrane. Now, Devanathan and Dupuis are conducting detailed studies to determine what influences protons bouncing between the sulfonate groups.  They are also examining how to transfer the protons effectively in the absence of water, perhaps with an ionic liquid.

“We’ve been using NERSC resources for a long time and it has been an invaluable resource for our research,” says Devanathan. “Our models require us to run in parallel on thousands of processors and we achieve this pretty easily on NERSC systems. Also, the NERSC tutorials and webinars have also helped us get our codes to run more efficiently on these supercomputers.”

This story was adapted from a research highlight published by the Pacific Northwest National Laboratory:
http://www.pnl.gov/science/highlights/highlight.asp?id=1194