Jump-starting the hydrogen economy

by J. William Bell

With computing resources from XSEDE, engineers at Ohio University explore ammonia as a source of hydrogen for tomorrow’s fuel cells.

For the fuel cell industry, ammonia has all kinds of things going for it. It’s abundant—the U.S. Geological Survey reports that more than 130 million metric tons were produced worldwide in 2009. The pipelines and trucks to ship ammonia around the country are already in place. It releases nothing but nitrogen and water vapor into the atmosphere when used for fuel cells. And, according to the Iowa Energy Center, it’s at least as safe as gasoline when used as a transportation fuel.

But many have considered ammonia a dead end. It requires a lot of heat to break the hydrogen, which is what most fuel cells generate their energy with, out of the ammonia. By-products of this process can also foul the fuel cells and reduce their efficiency.

With XSEDE resources at NCSA and the Pittsburgh Supercomputing Center, a team of engineers at Ohio University’s Fritz J. and Dolores H. Russ College of Engineering and Technology is jump-starting the prospects of ammonia-based fuel cells for cars and other applications. XSEDE is a collection of advanced digital resources around the country, a single virtual system that scientists can use to interactively share computing resources, data, and expertise. XSEDE is led by NCSA.

The team sees ammonia-based fuel cells as a way of shifting the United States, and the world, away from petroleum and other fossil fuels and toward a hydrogen economy. That economy would require hydrogen that is “cheap and produced at the point of use. Ammonia can provide that,” according to Damilola Daramola, a post-doctoral research associate working with Russ professor Botte.

Matt Simmons, who before his death was a member of the Council on Foreign Relations, agreed. Simmons was a long-time advocate for solutions to “peak oil,” the point at which the world begins consuming more oil than it can extract and refine.

In a 2010 report, he said “NH₃ [or ammonia] is the only realistic energy solution that makes sense.”

The Ohio team’s results have been published in Computational and Theoretical Chemistry. Their approach to breaking the ammonia down for use in fuel cells has also been licensed by the company E3 Clean Technologies for possible use in cleaning up wastewater. (Ammonia, by way of urine, is a major component of wastewater that has to be removed.)

1.5 gets you 33

An ammonia-based fuel cell relies, like the vast majority of fuel cell designs, on hydrogen. Using a small electric current and a catalyst to drive the chemical reaction, the hydrogen atoms are split into protons and electrons. The protons pass through a selective membrane, while the electrons are forced through a separate external circuit to generate electricity. On the far side of the membrane, the electrons and protons combine with oxygen to form water. In the case of ammonia, a harmless gas like nitrogen is a by-product.

In the case of an ammonia-based fuel cell, however, the hydrogen is created by “cracking” the ammonia into its component parts. Unlike hydrogen, ammonia does not have to be transported under pressure, reducing the risk of explosions and other mishaps when carrying the fuel on board a car.

Older designs for cracking ammonia were inefficient.

However, Madhivanan Muthuvel, a research assistant professor in Botte’s lab at Ohio University’s Center for Electrochemical Engineering Research, told The Columbus Dispatch that the team’s approach addresses that issue. 1.55 watt-hours of electric energy, using Botte’s patented method, yields a gram of hydrogen from ammonia. That hydrogen can then produce 33 watt-hours of electricity from a fuel cell.

One kilogram of hydrogen produced this way using ammonia costs about 90 cents. To produce that kilogram of hydrogen using water runs about $7. And that 90-cent kilogram of hydrogen can deliver roughly the same amount of energy as a gallon of gas.

A Goldilocks reaction

The Botte team’s work has an experimental element. They actually build “electrolyzers” that zap ammonia into its component parts. Those are used to test the overall efficiency of the system and how changes to the system—changing the elements used to make the catalyst that drives the reaction—impact that efficiency.

The computational element on XSEDE supercomputers is crucial as well.

“The computation strengthens our arguments by telling us why the reactions behave as they do,” explains Daramola. Not to mention the fact that their computational experiments are much less expensive to run than the equivalent lab-based experiments would be.

The team models the molecules that form and are then further broken down in the process of sheering hydrogen from ammonia molecules. Specifically they consider the strength of the bond between these intermediates and the platinum catalyst that drives the interaction. They also look at the orientation of those molecules as they interact with the catalyst, as well as how much and in what ways individual atoms within the intermediates are moving.

They then rank the intermediates in terms of which are most likely to adhere to the platinum, which in turn tells them which are most likely to pollute the catalyst over time and impinge on how well it works.

Understanding these features allows the team to look for a Goldilocks reaction that produces the most hydrogen while degrading the catalyst as little as possible. It also means they can begin to explore platinum alloys—platinum mixed with elements like iridium or rhodium—that might be more efficient or cost less money.

“You need the intermediates to stick to the surface of the catalyst and become converted to the desired product. But you don’t want to make it too sticky such that it adheres too strongly to the surface or not sticky enough such that the conversion will not occur,” Daramola says.

The Ohio University team runs their simulations on the Blacklight system at the Pittsburgh Supercomputing Center and the Ember system at NCSA.

Project at a glance

Team members
Gerardine Botte
Damilola Daramola
M. Muthuvel
Alex Miller
John Goettge
R. Palaniappan
Luis Diaz-Aldana
Dan Wang

Funding
U.S. Army Construction Engineering Research Laboratory

For more information
http://www.ohio.edu/ceer/

Disclaimer: Due to changes in website systems, we've adjusted archived content to fit the present-day site and the articles will not appear in their original published format. Formatting, header information, photographs and other illustrations are not available in archived articles.