A new catalyst based on iron works as well as platinum-based catalysts for accelerating the chemical reactions inside hydrogen fuel cells. The finding could help make fuel cells for electric cars cheaper and more practical.

Fuel cell researchers have been looking for cheaper, more abundant alternatives to platinum, which costs between $1,000 and $2,000 an ounce and is mined almost exclusively in just two countries: South Africa and Russia. One promising catalyst that uses far less expensive materials–iron, nitrogen, and carbon–has long been known to promote the necessary reactions, but at rates that are far too slow to be practical.

Now researchers at the Institut National de la Recherche Scientifique (INRS) in Quebec have dramatically increased the performance of this type of iron-based catalyst. Their material produces 99 amps per cubic centimeter at 0.8 volts, a key measurement of catalytic activity. That is 35 times better than the best nonprecious metal catalyst so far, and close to the Department of Energy’s goal for fuel-cell catalysts: 130 amps per cubic centimeter. It also matches the performance of typical platinum catalysts, says Jean-Pol Dodelet, a professor of energy, materials, and telecommunications at INRS who led the work.

The improvement, reported in the latest issue of the journal Science, is “quite surprising,” says Radoslav Adzic, a senior chemist at Brookhaven National Laboratory in Upton, NY, who also develops catalysts for fuel cells. The new material meets a benchmark for hydrogen fuel cells set five years ago that “we thought nobody would ever meet,” adds Hubert Gasteiger, a visiting professor of mechanical engineering at MIT. “For the very first time, a nonprecious metal catalyst makes sense.”

The INRS researchers’ key insight was finding a way to increase the number of active catalytic sites within the material–with more sites for chemical reactions, the overall rate of the reactions in the material increases. In previous work, the researchers had shown that heating carbon black (a powdery form of carbon similar to graphite) to high temperatures in the presence of ammonia and iron acetate created gaps in the carbon that are just a few atoms wide. Nitrogen atoms bind to opposite sides of these tiny gaps, and an iron ion bridges these atoms, forming an active site for catalysis.

To increase the number of these sites, the researchers used a commercially available form of carbon that already has a large number of similarly narrow pores. Filling these pores with a nitrogen-and-iron-containing material and then heating up the mixture resulted in the much improved reaction rates.

The catalyst is designed to work in proton exchange membrane (PEM) fuel cells, a type of fuel cell favored by automakers because it operates at relatively low temperatures and has high power density–that is, a relatively small fuel cell can produce enough electricity to propel a car. PEM fuel cells use catalysts at two electrodes. One catalyst splits hydrogen and the other promotes a reaction that combines protons and oxygen to produce water. The second reaction is more difficult to perform: in conventional fuel cells, platinum is used in both electrodes, but 10 times as much is needed on the water-producing side. The new catalyst replaces platinum on the water-producing side, eliminating almost all of the platinum in the fuel cell.

Recently, other nonprecious metal catalysts have been demonstrated in another type of fuel cell, called an alkaline cell, but these may not work in the acidic environment in PEM fuel cells. At the same time, many researchers are finding ways to reduce the amount of platinum needed, rather than replacing the material altogether. This could make fuel cells more affordable in the short term, although eventually, if fuel cells are to be used widely, a nonprecious metal catalyst will be needed, Adzic says.

Dodelet believes that while his group has “solved the problem” of increasing the activity of the catalyst, two more significant hurdles remain before it can be practical in fuel cells. First, the catalyst’s durability needs to be improved. After 100 hours of testing, the reaction rates decreased by half. Second, because the catalyst can only work as fast as the reactants are provided, the transport of oxygen and protons into the material needs to be improved, something Dodelet plans to leave to fuel-cell engineers. Adzic says that the first step toward addressing the materials’ durability will be closely studying the catalyst to better understand how it works.