This was no ordinary hidden treasure. It involved nanotubes of gold rather than pots and was more or less hidden in plain site. But oddly enough, for reasons that researchers debated for years, tiny bits of the precious metal in careful arrangements would display hidden value, or extra power. The gold, in this case, is used as a catalyst to clean carbon monoxide (CO) from hydrogen fuel.

Fuel cells make energy from hydrogen. Unfortunately, hydrogen produced by the usual process contains large amounts of CO which, if not removed from the hydrogen, impairs the function of a fuel cell. Research has shown that gold nanoparticles on a support with a high surface area are good catalysts for the room-temperature oxidation of CO to CO₂, but there has been controversy surrounding the function of the gold in this process and the role of the support. Now, researchers at the UW-Madison College of Engineering have developed a "membrane reactor" that unravels the mystery by allowing examination of the gold catalyst without its support.

The team led by Chemical and Biological Engineering Steenbock Professor James A. Dumesic had a clever idea. The researchers took a whisper-thin plastic membrane made of polycarbonate containing pores with a diameter of 220 nm. After the surface was specially prepared, gold was deposited onto the membrane. When the precious metal settled onto the walls of the tiny pores, it formed pure gold nanotubes. A subsequent etching process selectively removed the upper layer of the polycarbonate membrane, so that the gold nanotubes protruded from the surface. The researchers stretched this membrane between two chambers, one of which is used to admit gases, the other liquids. Indeed, just like gold nanoparticles, the gold nanotubes catalyzed the reaction of CO and O₂ to form CO₂. Systematic examination of the reaction revealed the following: The catalytic activity is increased by the presence of water in the tubes, and is raised still further if the pH level is raised (the solution is made more alkaline). It is clear that hydroxyl groups (OH-), which come to the gold surface from basic materials or the dissociation of water molecules, facilitate the interaction between CO and O₂, which seems to result in CO₂ and peroxidic intermediates.

"We showed that the reason they work so well is, if they are in the presence of water, or OH groups, then they really have high rates for CO oxidation," Dumesic says. "So we think then that the role of these supports is to provide OH hydroxyl groups to the gold and that is why they work."

The theory is supported by the fact that the reaction rates for supported gold nanoparticles strongly depend on the type of material used for the support. Gold nanoparticles on oxide-containing supports in a damp atmosphere are most active, which fits the theory, since hydroxyl groups also occur under those conditions.

With hydrogen peroxide instead of oxygen as the oxidizing agent, the reaction runs better still, presumably because the bond between the two oxygen atoms in the former is easier to break.

As the U.S. and world move toward hydrogen as a source of energy, the new knowledge points the way toward making more efficient catalysts for hydrogen production.