Start-up zeros in on hydrogen fuel cells

Michael Lefenfeld and James Dye of Signa Chemistry wanted to make rooms smell better. Instead, they stumbled on a way that could make hydrogen fuel cells a practical reality.

New York City-based Signa says it has come up with a new--and fairly efficient--way to produce hydrogen, one of the vexing problems for boosters of the hydrogen economy.

Conceivably, the company's technology could be incorporated into fuel cells that could generate enough electricity to run a cell phone for a week, or a car in emergency situations. The company's techniques could also reduce cost and complexity for pharmaceutical manufacturers and petroleum refiners.

The key is sodium, the ornery alkali metal that bursts into sparks when dunked in water. The sodium/water reaction can generate hydrogen (along with other byproducts). But, because of the sparks and heat, industrial companies shy away from it.

Signa has devised a way to mix sodium with silica gel or crystalline silicon to create a powder that essentially strips electrons from the sodium molecules in advance and stores them. When water is introduced, the chemical reaction proceeds calmly. (The harvested hydrogen molecules in turn undergo a second reaction: Electrons are stripped from the molecules and get channeled into electrical power.)

Just as important, the powder generates hydrogen efficiently. More than 9 percent of a kilogram of the powder gets converted to hydrogen and little energy is lost through heat.

"You toss it into water and it just bubbles," said Lefenfeld in an interview. "It frees up the electron to make it readily available for the reaction. A lot of that heat (in a normal sodium-water reaction) comes from the stripping away of that electron."

Although it's a small company--it only has three full-time employees--Signa has begun to get its name around. It has delivered powders to chemical and drug manufacturers and is working with a fuel cell manufacturer to develop prototypes. It will announce a deal with a major chemical distributor soon. While the company's offices are in New York, a contract manufacturer in Buffalo, N.Y., is producing the materials--about 10 kilograms a day.

Hydrogen is the fuel of the future, or the next failed promise, depending on who you ask. Panasonic has started to conduct trials with hydrogen home-heating systems in Japan and Honda has obtained certification for a hydrogen car there.

Others, however, note that the expense and energy involved in making and storing the gas can outweigh the benefits. The most common method now involves electrolyzing pure water. Other methods, like the proton exchange membrane (PEM) fuel cell, one of the leading hydrogen vehicles, have yet to be perfected. But alternative vehicles for burning hydrogen, namely the solid oxide fuel cell, show promise.

"I have been and I'm still enormously skeptical about most of the solutions for alternatives. People say when hydrogen burns, it produces only water. Did you know that hydrogen is a greenhouse gas? Nobody thinks about it, right?" Arno Penzias, a partner at New Enterprise Associates, said in a recent interview. "But then the bigger problem is, how do you make the hydrogen?"

To gain a foothold in the market, Signa is taking the path of least resistance. It will first target a product--a powder that consists of sodium and crystalline silicon--at industrial chemical manufacturers who consume large quantities of materials, are intimately familiar with industrial chemical processes, and understand the promise (and pitfalls) of sodium.

"Pharmaceutical companies will take several steps to get around using alkali metals. Petrochemical manufacturers are the same," Lefenfeld said.

Fuel cells will follow later. Mostly, the company will partner with fuel cell makers to devise cells for smaller devices, such as phones or MP3 players. Currently, several companies have developed prototypes of methanol fuel cells and fuel cells that generate electricity by combining hydrogen with solid oxides.

Lefenfeld, however, says hydrogen will work better. Methanol is flammable, and oxide fuel cells require a catalyst, which invariably reduces the efficiency of a reaction. Hydrogen fuel cells can also deliver larger amounts of energy, say some. One company, Millennium Cell, is working on a prototype hydrogen fuel cell that could ship to notebook makers by 2007.

Methanol fuel cell advocates, on the other hand, have asserted that their products are safe and have been tested for several years now.

Hydrogen fuel cells produced with the company's powers could also run a car, although not particularly economically in the foreseeable future.

"I can see this being used when you run out of gas, for that emergency 50-mile drive," he said. "The material is not expensive, but it is not as cheap as gasoline."

The powder for fuel cells will consist of sodium mixed with a silica gel. While this mixture produces less overall hydrogen than the sodium/crystal silicon mix, the potential for impurities is reduced.

If it's so efficient, and the reaction uses well-known and understood materials, why did no one come up with this years ago? Lefenfeld, who is working on a PhD in chemistry at Columbia, admits that luck, and a general reluctance to work with these materials, helped.

"That side of the periodic table people tend to ignore," he said.

The two were trying to come up with an aerosol substitute for spraying things like fragrances, which are a combination of oils and water. "When you get fragrant oils on water, the difficultly is getting the fragrant oils off the water," he said. Alkali metals can do the trick, but the violent release of energy is a problem.

The first idea was to mix sodium with zeolites, a class of crystalline solids. Unfortunately, Zeolites are expensive. The group then came up with the idea of using porous silicon gels, which have similar attributes but are cheaper. The amount of hydrogen generated from the reaction prompted the group to start to examine fuel cells and industrial applications.