Imagine you ran a power plant from your basement. It would be small enough to fit in one room, powerful enough to supply the whole neighborhood. It would be super-quiet so you could sleep at night. It would run pollution-free because, of course, you wouldn't want noxious fumes seeping into the den.
That's the promise of fuel-cell technology. In the next decade, blocks of these small units could be fitted together to run the electrical systems of office buildings and schools. Instead of relying on utilities for power, large companies could provide their own.
''It's becoming very tricky and expensive for the electric industry to increase its capacity,'' says Kevin Krist, principal technology manager at the Gas Research Institute in Chicago. ''It's expensive to commission central power plants.'' They require long lead times. ''It's easier for people to have these small facilities.''
Laboratories around the world are trying to find the design that will make fuel cells a commercial success. ''If it's going to happen, it's going to happen in the next five to 10 years,'' says Scott Barnett, a fuel-cell researcher at Northwestern University in Evanston, Ill.
Unlike, say, cars and coal plants, which burn fuel to make power, fuel cells produce electricity directly by combining the fuel with air electrochemically. Think of a fuel cell as a special kind of battery. It has a positive and a negative electrode. One is fed with non-polluting hydrogen-based fuel, while the other is supplied with air.
In between is an electrolyte or membrane. Oxygen from the air carries electrons across the membrane and then reacts with the fuel. Electrons are released and create an electric current. Connect enough of the cells together and you have a power plant.
Cutting cost and size
Dr. Barnett is working on one of the newer versions of the technology known as solid-oxide fuel cells. Barnett's innovation has been to find ways to make the cells' ceramic-based membranes thinner and more efficient. His technique involves electrically charging argon gas atoms and smashing them against a piece of yttria-stabilized zirconia. (That's the same zirconia used to make fake diamonds.) Atoms knocked off the zirconia by the argon are collected to form the thin membrane. The technique allows Barnett to make fuel cells that are 10 to 20 times thinner than older solid-oxide fuel cells.
This membrane slimming is important, because it allows the fuel cell to operate at 600 to 700 degrees C. That's considerably less than the 1,000-degree temperatures of first-generation solid-oxide fuel cells. More importantly, it's in the range where several kinds of steel could withstand the heat. Thus, researchers foresee the time when the parts connecting the fuel cells could be made out of steel rather than the much higher-priced materials used today.
''The main thing right now is the cost,'' says Nguyen Minh, lead scientist for AlliedSignal's fuel-cell program in Torrance, Calif. Unless scientists can bring it down significantly, the technology won't make it out of the lab.
Barnett estimates that it costs the average utility $1,000 to install a kilowatt of electrical generating capacity with conventional technology. With today's fuel cell, it costs three times that amount. Thus, even proponents of the technology believe that commercialization is at least five years away.
Experiments continue with other types of fuel cells, such as phosphoric acid, alkaline (used on US spacecraft), molten carbonate, and various polymers.
For example, researchers at NASA's Jet Propulsion Laboratory announced recently that they have developed a fuel cell with a polymer membrane that uses a solution of water and methanol as fuel. The electrodes are made of carbon and a metal such as platinum, which acts as a catalyst for the electrochemical reactions. The researchers say using a liquid fuel greatly simplifies the fuel cell's design and reduces its size.
Costs can be squeezed further by developing new polymers to use as membranes. Researchers at Rennselaer Polytechnic Institute (RPI) in Troy, N.Y., have just received a patent for a new polymer membrane that could substantially reduce fuel-cell costs.
Currently, the most widely used membrane for fuel cells is made by DuPont. Known as Nafion, the material costs $70 a square foot. For a 10 kilowatt fuel cell, the cost of the membrane alone would run $2,800.
With their new polymer, Gary Wnek and his team at RPI could bring the membrane price down to around $1 a square foot. The problem: The new material isn't compatible with commercially available electrodes. ''Now we're focusing on new electrodes,'' says Dr. Wnek, chairman of the school's chemistry department.
In addition to membranes and electrodes, another approach to cutting costs is to find new catalysts for the cells' chemical reactions. ''Platinum is about the best material to do these kind of reactions,'' Wnek says. But as any jeweler will confirm, platinum isn't cheap.
Westinghouse has set up small, 25-kilowatt test plants around the world using high-temperature solid-oxide fuel cells. This is older but proven technology. One Japanese utility operated a Westinghouse plant more than 7,000 hours, a world-record for high-temperature fuel cells. Two Westinghouse single cells continue to operate after six years of service.
''We are aiming toward commercialization at the end of this decade,'' says Stephen Veyo, a manager of solid-oxide fuel-cell projects at the Westinghouse Science and Technology Center in Pittsburgh.
Companies are interested in the technology because it is more than twice as efficient as automobile gasoline engines, causes virtually no pollution, and runs very quietly.
If researchers find a way to cut costs, fuel cells could power a 21st-century school from the basement with no noise, no noxious fumes.