If fuel cells are an inevitable addition to the U.S. electric-generation landscape -- and if through unitized, or distributed power, they have the potential to reshape the $225 billion generation and distribution industry -- then it's worth understanding basic technological and economic issues.

On the Wings of Power

Early Positions in Fuel-Cell Makers May Be Cheap, but Risky

The Wind -- and Water -- Beneath Power's Wings

Fuel Cells' Technology, Economics Shape Coming Market

Rushing to Get Off the Grid -- and Make Some Money

How fuel cells work isn't hard to understand, though for producers, refining the technology is still daunting. Think of fuel cells as batteries that keep producing energy as long as you supply them with a continuing, external source of fuel. They rely on electrolysis, which we've known about since Sir William Grove, the English scientist, used reverse polarity in 1839 to produce electricity.

Essentially, fuel cells rely on hydrogen for fuel, plus an anode and a cathode, and an electrolyte. But within that broad definition, there are five major competing technologies. There won't be a single winner; several will survive and prosper to meet specific generation needs. The big five:

• Phosphoric Acid: This is the best-established technology today, the method used in more than 200 commercial fuel-cell systems worldwide. It is relatively efficient (see more on efficiency, below), but is expensive to produce, in the neighborhood of $5,000 per kilowatt-hour, installed, and thus suitable only for large commercial installations. Think of a minivan or a medium-sized Dumpster parked next to your office, and you get a sense of scale. Researchers are pushing phosphoric acid technology hard and hope to get both the size of systems and their cost down sharply -- perhaps cutting the per-kilowatt-hour cost by two-thirds. That could move phosphoric acid systems into the reach of affluent homeowners.

• Alkaline: This is the granddaddy. NASA used onboard fuel-cell electricity generation in its Gemini and Apollo programs. It can be extremely expensive: The fuel-cell systems used in the space shuttle cost about $600,000 per kilowatt-hour. Those price tags mean alkaline systems are likely to remain interesting only to the military and other government agencies.

• Proton-Exchange Membrane, or PEM: Bingo! With more research focused on PEM systems than any other today, this is the hot bet. PEM systems are already approaching the size necessary for residential service and are expected to drop to the $5,000-$7,000 range for home-sized units. PEM units are also small and light enough to offer hope for powering cars and trucks, offer quick start up, and run cooler (60-10 C) than other systems. The high cost of the platinum catalyst needed by PEM systems was once a problem, but current units require less than a tenth as much platinum as earlier models.

• Molten Carbonate: Silent but hot, molten-carbonate fuel-cell systems are at the industrial-size applications end of the scale, thanks to both their capital cost and their safety requirements. They also produce, as a result of that high running temperature, steam and heat, which can be recycled in cogeneration to add efficiency. Indeed, in the end, molten-carbonate systems are expensive -- as much as $8,000 per kilowatt-hour -- but are likely to win the high-end market share.

• Solid Oxide: The most technically challenging of the five main fuel-cell technologies, these rely on zirconium oxide as an electrolyte. A sandwich of high-tech ceramics and metal foam, running at 1,500F, a solid-oxide system can achieve efficiency greater than 80%. Like molten-carbonate technology, it throws off a lot of steam and heat, which if recaptured can be used for cogeneration, further increasing efficiency, maybe to as much as 90%. For now, though, solid oxide looks like a future bet.

Beyond the technology used, other issues remain before widespread commercial adoption of fuel cells. Safety, for example: Systems that run hot are suspect in residential use. And given their fuel source, hydrogen, there is apprehension about indoor installations of fuel cells.

Major Fuel Cell Technologies
TYPE	POSITIVES	NEGATIVES	LIKELY APPLICATIONS
phosphoric acid	proven, reliable, relatively efficient	large, heavy, high capital cost	offices, industrial applications
proton-exchange membrane ("PEM")	small, light, potentially low capital cost, probably most promising	less proven, relatively low efficiency	homes, automobiles; portable
solid oxide	highly efficient, also produces re-usable heat, steam ("cogeneration")	large units, run very hot, for large-scale uses only	offices, industrial applications
alkaline	well-proven, well-understood, 30 yrs. of experience, high efficiency	aging technology, expensive	military, space; NASA loves 'em
molten carbonate	silent, extremely efficient, also produces re-usable heat, steam	high capital cost, runs very hot, large	industrial applications, ships
Source: TheStreet.com research

Efficiency and the Environment

Environmentally, all fuel-cell technologies are winners. At the tail end of the generating process, fuel cells produce only drinkable water, or sometimes a little carbon dioxide, as waste.

Not only is this benign waste output a huge improvement over the pollutants pumped into the sky by, say, a coal-fired power plant, their relatively high efficiency also means less thermal pollution, as well -- a big advantage in times of concern about global warming. And by moving away from large-scale coal mining, or extraction of oil and gas, and the subsequent refining and transport to traditional power plants, fuel cells offer hope of a much cleaner power-generation cycle.

Efficiency is increased. Our present grid is remarkably inefficient, running on average about 35%. That means only 35% of what starts out ever reaches customers. Not only is the conversion of stored energy in fuels such as coal and gas an inherently low-efficiency process, but also running electricity long distances over the grid is extremely wasteful, with substantial transmission losses. Fuel cells can offer about twice that overall efficiency, ignoring possible cogeneration gains but including transmission-efficiency improvements.

There are obvious reliability issues here, too. When the grid goes down, everyone's power goes down. In a unitized or distributed-generation system, the failure of a single unit affects only the user or users of that unit. That's "unitized" reliability -- localizing power interruptions.

For every location, moment-to-moment reliability is important, too. The newer fuel-cell technologies don't have long track records, but it looks as if observed system reliability on the order of losing power, on average, of only a few seconds a year is achievable. That's far better than the grid delivers today -- and likely, much better than the grid will deliver tomorrow.

Cost and the Yellow Sticker

Most of the cost of fuel-cell systems is "first cost," or capital cost. Operating costs are much lower: for a large residential fuel-cell system, including fuel, probably about $50 per month.

Bringing down that high first cost is the primary goal of the fuel-cell industry. The industry's Holy Grail is a $5,000-$10,000 five- to 10-kilowatt residential system that is no larger than an outside air conditioning compressor. I've seen several announcements of plans to ship such a unit in 2000 and 2001, but not much on the delivery side.

One big remaining obstacle is Underwriters Laboratory approval. The UL sticker is a de facto requirement for large-scale commercialization. UL won't comment on ongoing testing work, but it is generally understood that no approved fuel-cell systems are about to appear.