Solving General Motors' problems may be nothing compared with storing hydrogen on automobiles.
Robert Stempel has experience with both.
The former General Motors chairman is president of Energy Conversion Devices in Troy, Mich., a small supplier of advanced batteries to GM's EV1 electric car. With patents on nickel metal-hydride batteries, solar cells and compact-disc coatings, the eclectic research company has set its sights on storing hydrogen gas as a solid.
Solid-state hydrogen storage is considered a key steppingstone toward superclean fuel cell-powered vehicles.
A fuel cell consumes hydrogen in an electrochemical reaction that produces electricity to power a drive motor.
In most hydrogen-powered prototypes shown to date, pure hydrogen is stored as either a supercooled liquid that must be kept chilled to below minus 400 degrees Fahrenheit or as a gas packed in at up to 5,000 pounds per square inch.
Either form requires exotic tanks built to withstand high pressures, heat transfer and puncture. Vehicle range is limited by the packaging constraints on these bulky tanks, while the fueling rigs must be equally complex to handle refills safely.
DaimlerChrysler plans to sidestep this issue in 2004 by powering its first production fuel cell vehicle with methanol. An on-board reformer separates the hydrogen from the methanol in a process that produces some pollutants.
Solid-state hydrogen storage removes one major hurdle to hydrogen-powered cars. The technique relies on hydrides, or metal alloys, that soak up hydrogen molecules in their crystalline structure. The hydrogen is released when heat is applied to the alloy. A tank displacing 30 gallons and containing 220 pounds of hydride could give a fuel-cell vehicle a 300-mile range, Stempel claims.
By comparison, 30 gallons of gasoline weighs 180 pounds and gives a 2000 Ford Taurus SE a maximum range of 840 miles.
But hydrides really are 'a challenge to the on-board reformer,' Stempel says. 'You don't have to have a miniature chemical plant on board the vehicle,' he said.
Energy Conversion Devices is developing a fuel tank design in which the hydrogen is pumped into a tank containing racks of canisters filled with powdered magnesium alloy. The system operates at relatively low pressures - around 350 pounds per square inch.
The gas is released from the hydride at a temperature of 572 degrees Fahrenheit, heat that is generated on startup by an igniter, or 'catalytic combuster,' in each canister. Up to 20 percent of the hydrogen is consumed in starting the release process. Later, the heat is supplied at no cost from the reaction in the fuel cell itself.
Stempel says the company has put its design through 600 fill-and-release cycles and is going for 1,000 cycles.
The auto industry has viewed hydride storage with skepticism because of its inefficiency. Existing hydrides can store 2 percent of their own weight in hydrogen. Stempel says fiddling with the recipe of the magnesium alloy has allowed Energy Conversion Devices to bump that figure up to 7 percent or about 12.8 cubic feet of hydrogen for every pound of hydride.
'When we hit 7 percent, that got the industry's attention,' Stempel says.
GM considers the technology promising enough that its Precept fuel cell concept car, shown at the recent Detroit auto show, featured a theoretical mock-up of a hydride storage system.
Byron McCormick, co-executive director of GM's Global Alternative Powertrain Center, says the automaker is working with suppliers to leap the remaining technical hurdles. They include trying to speed up the rate at which the hydride soaks up and releases the hydrogen to cut the wait at the pump and shrinking the size of the tanks.
'The only unknown is whether we have materials that can do that,' McCormick says.
Cost is another issue. At $4 per pound, the magnesium hydride alone would add up to $1,000 to the vehicle's cost. That helps explain why Stempel wants to be a hydride supplier if the technology takes off. But that role carries with it another technical challenge: how to safely handle the magnesium powder.
On-board tanks would be designed to resist puncture and fires, but the magnesium dust would be dangerously explosive during processing.
Says Stempel: 'We're very, very cautious.'