HURDLE 2: Storage
At room temperature and pressure, hydrogen's density is so low that it contains less than one-three-hundredth the energy in an equivalent volume of gasoline. In order to fit into a reasonably sized storage tank, hydrogen has to be somehow squeezed into a denser form.

LIQUEFACTION: Chilled to near absolute zero, hydrogen gas turns into a liquid containing one-quarter the energy in an equivalent volume of gasoline. The technology is well-proven: For decades, NASA has used liquid hydrogen to power vehicles such as the space shuttle. The cooling process requires a lot of energy, though-roughly a third of the amount held in the hydrogen. Storage tanks are bulky, heavy and expensive.

COMPRESSION: Some hydrogen-powered vehicles use tanks of room-temperature hydrogen compressed to an astounding 10,000 psi. The Sequel, which GM unveiled in January 2005, carries 8 kilograms of compressed hydrogen this way-enough to power the vehicle for 300 miles. Refueling with compressed hydrogen is relatively fast and simple. But even compressed, hydrogen requires large- volume tanks. They take up four to five times as much space as a gas tank with an equivalent mileage range. Then again, fuel cell cars can accommodate bigger tanks because they contain fewer mechanical parts.

SOLID-STATE: Certain compounds can trap hydrogen molecules at room temperature and pressure, then release them upon demand. So far, the most promising research has been conducted with a class of materials called metal hydrides. These materials are stable, but heavy: A 700-pound tank might hold a few hours' fuel. However, exotic compounds now being studied could provide a breakthrough to make hydrogen storage truly practical. "High-pressure tanks are a stopgap until we can develop materials that will allow us to do solid-state storage efficiently," says Dan O'Connell, a director of GM's hydrogen vehicle program.

HURDLE 3: Distribution
Even in portable form, hydrogen is a tough substance to move from place to place. It can embrittle steel and other metals, weakening them to the point of fracture.

CLEAN FUEL: This fueling station in Burlington, Vt., uses electricity to convert water into hydrogen for powering fuel cell cars. It is part of a Department of Energy program for testing alternative fuels in colder climates.

TRUCKING AND RAIL: Currently, most hydrogen is transported either in liquid form by tankers or as compressed gas in cylinders by trailers. Both methods are inefficient. Trucking compressed hydrogen 150 miles, for instance, burns diesel equivalent to 11 percent of the energy the hydrogen stores. It also requires a lot of round trips: A 44-ton vehicle that can carry enough gasoline to refuel 800 cars could only carry enough hydrogen to fuel 80 vehicles.

PIPELINES: One way to avoid this endless back-and-forth would be to send the hydrogen through a pipeline. About 700 miles of hydrogen pipelines now operate in the States, generally near large users such as oil refineries. The longest in the world is a 250-mile line between Belgium and France. Treating pipelines to protect them from embrittlement and high pressure makes them expensive up front-about $1 million per mile. But once built, they are the cheapest way to deliver high volumes of hydrogen.

LOCAL PRODUCTION: Given the difficulty of transporting hydrogen, why not just make it where you need it? That's what's done at roughly half the 36 hydrogen fueling stations currently operating in the U.S. Four rely on natural gas; the rest use electrolysis. In 2003, Honda introduced a Home Energy Station that performs steam reformation right in the owner's garage-but because natural gas is the feedstock, it still releases carbon dioxide to the atmosphere.

A greenhouse gas-free approach would use on-site wind or solar power to produce hydrogen through electrolysis. Honda also designed a solar-powered hydrogen refueling station, which has been operating at the company's California lab since 2001. If the national power supply becomes more eco-friendly, clean electrolysis could run off the grid.

ON-BOARD PRODUCTION: Several prototype vehicles make their own hydrogen from stored hydrocarbons, eliminating the question of distribution altogether. The DaimlerChrysler NECAR 3, for example, produces hydrogen from methanol. Researchers are also experimenting with more futuristic on-board production technologies, which combine ordinary water with reagents like boron or aluminum to produce hydrogen, oxygen and a metal oxide residue. These, however, are still a long way off.

HURDLE 4: Use
Once hydrogen reaches consumers, is there anything they can do with it except drive vehicles? Home energy generation is one other option. The question is whether hydrogen would be more practical than current methods. Hydrogen produced by steam reformation or by electrolysis loses energy when it is converted into electricity. The resulting efficiency is roughly equal to that of today's power plants — which pay a lot less for raw materials. Direct generation of electricity through wind and solar power will also be more efficient for most stationary applications. That leaves transportation as the most promising use for hydrogen.

INTERNAL COMBUSTION: The most straight-forward approach is to burn hydrogen in an adapted model of your garden-variety internal-combustion engine (ICE). Since little modification is required, these engines are relatively cheap, and 25 percent more efficient than gasoline-powered engines. BMW built its first hydrogen ICE back in the 1970s, and the concept still has legs: Ford began production of a hydrogen ICE shuttle bus last July.

GREEN BEER: The four 250-kilowatt hydrogen fuel cells at the Sierra Nevada Brewery in Chico, Calif., run on a combination of natural gas and methane. They generate enough electricity to power the entire production plant.

FUEL CELL: First invented in 1839, a fuel cell combines hydrogen and oxygen to generate electricity without any moving parts. Several different varieties exist, but only the proton exchange membrane (PEM) fuel cell is lightweight and responsive enough to be practical for vehicle use. Though twice as efficient as ICEs, PEM fuel cells are hindered by high prices — even in mass production, they would currently cost about $36,000 each.

Once the technical hurdles are crossed, hydrogen's huge price tag may still make the technology prohibitive. A recent analysis by the Department of Energy projected that a supply network adequate for even 40 percent of the light-duty fleet could cost more than $500 billion. And that leads to a classic chicken-and-egg problem: How do you get millions of Americans to buy hydrogen-powered vehicles before there's an infrastructure in place to refuel them? And how do you get energy companies to build that infrastructure before there's a potential customer base?

"Companies are not willing to invest if they don't think there's going to be a market," says Daniel Sperling, director of the Institute of Transportation Studies at UC Davis. "The government has to be behind it. There has to be leadership."

There's reason to hope the technology will advance even without much government involvement. Hydrogen fuel cells already replace batteries in niche equipment, such as TV cameras and forklifts, and provide power at remote locations, such as at cellphone towers. They even power the police station in New York's Central Park. As these applications continue to develop, they will force advances in technology that will make hydrogen vehicles more feasible. Even then, hydrogen might make the most sense for fleet vehicles that don't require widespread infrastructure for service and refueling.

Ultimately, hydrogen may be just one part of a whole suite of energy alternatives. Any one of them will involve investing heavily in new infrastructure. Though the price tag will be steep, we can't afford oil's environmental, economic and political drawbacks any longer.

	NATURAL GAS	NUCLEAR	SOLAR	WIND	BIOMASS	COAL
	Gas station-size facilities using steam reformation	Very High Temperature Reactors providing heat for electrolysis or for thermochemical cycles	Photovoltaic systems providing electricity for electrolysis with 10% efficiency	Turbines producing electricity for electrolysis, assuming they operate at 30% capacity	Gasification plants using steam reformation	FutureGen plants using coal gasification then steam reformation
Raw Materials Required	15.9 million cu. ft. of natural gas — only a fraction of current U.S. annual consumption	240,000 tons of unenriched uranium, five times today's global production	2500 kilowatt-hours of sun per square meter per year, found in the Southwestern states of the Sun Belt	7 meters per second average wind speed, typically found in many parts of the country	1.5 billion tons of dry biomass (initially byproducts such as peanut shells, then concentrated crops)	1 billion tons of coal — which would require doubling current U.S. domestic production
Infrastructure	777,000 facilities; though a more likely scenario would include a mix of larger central production plants	2000 600-megawatt next-generation nuclear power plants; only 103 nuclear power plants operate in the States today	113 million 40-kilowatt systems, covering 50% of more than 300 million acres — an area three size the size of Nevada	1 million 2-megawatt wind turbines, covering 5% of 120 million acres, or an area larger than California	3300 gasification plants, and up to 113.4 million acres — or 11% of U.S. farmland — dedicated to growing the biomass	1000 275-megawatt plants; only 12 sites were proposed for a DOE demonstration plant — not all met the requirements
Total Cost	$1 trillion	$840 billion	$22 trillion	$3 trillion	$565 billion	$500 billion
Price Per GGE (Gallon of Gas Equivalent)	$3.00	$2.50	$9.50	$3.00	$1.90	$1
CO2 Emissions measured in tons	300 million	0	0	0	600 million*	600 million**
	*Zero net emissions because crops pull CO2 from the air. **90% will be captured and stored underground.
Time Frame	There are four fueling stations that now produce hydrogen from natural gas.	The first Very High Temperature Reactor in the U.S. will be built at Idaho National Laboratory in 2021.	Honda built an experimental solar-powered hydrogen refueling station at its lab in California in 2001.	A 100-kilowatt turbine is now being built at the National Renewable Energy Lab in Colorado.	Government funded bio-mass research will be transferred to private industry in 2015.	By 2012, the first FutureGen demonstration plant should be running at 50% capacity.