Among other key accomplishments, DOE has reduced the cost of producing hydrogen from natural gas; developed a sophisticated model to identify and optimize major elements of a projected hydrogen delivery infrastructure; increased by 50 percent the storage capacity for hydrogen, a key factor for increasing the driving range of vehicles; and reduced the cost and increased the durability of hydrogen fuel cells, GAO said.

But meeting the 2020 goal would require the agency to solve a number of additional knotty technical problems by 2015, DOE's so-called "technological readiness" deadline, GAO said.

The difficulty in meeting these challenges, along with fiscal constraints, has led DOE to push back some of the interim goals of the hydrogen program, GAO concluded. The government watchdog said the department should update its 2006 Hydrogen Posture Plan's assessment of what DOE reasonably expects to achieve by the 2015 technology readiness deadline and how this may differ from earlier posture plans.

In addition, GAO said, deploying the support infrastructure needed to commercialize hydrogen fuel-cell vehicles across the country will require an investment of tens of billions of dollars over several decades after 2015.

Using hydrogen to displace gasoline as the primary fuel for motor vehicles has been a dream of many scientists for decades, but the dream has been slow to realize because of the daunting technological problems that must be surmounted.

President Bush in 2003 unveiled the initial phase of a five-year, $1.2 billion hydrogen fuel initiative aimed at tackling the four biggest technical challenges: cost-effective hydrogen production, storage and delivery and fuel cell durability and cost. Congress in the 2005 Energy Policy Act extended the program beyond its initial five-year period by authorizing funding through 2020.

Currently most hydrogen is produced by cracking natural gas, but the process is expensive--a cost exacerbated by rising natural gas prices--and requires large amounts of energy. Research is focusing on economically extracting hydrogen from other compounds using fossil, renewable and nuclear energy. DOE established 2015 as a target date for extracting hydrogen from natural gas at a cost equivalent to $2 to $3 per gallon of gasoline.

GAO reported that DOE has succeeded in meeting its target of extracting hydrogen from natural gas through a process known as steam reformation, reducing the cost to less than $3 per gallon of gasoline equivalent. But the department has pushed back its target dates, from 2015 to 2017, for extracting hydrogen from biomass and water using wind energy.

Storing hydrogen requires it to be compressed in gaseous form under very high pressure or super-cooled to convert the gas into a liquid. Both these technologies require significant amounts of energy and are currently too costly. DOE's 2015 goal calls for developing a vehicle that can travel at least 300 miles using only the hydrogen stored on board the vehicle.

Scientists at Los Alamos National Laboratory have succeeded in developing materials that have the potential to meet DOE's 2010 technical deadline for chemically storing hydrogen, but it remains unclear if the process, which employs a recyclable liquid boron-based compound to bind the hydrogen, will meet cost targets, GAO said.

Current truck delivery technologies cannot compete with gasoline delivery technologies because of the cost of compressing or liquefying the gas. Although delivery by pipeline may be more economical, hydrogen causes pipelines to become brittle, raising safety concerns. Brittleness in pipes is not well understood, and DOE is conducting research to develop new composite materials for pipes or to develop pipe liners to prevent leaks and pipe failures due to brittleness, GAO said.

Finally, the type of fuel cell considered the most promising for vehicles has cost and durability limitations. Current systems cost about $8,000 per unit to produce at high volumes, compared to $2,000 to $3,000 to produce a conventional gasoline engine. DOE has set a target date of 2015 to develop a fuel cell with a life span of about 5,000 hours, or 150,000 miles, making it competitive with internal combustion engines.

The fuel cells also must be capable of operating in temperatures ranging as low as minus 40 degrees Fahrenheit (F) and must be able to start up quickly at low temperatures with minimal energy consumption.

DOE has achieved a lifespan of about 1,600 hours for vehicle fuel cells, but has not yet demonstrated start-up in sub-freezing temperatures, GAO said. In addition, although DOE has reduced the cost of fuel cells, "significant gains in cost remain to be achieved," GAO said.

Currently platinum, already in high demand for jewelry production and for use in automobile catalytic converters, is the only catalyst that can generate enough power at low operating temperatures to operate a vehicle. DOE's cost- reduction target is focusing on decreasing the amount of platinum used in designing a fuel cell in 2005 by 80 percent in 2015, GAO said.