Under the bonnet of the HydroGen3 minivan, a General Motors' creation, 200 fuel cells inhale hydrogen molecules, strip off their electrons and feed current to the electric engine. The only emissions: a little extra heat and humidity. The result is a smooth, eerily quiet ride - although one that, with H3s priced at $1m (£560,000) each, remains out of reach of most drivers.
Hydrogen-powered vehicles may be rarities on the street, but in the world's capitals they and their technological kin are very much on people's minds. Switching from fossil fuels to hydrogen could dramatically reduce urban air pollution, lower dependence on oil and reduce the build-up of greenhouse gases that threaten to trigger severe climate change.
With those perceived benefits in view, the US, the European Union, Japan and other countries have sunk billions of government dollars into hydrogen initiatives aimed at revving up the technology and propelling it to market.
Car and energy companies are pumping billions more into building demonstration fleets and hydrogen fuelling stations.
The only problem is that the bet on the hydrogen economy is at best a long shot. Recent reports from the US National Academy of Sciences and the American Physical Society conclude that researchers face huge challenges in finding ways to produce and store hydrogen, convert it to electricity, supply it to consumers and overcome safety concerns. The transition to a hydrogen economy, if it comes at all, will not happen soon. Almost everyone agrees that producing a hydrogen economy is a worthy long-term goal. When burned, or oxidised in a fuel cell, hydrogen emits no pollution, including no greenhouse gases. Gram for gram, it releases more energy than any other fuel. And as a constituent of water, hydrogen is all around us.
Yet many researchers and energy experts say current hydrogen programmes fall pitifully short of what is needed to bring a hydrogen economy to pass. The world's energy infrastructure is too vast, they say, and the challenges of making hydrogen technology competitive with fossil fuels too great unless substantially more funds are added to the pot.
The current initiatives are just "a start", says Mildred Dresselhaus, a physicist at the Massachusetts Institute of Technology. "None of the reports say it's impossible," she says. However, she adds: "The problem is very difficult no matter how you slice it."
Top of the list of difficulties is finding a simple and cheap way to produce hydrogen. Nearly all the hydrogen on Earth is bound to other elements in molecules, such as hydrocarbons and water. Hydrogen atoms must be split off from these molecules to work in most fuel cells. These devices then combine hydrogen and oxygen to make water and liberate electricity in the process.
Today, by far the cheapest way to produce hydrogen is by using steam and catalysts to break down natural gas into H2 and CO2. But 15 per cent of the energy in natural gas is lost as waste heat during the re-forming process. The upshot, according to Pete Devlin, who runs a hydrogen production programme at the US Department of Energy, is that it costs $5 to produce the amount of hydrogen that releases as much energy as a gallon of gasoline.
In addition to stripping hydrogen from fossil fuels, DOE and other funding agencies are backing innovative research ideas to produce hydrogen with algae, use sunlight and catalysts to split water molecules directly and siphon hydro gen from agricultural waste and other types of "biomass". Years of research in all these areas, however, have yet to yield decisive progress.
If producing hydrogen cheaply has researchers scratching their heads, storing enough of it on board a car has them positively stymied. Because hydrogen is the lightest element, far less of it can fit into a given volume than other fuels.
At room temperature and pressure, hydrogen takes up roughly 3,000 times as much space as gasoline containing the same amount of energy. That means storing a useful amount in a fuel tank requires either compressing it, liquefying it or using some other form of advanced storage system. Unfortunately, pressurised gas tanks are bulky, taking up to four times the volume of a conventional fuel tank to match the driving distance of a gasoline engine.
Liquid hydrogen takes up much less room but must be chilled to just a few degrees above absolute zero, requiring large amounts of energy. Other storage materials, such as carbon nanotubes, have shown some promise. But for now, each still has big drawbacks.
Another area in need of progress is the fuel cells that convert hydrogen to electricity. Fuel cells have been used to power spacecraft, but their high cost and other drawbacks have kept them out of everyday applications such as cars. Various technical challenges - such as making them rugged enough to withstand the shocks of driving and ensuring the safety of cars loaded with flammable hydrogen gas - are also likely to make hydrogen cars costlier to engineer and slower to win public acceptance.
Hydrogen fuel cell cars also face an obstacle from outside: the infrastructure they need to refuel. If hydrogen is generated in centralised plants, it will have to be trucked or piped to its final destination. But because of hydrogen's low density, transporting the gas over long distances is too inefficient and expensive to be realistic - at least with current technology.
On-site production would not be cheap, either. Baldur Eliasson and Ulf Bossel, two Switzerland-based energy researchers, have calculated that to supply hydrogen for between 100 and 2,000 cars a day, an electrolysis-based fuelling station would require between 5 and 81 megawatts of electricity. And at least for the foreseeable future, that extra electricity is likely to come from fossil fuels.
Whichever approach wins out, it will need a massive new hydrogen infrastructure to deliver the goods. For a hydrogen economy to catch on, the fuel must be available in 30 to 50 per cent of filling stations when mass-market hydrogen cars become available, says Bernard Bulkin, the former chief scientist at BP.
Energy and car companies are unlikely to spend such sums unless they know mass-produced hydrogen vehicles are on the way. "We face a chicken and egg problem that will be difficult to overcome," says Michael Ramage, a former executive vice-president of ExxonMobil Research and Engineering, who chaired the NAS hydrogen report.
Because many of these problems require fundamental breakthroughs, many US researchers question their country's early heavy emphasis on expensive demonstration projects of fuel cell cars, fuelling stations and other technologies.
Concentrating the hydrogen programme on basic research will naturally give it the appropriate long-term focus it deserves, according to Joseph Romm, former acting assistant secretary of energy for renewable energy in the Clinton administration. In the meantime, Dr Romm and others say the focus should be on slowing the build-up of greenhouse gases.
"If we fail to limit greenhouse gas emissions over the next decade - and especially if we fail to do so because we have bought into the hype about hydrogen's near-term prospects - we will be making an unforgivable national blunder that may lock in global warming for the US of 1 degree Fahrenheit [0.56 degree Celsius] per decade by mid-century," Dr Romm told the House Science Committee in written testimony in March.
To combat the warming threat, funding agencies should place a short-term priority on promoting energy efficiency, research on renewables and development of hybrid cars, critics say. Short-term help may also come from capturing CO2 emissions from power and industrial plants and storing it underground, a process known as carbon sequestration.
In response to the litany of concerns over making the transition to a hydrogen economy, JoAnn Milliken, who heads hydrogen storage research for DOE, points out that DOE and other funding agencies are not promoting hydrogen to the exclusion of other energy research.
She says the inescapable truth is that "we need a substitute for gasoline: gas hybrids are going to improve fuel economy, but they can't solve the problem".
Yet, if that is the case, many energy experts argue, governments should be spending far more money to lower the technical and economic barriers to all types of alternative energy - hydrogen included - and bring it to reality sooner.
This article was provided by AAAS and Science, its international journal.