The new process, which has prompted the University of Windsor chemists to file for patents in Canada and the U.S., might also reduce metal contamination in foods and medicines.

"This is a real fundamental advance, it is a completely different way to think about dealing with hydrogen," says chemist Douglas Stephan, head of the team which describes the chemical reaction in the journal Science Today.

The new process takes up hydrogen, hangs on to it and releases it on demand, much like what happens in a rechargeable battery, he says. This might prove useful in development of fuel cells to power vehicles.

"What you want to be able to do is burn the hydrogen in your car and go to a station and recharge your storage tank with hydrogen," says Stephan.

Hydrogen is also at the heart of one of the most commonly used chemical reactions in the world, the "hydrogenation" that adds hydrogen to fertilizers, pharmaceuticals and oils widely used in foods.

Until now metals such as platinum have been needed to drive hydrogenation.

A compound called phosphonium borate is central to the new process. It can be created in the lab, unlike the metals now used in hydrogenation -- which can be costly, in short supply, contaminate the environment and leave low levels of metal contamination in drugs and food.

In a commentary accompanying today's report, Gregory Kubas of the Los Alamos National Laboratory in New Mexico said the finding may be important for hydrogen fuel storage. It opens the door to creating lightweight hydrogen fuel cells, which so far have eluded researchers.

Phosphonium borate takes on hydrogen at room temperature and liberates it when temperatures climb above 100 C. What makes it stand out is the reversible nature of the process. But Stephan says it is not perfect.

"The problem with our molecule, in terms of commercial applications, is it doesn't store very much hydrogen by weight," says Stephan.

It might be possible, however, to use it to modify existing technologies to store and release hydrogen much more efficiently.

Stephan says he is looking for industrial partners to collaborate on the research, which is funded by the federal Natural Science and Engineering Research Council.