By Dr David Whitehouse BBC News Online science editor

Claims that a team of researchers have achieved nuclear fusion in a small, table-top experiment have split nuclear physicists - with some expressing enthusiasm and others grave concern.

The premature critics of the result, and those who believe in it, would both do well to cool it, and wait for the scientific process to do its work

Science magazine

The research, to be published in Friday's issue of the journal Science, was done at the Oak Ridge National Laboratory in Tennessee and at the Rensselaer Polytechnic Institute in New York (both US).

The scientists fired sound waves through acetone, causing minute bubbles in the liquid to collapse at temperatures of millions of degrees to produce small flashes of light.

It is within these collapsed bubbles, the researchers say, that the nuclei of atoms fuse, releasing energy in the same way that the Sun does.

If science ever does find a way to build a practical nuclear fusion reactor on Earth, it would transform society as the "fuel" could be derived from water.

Plenty of scepticism

"If the results are confirmed, this new, compact apparatus will be a unique tool for studying nuclear fusion reactions in the laboratory," Professor Fred Becchetti, of the University of Michigan, US, wrote in a commentary on the findings.

"But scientists will - and should - remain sceptical until the experiments are reproduced by others.

The latest edition of Science will spark debate

"Many... could not reproduce past claims made for table-top fusion devices," he pointed out. However, Professor Becchetti said the latest claims were "credible until proven otherwise".

Many physicists will be wary. Science is still smarting from the 1989 "cold fusion" fiasco in which Stanley Pons of the University of Utah, US, and Martin Fleischmann of the University of Southampton, UK, held a press conference to announce that fusion had been achieved with the metal palladium.

When other scientists checked the work, the claims were found to be groundless.

The whole affair left physicists with a deep suspicion of table-top fusion experiments.

'Premature' publication

In fusion, the nuclei of two atoms of deuterium - a heavy form of hydrogen - are pushed together to form helium or tritium - another form of hydrogen - with the release of energy.

For decades, scientists have been trying to harness this form of energy, using superhot gas conditioned to mimic the conditions at the heart of the Sun.

The work was done with an acetone liquid in which the normal hydrogen atoms had been replaced with deuterium

The experiments have been done in huge, complex reactors and have had only limited success.

Some physicists have criticised the journal Science for publishing the "bubble fusion" research. One of the magazine's reviewers, Dr Seth Putterman, of the University of California, US, said it would be premature to go into print given the uncertainty that surrounded the experiment.

"I think the paper is wrong," he said.

This has prompted Science to issue some stern advice for those arguing about the research paper. "The premature critics of the result, and those who believe in it, would both do well to cool it, and wait for the scientific process to do its work," it said.

Background noise

According to Professor Richard Lahey, one of the researchers at Rensselaer Polytechnic Institute, the experiment had been repeated many times over the past year and been subjected to an exhausting process of peer review.

He said that as the bubbles in the acetone collapsed due to the ultrasound pulse, they produced miniscule amounts of energy. It is believed that the bubble collapse causes a momentary shock wave that creates high pressures, high temperatures and a flash of light, which scientists call sonoluminescence.

We and other people will do the experiments again and the truth will out

Dr Lee Reidinger

"It's hard to know at this point what the ultimate importance of this discovery will be. However, at this time, it looks promising," said Professor Lahey.

But Dr Lee Reidinger, deputy director for science at Oak Ridge National Laboratory, said that whilst the effect might be interesting it was going too far to think of it as an exploitable energy source.

Indeed, such was the concern at Oak Ridge for the claims of the experiment that two other scientists at the institution attempted to carry out the same experiment themselves to see if they could verify the results. They could not.

More out than in

They said telltale neutrons of a certain energy that must be given off during such a fusion reaction were not seen. Some commentators speculate that the original researchers could have detected random particles against a background of emission.

Despite not being officially published until Friday, copies of the "bubble fusion" research paper are freely circulating on the internet, meaning that more and more scientists will read it and decide where they stand.

Oak Ridge will also continue to investigate.

"We and other people will do the experiments again," Dr Reidinger told the BBC. "The scientific community will sort out the differences."

And British scientist Thornton Greenland cautioned that even if there were verification there would still be huge difficulties in generating power from nuclear fusion.

"If this claim is confirmed, the problem will still be to develop engineering devices that will produce more power from fusion than is needed to sustain the conditions that make the fusion possible," he told BBC News Online.