It is three years since Professor Rusi Taleyarkhan made the controversial claim that he had achieved one of the holy grails of science - nuclear fusion.
Nuclear fusion is nature's atomic power
Since then, he has grown tired of the scepticism of his fellow scientists.
"My lab has been audited, my instruments have been audited, my books have been audited, the data speaks for itself.
"The data has to speak for itself - I mean how can I answer that I know absolutely 100% sure that it is what I think it is? I just have to look at the data and the data have been looked at very carefully.
"In the history of publication I probably will not be able to find one that has gone through this level of scrutiny - if you do, let me know," he said.
Nuclear fusion is nature's atomic power - it is what powers the Sun and, if it can be made to happen here on Earth on a large enough scale, it promises to solve all of mankind's energy problems in one go.
It would be clean, last for ever and create no long-term nuclear waste. And Rusi Taleyarkhan claims to have achieved it using simple sound waves.
His breakthrough is based on something called sonoluminescence. It is a process that transforms sound waves into flashes of light, focusing the sound energy into a tiny flickering hot spot inside a bubble.
It has been nicknamed "the star in a jar" by researchers in the field.
The star in a jar effortlessly reaches temperatures of tens of thousands of degrees, which is hotter than the surface of the Sun. It was able to do all this by simply focusing the energy of the sound wave into a tiny hot spot.
In order to get fusion, temperatures inside the bubble had to be in the region of 10 million degrees. It seemed improbable that the tiny hot spots could be this hot. But if they were - or if a way could be found to make them so - then a new route to nuclear fusion would be opened up.
In 1999, the US government made some research funds available and across America a few laboratories started to explore ways to try to turn their star in a jar into fusion. And Rusi Taleyarkhan got there first.
But there was one major criticism of Rusi Taleyarkhan's work.
When fusion takes place, particles called neutrons are given off. These are considered by scientists to be the key signature of nuclear fusion - but measuring neutrons on a small, laboratory scale had proven notoriously difficult in the past because neutrons also occur naturally in the Earth's environment.
Professor Taleyarkhan was also using them in part of his experiment. Many scientists were unconvinced that Rusi Taleyarkhan's neutron detection was as accurate as it needed to be to prove such a big claim.
To try to get to the bottom of the issue, the experiment was re-run by Mike Saltmarsh and Dan Shapiro, colleagues of Taleyarkhan's at the Oak Ridge National Laboratory. But, when they repeated the experiment, they couldn't find any evidence of fusion.
"If there had been fusion going on at the sort of rate that Taleyarkhan's paper was claiming we should have seen an enormous increase in the neutron detection and we didn't," said Mike Saltmarsh.
Most of the key figures in the field lined up on the side of Mike Saltmarsh but they could not dispute that Rusi Taleyarkhan had found what he said he had found. It seemed to be a scientific stalemate.
Then two years later, in March 2004, Rusi Taleyarkhan came out with a new paper, showing even more fusion and more neutrons. This paper was thoroughly reviewed and published in another respected journal.
Taleyarkhan has had to battle the sceptics
But some sceptics still were not satisfied.
Nuclear fusion from sound waves would be a huge scientific breakthrough. But to be convinced of it, many scientists wanted to see better evidence; evidence that was absolutely incontrovertible.
They wanted to look very carefully at the timing of the neutrons to see just how closely they were related to the flashes of light.
If they occurred at the exact same time, they would finally be convinced that fusion was taking place. The question was - just how exact did the measurements need to be?
The sceptics wanted to time it with incredible accuracy - that of a nanosecond, or a billionth of a second. This was one measurement that, though possible to do, still had not been carried out by Rusi Taleyarkhan and his team.
The BBC Horizon programme decided to try to sort out the issue once and for all. It commissioned an independent team led by Seth Putterman to conduct a unique experiment.
Working from the instructions set out in Rusi Taleyarkhan's paper, it assembled the same key scientific conditions necessary to create nuclear fusion from sonoluminescence.
But to see if it could find fusion, we measured the neutrons and the flashes of light simultaneously with nanosecond accuracy, something that had never been done before.
Recording data nanosecond by nanosecond, Seth Putterman did not find a single neutron close enough to a flash of light for it to be considered the result of nuclear fusion. So the conclusion was negative.
Horizon put this conclusion to Rusi Taleyarkhan who said that several differences in the equipment could have affected the results.
It is very possible that other laboratories around the world will reproduce Rusi Taleyarkhan's fusion results but until then, the claim will attract great scepticism from the wider scientific community.
Horizon was broadcast on BBC Two on Thursday, 17 February, 2005.