Taking the basic apparatus used in 2002, two Purdue researchers refined the experiment and published new results that once again seem to prove that nuclear fusion was taking place. If it proves to be real, the new approach might lead to a genuine new source of energy.

An inexpensive, practical method of controlling nuclear fusion could revolutionize energy production, so any hint of a breakthrough in that direction generates high interest among both the technical community and the mainstream media. But hard-headed physicists have grown wary of "fusion in a jar" experiments.

The physics community was lukewarm to n approach to tabletop fusion that originated with Rusi Taleyarkhan at Oak Ridge National Laboratory in 2002. Using acoustic cavitation generated by ultrasound waves in a solution doped with deuterium, Taleyarkhan and his colleagues published results that they considered an airtight case for nuclear fusion. But criticism followed. When Taleyarkhan replied with a follow-up experiment to address those concerns, the reaction was muted.

The Purdue team began its work independently two years ago. "Sonofusion is thermonuclear fusion and is scalable," said Yiban Xu, who performed the experiment with fellow researcher Adam Butt. "However, much research and development needs to be done before reaching so-called energy break-even."

In the language of nuclear fusion researchers, break-even is the point beyond which a reaction produces more energy than it consumes, the minimal requirement for success. Xu, more concerned with proving that any nuclear fusion occurred, cannot say whether the reaction produces energy efficiently.

Xu said a small-scale apparatus like his experimental setup could have other important applications. "Neutrons seed cavitation in the test fluid, and so do the other nuclear particles. Therefore, in principle, cavitation occurrence indicates the presence of radiation activities if appropriate conditions are provided," he said.

Possible applications could be a simple and portable neutron source or a way to generate tritium, a helium isotope produced by the reaction.

Various names have been coined for the technique — bubble fusion, sonofusion, fusion in a jar — but the underlying physical principle is simple: Ultrasound-induced bubbles in a liquid can collapse, generating intense pressure in a tiny volume. If fusionable elements are present, cavitation might produce the temperature and pressure needed to drive nuclei together.

Inside the sun, extreme temperatures and pressures slam hydrogen atoms together with enough force to overcome the electrical repulsion of their electron shells, simultaneously producing helium and liberating energy. The sun also contains deuterium, a hydrogen isotope with a neutron in its nucleus. Fusing these atoms liberates neutrons and produces tritium, a form of hydrogen with two neutrons in its nucleus.

It seems counterintuitive that bubbles inside a room-temperature liquid can reach the temperatures and pressures that occur inside the sun. But physicists have known for some time that ultrasound can produce tiny bubbles that collapse rapidly in a process called cavitation, generating high temperature and pressure, which occasionally produces photons. Whether the temperature and pressure are high enough to drive nuclei together is a controversial question.

Criticisms of the original experiment focused on the details of measurement. Nuclear fusion between deuterium atoms produces two unmistakable effects: neutrons in a certain energy range and tritium. The neutrons escape in every direction and are difficult to detect. Small amounts of tritium occur in the acetone, so the experimental results have to show an increased amount of tritium — something beyond the background level.

The basic experiment designed by Taleyarkhan introduced deuterium by synthesizing acetone using deuterium in place of hydrogen. To amplify bubble expansion, he irradiated the solution with neutrons. When the neutrons strike an acetone molecule, they bump up the energy level and trigger bubble formation. He discovered an enormous amplification effect, with the bubbles growing to a size where they were visible. The theory was that such giant bubbles, when collapsing, would push the temperature and pressure up to fusion level.

Xu and Butt took the same basic setup and ran identical experiments, one with normal acetone and another with the deuterated variety. In their report, which was published in the peer-reviewed journal Nuclear Engineering and Design, they maintained that the parallel experiments clearly resolved that measurement issue.

"The control experiments didn't show anything," Xu said. "We changed just one parameter, substituting the deuterated acetone with normal acetone."

The experiment also suggested more reasons to believe that cavitating bubbles could reach high enough temperature and pressure for fusion. The pair found many of the bubbles were highly spherical, which produces higher forces than irregular bubbles during collapse. Indeed, the evidence suggests that only highly spherical bubbles can generate the forces required for fusion.

The new experiment also used a simpler, lower-cost method for the neutron source. Oak Ridge used a large pulsed-neutron source. Xu's experiment used a radioactive source emitting neutrons. The simplification should make it easier for other physicists to reproduce the experiment.

Taleyarkhan has now moved to Purdue and is working with Xu and Butt.