American Scientist

Newsflash: A simple method for achieving nuclear fusion, the process that powers the sun, has been devised. The requirements are so modest that fusion reactions between isotopes of hydrogen can readily be brought about with tabletop equipment.

The popular and scientific press carried much coverage of this achievement after the March 8, 2002, publication of a peer-reviewed report in Science by Rusi P. Taleyarkhan of Oak Ridge National Laboratory and five colleagues, which described the means for generating fusion in a small glass vessel. Reporters from many newspapers and magazines—including The New York Times, Business Week, Scientific American, and the news bureau of Science itself—immediately probed the controversial question of whether the evidence was sufficient to conclude that fusion reactions truly took place. They described the observations and the counter-observations. They evoked memories of the cold-fusion debate of 1989. They examined the judgment of the editors of Science in accepting the paper. But even if one takes the above "newsflash" at face value, this information is, in fact, quite dated: A tabletop method for bringing about fusion in a small vessel became public knowledge on November 29, 1949, in a patent issued to Winfield W. Salisbury of Cedar Rapids, Iowa. Commercial devices based on his patent have been in widespread use for decades.

Huh? If this is true, why so much excitement? The answer is that most people confuse nuclear fusion with power generation. So when a new method for bringing about fusion reactions is presented, as was the case in March, it is easy to start musing that it could be harnessed to generate power. Taleyarkhan's article reported no such achievement. Its importance lies in its claim to have generated fusion in a novel way: by imploding a bubble with sound waves, a technique known to generate high temperatures and brief flashes of light.

Although this phenomenon, called sonoluminescence, has been known for many decades, in recent years specialists have postulated that perhaps temperatures in the collapsing bubble can be made high enough to induce fusion. At least one private company, Impulse Devices of Grass Valley, California, is investigating ways that such imploding bubbles might be harnessed in a fusion power reactor. But the work of this small startup is still limited to computer modeling and trying to detect fusion neutrons from sonoluminescence. Although much hope may exist, no one has yet shown how to generate power this way.

Power generation from imploding bubbles would indeed be a phenomenal breakthrough. But that feat is considerably more demanding than just coercing hydrogen isotopes to fuse, which has been a relatively straightforward technical exercise since Salisbury's patent issued. In essence, all one needs is to arrange for ions of a heavy isotope of hydrogen, say deuterium, to be accelerated by an electric field so that they strike a target also containing a heavy hydrogen isotope. The required apparatus is not a liquid-filled flask but something more like an old-fashioned vacuum tube, which probably has something to do with why Salisbury assigned his patent to the Collins Radio Company. Whereas the experiment reported in March is controversial because the number of neutrons given off, the telltale marker for fusion, was relatively small (some skeptics would argue it is zero), commercial devices based on Salisbury's technique are called "neutron tubes" precisely because they give off these particles in copious amounts.

The chief use for neutron tubes is in the oil and gas industry. Lowered into an exploratory borehole, they emit neutrons, which interact with the various minerals and fluids present, giving off gamma rays. Analysis of this radiation in turn reveals information about the surrounding geological formation. One of the companies manufacturing neutron tubes for such purposes is Thermo MF Physics of Colorado Springs. Their offerings include a portable unit that sells for about $55,000. Asked whether tabletop fusion is a remarkable thing, Jack Reichardt, general manager of Thermo MF Physics, says, "Heavens, no; that's clearly not the case and never has been."

Interestingly, most producers of neutron tubes shy away from mentioning that their devices produce fusion reactions. Perhaps they are afraid their customers will subconsciously worry that pushing the wrong button will turn the instrument into an H-bomb. But not all companies are so reticent. One European firm, EADS, is advertising its "FusionStar FS-NG1 neutron generator," which, like standard neutron tubes, relies on electric fields to force heavy hydrogen isotopes together within a vacuum chamber. This device, however, uses a spherical configuration rather than a cylindrical tube.

How do the various techniques for tabletop fusion compare? In their attempt to fuse deuterium, Taleyarkhan and his colleagues reported that the number of 2.5-mega-electron-volt neutrons, an energy level characteristic for the fusion of deuterium, rose about 4 percent above background, with an estimated emission rate of some 10⁴ or 10⁵ neutrons per second. By contrast, the neutron tubes manufactured by Thermo MF Physics put out from 10⁸ to 10¹¹ neutrons per second. The specifications on the EDS device are for 10⁷ to 10⁸. Given such statistics, one might reasonably wonder why electric methods haven't been pursued in the quest for fusion energy. The answer is that they have been investigated, but so far all designs have fallen far short of what would be needed to produce more power than is consumed.

The problem with all these tabletop techniques is that they do not produce sufficient energy density. "That's the hard part," says Todd Ditmire, a physicist at the University of Texas at Austin. Ditmire and colleagues developed yet another way to achieve tabletop fusion in 1999 using laser pulses to heat deuterium clusters. At the time he was careful to stress that his method did not open the door to fusion power generation. Ditmire believes that it is important to study the possibility of producing fusion with imploding bubbles, but he notes, "I don't see any obvious way that this could be scaled up to a scheme that would be interesting for fusion energy."

It's understandable that some of the newspaper and magazine reporters caught up in the frenzy were drawn into the supposition that any tabletop device that can achieve fusion must be a breakthrough. But it is curious that the editors of Science were not more circumspect in their coverage: A news article in the same issue that carried Taleyarkhan's controversial report indicated that although some labs are trying to achieve conditions for fusion using enormous lasers or powerful magnetic fields, "small-scale 'tabletop' fusion reactions, meanwhile, have remained far out of reach." Another commentary in that very issue, written by Fred D. Becchetti, a nuclear physicist at the University of Michigan, corrects that misstatement, noting that tabletop fusion devices are in routine use.

Becchetti's piece did, however, also somewhat confound facts in a way that exaggerated the import of the new work. He indicated that existing methods require either megavolt accelerators to fuse deuterium or "a special radioactive tritium (t) target and hence government licensing." In fact, deuterium fusion reactions can be easily achieved using a few tens of thousands of volts. This has been well demonstrated, even by amateur scientists, including Joe Zambelli, Jr., an enterprising undergraduate who constructed such a device recently over a summer break. When I spoke with Zambelli, he was attempting to increase his working level from 30,000 to 50,000 volts, which he expects will allow his homebrew machine to put out about a million fusion neutrons per second. But even at the lower level, his device, which sits on a wooden table in the family room of his parents' house, can generate 250,000 neutrons per second. The fact that fusion can be carried out on a small scale is, Zambelli notes, "nothing really amazing."—David Schneider