Last summer, Robert T. Bush came up with the kind of theory that will brand him as either a genius or a crackpot. Bush, a physics professor at California State Polytechnic University, claims to be producing energy by means of a nuclear reaction in an apparatus filled with nothing more exotic than distilled water and a few common electrolytes. Further, he says, the technology could be made practical within a decade for generating electricity. If so, Bush declares, the upshot will be "even more important than the breakthrough a few years ago in warm superconductors."

If it all sounds familiar, it should. Three years ago, chemists B. Stanley Pons and Martin Fleischmann triggered a scientific frenzy by announcing that they had achieved fusion--the nuclear reaction that powers the sun--in a jar. In their experiments at the University of Utah, they claimed, deuterium atoms fused inside a palladium rod, giving off enough power to promise a new era of cheap energy. After a roller-coaster ride from all the rage to outrage, their discovery of "cold fusion" was branded as dubious at best. The subject largely disappeared from such journals as Nature and Scientific American.

STUMPED. But faith in cold fusion--or something like it--survives among a determined band of researchers. Around the world, more than 200 of these Don Quixotes are pursuing their quest and claiming to make progress. Some remain fixated on the idea of cold fusion: Pons and Fleischmann have retreated to France, with unnamed backers funding their work. Others, such as Bush, suggest that the power they claim to obtain is coming not from traditional fusion of forms of hydrogen but from more esoteric reactions that also violate conventional theories of physics.

What all this research will produce remains to be seen. Critics say it adds up to nothing resembling classical fusion. But even a few skeptics agree there is tantalizing evidence that experiments are turning up something unexplained. And the Electric Power Research Institute in Palo Alto, Calif., the utility industry's research arm, wants to crack the secret of whatever-it-is. After plowing $2 million into the effort since 1989, EPRI has budgeted $3 million this year alone.

What it's looking for may solve a mystery at SRI International in Menlo Park, Calif., which gets most of the utility industry's money for what EPRI now calls "heat production phenomena" research. On Jan. 2, a flask the size of a thermos bottle exploded in the face of SRI researcher Andrew M. Riley, one of Pons's original associates, killing him instantly. Seven weeks later, "the in-vestigators are still stumped," says John Taylor, EPRI's vice-president for nuclear research. Cold-fusion skeptics think it was merely an unfortunate lab accident, having nothing to do with a nuclear reaction.

TURNABOUT. The current claims fall into at least two distinct camps. Some researchers, such as Pons and Fleischmann, still seek their miracle in experiments in which an electric current is passed between a platinum wire and a palladium rod immersed in water made from deuterium, a heavy form of hydrogen. Theoretically, the deuterium atoms then fuse, giving off more energy, in the form of heat, than is contained in the electric current.

From his lab in Nice, Fleischmann says these experiments are still giving results that can't be explained away as some kind of chemical reaction. "For these very high levels of heat," he insists, "it's got to be nuclear." Even some former skeptics are impressed. Last October, Heinz Gerischer, an electrochemist at Berlin's Max Planck Fritz Bader Institute, called on scientists to reconsider cold fusion. Prior to that, he had been one of the theory's biggest critics in Europe. His change of heart stemmed from a meeting last summer of some 200 cold-fusion stalwarts who had retreated to Lake Como in the Italian Alps to share their latest findings. Gerischer was particularly intrigued by reports from 50 labs of neutron radiation (table), which should be a byproduct of a traditional fusion reaction. "I don't see any absolute proof," he says. And indeed, the level of neutrons observed so far doesn't account for all the heat that researchers are claiming to get. Still, Gerischer adds: "I believe that another systematic investigation should be made."

If most scientists find Pons's and Fleischmann's claims hard to swallow, wait until they hear what Cal Poly's Bush and others are saying. This new round of theories started with Randell L. Mills, president of HydroCatalysis Power Corp. in Lancaster, Pa., who claims to produce heat using plain distilled water. When Bush was asked about that at Lake Como, he had an inspiration. "It suddenly dawned on me," he recalls, "that the major reactant that Mills is using (in the water), potassium-39, plus a proton from light water, would give you calcium-40." His theory is that potassium in the water combines with a proton at the edge of the palladium rod in a nuclear reaction that may not even be fusion.

OUT OF TOUCH? Bush and his collaborator, Robert D. Eagleton, also a physics professor, claim that when they set up a Mills experiment it produced even more heat than their Pons-Fleischmann cell. And when they looked for calcium-40, it was there roughly in the predicted amount--though Bush acknowledges he could just be measuring contaminants in the water. Next they used sodium, which Bush's theory predicts could combine with a proton to make magnesium. It produced nearly twice as much heat as potassium-laced water, Bush says.

To John R. Huizenga, a professor of physics and chemistry at the University of Rochester and co-chairman of a government panel that reviewed the original cold-fusion experiments, this is just one of many bizarre theories that cold-fusion advocates have come up with. He says, for instance, that the laws of physics make the fusion of potassium atoms with protons many times less likely than fusion of deuterium atoms. "These guys have lost all reality," he adds. Bush concedes that his explanation is outside the realm of accepted physics. "I can't tell you how the nuclear trick works," he says. But he insists that he has stumbled onto "modern alchemy."

EPRI's Taylor is thinking along similar lines. The energy gains he says EPRI-funded researchers are getting--10% to 20% more than is pumped into the cell--are inexplicable in chemical terms. So it would appear to be nuclear, Taylor says, "but it's not fusion as we know it."

Another unorthodox theory comes from Akito Takahashi, a physicist at Osaka University. He aims to explain why heat seems to be generated in bursts of varying duration during experiments, as some researchers have reported. The answer, he believes, isn't the fusing of two nuclei but of three--or perhaps even four--producing tritium and neutrons as byproducts. Huizenga debunks this theory, too. Getting two nuclei to fuse is tough enough, he says, "and when you require three or four to bump together, the chances of success go down." Takahashi agrees but insists that his findings are possible within the confines of a palladium rod.

Professional animosity is becoming a way of life for cold-fusion advocates and their detractors. For instance, Vesco C. Noninski, a Bulgarian electrochemist, and Melvin H. Miles, a chemist at the Naval Weapons Center in China Lake, Calif., say one series of often cited negative experiments is flawed by a mathematical error. And they accuse the prestigious journal Nature of refusing to publish a letter that they claim exposes the mistake. Baloney, retorts David Lindley, Nature's associate editor for physical sciences. The letter, he says, just didn't hold up to scientific scrutiny.

`MENTAL CONSTIPATION.' In fact, there's more bad news ahead for cold fusion. A new study by scientists at General Electric Co., to be published this spring in the Journal of Electroanalytical Chemistry, is "devastating," says Robert L. Park, a University of Maryland physicist who has reviewed it. GE was one of the few companies to sign confidential agreements with the University of Utah, giving the company access to details of the Pons-Fleischmann experiments. The researchers at GE say the Utah work was flawed.

"We find it difficult to explain the results they get if the procedures they use are what they say they are," says GE physicist Ronald H. Wilson. In fact, no experiment that Wilson set up yielded any excess heat. Pons and Fleischmann asked the journal to let them write a rebuttal--but they haven't done so yet. That may be because Fleischmann feels it's not worth replying to critics: "Those people have got mental constipation about this thing," he says. That's an odd way to win over doubters, counters Kelvin G. Lynn, a senior physicist at Brookhaven National Laboratory. "These people have to stop feeling they are outcasts of science. They need to show how their experiments are done."

That's the spirit in which Thomas F. Droege is working. An electrical engineer at Fermi National Accelerator Laboratory in Chicago, Droege has built in his basement what may be the world's most elaborate calorimeter, the tool used to measure heat produced in cold-fusion tests. Since December, it has been taking data from a Mills experiment, with Droege watching until the wee hours to see if he can "prove the idea, disprove it, or decide it's due to experimental error." For Droege, it's all less quest than adventure. "I'm tired of science that's conservative and safe," he says. "It's just fun to be out there on the edge."

      When atoms fuse, energy is released and certain byproducts result. Finding 
      telltale byproducts, then, is one key to whether a reaction is fusion. No 
      researcher has seen enough byproducts to account for the heat that cold fusion 
      experiments are producing. And only some labs doing fusion research have found 
      any byproducts. Here are the latest results:
      Fusion byproducts   Number of labs detecting  Reaction among skeptics
      NEUTRONS (N)                      50           Gaining some acceptance
      TRITIUM (T)                       40           Attributed to error or
      HELIUM-4                           6           Attributed to error or
      N+T                          4 to 10           Strongly doubted or dismissed
      N+T+GAMMA RAYS                     0           If verified, should be
Before it's here, it's on the Bloomberg Terminal. LEARN MORE