Sun in a Bottle Page 18

  The first thing that would strike a visitor to the Princeton facility in the early 1990s would be the circles. There were circles everywhere. In the lobby, an office assistant swiveled about behind a large ring-shaped desk. A circular sofa surrounded a donut-shaped model of the TFTR. Other models of ringlike tokamaks were displayed in the waiting room. Even the auditorium was semicircular. And of course, the heart of the whole facility was the donut-shaped TFTR tokamak.

  The second thing that would strike a visitor was the air of quiet desperation that hung about the lab. The staff was trying to sell fusion to the public, and while the TFTR was setting temperature records almost daily, nobody seemed to be buying. Budgets were still dropping, and the taxpayers didn’t protest. The lab, quietly, tried to change that attitude. Along each wall of the laboratory’s lobby, colorful posters exhorted the taxpayer to back fusion research. “Why Fusion?” read one. “Do We Really Need To Spend This Much On Energy Research?” asked another. Rush Holt, a physicist and the spokesman for the TFTR project, promised great things for TFTR—6 watts out for every 10 put in, within spitting distance of breakeven—but most of all, he conjured a future with fusion energy. Without it, he said, humanity would be in trouble.64

  Where can we as a society get our energy? Fossil fuels pollute, cause global warming, and are running out. Renewable sources—solar, geothermal, wind—can’t provide nearly enough energy for an industrial society.65 That leaves nuclear energy: fusion or fission. Holt argued that fission is messy: a fission reactor uses up its fuel rods and leaves behind a radioactive mess that nobody knows how to dispose of. Fusion, on the other hand, leaves no harmful by-products. It runs on deuterium and tritium, he said, and leaves only harmless helium behind. Clean fusion energy would be a much better choice.

  This is the sales pitch of faithful magnetic fusion scientists everywhere. Fusion provides unlimited power—clean, safe energy without the harmful by-products of fission. But there is a dirty little secret. Fusion is not clean. Once again, it’s the fault of those darn neutrons.

  Magnetic fields can contain charged particles, but they are invisible to neutral ones. Neutrons, remember, carry no charge and do not feel magnetic forces. They zoom right through a magnetic bottle and slam into the walls of the container beyond. Since a deuterium-deuterium fusion reaction produces lots of high-energy neutrons (one for every two fusions), the walls of a tokamak reactor are bombarded with zillions of the particles every moment it runs.66

  Neutrons are nasty little critters. They are hard to stop: they whiz through ordinary matter rather easily. When they do stop—when they strike an atom in a hunk of matter—they do damage. They knock atoms about. They introduce impurities. A metal irradiated by neutrons becomes brittle and weak. That means the metal walls of the tokamak become susceptible to fracture before too long. Every few years, the entire reactor vessel, the entire metal donut surrounding the plasma, has to be replaced.

  Unfortunately, neutrons also make materials radioactive. The neutrons hit the nuclei in a metal and sometimes stick, making the nucleus unstable. The longer a substance is exposed to neutrons, the “hotter” it gets with radioactivity. By the time a tokamak’s walls need to be replaced, they are quite hot indeed.

  Though fusion scientists portray fusion energy as cleaner than fission, a fusion power plant would produce a larger volume of radioactive waste than a standard nuclear power plant. It would also be just as dangerous—at first. Much of the waste from a fusion reactor tends to “cool down” more quickly than the waste from a fission reactor, taking a mere hundred years or so until humans can approach it safely. But it means that humans will have to figure out where to store it in the meantime, as well as the rest of the waste that, like spent fission fuel, will remain untouchable for thousands of years. Fusion is a bit cleaner than fission, but it still presents a major waste problem.

  Fusion scientists recognize this, of course. They are working on exotic alloys that are less affected by neutron bombardment, materials made of vanadium and silicon carbide. However, developing those materials is going to cost a lot of money, and they will still present a waste problem, albeit a reduced one.

  It’s an open secret. Fusion isn’t clean, and it probably never will be.

  CHAPTER 8

  BUBBLE TROUBLE

  Hegel observes somewhere that all great incidents and individuals of world history occur, as it were, twice. He forgot to add: the first time as tragedy, the second as farce.

  —KARL MARX, THE 18TH BRUMAIRE OF LOUIS NAPOLEON

  The mere mention of cold fusion made everyone bristle. The scientists, the press office, the editor of the magazine all objected to anyone’s using the term. But the phrase was soon echoing across the nation. It was on the front pages, in the evening television broadcasts, and plastered all over the press. Cold fusion rides again. History seemed to be repeating itself.

  The controversy seemed familiar from the start. Scientists at Oak Ridge National Laboratory and Rensselaer Polytechnic Institute, both well-respected institutions, claimed that they had created fusion in a little beaker of acetone not much bigger than the original Pons and Fleischmann cell. In many ways, though, this situation was very different from cold fusion. Rather than announcing their results at a press conference, the scientists sent them to Science magazine, the most prestigious peer-reviewed journal in the United States, and their paper had been accepted. The scientists weren’t saying they had discovered dramatically new physics, as Pons and Fleischmann’s palladium-catalyzed fusion would have required. These bubble fusion reactions were supposedly happening at tens of millions of degrees, rather than at room temperature.

  But like cold fusion, the bubble fusion researchers believed their work could lead to an unlimited source of energy. And like Pons and Fleischmann, the bubble fusion scientists quickly came under attack by some of the leading fusion physicists in the nation. Even before their paper had been published in Science, the bubble fusion scientists were labeled as incompetent. It got worse after publication. Increasingly isolated, they were forced toward the fringe, and before long they were fighting accusations of scientific misconduct and a fraud investigation that led all the way to Capitol Hill. History had repeated itself.

  The bubble fusion imbroglio was a twisted reflection of the cold-fusion affair. The second time around, the tale would be a tragedy as well as a farce. Researchers, peer reviewers, editors, journalists, press officers, and all the other players in the drama were caught in a colossal web of mutual misunderstanding. It was a story of good intentions gone wrong, of paranoia and mistrust, and of hubris that led to the downfall of a scientist.

  When the bubble fusion story broke in 2002, I was a reporter for Science, ground zero for the controversy.

  Science is famous because it is arguably the premier peer-reviewed scientific journal in the United States. For many scientists, a publication in it (or its British rival, Nature) would be considered a major coup, perhaps even the crowning achievement in an average scientific career. Researchers from around the world submit manuscripts to Science, and it is an enormous task to examine the submissions and select those worth publishing.

  I had nothing to do with the peer-reviewed section of Science. I worked for the news pages at the front of the magazine. News reporters at Science are deliberately isolated from the peer-reviewed section. We weren’t told about manuscripts in the pipeline, or about the status of a paper undergoing peer review. We weren’t even allowed to know who the peer reviewers of a given manuscript were.67 So I was quite surprised when, on February 5, 2002, my editor, Robert Coontz, e-mailed me a paper entitled “Nuclear Emissions during Acoustic Cavitation.” It had already been through the peer-review process, but it wasn’t an ordinary manuscript. “Here’s a Science paper that’s likely to be very controversial,” Coontz wrote. “First task is to decide whether we want to cover it.” Within a few seconds, I knew it was going to be explosive.

  The manuscript was couched in the typical cold, technical la
nguage of the scientific paper, but its authors, a team led by Rusi Taleyarkhan at the Oak Ridge National Laboratory, were making a claim that seemed eerily reminiscent of cold fusion. They claimed to have induced fusion reactions on a tabletop using a process that might lead to energy production. More important, they did it in an ingenious, and seemingly plausible, way. They did it with a technique linked to a mysterious phenomenon known as sonoluminescence.

  As early as the 1930s, scientists had discovered a bizarre method to convert sound into light. If you take a tub of liquid and bombard it with sound waves in the correct manner, the tub begins to generate tiny little bubbles that glow with a faint blue light. This phenomenon is not perfectly understood, but scientists are pretty sure they know what is going on, at least in gross terms.

  If you have ever belly flopped off a diving board, you know that a liquid like water doesn’t always behave quite like a fluid. Hit it hard enough and fast enough, faster than the water can flow out of your way, and it feels almost like concrete. It behaves more like a solid than like a liquid. This is more than a mere metaphor. Under certain circumstances—if you hit a liquid in the right way—it will “crack” just as a solid would. The liquid ruptures, creating tiny vacuum-filled bubbles that instantly fill with a tiny bit of evaporated liquid. This phenomenon is known as cavitation, and it occurs in a number of different places. Submarine propellers, for example, cause cavitation if they spin too fast. Sound waves rattling through a fluid can also create these bubbles.

  Under the right conditions, the sound waves reverberating through the liquid also cause these bubbles to compress and expand, compress and expand. Each time the bubbles are squashed by the sound waves, they heat up. If the sound waves are just right, the bubble can collapse to roughly one-tenth its original size, heating up to tens of thousands of degrees and emitting a flash of light. This is sonoluminescence.

  Taleyarkhan wondered what was happening at the center of those collapsing bubbles. What would happen if you replaced water with a deuterium-laden liquid? If those bubbles got squashed far enough and became hot enough, could they induce the little bit of deuterium vapor in the center of the bubble to fuse? Could they induce a fusion reaction in a beaker?

  The first problem he encountered was that tens of thousands of degrees isn’t nearly enough to induce fusion, so ordinary sonoluminescence didn’t have any hope of getting deuterium nuclei to stick together. For fusion, Taleyarkhan needed to heat deuterium and to tens of millions of degrees, a thousand times hotter than what traditional sonoluminescence could achieve. The only way to get those temperatures was to compress the bubbles far more than had ever been done before, either by squashing them tighter or by starting with bigger bubbles. Taleyarkhan had figured out an innovative way to do the latter.

  His research team started with a solution of deuterated acetone, the same molecule that’s in nail polish remover, except for the fact that its six hydrogen atoms have been replaced with deuteriums. Then they irradiated the liquid with energetic neutrons and exposed it to sound waves. The energetic neutrons poured their energy into the solution and birthed very large bubbles—tens or hundreds of times larger than the ordinary bubbles in sonoluminescence—and, according to Taleyarkhan and his colleagues, the sound waves compressed them by a factor of ten thousand. This was a much higher compression than had ever been observed before. Taleyarkhan’s calculations implied that this extreme compression led to a temperature in the range of millions of degrees. This, in turn, supposedly led to fusion.

  To all appearances, Taleyarkhan and his colleagues did all the right things when they went looking for deuterium-deuterium fusion. The paper told of how the researchers looked for neutrons—and found them. Tritium? Found it. They also avoided many of Pons and Fleischmann’s mistakes. They ran the obvious control experiments, substituting ordinary acetone for the deuterated variety. The neutrons and tritium disappeared. Finally, the paper convinced a science editor and a group of peer reviewers who, presumably, were satisfied with its quality.

  But I was skeptical. For one thing, I knew Taleyarkhan, and while I held him in reasonably high esteem, I didn’t think of him as a fusion expert. A few years earlier—in 1999, when I was a reporter for New Scientist magazine—I had written about one of his inventions. He had figured out a clever way to make a gun that would shoot bullets at different speeds. In theory, you would be able to turn a dial on a gun and set it to “stun” with low-velocity bullets or to “kill” with high-velocity ones. (It used an aluminum-based propellant that could do things ordinary gunpowder couldn’t.) Interesting stuff, but not the sort of thing a fusion expert would invent. Taleyarkhan was a nuclear engineer, and I associated him with steam explosions and propellants and reactor safety, not fusion physics. What really bothered me, though, were the neutrons.

  The bubble fusion paper was going to live or die by the neutrons Taleyarkhan was claiming to see. Neutrons were what killed Pons and Fleischmann. Neutrons were what killed ZETA. Without a nice, clear demonstration of neutrons of the proper energy—2.45 MeV—streaming from the experimental cell, nobody would take Taleyarkhan seriously for a minute. So the first thing I looked at was the paper’s graph of neutrons. I was surprised.

  Skeptical physicists would only be convinced by a detailed graph showing how many neutrons were detected at what sorts of energies. Taleyarkhan’s paper had a few graphs, but they were far from detailed. The main one only had four points—two for the deuterium experiment and two for the control experiment—telling how many neutrons were detected above and below 2.5 MeV. That wasn’t nearly enough, at least in my opinion. I expected a neutron spectrum to have tens of points, not two. Without that level of detail, I didn’t think that there was enough information to determine whether the experimenters were seeing something real.

  I was uneasy. The content of the graph did not rule out the claim of fusion. Taleyarkhan’s team may well have seen neutrons that were drop-dead evidence of fusion. But if they did, I couldn’t tell from the graph. If they had confirmatory data, they were not presenting it in a convincing way. If they didn’t know how to convince other scientists of their claims, I suspected that they didn’t know enough about the field to make such claims in the first place.

  That was my initial impression. But as a journalist, I’ve learned that first impressions are very often wrong. In fact, I wanted to be convinced that I had erred in my snap judgment, if for no other reason than I thought it would make a better story if Taleyarkhan was correct. Furthermore, I knew that the manuscript had gone through Science’s peer-review process. The editor who had handled the manuscript—I presumed that it was our physics editor, Ian Osborne—did not laugh it out of the room when he read it. The peer reviewers (whose identities I didn’t know) had also, presumably, vetted the manuscript and found it worthy of publication. This certainly did not ensure that Taleyarkhan and his colleagues were right, but it did theoretically mean that there were no obvious flaws.

  I wanted to get to the bottom of it. I wanted to figure out whether bubble fusion was real. If it was, it could be the biggest science news to come around in a long time. I wrote back to Coontz. Of course I wanted to cover the story.

  When Coontz first sent me the paper on February 5, he told me of an additional complication. The editors were afraid of an embargo break.

  The embargo system is the dirty little secret of science journalism. Over the past few decades, science journalists entered into a compact with peer-reviewed journals like Science, Nature, and the New England Journal of Medicine. The journals provide copies of manuscripts to reporters a few days ahead of publication; these journalists, in return, agree not to tell the public about the manuscripts until the embargo expires, usually the evening before the peer-reviewed journal is published. Journalists who break the embargo, publishing ahead of the set time, are threatened with the loss of access to advance manuscripts, putting them at a great disadvantage with respect to their peers who abide by the rules. Nonetheless, some stories are so j
uicy that reporters can’t resist; word inevitably leaks out before the embargo expires (often the fault of British newspapers, whose reporters are particularly jumpy). The embargo breaks, and it’s a free-for-all.

  Bubble fusion was obviously a juicy story, so the slightest word leaking to the press could trigger a media feeding frenzy. It was crucial to the editors that nothing be reported in the newspapers until the final version of the manuscript was ready. If the press started talking about the experiment before the paper was available, it could easily be a repeat of the cold-fusion disaster—science by press conference. It would not be fair to Taleyarkhan and his colleagues, who went through the peer-review process, to open them to accusations of subverting the system. Security had to be extraordinarily tight.

  The paper was to be published on February 14. I was asked not to contact any scientists other than the authors until February 8, so that word of the paper wouldn’t spread. It wasn’t an extraordinary request, and I could certainly hold off on some of the phone calls for three days. But I had to contact Taleyarkhan right away. After I digested the paper, I sent him an e-mail to set up an interview; I also asked whether he was the same Taleyarkhan of the variable-speed bullets.

  He was. He remembered the story I had written in 1999 and hinted darkly in his e-mail reply that the government might try to keep bubble fusion a secret, just as they had done with his earlier research. “Right after you did your story my project got classified. I hope something like this does not happen to this area. That would be a shame.” In the rest of the note Taleyarkhan clearly showed that he thought he had made an important discovery. “This current area could have somewhat revolutionary and far-reaching consequences with very significant impacts on everyday life and a variety of disciplines (ranging from materials synthesis to medicine to food sterilization to counter-terrorism to power production and the like).” Power production. There it was. He hedged, and he put it last in his list, but it was there. Taleyarkhan thought he had found a path to fusion energy. This was going to be a big story, one way or another. Taleyarkhan and I made an appointment to speak on Friday, February 8. On the evening of February 6, I learned that the bubble fusion article was on hold. Taleyarkhan’s employer, Oak Ridge National Laboratory, was trying to apply the brakes. Gil Gilliland, an associate director of Oak Ridge, apparently called Science and complained that the paper had not yet passed Oak Ridge’s internal review process, which had been ongoing since November. (This was an unusually long time to spend on a review.) Gilliland promised to get the review finished as soon as humanly possible, and the manuscript was rescheduled for publication on March 8. I was asked to hold off on the interviews until the paper was back on track. I didn’t know it at the time, but the scene at Oak Ridge was getting ugly.