Stone Cold

In the last few years the University of Utah has bestowed on the world two much-trumpeted scientific achievements, the artificial heart and cold fusion. That two such seriously cracked ideas should sprout on the same ground is a matter that should worry the State of Utah considerably. Indeed, there’s enough embarrassment left over for others to share, especially from the hilarious tale of cold fusion. The artificial heart proved an ideal mechanism for driving a Robocop but too fierce for the mere physiology of the human frame. After only a handful of patients had suffered the monstrous mechanism, its use was mercifully curtailed by Federal regulators. Yet the heart caper made the University of Utah look like a sober scientific institution compared with the episode that followed.

On 23 March 1989 Stanley Pons, chairman of the department of chemistry at the University of Utah, and Martin Fleischmann, a collaborator who had recently retired from his professorship at the University of Southampton, held a press conference under the University’s auspices. They announced that they had made nuclear fusion occur in an electric cell at ordinary room temperatures. This was an astonishing claim, first because it implied that a new and cheap source of electrical power was at hand, and second because the two chemists seemed to have cheaply and elegantly trumped the conventional efforts to achieve nuclear fusion with multi-million dollar machines that operate at vast temperatures. The Utah type, in comparison, is therefore known as cold fusion.

A period of considerable agitation ensued as scientific teams around the world attempted to follow up on the Pons and Fleischmann claim. Two laboratories, at Texas A & M University and the Georgia Institute of Technology, reported that they, too, had seen evidence of nuclear fusion. The bandwagon began to roll. The Utah State legislature voted 96-to-3 to grant $5 million for cold fusion research, after being told by the University of Utah that should it fail to do so, ‘the discovery of the century will be developed by Mitsubishi.’ Chase Petersen, the University’s President, then took his case to Washington, intimating to a Congressional committee that he expected the nation’s taxpayers to come up with $125 million eventually. Asked how much he could get along with in the meantime, Petersen responded with the delicious reply: ‘The figure that comes to mind is 25 million.’

Disappointingly for connoisseurs of high farce, this insouciant claim on the Federal treasury marked the highest point attained by the hot air balloon of cold fusion. The strange vehicle, swept aloft solely by the power of human folly, frailty and greed, then entered a long period of subsidence and even now resists complete deflation.

Frank Close is an English physicist and writer about science who followed the cold fusion story as it unfolded and has now presented his analysis. He tells a good yarn, and his firm grasp of the science enables the reader to savour the blunders of Pons and Fleischmann exquisitely. This is probably the first time the technical background of cold fusion has been laid out so clearly. The exposition provides a fascinating example of all the little details that can go wrong in designing a scientific experiment and how, by systematically misinterpreting all these errors, the Utah chemists were led down the primrose path of believing exactly what they wanted to believe. Despite the value of his technical description, however, Close is unable to set aside his professional outlook sufficiently to analyse the cold fusion episode satisfactorily. Defensive of science’s good name, he is at pains to assure the reader that this is not how science is usually done, since ‘if these events became regarded as a norm of science then public confidence would be threatened’ – scientists’ code for ‘the money will dry up.’ But the episode is far more interesting than Close’s protective judgment implies. His own book shows how the episode was deeply normal in some ways, abnormal in others.

He has missed the opportunity, moreover, to pry into the motives that drove the University of Utah’s administrators so badly off the tracks. That’s an essential part of the story, for had they not pressured Pons and Fleischmann into premature announcement, the whole embarrassment might have been averted. Also too lightly glossed over is the central mystery, the motives and behaviour of Pons and Fleischmann, two well-regarded scientists who gambled away their professional reputations in a mad pursuit.

Declaring a professional bias of my own, I would say Close has also neglected the chance to analyse the role of newspapers and television in cold fusion. He refers often to ‘the media’, generally in tones that imply that it uncritically fanned the frenzy, but fails to distinguish between its different parts, or even to analyse the performance of any individual newspaper. Yet on 28 March 1989, a mere four days after the first press conference, my colleague Malcolm Browne wrote in the New York Times: ‘Scientists at Federally-financed laboratories who have worked for decades toward achieving controlled hydrogen fusion are profoundly sceptical of [the two chemists’] assertion. In dozens of interviews, most said that although they believed that the two are honest researchers, their results are most probably flukes or errors.’ That’s a fair anticipation of Close’s own conclusions, as he would have done better to acknowledge.

The most startling finding in Close’s book, however, is one that goes beyond mere error. It is his conclusion that Pons and Fleischmann misreported a piece of data critical to their case. Nuclear fusion liberates the subatomic particles known as neutrons. When these slow down enough, they are captured by hydrogen nuclei, which then release gamma rays of a characteristic energy or frequency. Pons and Fleischmann recorded gamma rays with a peak of 2500 kiloelectron-volts, which they cited as evidence of cold fusion, in a manuscript submitted to the scientific journal Nature on 24 March, the day after their press conference. But then an unpleasant thing happened. In the course of the next few days both Pons and Fleischmann, who are chemists, were each told separately by knowledgeable physicists that their gamma-ray peak had the wrong energy. Fusion neutrons on capture yield gamma-rays with a 2200 kiloelectron-volt energy.

Instead of redoing the experiment, or withdrawing their claim, Pons and Fleischmann simply moved the 2500 peak to 2200, Close says, and published it in another journal without showing the accompanying spectrum from which it was removed, and without explanation of the change. Mike Salamon, the Utah physicist who pointed out the problem of the 2500 peak to Pons, gave Close the following account: ‘I told Stan and he said that he knew that and that it was a mistake – the figure had to be corrected. I asked him how did he know that the peak really was at 2.2; had he recalibrated his detector, and he said that he had and that the peak was now at 2.2 MeV [2200 kiloelectron-volts].’

The matter of the peak is important because if the original recording showed nothing at 2200, it proved no fusion was occurring on that occasion. By moving the peak, and claiming it as evidence for fusion. Pons and Fleischmann were not just recalibrating their detectors or smoothing out data: they were reporting black as white, and committing a very serious breach of scientific ethics.

The pair committed other assaults on conventional scientific procedure. Their blunders are almost enough to evoke sympathy, were it not for the fact that to this day they have not recanted for leading everyone on a wild goose chase. The root of the problem was that the experiment they were trying to do was far from straightforward, and indeed lay beyond their immediate competence and resources. Given time, Pons and Fleischmann might have corrected their oversights, but their zeal for publicity lost them that chance.

As Close tells the tale, the two electrochemists began with the idea that they could make deuterium nuclei fuse by electrolysis. Usually it requires temperatures and pressures like those in the interior of the Sun to get deuterium nuclei close enough to fuse. Pons and Fleischmann hoped that by setting up electric cells with electrodes made the metal palladium, they could force deuterium into the metal and squeeze the atoms so tight that a few would fuse. This seems to have been a thorough misconception. Deuterium atoms forced into palladium in fact take up specific positions in its crystal lattice that leave them further apart than when they are in solution. In any event, early in their experiments Pons and Fleischmann left a small block of palladium overnight to be charged with deuterium. In the morning, they were vastly awed and excited to find the block had vaporised, leaving traces of a spectacular explosion. They seemed to have believed they had caused a mini-nuclear explosion and from this moment on, Close suggests, they laboured under ‘an idée fixe about test-tube fusion being real’.

In this positive frame of mind, unfortunately, they neglected to avoid the pitfalls that nature invariably sets for the unwary experimenter. They believed they had found their principal proof of fusion when they measured excess amounts of heat emanating from cells operating with heavy water, also known as deuterium oxide. But they apparently neglected to perform until much later the mandatory control experiment – in this case, the replacement of heavy water with the ordinary kind. Yet a subsequent investigation by a committee of the US Department of Energy concluded they had failed to calibrate all their cells correctly and may have overestimated the amount of heat really produced. Strangely, whenever the committee was around, the cells failed to produce heat. ‘In none of our visits to the different sites did we see an operating cell that was claimed to be producing excess heat at that time,’ the panel drily noted.

Pons and Fleischmann believed they had detected the neutrons that accompany fusion, but the measurement of such neutrons is a delicate matter because it is easily confounded by the regular bath of neutrons created by cosmic rays reaching Earth. Since this invisible bath can vary sharply from one spot to another, it is essential to record the background drizzle of neutrons in exactly the same spot as the experiment expected to engender neutrons. Pons and Fleischmann took neutron measurements 50 metres away from their test cells, an invalid procedure. The quantity of neutrons they measured was in any case far too small to originate from nuclear fusion reactions. Given the amount of heat they claimed was being generated by their cells, there should have been a flux of neutrons quite strong enough to cook the two intrepid experimenters, or at least rather seriously parboil them. Strangely enough, they took no precautions to shield themselves from the radiation they should have expected had they actually obtained cold fusion.

The strong likelihood of methodological errors was, in fact, evident from the moment of their press conference. Scientific etiquette requires that before announcing one’s breakthroughs to the world one should have written up one’s results and had them either published, or at least accepted for publication, in a reputable scientific journal. The idea is that the journal’s referees will have screened the paper for obvious errors. Though this assumption is far from foolproof, it is a sure bet that scientists who jump the gun are likely to be shooting themselves in the foot. The Utah duo gave their press conference a day before they had even submitted an article for publication in Nature. When the journal’s referees offered suggestions for improvement. Pons declined to revise the article, offering the ludicrous excuse that he didn’t have time.

Given the sheer unlikelihood of cold fusion in the first place, and such strong hints that something was amiss with the Utahns’ methodology, the fervid pursuit of their ideas in other laboratories is perhaps surprising. So many scientists tried their hand at cold fusion experiments that the price of palladium rods soared. Harwell, a leading government physics laboratory in England, spent the equivalent of half a million dollars before concluding cold fusion was a crock. Close does not attempt to estimate the total spent by American laboratories in pursuit of the same chimera, but it must have been many times this sum.

Setting aside Close’s anxious assurance that the cold fusion is wholly untypical of normal science, the issue can now be raised of how well the institutions of science responded to the challenge of the Utah claim. The first question is why respond at all? For a whole month many physics labs diverted enormous effort into studying a claim, which, given time, would have collapsed of its own weight. It is reasonable to suppose such effort would not have been expended if scientists had expected merely to disprove the claims of Pons and Fleischmann. Most started out as believers in cold fusion, who hoped to claim a piece of the action and glory for proving it worked. Didn’t that reaction display rather more credulity than might have been expected of a community of professional sceptics?

The deluge of attention nevertheless had the effect of shooting down Pons and Fleischmann far more quickly than would otherwise have happened. Teams at the Massachusetts Institute of Technology and Caltech took only a month to assure themselves that the claim of cold fusion could not be right. At a meeting of the American Physical Society on 1 May 1989, the Caltech team denounced the ‘incompetence and possible delusion’ of Pons and Fleischmann.

The attempt to replicate a claim is meant to be a fixture of normal science. By this criterion, the response to cold fusion was massively normal and science worked as it is meant to. The laboratories that reported initial confirmations of the Utah claim soon found and confessed their errors, also part of the expected routine. Why then should the cold fusion episode present any embarrassment to the scientific community?

Perhaps because Pons and Fleischmann fell so easily for the golden lure of cold fusion, and for a few heady days dragged too many other sober physicists into sharing their deluded dreams. They were not a pair of graduate students who made a misjudgment. Pons was chair of his department at Utah and the author of many scientific papers. Fleischmann was an even more distinguished scientist with an international reputation in his field. For either to throw the training of a lifetime to the winds and get sucked into the mirage of cold fusion would have been bizarre. For the pair to commit this act of professional hara-kiri together was an extraordinary scandal.

The only thing that could be a worse embarrassment than their initial claim was their behaviour thereafter. Against all scientific etiquette, they declined to publish their full data, to help others to duplicate their equipment, or to answer questions fully at scientific conferences. When the facts turned against them, they refused to admit they were wrong. Pons even had a lawyer threaten to sue scientists who criticised the cold fusion claim, a truly breathtaking affront to the norms of academic debate. If senior scientists like Pons and Fleischmann could suddenly kick over the traces in pursuit of their private chimera, how many other seemingly sober researchers might be off on similar wild kicks? And if so, who is reigning them in? Claims less spectacular than the siren call of cold fusion might well be ignored and allowed to fester.

Close’s book doesn’t carefully explore these wider issues, but his excellent technical narrative lays the basis for others to do so. Cold fusion may not necessarily be ‘the most bizarre 500 days in the history of modern science’, as Close claims, but his book certainly establishes it as a leading candidate.