A few days ago, the internet lit up with news of a new paper on cold fusion in Nature.
Google has been funding cold fusion research for the last several years. This project, though, was not publicized. The CMNS (Condensed Matter Nuclear Science) research community in general knew little about it, though there were hints and leaks. There is a National Geographic page that tells the story.
However, I’m going to start this series by revisiting an old editorial, 29 March, 1990, by David Lindley, then an associate editor of Nature. He wrote:
This is best known for its last words:
Would a measure of unrestrained mockery, even a little unqualified vituperation, have speeded cold fusion’s demise?
This editorial was rife with the characteristics of pseudoskepticism, and even disparages real skepticism, essential to science. Real skepticism is open-minded, merely not easily convinced about “extraordinary claims.” But it does not reject those claims based on existing theory, because it is also skeptical that existing theory is universally true. (It is not so open-minded that we find brains on the floor. It will point out the obvious, but it is not a “believer” position.)
This was a year after the announcement by Fleischmann and Pons. By that time, there had been some reports and confirmations of nuclear effects, but it was all still very unsettled. However, Lindley writes as if cold fusion were preposterous, blatantly impossible.
But . . . what is “cold fusion?”
Pons and Fleischmann had actually claimed an “unknown nuclear reaction,” and their claim of “nuclear” was reasonable if they had made no major errors in their calorimetry, and they believed they had seen radiation (which was apparently artifact, error.)
Nevertheless, what they had seen, clear to them, was anomalous heat, at levels that they, as highly skilled chemists, could not explain with chemistry. That would remain a mystery and it still is a mystery, though aspects are now understood. It is not what Lindley imagined “cold fusion” would be, in many ways.
It was not until 1991 that Miles announced that he had found helium correlated with anomalous heat, which was stunning, as Huizenga noted. If this was confirmed, Huizenga wrote, it would explain one of the major mysteries of cold fusion, the nuclear product. However, Huizenga expected that this would not be confirmed, because “no gammas.”
And this shows how mind-locked Huizenga and many at the time were. Gammas are found with two-deuteron fusion, very strong gammas, if helium is the product, but two-deuteron fusion only rarely produces helium, and is a very well-understood reaction (though not entirely, and part of the new paper explores that).
If helium is the main product — it seems obvious in hindsight — the reaction is not two-deuteron fusion! What is it?
Lindley looks at some theories, but simply assumes, as Huizenga, that if this is fusion, it is fusion of two deuterons. That assumption was common, including probably with Pons and Fleischmann and others who supported “cold fusion.”
There is another reaction which may be possible that does not generate that very hot gamma. Cold fusion is taking place in condensed matter, not in a plasma, so more complex structures, including electrons, are possible. Lindley does consider Bose-Einstein Condensates, but only with two deuterons. Not with two deuterium molecules. If two molecules were to fuse, the product expected would be an isotope of beryllium, 4Be8, which will decay into two helium nuclei (2He4). No very hot gamma. While there are other problems to be solved with this theory, I won’t go into them, this may well be on the right track to the actual mechanism behind cold fusion.
But all this focus on theory lost the most important principle in science: Experiment is King, not Theory. The first question to have properly asked (and some did ask it) was not, “Is this fusion?”, but “Is there a real heat effect?” And then, what conditions cause the effect, what are associated and especially correlated effects, what data can we collect?
By focusing on fusion, and looking for “fusion products,” meaning neutrons and tritium, and then concluding, when these were not found, that the heat must be an error, scientists fooled themselves. And where they were considered experts, they also fooled others who trusted them.
Truly ironic is what Lindley remembered before making the vituperation comment:
Perhaps science has become too polite. Lord Kelvin dismissed the whole of geology because his calculations proved that the Sun could be no more than a few million years old; Ernest Rutherford is still remembered for his declaration that talk of practical atomic energy was “moonshine” — but the stature of neither man has been noticeably diminished by their errors, which were as magnificent as their achievements. Kelvin and Rutherford had a common-sense confidence in the robustness of their judgements which the critics of cold fusion conspicuously lacked.
This is odd, looking at it now, knowing the history of cold fusion, and the very early comment of Steve Koonin at the APS conference in Baltimore, May, 1989:
My conclusion, based on my experience, my knowledge of nuclear physics, and my intuition, is that the experiments are just wrong. And that we’re suffering from the incompetence and perhaps delusion of Drs. Pons and Fleischmann.
It has been known for many years that the famous replication failures, that led to conclusions like that of Koonin, were based on a failure to set up the necessary conditions for the effect to be seen. That work is part of the corpus of evidence that is accepted as demonstrating how not to see the Fleischmann-Pons Heat Effect. The negative work was not experimentally “wrong.” They correctly reported that under the conditions they set up, no significant excess heat was observed, nor any nuclear product.
Lewis et al (Nature, 1989) reached a maximum “stoichiometry” (D/Pd ratio) of 80%, and there is no report of the FPHE below roughly 90% at initiation. The current report in Nature is very similar, except that the new authors are quite aware that they did not reach adequate loading, hence their call for more research.
Even reaching adequate loading is not enough. In SRI P14, a Fleischmann-Pons type cell was loaded for months to high loading, and a current protocol (ramping current up and then down) was run, while measuring “excess heat.” The same protocol was run three times. The first two times, nothing happened except a little more noise. The third time, there was clear excess heat, unmistakeable. All other conditions were the same. (And there was a hydrogen control in series, which shows no excess heat in all three runs.)
Something must happen to the material to change it. Loading and deloading palladium with deuterium puts it under stress, it can crack, and the latest thinking is that a new phase of the metal can form at high loading plus stress: super abundant vacancy (SAV) material, which can also load to a higher ratio.
Not all palladium is the same. Nobody has yet found a way to reliably create material that works immediately, or even that works at all. Some protocols are better than others, though, some show excess heat most of the time, but highly variable in amount. The evidence is strong that that the famous unreliability is due to not-understood material conditions.
Add to this the difficulties of calorimetry and the possibility of the file-drawer effect, and we have the Scientific Fiasco of the Century (Huizenga).
What is constant, though, where it has been measured, is that helium is found commensurate with anomalous heat.
That is so strong as evidence for the reality of the reaction that a jury could be convinced in a civil case with it, and possibly even in a criminal case.
I can think of no way that the helium could be consistently correlated with heat, across different protocols and conditions, in many experiments, other than being produced by the same reaction, nor have I seen any proposed that are consistent with the experimental conditions.
Heat is not going to make helium and helium is not going to make heat, if the heat is artifact (or even if not!) and if the helium were leakage or error, it would not be clearly correlated with heat, and the ratio would not so nicely approach that very special value, 23.8 MeV/4He, which is the thermodynamically necessary ratio for any reaction that converts deuterium to helium, regardless of mechanism, as long as there is no radiation loss, and there apparently is not anything significant.
I will examine the Lindley analysis in detail on a page, Lindley 1989.
This series will continue with Cold fusion is in our geography now.