Wikiversity/Cold fusion/Skeptical arguments
An article in Scientific American, July, 2010, written by Michael Shermer, publisher of Skeptic magazine, mentioned a recent book by David Goodstein:
- "David Goodstein in his new book On Fact and Fraud [writes] . . . Other cases are not so clear. Martin Fleischmann and Stanley Pons's "discovery" of cold fusion, Goodstein concludes, was most likely a case of scientists who "convince themselves that they are in the possession of knowledge that does not in fact exist." This self-deception is distinctly different from deliberate deception."
From the book itself we learn that Goodstein is ambivalent about cold fusion, and from the sociologist Simon we learn that "pathological science" can be used to describe the reaction to cold fusion as well as to describe belief in it.
- Shermer's comment aroused responses that are familiar to those who follow this issue. Here we will look at skeptical responses to assertions that there is validity in cold fusion research and examine their relevance today, in the light of what is now known, as shown through research reports; as part of this, we will examine the quality of involved publications. This seminar page will likely change as we undertake this research; the initial goal is to list the arguments and then proceed to compare them with what is known or asserted in primary and secondary sources.
For example, statements are made by writers about the significance of the 2004 United States Department of Energy review of Low Energy Nuclear Reactions. The report and the individual comments of the reviewers, as well as the review of the field that was the submitted material to review, are all available and can be compared with what is said about it. Are statements being taken out of context or interpreted differently than shown by the overall context?
This seminar is begun with arguments collected from responses to the Shermer article, comments about it begin on this page. After each question, there is some preliminary comment, which may be edited later. The collection of evidence relating to each question should begin on the subpages linked, and discussion should be on the attached Talk page, with discussion of this overall process being on the Talk page attached to this page, Talk:Cold fusion/Skeptical arguments.
As this page and the subpages grow, we may refactor the arguments entirely into more generic statements that are found in various sources.
- 1 Many papers, but results all bad
- 2 Did the 2004 DoE review reject cold fusion?
- 3 Publication has decreased to a trickle
- 4 J. Electroanal. Chem hasn't published a positive result since 2005
- 5 Are experimental results in peer-reviewed journals "delusions"?
- 6 It should be simple to demonstrate cold fusion if it's real
- 7 If this was real, everyone would be working on it
- 8 Excess heat may be real, but it's not from fusion.
- 9 Is a leap of faith involved?
- 10 Shanahan
- 11 Wikipedia talk
- 12 /Rothwell
From the Scientific American blog comments, signed "Kemo Sabe," 01:21 PM 6/22/10.
"Cold fusion has been replicated in over 180 major laboratories, by roughly 1,500 professional scientists. These replications have been published in roughly 800 papers in mainstream, peer reviewed journals certainly sounds impressive until you discover that [...] they are all bad:
So bad that none of the CF claims survive peer review in main-stream *nuclear* physics journals -- the most relevant field. (If a single result had any credibility, you couldn't keep it out of Phys Rev or PRL or Science or Nature.)
This comment was also essentially found at , signed by "Joshua Cude," February 27, 2013 at 3:34 pm.
What publications are there in peer reviewed journals? It has been claimed that there has been, with some journals, a blackout on publication in this field, with editors rejecting papers based on general policy without submitting them for peer review. Is there evidence for this? How is mainstream science currently treating this field?
Cold fusion is a specific hypothesis regarding the source of the excess heat found by Pons and Fleischmann in 1989, and reportedly confirmed by many others. The hypothesis is nuclear, but the experimental work is chemistry. One of the general terms for the field is Chemically Assisted Nuclear Reactions. This is cross-disciplinary, because some of the experimental techniques are those employed by chemists, not by physicists, and, in fact, when Pons and Fleischmann announced they had measured nuclear radiation coming from their cells, they were in error, they had been misled by an artifact that would probably not have misled a nuclear physicist. On the other hand, Fleischmann was one of the world's foremost electrochemists, and his excess heat findings were in a field where he may have been better qualified than anyone else in the world. He and Pons had worked for five years to obtain results that were still chaotic, most cells did not show excess heat at all, others showed substantial heat, very far above normal measurement error. Nuclear physicists tried to replicate his work and failed. Is it possible that this was because they were not skilled in the necessary arts?
The skeptical claim cited is that there are none at all. It is traditional in science to afford an assumption of accurate reporting to scientific research. Where doubt comes in is in the interpretation of the results. When a scientist reports results that were not actually obtained, that's scientific fraud. Goodstein is observing that fraud is not an issue with cold fusion (generally). So when a researcher in this field reports that such and such an experiment was undertaken, and such and such a result was found, the reported result is always presumed credible. But the interpretation of it may be entirely a different matter. The range of experimentation lumped into the field of cold fusion is now huge, there is a vast body of data. But, according to Storms (2007), no hypothesis as to the mechanism of cold fusion is adequate to explain all the observed effects. What is known are experimental results. It is reasonably concluded from the available data that certain experiments will show, with statistical reliability, excess heat, not found with controls, provided that certain known conditions are set up. Is this a reliable conclusion? Given that this general phenomenon has been replicated by hundreds of research groups, and it is a chemical phenomenon, is it correct to reject it based on the practice of nuclear physics journals?
One of the problems in the field is that there has been great incentive and pressure to increase energy generation and to increase the reliability of results. So researchers have continually manipulated the conditions of their experiments in order to seek some "improvement." The general rejection of the field in 1989 cut off both research funding (eventually, until recently) and the supply of graduate students, who are the workers who normally do "replications." There is no Nobel Prize and no patent wealth in replicating someone else's experiment. So, with the thousands of experiments that have been performed, relatively few are exact replications. However, there are some. What are they? What kind of work has been reproduced that shows reliable results, either directly reliable or statistically reliable?
A great deal of publication in this field is through conference presentations and papers. Some of this work goes on to be published under peer review. What is "bad" research? Cold fusion is an experimental science at this point, with data being collected that may support one or another hypothesis as to the mechanism. We will explore the concept of "bad research" and look for possible examples. Or is there no significant bad research in this field, only research from which, say, premature conclusions were drawn?
... So bad that an expert panel hired by the DOE in 2004 concluded (for a second time) that CF is a bust, after reviewing the best of the results.
We will look at the 1989 DoE review as an historical document, covering what is known about the process, reading it and the reports about it. We will then examine the 2004 review, and how the opinion has arisen that this review was a second rejection. Was it?
There was some coverage of what some people in the field consider the best results, but others in the field consider the report itself to be inadequate. There is evidence that one reviewer, at least, misread a crucial part of the report, and then the summarizing bureaucrat at the DoE further garbled the evidence, such that the strongest evidence that, in fact, fusion is taking place in a CF cell, which is the very strong correlation of excess heat with helium production, at the value predicted by the fusion of deuterium to helium, was misrepresented as an anti-correlation. We will look at the report itself to determine if this is the case, and we will also look at other presentations of evidence on this issue.
So bad that instead of the publication rate growing exponentially as it does with real discoveries, the rate has decreased to a mere trickle of a few (bad) papers a year. And instead of reporting ever more progress, the size of the effect (which bears no systematic relation to the amount of the "fuel") becomes ever smaller. Two analyses of the publication pattern have pointed to the similarities between the CF pattern and other pathological sciences.
In 1989 negative publications on cold fusion exceeded positive. In 1990, positive and negative publications matched roughly. In succeeding years, positive publications continued, to the point that, if we only consider peer-reviewed journals, positive publications came to greatly exceed negative ones. However, total publication did decline, until about 2004, when it started to increase. It has recently increased greatly. Analyses claiming similarity to the polywater and N-ray situation were shallow, and were done before the shift in the field.
We will consider possible explanations for a recent expansion of publication.
The argument that the more precise the measurements, the smaller the effect, is a standard anti-pathological science argument. Is it true for Cold fusion? What evidence is there for this?
Cold fusion effects are not well-understood, and there can be two cells that were made identically, as far as the experimenter knows, that show different results. Some physical processes are chaotic like that, but some of the reasons for this variability have been uncovered. The physical structure of solid palladium can vary greatly from batch to batch, and from sample to sample within a batch. That the results from usage of a drug may vary unpredictably does not mean that drug efficacy cannot be established; rather, it means that the human body is complex and people are different. We may think of palladium metal as being one thing, but it isn't! We will examine the evidence for this, and we will look at some kinds of possible cold fusion effects with apparently predictable results.
So bad that of the 2 selected journals, the J. of Electroanal. Chem. has not published a positive result since 2000, and nothing at all related to CF since 2005.
Some of the current work does not use electrolysis, which was, in the original work, possibly only used to generate deuterium gas, so other researchers began studying gas-loading techniques. Some of the important recent publications are in physics journals and one of the most significant cross-disciplinary journals, Naturwissenschaften.
In short, it is reasonable to conclude that cold fusion researchers are, to a man (or woman), delusional.
Many published papers related to cold fusion do not mention the "cold fusion hypothesis." Rather, they report their results. As is common, they present some reason why one might think their results significant. However, the core of any experimental report is what they did and what they observed when they did it. The journal determines if the work is likely to be of interest. Thus, if a peer-reviewed journal publishes a paper in the field, they have judged the results are of interest, and they have rejected the idea that the report is "delusional." That does not mean that error is not published, it is sometimes. But, typically, the error is only visible later. The opinion of this skeptic is important, because underneath some of the skepticism is an unexamined belief that all cold fusion researchers are naive and gullible and deluded, in spite of the great caution some exercise to avoid coming to premature conclusions.
Many people have strong opinions about cold fusion and cold fusion research. What is the basis for these opinions?
You're claiming something as simple as *heat* from water on a table top. If it's real, one good demonstration is all that it would take.
It is very obvious that if cold fusion occurs in nature, it occurs at extremely low rates, rates so low that they might be impossible to detect directly. The conditions of cold fusion are difficult to obtain. For example, some of the early negative replications apparently did not reach a loading ratio of over 70%, whereas the effects start to be seen, from later studies, somewhere around 88%. Obtaining this level of loading of a palladium lattice is not simple in "water on a table top," experts who did eventually find excess heat have said that this was the most difficult experiment they ever performed. One series of results obtained by Storms showed that, if the D2O used was 98.5 instead of 99.9 atom% (i.e., the H percentage was up to 1.5%), the effect was completely quenched. D2O is hygroscopic, it absorbs water from the air, gradually increasing in H percent. So some carelessness in handling this material can easily cause replication failure. A successful CF experiment is not spectacular (usually!). Excess heat of a few watts isn't impressive, even though excess heat on the level of milliwatts can be measured.
We will look for examples.
If a credible demonstration of CF existed, you'd have to beat the young (and old) researchers off with a stick. We know this because when the world thought (briefly) that a credible demonstration of CF existed in 1989, there were D/Pd electrolysis experiments in physics and chemistry labs around the world, and at the following ACS meeting, the delegates were hanging from the rafters at the CF sessions. Then, reality hit...
In 1989, it's said that half the discretionary budget of U.S. research was diverted to cold fusion research, briefly. Unfortunately, the press conference had left the impression that this was an easy experiment to perform. Almost all researchers gave up before setting up the necessary conditions. Miles, one of the researchers who had communicated negative results to the 1989 DoE panel, has written that when he later started seeing positive results, he telephoned them to tell them, they did not return his phone calls. Miles went on to be the first to discover the correlation of measured excess heat with measured helium.
We will examine what evidence exists that could indicate this, and we will look at alternate explanations.
In any case, the result that has been widely replicated is, for the most part, anomalously high temperature readings in D/Pd electrolysis. Most scientists probably believe that those readings were made (are real), but for many, that's where credibility ends. The connection between temperature and excess heat requires assumptions about calibrations and chemistry that are not so simple. Calorimetry is notorious for artifacts.
Temperature is not what is measured, as such, in some forms of calorimetry. Rather, the energy necessary to maintain a constant temperature may be measured. Many different forms of calorimetry are used, but calorimetry, by definition, measures heat, not temperature. The field where the best expertise in chemistry and calorimetry would be expected is electrochemistry (certainly not physics). The rejection of the excess heat conclusions by experts in the field is based on....
We'll look at why the excess heat measurements and calculations are rejected, when the norm would be to accept them and look for the cause. There was reference to pathological science. In the famous cases of polywater and N-rays, the consensus arose that these were errors when the mechanism for the original reports was demonstrated, experimentally. So...
The Miles-Fleischmann Calorimetry Model is the central technical model of isoperibolic Cold Fusion calorimetry. Their precision mathematical model exhibits the five epistemic characteristics of scientific knowledge, namely that ideas represented in the form of scientific models are 1) testable, 2) revisable, 3) explanatory, 4) conjectural, and 5) generative. Every term in their detailed mathematical model corresponds to some "mundane" process in physical chemistry that is liable to contribute to the overall energy budget of the system. A great deal of tedious technical work has gone into reviewing, testing, and revising the Miles-Fleischmann Calorimetry Model. Other techniques, inspired by the pioneering work of Antoine Lavoisier, have been adopted to ensure that nothing escapes the system without being detected, carefully measured, and fully accounted for in the overall energy budget. Have the critics been satisfied that all conceivable errors have been wrung out of the Miles-Fleischmann Calorimetry Model? And are the critics also satisfied that the experimenters are correctly measuring all relevant experimental variables that must be plugged into the Miles-Fleischmann Model to reliably compute excess heat?
Is there any way to independently confirm the calorimetry? If heat is being generated through the consumption of a fuel, there should be commensurable ash. Has an ash or have ashes been identified, and are they found in quantities commensurate with the heat, so that conservation of mass/energy is satisfied?
By whom? What theories have been advanced to explain the reported excess heat, and is this consistent with the experimental reports?
But the next step from excess heat to fusion, or anything nuclear, is nothing more than a leap of faith.
In some of the experiments, the level of excess energy is apparently much greater than possible from any known chemical process involving the constituents of the cell, even if they somehow were to function as a battery (storing energy then releasing it). There is an alternate chemical explanation that involves entirely new chemistry, hydrino theory, but this is not currently accepted as a theory within the field, and that status might remain even if hydrinos turn out to exist. Fusion was an obvious hypothesis, given that the apparent fuel was deuterium. However, almost all critics assumed that "fusion" would be "d-d fusion," which is one particular reaction, with known characteristics, and clearly those characteristics were absent. Fleischmann claimed, in his first paper, an "unknown nuclear reaction." Was it reasonable to reject an "unknown nuclear reaction" if the reaction probably was not "d-d fusion"?
We will look at the reasons why theoretical considerations led to strong rejection of cold fusion. What theory was involved? How well was this theory known to apply to the conditions of the experiment? Had the theory's application ever been attempted for that environment? What mechanism was examined from a theoretical standpoint and why that one? Are there others possible?
Fleischmann later stated that he was not attempting to discover some cold fusion mechanism to generate energy. He was, in fact, testing the theory that the predictions of the 2-body approximations of quantum mechanics would apply to the condensed matter environment. He was doing basic science, to test the theory that was being used to condemn his results and conclusion. Later published work, using the far more computationally difficult quantum field theory, has predicted fusion under certain physical conditions that might exist transiently in the palladium lattice.
Huizenga saw it in 1993 and reported it, but he expected that what he reported would not be confirmed, because, apparently, he had a firm belief that low-temperature fusion was impossible. This was the early work by Miles, showing excess heat correlated quantitatively with measured helium, in a substantial series of experiments. (From Storms, 2007, as part of the data, in 12 CF cells which generated no excess heat, no helium was found. In 21 cells where excess heat was found, helium was found in 19 of them. The quantities of helium found were roughly what would be expected from deuterium fusion.) Huizenga knew that helium was evidence of nuclear origin, and a correlation would validate both the excess heat measurements and the helium measurements. Was this finding confirmed by other researchers? Is there other evidence, besides heat, that nuclear reactions are taking place in CF cells? We will examine the evidence and the criticism of it.
Kirk Shanahan, who has published criticism of cold fusion, under peer review, especially of the calorimetry, has written a suggested section on "Conventional explanations". The subpage here reproduces that, as written by him, for study and analysis. The original, as submitted by him, is at Conventional explanations.
A Letter from Shanahan was also published in the Journal of Environmental Monitoring in 2010, and he covered his theory in a blog post at , which refers to a white paper that appears to not be accessible.
As this is written, there is an RfC on w:Talk:Cold fusion and an unusual level of discussion. Various skeptical arguments (and some questions that imply an argument) have been raised. This page will address those arguments (not those who made them). --Abd (discuss • contribs) 23:09, 27 June 2014 (UTC)
In 2006, lenr-canr.org librarian Jed Rothwell created a Wikipedia page, w:Cold fusion controversy. The page was deleted. A copy was posted on Ludwik Kowalski's pages. This page was largely a compendium of skeptical arguments and history. The Wikipedia portion of that page is copied onto our subpage, for review and commentary and further study.