Cold fusion/Skeptical arguments/Shanahan

< Cold fusion < Skeptical arguments

The following was proposed by Kirk Shanahan for the Wikipedia article w:Cold fusion. Comments, which should be signed, may be added to paragraphs, indented, as a study of the evidence and arguments presented. Dr. Shanahan has been invited to participate here. Sections have been given section headers. If someone cares to do some wikignoming, the references could be converted to in-line references, which are easier to follow.

Alternative Chemical Explanations of 'Cold Fusion' Observations

A variety of observations have been claimed over the years to support the contention that a new and novel nuclear reaction (or set thereof) had been discovered initially by Fleischmann and Pons (F&P), and separately, by Jones. The Jones claims have never been as strongly pushed as the F&P ones, primarily since the Jones observations are based in low level nuclear radiation counting techniques, which were recognized as being susceptible to various problems. Replication difficulties led Jones to take a more circumspect approach to the subject. The F&P claims however, have found vociferous support in a variety of forums, including the scientific literature and popular books. Two recent examples of this are the aforementioned 2007 book by Storms[1], and a review article by Krivit and Marwan (K&M) [2]. The following will roughly follow the outline of the latter, and refer heavily to the Comment on it published by Shanahan[3]. A Response to the Comment has been published [4], which will also be discussed briefly. The Storms book provides more detailed examinations of the extant CF literature on these topics through 2007, although in two cases key publications suggesting problems with the presented results were not discussed in either source[1,2].

The language here is appropriate for a polemic review (which is how Britz describes Shanahan). Lots of words here would be highly inappropriate for Wikipedia, but they are fine at Wikiversity as a presentation of Shanahan's opinion. "Key publications," however, apparently refers to Shanahan's isolated publications. He didn't specify which they were. However, we can see below. --Abd 15:53, 23 September 2010 (UTC)

Calorimetry

In 2002, Shanahan published a simple explanation for how apparent excess heat signals could be produced from F&P electrolysis cells[5]. This explanation was composed of 3 parts: (1) the basic problem, called the calibration constant shift (CCS) problem, (2) a contention that the calorimeter/cell needed to be considered at a minimum as a two-zone entity, which allowed an understanding of how a CCS could occur via a change in heat distribution, and (3) a postulated physical/chemical mechanism that would lead to the changes required to shift the cell steady-state configuration, which could produce a heat distribution shift between the two zones.

The basis of (1) was the recognition that calibration constants in calibration equations used to translate experimental observables into power signals were experimentally determined and thus subject to some error. The 2002 publication took real data published by E. Storms [6] and reanalyzed it under the assumption that no excess heat had been produced. It was found that this assumption forced one to change the calibration constants to drive the apparent excess heat signal to zero. What was noted was that the changes required were minor, being ~1-3% of the originally determined values. Further, it was recognized that there was a consistent pattern in the shifts with time and experimental operations that clearly showed the problem to be systematic, i.e. non-random. It was also pointed out that the estimated shifts were of approximately the same size as the variations in constants obtained by calibrating with different methods and at different times. This led Shanahan to conclude the CCS was a feasible explanation for the apparent excess heats.

Shanahan went further and postulated that a shift in heat distribution in a cell might cause the CCS, showing this mathematically would lead to a CCS. Shanahan also proposed a physical/chemical mechanism for how this shift might occur. This involved the formation of an unspecified 'special active surface state' which would promote at-the-electrode hydrogen + oxygen recombination, which would cause extra heat to be deposited at the electrode instead of at the recombination catalyst in a closed cell or being lost out the vent tube as unreacted H2 + O2 in an open cell. A surface state was proposed since (a) the cathode in these studies was Pt, which is not known to hydride under any conditions (thus no bulk hydride could be involved), and (b) the free metal surface under the growing hydrogen bubble would serve as the recombination catalyst.

The primary impact of the CCS concept is that it negates the idea that baseline noise represents the only important error in these experiments. The Storms data showed a 780 mW peak, but this was now explained as an ~3% shift in calibration constant. It was suggested that simple sensitivity analysis of what apparent excess power signals could be expected from 1-10% changes in calibration constants should be part of every subsequent cold fusion calorimetric study.

Unfortunately the CF community has rejected the CCS concept out of hand, as illustrated by the recent publication[4]. However, the CF community has apparently failed to grasp the CCS concept, since the recent primary thrust was to discredit “Shanahan's random hypothesis”[4], which indicated a complete misunderstanding of the systematic nature of the CCS as presented by Shanahan [3,5,7,8]. This has led to additional issues when they consistently apply their erroneous rejection to the other aspects of CF observations easily explained by Shanahan's proposed mechanism. For example in reference to HAD events, Shanahan contends these are events interpreted based on the same calibration equation while the severe change in cell conditions would in fact require recalibration. But since the CF community rejects the CCS, they reject this contention as well.

Likewise, Storms attempted to discredit the proposed Shanahan mechanism[9], but Shanahan rebutted this attempt[8]. Unfortunately Storms [1] stated that he had successfully discredited the mechanism but failed to mention the Shanahan rebuttal. This problem was described [3], however, in the response[4], the authors repeat this mistake, again indicating their lack of understanding of the Shanahan proposals.

Transmutation

Two types of transmutations have been discussed in the literature based on the observation of He in gas samples and on heavy metal content in/on solids. The original F&P claims [10] suggested deuterium fusion as the source of the apparent excess heat, which immediately suggested that 4He should be observed. Subsequently, various groups reported detecting 4He. It was pointed out in the 1989 DOE Review [ref] that this had to be shown to not be due to leakage of external 4He into the apparati. Various CF researchers have claimed to have shown this. However, the SRI group supplied samples to B. Oliver of PNNL, who, working with W. B. Clarke of McMaster University, a recognized expert in low level He measurement, found the samples grossly contaminated with air [11]. At this point no replication of this experiment has been attempted, so it remains an open question as to when the air was introduced. But if it was from the SRI phase of the work, then clearly the SRI group would not have solved the in-leakage problem recognized in 1989. No CF researcher has presented adequate documentation that environmental He has been successfully excluded from their apparatus, thus the use of He results to conclude active nuclear reactions is unsupported at this time. (This documentation would include full disclosure of analytical methods and results, including from calibrations and background measurements, and replicated results with those methods (preferably at other laboratories).)

Heavy metals were found on active F&P-type cathodes very early on and their presence in general is not contested. Some time later these results began to be claimed to be from transmutation reactions, as opposed to the commonly accepted contamination processes, and as well were claimed to often have anomalous isotope distributions. No substantial proof of the nuclear source of these metals has been offered, some unpublished results clearly show contamination as the identified source (discussed in [3]), and several data misinterpretation examples can be identified, leading to the conclusion that contamination still remains the most likely cause of the appearance of these metals.

The evidence for transmutation is generally difficult to analyze, in my opinion. Helium is the most important result of transmutation, however, and the evidence regarding helium is conclusive that it is the primary nuclear ash. Shanahan does not, here, seriously address this, yet it is the most important result, confirmed by many research groups, see Storms (2010) in his review of the field, showing that there is fusion taking place. If fusion is taking place, minor branching or secondary reactions could produce transmutation. Thus, once we know that it is possible for the condensed matter environment to bypass, in some way, the severe problem of Coulomb repulsion, we should not be surprised to find some level of other nuclear transmutations. The alleged impossibility was never actually known to be true, it was an assumption, based on approximations known to be inaccurate beyond the two-body problem, but the degree of inaccuracy wasn't known, and that, in fact, was what Pons and Fleischmann were investigating. Not "free energy," as some have though. It was basic scientific research, and it's tragic that it was rejected for superficial political reasons.
What was probably, what remains probably, impossible, is two-deuteron direct fusion at low temperatures, though even that statement must be qualified. (Muon-catalyzed fusion being a notable exception, and is some other form of catalysis possible? But MCF produces the same branching ratio as hot fusion, and thus strong radiation, and so we can expect that simple two-deuteron catalyzed fusion is likely not the explanation for CF phenomena.) --Abd 15:53, 23 September 2010 (UTC)

CR-39

The Shanahan mechanism proposes that the hydrogen + oxygen recombination in bubbles produces explosions (also contended to be present by Szpak, et al [3]). These explosions could potentially induce mechanical damage in the CR-39 material used to search for nuclear radiation, which in turn would lead to the observed copious pits in the etched material as mechanical damage is known to produce pits in etched CR-39 material [3]. The CF community rejects this hypothesis since they believe no at-the-electrode recombination is occurring, thus they are forced to conclude actual nuclear particles have been formed. The CF community had made much of the apparent similarity of so-called 'triplets' found in the sparsely pitted areas of CR-39 plates to those found in plates exposed to D-T fusion neutron sources [4]. However, until such time as the details of how the copious explosions in SPAWAR cells might or might not produce pits are worked out, it is impossible to say if the pits and 'triplets' in the CF CR-39 materials are due to nuclear particles or not. Essentially until the parameters of how and when mechanical shock can induce pitting are worked out, the CR-39 method is not exclusively diagnostic of nuclear reactions.

Shanahan cherry-picks the presented evidence, to exaggerate what he believes he can criticize, and he neglects aspects of that evidence for which he has no alternate explanation. Let's go over this:
  • hydrogen + oxygen recombination. Experimenters always consider hydrogen + oxygen recombination. However, the large bulk of the oxygen generated in the cell will bubble out. The flow from the anode, where the oxygen is generated, is up to the surface of the electrolyte, but it is certainly reasonable to suppose that some oxygen will be dissolved in the electrolyte, and some oxygen bubbles will be swept about in the circulation, and some might make it to the cathode. Some very small number of such bubbles. The "explosions" noted by Szpak et al are shown in two ways: through acoustic effects, there are occasional "pops," on the order of a few per second up to perhaps dozens per second, that show up in oscillographs of the voltage from a piezoelectric sensor which was made the substrate for the codeposited cathode. Then, in infrared imaging of the cathode while active, hundreds or thousands of small bright spots transiently appear. In addition, SEM images of a codep cathode, after removal from the cell, show holes in the deposited palladium, on the order of ten microns across. Some images show what appears to be molten ejecta, that froze around the hole, and elemental analysis of the material surrounding the hole shows isotopic anomalies that could indicate a nuclear reaction.
  • Shanahan posits this as due to D/O recombination. Problems with this:
  • There is no evidence of this reaction, as an explosion, taking place. We would expect there to be spectrographic evidence, characteristic emissions would occur. Shanahan is a theoretical critic, proposing alternate explanations, which is fine, in itself. But we need to keep in mind that these are only theoretical proposals, and sometimes they involve quite a stretch, so far that experimenters are probably not motivated to spend their time confirming them. I will be running a codep cell which might be expected to generate these phenomena, and I'll be watching for sound with a piezoelectric detector, though not fabricated as part of the cathode. Still, with filtering, I ought to see something. I will also be watching the cathode with a digital microscope, perhaps recording video. Again, I might or might not see something, even if there is a reaction. But having the spectrographic equipment to look for characteristic emissions from H2/O2 explosive recombination, that's completely beyond me unless someone supplies it. The experiment, to be sure, is designed to allow direct observation of the cathode, it sits against the cell wall, so until and unless the cell wall is seriously fogged, it should be possible to see this.
  • To create a ten-micron hole, melting the palladium surface, the bubble must be on the order of ten microns in size itself. There would be no mechanism to concentrate the heat from a larger explosive bubble to a ten-micron region. A ten-micron H2/O2 bubble would not release sufficient heat to melt palladium. (That's my strong impression, but I'll certainly correct this if I see math indicating otherwise.)
  • The apparently explosive phenomena and CR-39 damage of the type claimed as resulting from nuclear radiation, do not appear when light water is used in place of heavy water. Shanahan ignores that there are light water controls.
  • Shanahan may not realize that "triple tracks" are not merely three tracks co-located, perhaps by coincidence. In some images, where the depth of the originating point is just right, you can see grooves in the bottom of the image, where three tracks come together at a single point. Further, triple tracks of this kind are seen, most frequently, on the back side of the CR-39, as would be expected. (If the incident energetic neutrons are coming from below, as we are looking down at the CR-39 surface, the larger, more etched, pits would first appear as separated pits, then would grow together as etching moves more deeply into the plastic, until the point of origin is exposed. A triple-track produced by a neutron incident from above would first etch as the point of origin, and by the time the large pits are developed for the individual proton tracks that resulted from C-12 breakup, the point of origin would be completely etched away. These pits would originate at all levels within the CR-39, as a neutron happens to hit a C-12 atom, just right to fission it into three alpha particles.
  • Shanahan is positing that mini-explosions, detectable with a very sensitive piezo sensor in intimate contact with the cathode material, but otherwise not observed by anyone, to date, are capable of penetrating a millimeter of CR-39 plastic, very hard stuff, it's used for eyeglasses, and causing precise pitting (10 micron region) on the other side, without causing massive damage on the front side, directly against the cathode. Very hard to imagine with the copious apparent proton tracks that they find on the back, impossible to imagine, for me, anyway, that it would cause the very characteristic triple tracks from C-12 breakup. These tracks look just like those known to be caused by neutrons, this is why that report was considered to have knocked the ball out of the park, for some physicists (if confirmed, of course, not done adequately to date.) Most of the back-side tracks seen are single, apparently from knock-on protons, this is the major radiation used to detect neutrons with Solid State Nuclear Track Detectors, triple-tracks are very rare.
  • I will be using LR-115, a detector material that relies upon a similar phenomenon as CR-39, but I will use a two-layer detector, being two sheets of this material with the sensitive material adjacent. That sensitive layer is 6 microns of red cellulose nitrate, which makes imaging really easy, and it shows very precisely that damage occurred at a very narrow depth, compared to CR-39. The films will be pin-registered so that the images, after development, can be compared, to see if tracks crossed that 2-layer interface. Because those films will not be placed together until the experiment is done, this should allow me to almost completely avoid counting background tracks. The films will be placed against the cell wall opposite to a cathode. I'm just looking for energetic neutrons, which is why I'm using a gold cathode substrate. I should be close enough that I'll see them, if they are being generated at the levels SPAWAR has claimed. But far enough away, and protected, so that no chemical or explosive damage hypothesis would be tenable. But it's not tenable with the SPAWAR work, either, which has shown back-side results, mylar-protected results, and, of course, lack of tracks on controls with light water or other conditions.
To summarize, Shanahan has neglected the reality of these experiments, in order to create a Rube Goldberg hypothesis, and he's done this in many areas, justifying it by the idea that "cold fusion is so utterly impossible, contrary evidence must be bogus, and we should cling to that until every contrary hypothesis has been completely ruled out. And, of course, the human mind can always invent "contrary hypotheses," but at some point, Occam's razor properly sets in, and we start to re-examine the original assumptions.
Yes, there were very good reasons to doubt that straight deuterium-deuterium fusion was talking place. But to doubt that any kind of nuclear reaction could possibly take place at low temperatures was a stretch, an error.
Exceptions were already known where the environment of condensed matter caused nuclear effects. And, now, that belief in impossibility has been directly countered by theoretical work by Takahashi (a hot fusion physicist), which predicts fusion from a possible -- but very rare (if it occurs) -- physical configuration, a tetrahedral formation (at low relative temperature, of four deuterons (two deuterium molecules, electrons included) which, similar to the Oppenheimer-Phillips process, as I explain it, apparently creates a combination of conditions, combined with a Bose-Einstein collapse, that allows the nuclear force to take over. Unlike the OP process, the BEC collapse apparently pulls the protons in as well, resulting in Be-8, which then fissions. There are unsolved problems with this hypothesis, but it is no longer tenable to assert impossibility based on quantum mechanics. The more complete quantum field theory now predicts fusion is possible. --Abd 15:53, 23 September 2010 (UTC)

An alternate theory

A student here proposed an alternate theory, of heat released by deuterium forming deuterium molecules, D2. This has been moved to Talk:Cold fusion/Skeptical arguments/Shanahan#An alternate theory. (The original section here can be seen in history at .

Temporal Correlations

The CF community has also emphasized correlations between apparent excess heat signals and measured He levels. Shanahan pointed out that such correlations are meaningless when the actual identity of the plotted parameters is unknown [3]. Since excess heat signals are not proven real, and since He signals are not proven to exclude air leaks, correlations between these variables are meaningless. This point was also not understood by the CF community.

Correlation with helium is independent evidence of the accuracy of the calorimetry as of the accuracy of the helium measurements, because it is difficult to assert, absent some very creative and totally unverified imaginations, that the very strong correlation known from multiple research groups and many individual cell studies, that one of these effects causes the other, nor that they have a common cause other than fusion, or some other nuclear phenomenon, because helium isn't created by any other kind of process. The heat involved here is small, as is the amount of helium generated, so the idea that, say, helium stored up in the cathode by some means, is being released by the heat, is preposterous, and would not show the temporal behavior Shanahan refers to, nor the overall correlation with excess heat, and there are other problems with this idea.
Shanahan likes the word "proven." What we actually look for is strong evidence. Nobody but Shanahan has opined under peer review that these results are meaningless. Huizenga, in 1993, noted just how important they were, he simply expected, because of his very strong prejudice, that the work would not be confirmed, even though what he was noting was work that itself was a confirmation, with more data, of earlier work. It was confirmed by others, quite sufficiently, well enough that Storms ventures a specific estimate of heat/helium: 25 +/- 5 MeV, which compares very well with the expected value for any process (fusion or otherwise) that converts deuterium to helium.
Shanahan emphasizes the temporal correlation, I suspect, because that is based on a single study of a single cell. The general heat/helium correlation is based on study of many, many cells, by many researchers. Miles' work, reported by Storms (2007 and 2010), covered 33 cells, of which 12 showed no excess heat. None of the no-heat cells resulted in detected helium. The helium measurements were done blind. Of the 21 remaining cells, 18 showed helium, in amounts commensurate with the heat measured. Thus we are seeing a possibility of error, in fact. But only in one direction. Storms notes that with the 3 exceptional cells, for one the calorimetry was suspect, and for the other two, there was a different composition of the cathode. Frankly, if we were just studying the FPE, those wouldn't have been included at all!
One possibility that was widely neglected, particularly by skeptics, is that there is more than one kind of nuclear reaction! But many, many results remain mysterious in this field. The three anomalous results, compared to the others, must be considered in determining the probably that the correlation is random. But they are drastic outliers, with plausible explanations for this.

Shanahan also noted that selective data presentation was occurring in the recent review [2], and referred to the same data in the Hagelstein report [12]. He asked the question of why the He levels appeared to decrease after some point, which should not occur. The response offered an undocumented assertion that the sample had absorbed some of it [4], which is an inadequate answer, experimental proof or literature references are required.

The phenomenon of increase in helium, followed by a relatively small decrease as cell conditions change, is not clearly explained in the literature, re-absorption is possible, or there could be leakage, accompanied by a lessening in generation, particularly if there is positive pressure. (If there is positive pressure, some kinds of leakage wouldn't affect the helium results, they would lower them generally. But helium will also diffuse through glass, which must always be considered, and I don't know how much this varies with pressure.) As to why the helium output would decrease, perhaps reaction sites closest to the surface have been exhausted. The greater the depth at which the helium is being generated, the less of it will escape to be detected.
"Experimental proof or literature references" are not required to propose a possible explanation of an anomaly. --Abd 15:53, 23 September 2010 (UTC)

A General Problem

As can be seen, conventional explanations are varied and might or might not be applicable in any specific given case. For example, the Shanahan CCS mechanism is clearly not applicable in a gas-solid only system, while a CCS caused by another mechanism might be. However, the CF community consistently lists a variety of experiments and results under the single broad heading of LENR and attempts to derive confidence from the sheer number of such reports. The problem with this is that this is an illegitimate approach to scientific inquiry. Before individual experiments can be included in the whole body of knowledge there needs to be confidence that the individual results are reliable. Reliability is always indicated by detailed reproducibility. At this time, no CF experimental protocol is capable of producing reliable results. Until such time as a set of reliable experimental results is obtained, different experiments cannot be shown to result from the same putative source (LENR).

This is a reasonably clear statement of a common objection. There are few exact replications, using the same protocol. However, the massive rejection of cold fusion in 1989-1990 insured that the normal labor available for exact replications, grad students, was cut off. So researchers focused, for the most part, on new explorations. However, exact (or close) replications do exist, such as the McKubre and ENEA replication of work by Energetics Technologies. Further, following the same basic protocol, i.e., loading palladium cathodes with deuterium, has produced similar results even with differing methods of measuring excess heat. The requirement set by Shanahan (and implied by others) is not intrinsic to "scientific inquiry." It is desirable, in some cases, but not essential.
From the paper, "Replication of Condensed Matter Heat Production," by McKubre et al, published in Low Energy Nuclear Reactions Sourcebook, published by the American Chemical Society, 2008 (p. 221-222):
Since 1989, SRI has attempted to replicate roughly ten experimental results reported in the field of cold fusion or condensed matter nuclear science (CMNS). A reproduction protocol has been developed to maximize the chances of success and useful learning in this process. The SRI reproduction protocol employed for the replication of Energetics' results described below involves three steps.
  1. A "host hands off" rebuild by the original experimenter allowing simple technical support, but with conscientious effort not to improve the experiment at this point.
  2. A complete transfer of procedures and performance characteristics from the guest to the host, complete when the experimenter is running to the satisfaction of the originator and the host scientists believe they understand what is being done and what is required.
  3. Host (SR)) operation of the experiment with added diagnostics as suggested or required from the results of Step 1. Only when the experiment is running in the same way and with the same results as for the originator is an attempt made to "improve." In the present program, such "improvement" was restricted to sophistication of the data analysis, as the underlying Energetics protocol was demonstrated to be sound.
Shanahan asserts that there were no detailed replications; assertions like this, which are simply false or misleading, are common in skeptical commentary on cold fusion.
Further, there is a "protocol" which does not depend on specific experimental procedure, and it has been followed many times, with varying degrees of care, but always with consistent results. This is the simultaneous measurement of excess heat and helium, in palladium deuteride experiments. Many research groups have measured both heat and helium, and this includes some of the early "negative replications." No heat, no helium. But if a group finds heat, it almost always finds helium, in amounts commensurate with the heat. This kind of result cuts through the claims of "no replication," and it does more than that. It establishes that the "ash," the product of the reaction, is almost entirely (or certainly substantially) helium. If helium is being produced, there is a nuclear reaction of some kind. This set of experiments, by twelve research groups, with the detailed work of four groups on which Storms bases his estimate of 25 +/- 5 MeV for the heat/helium ratio (23.8 MeV being the theoretical value for heat produced when deuterium is fused to helium), is considered by some the most conclusive evidence for cold fusion that exists. It validates both the heat and helium results.
Shanahan has claimed that if the heat measurements are not accurate, heat/helium is meaningless. This is wildly at variance with standard scientific procedures, where the correlation of variables is used to validate the measurements and expose possible artifacts. Shanahan -- and practically nobody else -- has proposed that heat from unexpected deuterium/oxygen recombination drives helium from the cathodes; however, that would not explain the actual experimental data, and is a prosaic explanation that would be the first considered by experimenters. The "unexpected deuterium/oxygen recombination" is a result that would not affect "excess heat" in many of the experiments, where deuterium and oxygen are recombined within the cell and would not produce an excess heat artifact. Further, the energy does not appear, and the helium does not appear, in hydrogen controls. --Abd 16:58, 4 November 2010 (UTC)

REFERENCES

This article is issued from Wikiversity - version of the Saturday, August 16, 2014. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.