Cold Fusion: A Study in Scientific Controversy

Dongwoo Chung
March 14, 2015

Submitted as coursework for PH241, Stanford University, Winter 2015


Fig. 1: A deuterium gas-based cold fusion cell used circa 2007 by SRI International, following in the footsteps of Fleischmann and Pons. (Source: Wikimedia Commons)

The United States Patent Office's Manual of Patent Examination Procedures mentions cold fusion alongside perpetual motion and other inventions of "incredible" utility or "asserted utility ... inconsistent with known scientific principles". [1] But regardless of the mainstream attitude towards the subject, a contingent of cold fusion researchers work in both public and private sectors today, continuing experimental efforts to produce excess energy via nuclear reactions at low temperature. (See Fig. 1 for an example apparatus circa 2007.) In February of 2012, for instance, the University of Missouri had "received a $5.5 million gift to study the subject." [2] To understand the current state of cold fusion research, it may help to understand the 1989 claims of Martin Fleischmann and B. Stanley Pons, as well as the widely publicized events surrounding these claims.

The Pons-Fleischmann Plans

Mechanisms for cold fusion had been contemplated as early as 1926. [3] In fact, as early as 1934, scientists were speculating about hydrogen dissolved in palladium in relation to nuclear fusion. [3] An alternate mechanism, muon-catalysed fusion, was studied by Andrei Sakharov and later observed by Luis Alvarez in 1956, but, according to Lewenstein and Baur's chronology of cold fusion, scientists decided that this particular mechanism was "not likely to be a panacea." [3]

By the mid-1980s, two cold fusion research efforts that would come to gain widespread attention in 1989 were underway: the collaboration between Pons and Fleischmann at the University of Utah, and the work of Steven Jones and colleagues at Brigham Young University. [3] Both efforts attempted to produce D-D fusion, with two possible reactions under consideration: [4]

2D + 2D → 3He + n (1)
2D + 2D → 3T + H (2)

These equations describe fusion of two deuterium nuclei either (1) into helium and a neutron, or (2) into tritium and common hydrogen. The above reactions will also produce excess energy on the order of 3 to 4 MeV. [5] But Pons and Fleischmann wanted to achieve these reactions through electrolysis of heavy water. How could these fusion reactions happen in an electrolysis cell (shown in Fig. 2) - an apparatus that the New York Times, in its coverage of Pons and Fleischmann's 1989 announcement, said "would seem familiar to a ninth-grade student of general science"? [6]

Performing electrolysis in the cell involves passing an electric current between the cathode and the anode, through the heavy water. This splits the heavy water into its constituent deuterium and oxygen atoms, as would happen in electrolysis of light water. [6] Pons and Fleischmann's 1989 paper writes out this splitting as the reaction

Fig. 2: Diagram of an electrolytic cell. In Fleischmann and Pons's apparatus, the cathode was palladium, and the anode was platinum. [4,6] (Source: Wikimedia Commons.)
D2O + e- → Dads + OD-

where the subscript "ads" denotes that the deuterium is adsorbed, i.e. accumulates on the surface of the cathode. [4] The deuterium atoms are subsequently absorbed into the lattice of the cathode, which in this cell is palladium, and Pons and Fleischmann's paper writes this out as its own reaction step: [4]

Dads → Dlattice

The paper notes a feature of absorbed deuterium in palladium that "can be explained only if the H+ and D+ in the lattice behave as classical oscillators [...] i.e. they must be in very shallow potential wells." [4] Pons and Fleischmann also note that the deuterium in palladium appears to have a high chemical potential, which they equate with high compression of deuterium. [4] Pons and Fleischmann then reason that if the deuterium atoms are so highly compressed within the palladium lattice, as well as highly mobile, with each confined only by a shallow potential, "there must therefore be a significant number of close collisions" and the question arises as to whether the deuterium can fuse into helium or tritium as a significant part of these significantly numerous collisions. [4] This question involves the quantum-mechanical interaction between the deuterium nuclei (also called deuterons), namely the possibility of the deuterons tunnelling through the energy barrier between them towards each other, which will be discussed in more detail shortly.

Unlike Fleischmann and Pons, Steven Jones's initial steps in cold fusion research were not related to the approach outlined above, but rather involved a renewed interest in the previously observed mechanism of muon-catalysed fusion. This mechanism involves the fusion reactants both becoming bound to the muon, which has an incredibly high mass (207 times the electron mass) and thus enhances the fusion rate and acts as a catalyst for the fusion reaction. [7,8] Later, however, Jones started converging independently on Fleischmann and Pons's ideas, and began collaborating with the chemistry department on the same electrolytic approach to cold fusion. [3]

As the two efforts neared publication in 1989, Pons, Fleischmann, and Jones initially reached an agreement to simultaneously submit their results to Nature on 24 March. [3] However, Fleischmann and Pons submitted their paper to the Journal of Electroanalytical Chemistry on 11 March, apparently due to pressure from university administrators over patent concerns. [2,3] Pons and Fleischmann subsequently announced their results at a university press conference on 23 March, leading Jones to submit his paper to Nature that day. [3] Pons and Fleischmann sent their submission to Nature the day after. [3,9] The New York Times coverage of the initial announcement noted that scientists were outraged by the opaque nature of the announcement, particularly bemoaning the fact that the press conference occurred before any part of the work had been formally published. [6] Pons, in an interview with the Times, claimed that the work had undergone peer review by a journal, but refused to specify details. [6]

Fig. 3: A calorimeter used at the New Hydrogen Energy Institute in Japan. Excess heat is produced as part of nuclear fusion, but other fusion products should also be measurable. (Source: Wikimedia Commons)

Their submission to the Journal of Electroanalytical Chemistry eventually appeared in April, reporting not only excess heat, but also tritium production and a γ-ray spectrum. [4] (Fig. 3 shows a diagram of a calorimeter of the sort that would be used to measure excess heat in this sort of experiment.) The D-D fusion reactions shown above do not directly produce γ-rays, but since the cell is placed in a water bath for experimental measurements, Pons and Fleischmann claimed that the neutrons would be captured by the hydrogen in the water bath in this reaction: [4]

1H + n → 2D + γ

In their description of this neutron capture process, Pons and Fleischmann claim that capture of a 2.45 MeV neutron, which would be produced in D-D fusion with helium (see reaction (i) from above), would result in a 2.5 MeV γ-ray. [4] However, the paper later shows the experimentally measured γ-ray spectrum to have a peak around 2.2 MeV energy. [4] This discrepancy was later corrected in an erratum, but allegations would later surface about key data related to this γ-ray spectrum being manipulated between the initial announcement and the publication of the paper, and even this paper had to be followed up by another one in the July 1990 issue of the journal - 16 months after the initial announcement - that presented accurate data. [3,9,10] Questionable data notwithstanding, the 1989 paper claimed observation of neutron flux of around 4 × 104 neutrons per second. [4] Furthermore, the paper claimed measurements of tritium production that put the rate of fusion reaction of Eq. (2) on the order of 104 atoms per second. [4] This rate, while extremely high for a fusion reaction, was not enough to explain the excess heat production observed, which would require reaction rates as high as 1014 atoms per second to explain. [4] Pons and Fleischmann explained this as indication that "other nuclear processes must be involved." [4] But even if other processes were involved, they could not have been neutron-producing nuclear processes at that sort of rate, as the neutron radiation produced would be overwhelmingly fatal to anyone present in the lab. The flux observed for 2.5 MeV neutrons corresponds to a radiation power of

4 × 104 neutrons per second × 2.5 MeV per neutron × 1.6 × 10-13 joules per MeV
= 1.6 × 10-8 joules per second

whereas had this flux matched the aforementioned 1014 atoms per second, the corresponding power would be

1 × 1014 neutrons per second × 2.5 MeV per neutron × 1.6 × 10-13 joules per MeV
= 40 joules per second

In the first case a person with 80 kg body mass would be absorbing 2 × 10-10 Gy per second of neutron radiation, whereas in the second case that same person would be absorbing 0.5 Gy per second. If some sort of neutronic fusion reaction (with the neutron energy on the order of a few MeV) had occurred at the rate suggested by the excess heat observed, the Utah group would likely not have been alive to hold a press conference on these findings.

Pons and Fleischmann withdrew their submission to Nature when its referees raised criticisms, saying "they were too busy to respond". [11] Nature did, however, accept Jones's submission, which was published in the 27 April issue. [12] While Jones's paper, describing work done with titanium electrodes rather than palladium electrodes, also reports detecting a flux of 2.5 MeV neutrons, the detection is reported to have been through a "neutron spectrometer" developed at BYU, rather than detection of γ-rays from neutron capture in a water bath. [12] Furthermore, the reported neutron flux is on the order of 4×10-3 counts per second - many orders of magnitude lower than what Fleischmann and Pons reported. [12] On the basis of this neutron flux, Jones claimed a fusion reaction rate for reaction (i) on the order of at least 10-23 fusions per deuteron pair per second, possibly as high as 10-20 per deuteron pair per second (a level that Jones deemed "readily measurable"). [12]

In the same issue as Jones's article, editor John Maddox noted a lack of basic control experiments in the Utah group's work, decrying this omission as a "glaring lapse" in scientific investigation. [13] Indeed, while Jones's paper notes that electrolysis runs were done with light water (H2O) in place of heavy water (D2O) to understand experimental backgrounds, Pons and Fleischmann disclose no such runs. [4,12] Maddox further cautioned that however positive the initial reaction from the general public may have been, it is "possible that the general reaction to the failure of attempts at replication will be more sour." [13]

A Scandal in Utah

Initially, the optimism seemed justified. Lewenstein and Baur, in their chronology of events, list a slew of confirmation announcements. Within days, scientists at Tokyo University and at Kossuth University in Hungary were reporting similar results. [3] Within a month, groups at Texas A&M and Georgia Tech claimed to have reproduced the results at least partially, with the Texas A&M group reporting excess energy (and later tritium production) and the Georgia Tech group reporting neutron and tritium production. [3] Reports of replication from all over the world followed - from Moscow, from Florida, from Italy, from Brazil, and from Stanford, among others. [3] Palladium futures prices reached an eight-year high in anticipation of demand arising from fusion research. [3,14] At the 1989 meeting of the American Chemical Society in mid-April, Pons received a "rousing ovation," as the New York Times put it, at a session attended by about 7000 scientists. [2,3]

But the day after, Georgia Tech retracted its initial confirmation, reporting a serious flaw in their neutron detection methodology. [15] Only a few weeks afterwards, a meeting of the American Physical Society in May severely condemned the original claims. While those in attendance sharply questioned Jones's work at BYU, noting that the measured neutron levels were only very slightly above background, they tempered their criticism of Jones, according to the New York Times, "acknowledging that Dr. Jones is a careful scientist." [16] On the other hand, teams at MIT, Lawrence Berkeley Laboratory, Brookhaven National Laboratory, and Yale University all reported failure to replicate the Utah group's results, with a CERN representative also reporting that "'essentially all' West European attempts to duplicate the Pons-Fleischmann experiment had failed", and denouncing the work as "pathological science." [16] Those who had attempted to reproduce these results also reported severe difficulty replicating the experimental setup in the first place, due to extreme secrecy and lack of cooperation from Pons and Fleischmann. A scientist from Oak Ridge National Laboratory noted that his team had been forced to estimate the size of the Utah group's fusion cell by comparing the size of the cell against the size of Pons's hand in a photograph where Pons was shown holding the cell. [16] A team at Caltech attempted "every possible variant of the Pons-Fleischmann experiment [...] without success." [16]

A press conference at the APS meeting saw eight of the nine leading speakers ruling the Utah claim "as dead," with one abstention. [16] In stark contrast to the ACS meeting, here it was Steven Koonin of Caltech that, according to the Times, received great applause for calling the Utah group's report "a result of 'the incompetence and delusion of Pons and Fleischmann.'" [16]

The Problem of Coulomb-Barrier Tunneling

Fig. 4: Quantum tunneling of a particle, depicted by the wave function of the particle traversing a potential barrier. (Source: Wikimedia Commons)

Koonin and fellow theorist Michael Nauenberg later wrote a letter to Nature quantitatively examining the claimed observations of both the Utah and BYU groups. [8] Specifically, Koonin and Nauenberg examine the details of tunneling, the previously mentioned quantum-mechanical effect between deuterium nuclei in a cold fusion reaction. Under everyday conditions, fusion between deuterium atoms is unlikely because the repulsive Coulomb interaction between the deuterium nuclei, which are electrically like-charged, gives rise to a potential barrier between the atoms. [8,12] For small internuclear separation, the potential is inversely proportional to the separation. [8] Classically, this divergence of the potential as separation goes to zero would mean that nuclei at low energies cannot even stay extremely close to each other, let alone take part in a fusion reaction. Fusion thus requires extremely high temperatures, so that the correspondingly high-energy nuclei can surmount the Coulomb barrier and react. However, in quantum mechanics, a particle may cross through a potential barrier higher than its energy by way of quantum tunneling (depicted schematically in Fig. 4 for a very simple rectangular potential). Therefore, when this effect is taken into account, the deuterium nuclei could conceivably tunnel past their mutual Coulomb barrier and fuse. [8,12]

The fact that this possibility exists does not mean it is a frequent occurrence. As an inspection of Fig. 4 may indicate, allowing tunneling through a potential barrier does not mean that the amplitude of the particle's wave function is the same on both sides of the barrier - the amplitude is suppressed across the barrier, and this suppression increases as the barrier width increases. Even without exactly solving the Schrödinger equation, the WKB (Wentzel-Kramers-Brillouin) approximation can provide an estimate of how much the amplitude of the particle's wave function is suppressed across the barrier. [8,17] This factor depends on the exact form of the barrier potential - for instance, for a constant rectangular potential of the sort depicted in Fig. 4, the suppression increases exponentially with barrier width. This suppression directly affects the fusion rate, as the latter is proportional to the probability of the deuterons being at a given internuclear separation, i.e. the squared amplitude of the wave function of the deuterons at that separation, as suppressed by the potential barrier. [8]

For their quantitative estimates of cold-fusion rates, Koonin and Nauenberg actually use numerical methods to directly solve the exact Schrödinger equation. [8] Any method of estimating the fusion rate would depend on the model used for the potential between the two deuterons, for which Koonin and Nauenberg take what they say is "the best available numerical calculation in the Born-Oppenheimer approximation" from publications in 1964 and 1968. [8] In a diatomic molecule of deuterium, the internuclear separation is 0.74 angstroms (7.4 × 10-11 m), and for that state of deuterium, Koonin and Nauenberg estimate the cold D-D fusion rate to be 3 × 10-64 reactions per second (actually ten orders of magnitude above what Jones had estimated). [8,12] Muon-catalyzed fusion was observable because in that process, the electron in each hydrogen atom is replaced by the catalyzing muon, substantially reducing the internuclear separation - by the ratio of the muon mass to the electron mass, which is approximately 207 (as mentioned above) - and thus enhancing the D-D fusion rate by many orders of magnitude, to about 1012 reactions per second. [7,8,12]

Jones, drawing an analogue between muon-catalyzed fusion and what he dubbed "piezonuclear" fusion, claimed that the fusion rate he observed could be explained if the internuclear separation were half of what it is in diatomic deuterium, this separation being maintained by the high compression of deuterium absorbed in the metal electrode. [12] However, Koonin and Nauenberg conclude that reduction of separation by a factor of 5 is actually necessary to explain the fusion rates claimed by BYU; they also claim that to explain the results of Pons and Fleischmann, they require a factor of 10. [8] Merely halving the separation would only push the fusion rate up to about 4 × 10-41 reactions per second. [8] Unfortunately, it is not clear what Jones and others thought to be a plausible number for internuclear separation between deuterium nuclei when absorbed in palladium or titanium. In his paper, Jones claims that "quasi-electrons" in the metal lattice could have an effective mass of a few times the electron mass and thus cause a reduction of that order in separation, much in the same way that a muon would. [12] However, while Koonin and Nauenberg do discuss the reduction in separation in terms of hypothetically "endowing the electron with a larger mass [...] than it actually has," they also attach a parenthetical warning:

(This enhanced mass should not be associated with any physical excitation in a solid material, as only the bare electron is relevant at the short length scales that are important here.) [8]

Regardless of Koonin's and others' criticisms, the Utah group continued what the New York Times described as "a pattern of non-cooperation", refusing to allow a panel visiting from the Energy Department because of hostile scientists being included in the panel, and refusing to be open with data even without these openly hostile members visiting the laboratory. [9] Afterwards, "the panel concluded that the prospects of producing energy with cold fusion were so remote that no new laboratories should be built by the Federal Government," with the advisory board's final report concluding that the evidence was not persuasive. [3,9]

Following this review, scientists continued to scrutinize Pons and Fleischmann's claims. In March of 1990, one year after Pons and Fleischmann's original press conference, Nature published an article co-authored by Michael Salamon, a physicist at the University of Utah - Pons's institution - that measured no significant emission of neutrons, gamma radiation, or any other indicator of fusion activity in the cells being used in Pons's laboratory. [5] The following June, the Texas A&M group found their palladium electrodes to be contaminated with tritium, explaining their earlier tritium results without having to claim fusion. [3] After these developments, Fleischmann and Pons left the United States for France, where they resumed their work under Toyota until the mid-1990s. [2] When Martin Fleischmann died at age 85 in 2012, the New York Times started its obituary by saying:

Martin Fleischmann made the greatest discovery since fire: replicating the furnace of the sun at room temperature in a jar of water, essentially solving the world's energy needs forever.

That is how Dr. Fleischmann might have liked his obituary to read. [2]

This may seem an unnecessarily cruel way to start an obituary, but perhaps reflects the sharp U-turn of the scientific community against his and Pons's claims.

A Case of Identity

Yet support for cold fusion efforts continued to some extent. In 1991, two years after the original Utah announcement, scientists at the Naval Weapons Center were still claiming "'strong evidence that nuclear processes are occurring' in apparatus similar to that used at Utah." [18] The Soviet Union was also very enthusiastic about cold fusion prospects, with the Soviet Academy of Sciences announcing "that 15 million rubles would be devoted to such work over the next four years." [18]

Fig. 5: Michael McKubre working on a deuterium gas-based cold fusion cell used by SRI International. (Source: Wikimedia Commons)

Support for the term "cold fusion" is less evident. A Popular Science article in 2012 noted that cold fusion research has had various aliases, such as "condensed matter nuclear science, lattice-assisted nuclear reaction, chemically assisted nuclear reaction," as well as "low-energy nuclear reaction (LENR)". [19] With this change of identity of their field, researchers in Japan, Italy, Israel, India, Russia, and China are "spending significant resources on LENR research" both in national laboratories and private companies, enough for the US Defense Intelligence Agency to be concerned about "the likelihood of a technology breakthrough - as well as the potential for technology surprise - by an international team, especially from those countries that are devoting more resources to this research than is the United States." [20]

The Department of Energy conducted a second review of cold fusion in 2004, but still found evidence inconclusive, leaving the Department's stance unchanged from its 1989 review. [21,22] There are nonetheless some institutional and industrial cold fusion researchers in the United States, among them Michael McKubre at SRI International (depicted in Fig. 5) and Peter Hagelstein at MIT. [21,23] Fusion experiments have also continued within the US military, at the Space and Naval Warfare (SPAWAR) Systems Center in San Diego, CA. The SPAWAR team has published two short communications in Naturwissenschaften and more papers in other peer-reviewed journals. [24-26] (On a historical note, Naturwissenschaften was also the journal in which Otto Hahn and Fritz Strassmann first reported transmutation due to nuclear fission processes. [27]) The SPAWAR team reported unusual shape changes, possible transmutation, and production of energetic charged particles in their setup, claiming these observations to be evidence of nuclear reactions. [24,25] These publications are light on theoretical details, however, and are unwilling to attribute these changes to specific nuclear processes.

Research has also continued in Italy, where the National Institute of Nuclear Physics retains a team of LENR experimentalists. [19] An example of a privately funded effort is Andrea Rossi's E-Cat device, which purports to use "nickel powder, a small amount of hydrogen gas, and Rossi's 'secret catalyst.'" [19] The device has been publicly demonstrated, but with very little transparency or external testing allowed. [19]

Transparent or not, cold fusion research continues to be critiqued for its lack of adherence to science. For instance, in the 2004 Department of Energy review of cold fusion, according to Science,

Several reviewers were indeed extremely critical of the research, saying that many of the experiments were poorly conducted, had results that were inconsistent with each other, and often weren't reproducible. One skeptical reviewer went further, opining that "[cold fusion] workers are true believers, and so there is no experiment that can make them quit." [22]

Following the review, James Decker, deputy director of the Department's Office of Science, made the point that the Department has "always been open to proposals that have scientific merit as determined by peer review." [21]

Regardless of the Department's stance, the study of cold fusion in the fashion of Pons and Fleischmann seems likely to remain largely excluded from mainstream physics, simply because the current understanding of mainstream physics does not allow D-D fusion from electrolytic compression in palladium. As seen in the above discussion of Koonin and Nauenberg's considerations, even with quantum-mechanical effects in play, enhancement of cold D-D fusion rates to readily measurable levels - let alone useful levels - requires significant reduction of the internuclear separation, and cannot rely on lattice excitations over length scales larger than the internuclear length scale. If this reduction is not provided by higher nuclear energies or catalyzing muons, the metal lattice must provide an extreme amount of compression of the deuterium to maintain small separation between the deuterium nuclei well below 10-10 m, thus allowing fusion to occur in spite of the extreme Coulomb potential barrier that exists between every deuteron pair. But maintaining such a small separation would still require overcoming that same Coulomb barrier, and numerical studies carried out in the aftermath of the 1989 claims have shown this to be a negligible possibility. [28] Further study of the properties of deuterium in palladium may well be warranted as part of chemical physics, but deeming any phenomenon in that system to be cold fusion is not well-substantiated at present.

© Dongwoo Chung. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.


[1] "Chapter 2100 Patentability", in Manual of Patent Examining Procedure, 9th Ed., United States Patent and Trademark Office, March 2014.

[2] D. Martin, "Martin Fleischmann, Seeker of Cold Fusion, Dies at 85," New York Times, 12 Aug 12.

[3] B. V. Lewenstein and W. Baur, "A Cold Fusion Chronology," J. Radioanal. Nucl. Chem., 152, 273 (1991).

[4] M. Fleischmann, S. Pons, and M. Hawkins, "Electrochemically Induced Nuclear Fusion of Deuterium", J. Electroanal. Chem., 261, 301 (1989).

[5] M. H. Salamon et al., "Limits on the Emission of Neutrons, γ-Rays, Electrons and Protons From Pons/Fleischmann Electrolytic Cells," Nature, 344, 401 (1990).

[6] M. W. Browne, "Fusion in a Jar: Announcement by 2 Chemists Ignites Uproar," New York Times, 28 Mar 89.

[7] S. E. Jones, "Muon-Catalysed Fusion Revisited," Nature, 321, 127 (1986).

[8] S. E. Koonin and M. Nauenberg, "Calculated Fusion Rates in Isotopic Hydrogen Molecules," Nature, 339, 690 (1989).

[9] W. J. Broad, "Cold Fusion Still Escapes Usual Checks Of Science," New York Times, 30 Oct 90.

[10] W. J. Broad, "Cold-Fusion Claim Is Faulted On Ethics as Well as Science," New York Times, 17 Mar 91.

[11] "The Utah Fusion Circus," New York Times, 30 Apr 89.

[12] S. E. Jones et al., "Observation of Cold Nuclear Fusion in Condensed Matter," Nature, 338, 737 (1989).

[13] J. Maddox, "What to Say About Cold Fusion," Nature, 338, 701 (1989).

[14] Associated Press, "Palladium Futures Fall From Highs", San Jose Mercury News, 18 Apr 89.

[15] W. J. Broad, "Georgia Tech Team Reports Flaw In Critical Experiment on Fusion," New York Times, 14 Apr 89.

[16] M. W. Browne, "Physicists Debunk Claim Of a New Kind of Fusion," New York Times, 2 May 89.

[17] D. J. Griffiths, Introduction to Quantum Mechanics (Pearson Prentice Hall, 2005), pp. 315-339.

[18] W. J. Broad, "There Still May Be Something Scientific About Cold Fusion," New York Times, 14 Apr 91.

[19] S. Featherstone, "Andrea Rossi's Black Box," Popular Science 281, No. 5, 62 (November 2012).

[20] "Technology Forecast: Worldwide Research on Low-Energy Nuclear Reactions Increasing and Gaining Acceptance," U.S. Defense Intelligence Agency, DIA-08-0911-003, November 2009.

[21] K. Chang, "Evidence on Cold Fusion Remains Inconclusive, New Review Finds," New York Times, 2 Dec 04.

[22] C. Seife, "Outlook for Cold Fusion Is Still Chilly," Science 306, 1873 (2004).

[23] B. Daley, "Heating Up a Cold Theory," Boston Globe, 27 Jul 04.

[24] S. Szpak et al., "Evidence of Nuclear Reactions in the Pd Lattice," Naturwiss. 92, 394 (2005).

[25] S. Szpak, P. A. Mosier-Boss, and F. E. Gordon, "Further Evidence of Nuclear Reactions in the Pd Lattice: Emission of Charged Particles," Naturwiss. 94, 511 (2007).

[26] B. Daviss, "Cold Fusion Rides Again," New Scientist 194, No. 2602, 32 (May 2007).

[27] O. Hahn and F. Strassmann, "Über den Nachweis und das Verhalten der bei der Bestrahlung des Urans mittels Neutronen entstehenden Erdalkalimetalle," Naturwiss. 27, 11 (1939).

[28] C. T. White et al., "Limits of Chemical Effects on Cold Fusion", J. Fusion Energy, 9, 363 (1990).