back to homeReturn to home page

[Published in Infinite Energy Vol. 6, Issue #32, page 52 (2000)]

A Critical Evaluation of the Pons-Fleischmann Effect (Part 2)

Edmund Storms
2140 Paseo Ponderosa
Santa Fe, NM 87501


Many new studies are available to make an objective evaluation of the Pons-Fleischmann effect possible. The phenomenon is conventionally known as "cold fusion,"or "chemically assisted nuclear reactions (CANR)" when the environment is emphasized, or "low-energy nuclear-reactions (LENR)" if emphasis is placed on the process. A wide range of observations involving anomalous production of energy as well as nuclear products have been published. While many of the claims are still open to interpretation, the general conclusion is that an important, novel phenomenon has been discovered which deserves renewed interest.


This is the second part of a review of the Pons-Fleischmann effect. The first part appeared in the previous issue of Infinite Energy and discussed claims for anomalous energy. This part will address claims for helium production as the source of this energy. Various explanations are briefly described to give the reader insight into the proposed mechanisms for the effect. I would like to apologize to the reader for having to use so many references to several proceedings of the International Conference on Cold Fusion (ICCF), which are not easily available. Unfortunately, much of this work has not been published in conventional journals in spite of attempts by various authors.


Profs. Pons and Fleischmann proposed that the source of energy was nuclear fusion. (1) This explanation was quickly rejected when the expected high neutron emission was not found. Tritium production would also be expected. This product has been detected occasionally at significant levels (2), but the amount does not account for the observed energy. Clearly, anomalous energy is not a product of conventional fusion, nor does it show the behavior found during "hot fusion."

Nevertheless, anomalous emissions have been detected on numerous occasions, including neutrons, X-rays, gamma-rays, charged particles, as well as radiation from radioactive products. While such radiation along with production of nonradioactive isotopes suggest anomalous nuclear activity, this paper will not attempt to assess these claims. Instead, only claims for helium production will be addressed.

Helium can be produced by several nuclear reactions besides fusion. This element was looked for and found by numerous investigators in several different environments including in the gas (3-8) dissolved in the materials (9-15) and emitted as charged particles (16-18) using palladium or titanium.(19, 20) Naturally, not all studies are definitive and some failed to find helium when it was sought. Recent studies of gas-loaded palladium powder also claim to show helium production.(14, 21-23) Cycling of LaNi5 (24) in D2 and allowing palladium-black (14) to remain in the gas are claimed to produce 3He as well as 4He. This 3He might actually result from decay of produced tritium.

While these observations are suggestive, only two independent measurements have provided a quantitative relationship between production of anomalous power and helium-4. Both studies used all-metal systems and measured helium in the flowing gas generated during continuous electrolysis. This method insures that air contamination and absorbed He are removed. Two different calorimeter types were used and helium was measured at two different laboratories. These two studies are compared in Figure 1. Three conclusions can be drawn from the figure. First, the two studies agree very well, given the difficulty of the measurement. Second, values for the ratio of the He production rate and power production are largely independent of observed anomalous power, as would be expected if the two quantities are functionally related. Third, the average values are within a factor of 2 of being consistent with an energy of 24 MeV/helium atom, the calculated value based on mass change. In addition, Bush et al.(4 ) as well as Gozzi et al.(6) found helium to be released slowly and only after a delay. Consequently, some of the helium might still remain within the solid palladium deuteride, which was not analyzed, thereby making the measured energy/atom potentially more consistent with a conventional fusion reaction. During the study by Miles et al.(3), seven cells using pure palladium produced no detectable excess energy in addition to the six successful experiments. In each case, no helium over the background value was found in these cells. Thus, anomalous helium was found only when anomalous heat was detected. Only one cell, which used a Pd-Ce alloy, showed heat but no released helium. Such behavior is strongly against chance alone as Miles noted.

A 23.8 MeV gamma emission has been detected when the helium producing branch of the fusion reaction has been previously observed using conventional fusion. This radiation is required because fusion of two deuterons produces only one product nucleus. Gamma emission allows momentum to be conserved. Because this gamma energy is not detected during anomalous energy production, most critics dismiss the claimed helium as being an artifact. The other two branches of the fusion reaction, which provide the main activity under conventional conditions, are apparently not the source of significant energy in this environment. This distortion of expectation also adds to the skepticism.

What could be wrong with the helium measurement that would justify such sketicism? 

1. The results are very scattered and represent a narrow range of excess energy and helium. An off-set error in both measurements could generate the proposed relationship by chance.

Answer: Error bars are shown which reflect the expected chance variation including any bias. All of the points agree within their expected error. While excess energy has a narrow range of plausible bias, the helium values can, in principle, range over several orders of magnitude, depending on the care used. Yet, the helium values are well clustered.

2. The helium concentrations are very small and well below ambient helium concentrations in the surrounding air. A small leak or direct helium diffusion might allow helium to enter the gas stream.

Answer: Metal systems were used which allow insignificant helium diffusion. However, some gasket materials, including a few metals, allow diffusion of helium. This problem was recognized and eliminated. Leaks would admit other elements from the air such as argon. When this gas was looked for in the mass spectrum, it was not found. In addition, leaks would be expected to be very erratic, thereby leading to a very wide scatter in helium values. This kind of scatter is not seen.

3. The masses of D2 and He are so close that a variable presence of D2 could introduce an error in the He peak.

Answer: In each case, D2 was removed by an absorption trap and, in addition, the mass spectrometers were able to completely resolve the two mass peaks.

4. The excess energy measurements were done using inaccurate calorimeters. While anomalous energy might be real, its measurement during the helium studies might be faulty. Therefore, the claimed correlation may not be real.

Answer: The studies by Miles et al. (3 ) were based on using a double-walled isoperibolic calorimeter that did not contain a recombiner. No recombination was detected. Although the amount of excess power is small, it is well within the claimed accuracy and stability of the device. In addition, failure to detect excess energy correlated with a failure to detect helium. In other words, when heat was present, helium was present, and when heat was absent, helium was absent. A very stable Seebeck-type calorimeter (25) was used during the work by Bush. Again, the claimed excess power was well within the sensitivity of the device.


3.1 Basic Properties

Before an explanation is attempted, several basic properties of b-PdD must be taken into account. The b-PdD phase has a face-centered cubic structure (fcc) with deuterium atoms occupying random positions within the deuterium sublattice under ideal conditions. However, composition gradients, impurities, and dislocations make actual occupancy very nonrandom. These atoms are sometimes described as occupying octahedral positions. Tetrahedral positions are not occupied except perhaps during diffusion. The lattice parameter vs composition extrapolates to 0.4084 nm with smaller values at lower compositions. (26) The lower limit of the b-phase is near PdD0.70, depending on the temperature and applied D2 pressure, while the upper limit approaches PdD1.0 as all vacant positions are filled. In addition to this nonuniformity in composition, palladium can also acquire many dislocations in the palladium sublattice, which can easily degrade to microcracks. Simply bending a wire will cause these dislocation to form in great abundance. In addition, the hydriding process will change the crystal morphology and introduce strain. When electrolysis is used to form the hydride, many impurities are also introduced into the lattice making the surface region a very complex alloy containing at least Li, Pt, Si, and B, with unknown characteristics. Such factors combine to make the material very nonuniform in all its properties and very unlikey to be described adequately by "ideal" models.

As the deuterium content is increased, the equilibrium pressure of deuterium gas also increases, as shown in Figure 2.(27) The enormous pressure required to form the required large deuterium concentrations is the main reason these compositions are so difficult to achieve by electrolysis. Deuterium can readily leak out of the sample through the many cracks known to inhabit such material. Notice that the pressure increases as temperature increases, with no indication of a reverse temperature effect near PdD0.85 as proposed by Fleischmann (28) to explain energy production at higher temperatures. Indeed, all available data as well as the theory of such compounds predicts that all compositions of b-PdD will have a higher equilibrium pressure of D2, hence lose deuterium more rapidly, as temperature is increased. Consequently, either excess energy production does not involve the b-phase as argued by Storms (34), or the effect does not require a high composition at higher temperatures, as argued by Waisman and Summeri.(29)

Occasional reports are published claiming average compositions in excess of PdD1.0. (30-32) These high compositions are proposed to be caused by deuterium ions occupying tetrahedral sites or by formation of a new phase based on D2 ions occupying octahedral sites.(34) The latter explanation would imply a two-phase region between PdD1-x and PdD2-y where x and y are small unknown numbers. Recent measurements indicate that this high composition forms in the near surface region and is required before anomalous energy is detected. In brief, the nuclear reactions do not occur in b-PdD, but in an unknown phase of higher deuterium content containing various impurities. Therefore all models which depend on the properties of pure b-PdD must be viewed with skepticism. This trans-beta phase, once formed, can be proposed to become more stable as temperature is increased, thereby explaining the claimed excess energy obtained at higher temperatures. The roles of the always present impurities, such as Li and Pt, are not known.

On the other hand, the nuclear-active environment may not involve palladium at all. Recent claims for heat production using gold (35) or platinum (36-38),neither of which absorb significant hydrogen (39), suggest the active environment is a deposited layer on the surface. While more than one active environment might exist, another possibility must be considered, i.e. that all such nuclear reactions occur in the same or similar chemical environments that are created as a surface layer on an otherwise inert substrate. When used with the electrolytic method, palladium may turn out to be a poor choice of substrate because of its tendency to crack in the presence of hydrogen.

3.2 Requirements of a Theory 

Theoretical objections to the claims are based on the following arguments as summarized by Huizenga (40) and many other skeptics:

1. The deuterium nuclei in PdD are not close enough to allow nuclear interaction.

2. Energy required to overcome a Coulomb barrier is not available in a chemical compound.

3. If a fusion reaction should occur, the nuclear products must be the same as those observed when fusion is initiated using high energy methods. These products are not detected with the expected magnitude or in the expected ratio.

4. Any helium produced by a fusion reaction must be accompanied by gamma emission, which is not detected.

5. The resulting energetic nuclear products should produce intense X-ray emission, which is not detected. 

The issue requiring resolution is whether unknown process exist which would allow a fusion or any other nuclear reaction to occur in the unique environment of a chemical lattice. Adding to the challenge are the many other types of nuclear reactions now being claimed. A few attempts to understand the process are discussed, but only to give the reader insight into the types of mechanisms being considered. A more detailed evaluation is not possible at the present time for various reasons. For example, the author finds that theoreticians seldom agree with each other, preferring instead to focus on their own favorite models. Consequently, agreement or at least objective discussion within the field is handicapped from the start. Partial evaluations have been undertaken by Preparata (41) and by Rabinowitz et al. (42), but with very little agreement. In addition, there is a tendency to choose only those experimental observations that fit the model while ignoring much data that are inconsistent. This makes an application of the model very limited. This review will only attempt to show the general approach and note whether the proposed models are consistent with experimental observation. Only models which have been developed to a high degree are cited. Many dozens of models, which involve only brief suggestions or address limited observations, are not included. The fractofusion model, although it has been given considerable attention, is not discussed because this process involves high energy and is expected to produce "normal" nuclear products, hence is not consistent with the anomalous observations.

Models generally fall into two general overlapping categories. The first is a mechanism that explains how a nuclear reaction can occur once the necessary conditions are achieved, and the second is the nature of the necessary conditions. The first category involves nuclear physics and the second one involves chemistry. Both must be addressed by a successful theory.

To be successful, a theory must answer at least five basic questions to explain the P-F effect and several other questions if the entire range of published observation is to be explained.

1. What mechanism allows the Coulomb barrier to be overcome? This question is fundamental. The model will have to explain the proposed fusion between deuterium nuclei as well as how nuclei as heavy as palladium can suffer a reaction with nuclei as heavy as oxygen.

2. What mechanism distributes the released energy throughout the lattice rather than requiring it to be focused on a few individual particles? This mechanism must also explain why some nuclear energy is retained by the nuclear products when these products are produced very near a surface. Otherwise, charged particles having significant energy to leave the material would not be detected.

3. How is the proposed mechanism related to the physical environment? Most present theories assume the nuclear reactions occur in b-PdD having a composition near PdD1.0. The model must explain why anomalous reactions occasionally involve other materials, why the required conditions are so difficult to achieve, and why the active regions are so locallized on the surface.

4. What nuclear reaction is the source of observed helium? Fusion is not the only conceivable source of helium as a nuclear product.

5. If helium results from a fusion reaction, what mechanism allows conservation of momentum and energy, and what mechanism distorts the reaction paths to produce helium rather than neutrons and tritium?

A single explanation seems impossible. The diverse nature of observed behavior and uncertain reality of the claims add to this problem. To simplify the challenge, this review will focus only on the environment claiming to produce energy when heavy-water and palladium are used in an electrolytic cell. Any successful general theory will have to explain how this anomalous energy is produced in other metals while using other methods, involving both light water and heavy water. In addition, the model would have to explain how the many anomalous heavy isotopes are produced.

A quantitative test of any model is impossible using experimental data, because the palladium samples are very nonuniform, hence the fraction of active material is unknown. Indeed, a calculated power density based on the gross physical dimensions of the sample will greatly underestimate actual local power density, and the ultimate power density of which this phenomenon is capable. Therefore, comparing samples on the basis of watts/cm2 or watts/cm3, as is frequently done, is not appropriate. Indeed, evidence of local melting suggests that very high local power density is possible.

The models discussed below are placed in eight general categories, but with the understanding that to be successful, a model may have to invoke more than one category. In addition, the examples chosen are intended to give the reader only a very general understanding of the range of models being proposed. The discussion is not meant to be a complete evaluation of all efforts to understand the phenomenon.

3.3 Proposed theories

3.3.1 Neutrons

The easiest way to solve the Coulomb barrier problem is to invoke the neutral neutron. However, free neutrons are unstable, hence are not present in sufficient numbers within a solid. Consequently, such theories are forced to identify a steady source of virtual neutrons.(43-45) This search has taken several forms. Kozima (43) proposes that trapped thermal neutrons catalyze the cold fusion reaction (TNCF Model). In an attempt to identify a source of such neutrons, he has proposed formation of metastable neutron clusters of various sizes. Several questions have not been answered including exactly what conditions cause the clusters to interact with the surrounding nuclear environment, why individual neutrons are not emitted from the material during such interaction, and why the clusters do not produce "normal" nuclear products. In addition, his efforts to compare the concentration of these neutrons to the observed effects is doomed to failure because nuclear activity is highly localized and impossible to relate to the measured volume of the sample, a variable used in his model.

On the other hand, several workers have proposed direct formation through interaction between a proton or deuteron and an electron. However, complete formation of a neutron requires energy as well as a neutrino or antineutrino, both of which are not present in sufficient amounts. In addition, reaction with a neutron should produce "normal" nuclear products which are not observed. To eliminate these problems, the neutron is proposed to exist only as a partially collapsed structure. The manner of this partial collapse has been addressed by several workers and the process is proposed to produce various structures called Hydrex (47), Itons (48), Hydron (49), and Hydrino (50). This "virtual" neutron can, under special conditions, presumably be absorbed by a nucleus in the same manner as a real one, while producing anomalous nuclear products in the process . The source of the required neutrino has not been clearly identified. Mills views the process of partial electron collapse to release energy without involving a nuclear reaction (50) while Dufour (47), using conventional theory, finds that the collapse requires energy. The region between partial collapse and complete reaction between an electron and a proton is a gray area in which many models can operate. As yet insufficient experimental information is available to decide which model or models are correct.

Dufour (47) has gone one step further in describing how his collapsed hydrogen, the Hydrex, interacts with the nuclear environment. Rather than causing fusion, the structure is proposed to attach itself through dipole interaction to a large nucleus such as palladium, a process which is proposed to reduce the energy barrier for release of helium-4. This process would predict energy production in the 5 MeV region which, so far, is at odds with observation. Dufour must assume that considerable helium still remains trapped in the Pd metal which has been missed.

Another approach involves the existence of multineutron structures which are stable. The existence of dineutrons has been proposed before the CANR phenomenon needed such a model. Fisher (Thomas Paine Assoc.) (51) has carried this idea one step further. He proposes that large, stable neutron clusters can form and that these can attach themselves to normal nuclei to produce super-heavy atoms. A small concentration of such atoms is proposed to be present in all matter. Under the right conditions, these neutron clusters are released, thereby causing novel nuclear reactions. Within the confines of the assumptions, the model can explain many observations of CANR. In addition, the work of Oriani (52) suggests the existence of super-heavy carbon in electrodes subjected to CANR processes.

3.3.2 Novel particles

Various exotic particles, which have the ability to catalyze nuclear interaction, are proposed to exist in nature. These are located within the structure as part of the normal constituents or are provided by a steady flux from cosmic sources. The models do not show why hard-to-create special chemical conditions are required for these particles to do their unique work. These particles are called Neutrium (53), Quark (54), Hemitrons (55), Muon neutrinos (56), or Erzion (57).

3.3.3. Electron structure

Various modifications to the electron structure (58-61) including impurities (62), stress (63), diffusion (64), oscillation (65, 66) applied current (67), or changes in the periodicity of the atomic structure (68, 69), are proposed to reduce the Coulomb barrier. The QED (33) theory addresses the latter process in some detail. In this case, normal electrons within a lattice are proposed to acquire a coherent structure, much like a superconductor, which can reduce the Coulomb barrier and carry away the resulting nuclear energy.

Work based on the swimming electron layer model is ongoing at the University of Illinois under the direction of Prof. G. Miley.(70) This model proposes that deuterons can exist as a bare nuclei and move freely throughout the lattice. Relatively unbound electrons located at a metal surface or between those metals which have much different Fermi energies can screen the Coulomb barrier between bare deuterons or between them and other atoms. The result is enhanced fusion and various transmutation reactions that produce energy and helium.

While these processes are proposed to initiate nuclear reactions, they do not address which reactions might result.

3.3.4 Phonons

Lattice vibrations are proposed to cause a few adjacent deuterons to get close enough to fuse (71, 72), especially in the double layer at the electrolyzing surface (73). The likelihood of this process is increased by the average atom position in the deuteride being closer than previously thought.(66, 74, 75) Kucherov (76), working at ENECO (Salt Lake City, UT), proposes that these vibrations can transfer incremental energy to the nucleus until sufficient instability is accumulated to produce alpha emission or other nuclear reactions. Hagelstein (MIT) (72) is also exploring lattice-induced alpha decay of various heavy-metal impurities as the source of energy and helium, but using a different model from Kucherov. His model has evolved to the stage where quantitative predictions can be made about the energy of emitted particles and radiation. Li et al. (77), working in China, propose a resonance between energy levels in the nucleus and those in the lattice allowing energy transfer to a d-d pair. Energy can be transferred into the lattice by the same process after fusion takes place.

3.3.5 Particle-wave transformation

Under special conditions, the deuterons dissolved in a periodic lattice are proposed to assume wave-like properties which permit fusion and deposition of the resulting energy throughout the lattice.(78-80) The Ion Band State Theory is explored in detail by Scott Chubb (Naval Research Lab.) and Talbot Chubb (Oakton International Corp.) (81). They assume that energy is transferred in steps from the wave-states of two deuterons until the wave-state of a helium nuclei is achieved. Once this energy is lost to the other wave-states, the helium wave can revert back to a particle and be detected as such. This idea addresses problems associated with d-d fusion, but it does not explain other claimed nuclear reactions. 

3.3.6 Nonfusion Nuclear Reactions

Passell (EPRI, ret.) (82) proposes that helium and energy are produced by the reaction 10B + 2H = 4He + 8Be = 34He giving 5.9 MeV/He atom. Although this reaction might occur under some conditions, the expected energy is not consistent with the measurements shown in Figure 1.

Takahashi (Osaka Univ.)(16) has proposed a reaction between deuteron clusters containing three or four atoms, two atoms of which fuse while the remaining atoms help carry away the resulting energy. Various reactions between these deuteron clusters and other elements, producing unstable isotopes of Li and Be, are also proposed to result in helium production. This model avoids the problem of momentum conversation during fusion which requires gamma emission.

Enhanced electron screening is proposed by Hora (Univ. New South Wales)(70) to allow reactions such as Pd + d = 103Rh + 4He which releases 10.5 MeV of energy. This and other nuclear reactions (83, 84), initiated by this process are proposed to have a large positive temperature coefficient, thereby explaining the greater effect observed at higher temperatures.

3.3.7 Nuclear structure

The nucleus is proposed to have a structure which facilitates nuclear interaction, being composed of clusters of nucleons which can be easily lost or gained.(85-87) These clusters are thought to leave or enter the nucleus more easily than expected once the barrier is overcome.

3.3.8 Reduced Barrier

The Gamow factor has been explored in different ways by Kim (Purdue Univ.) and McNeil (Colorado School of Mines) to see if unanticipated conditions can cause a reduction (88, 89) including a tunneling mechanism.(90-92) Apparently, the barrier is lower than previous calculations would predict, but not as low as required to explain the claimed observations. Measurements of the branching ratio and the fusion rate when very low D+ energies are used to bombard various metals indicate that electron screening can, indeed, lower the barrier in some metals (93).


Before the claimed energy production can be attributed to fusion, nuclear products need to be detected. The expected neutron emission has been detected as large bursts and as a small steady rate, but in too small an amount to account for anomalous energy. The expected tritium has also been detected, but again in very small amounts. Only helium has been found in sufficient quantity to account for the claimed anomalous energy. In addition, both tritium and helium seem to be born with very little translational energy, again in contrast to hot fusion. While this helium might result from various prosaic sources, the list of observations is growing and the detected amounts are increasing, making such explanations increasingly unlikely.

Although production of various heavy isotopes (transmutation) has not been addressed here, this is another growing aspect of the phenomenon which needs to be understood. Not only are such reactions a potential source of energy, but they present a significant challenge to any explanation.

If energy does result from a novel form of fusion, how can this be explained? Unfortunately, we are presented with too many explanations rather than too few. To make matters worse, the nature of the nuclear-active regions has not been identified. Without such information, duplication of the effect is difficult and any theory is based more than usual on imagination. Identification of the active region is handicapped because these regions occupy only small areas on the surface and are unstable. Their existence is seen only after a few have destroyed themselves by melting the surrounding material. This problem does not have an easy solution. In any case, the conventional focus on b-PdD as the nuclear-active material needs to be re-examined because the properties of this material are not consistent with observed behavior.

In general, the various explanations have not been very useful because they fail to identify the unique conditions which must exist within the material to make it become nuclear-active. It does no good to say, for example, that a source of virtual neutrons must be present, that the electrons must be coherent, or that a particle-wave conversion must take place. The question is, How can these events be made to occur? Without such knowledge, efforts to make the effect more reproducible will be based more on luck than on knowledge. At the very least, to be useful an explanation must identify those real-world variables that influence the effect and show how the necessary conditions can be made less rare.

Once the necessary conditions have been identified, a successful theory must answer several difficult questions about the nature of the nuclear reactions &emdash; in brief the three miracles proposed by Huizenga (40) How can a Coulomb barrier be overcome using the very limited energy available within a lattice? How can the released energy be communicated to the lattice in the brief time frame typical of such events? What process distorts the fusion branches to favor helium production? What causes different nuclear products to be produced under what appear to be similar conditions, including those resulting from transmutation reactions? In addition, any proposed novel process must be tested outside of the CANR field, i.e. it must be shown to work under other conditions. As yet, no proposed theory has met the challenge, although a few are getting close.


The claims for anomalous energy production using electrolysis of heavy water have been evaluated and found to have a high probability of being caused by a novel phenomenon. In addition, the most likely source of the heat is a nuclear reaction which produces helium. This nuclear reaction is not normal fusion and does not follow the rules required by conventional theory. Numerous models have been proposed to explain the observations, but none at the present time can account for all of the reported behaviors. More work is required to determine which of the behaviors are part of this novel phenomenon and which can be explained by ordinary processes. However, the claims have now reached a level of understanding which justifies a reexamination of the published work and serious attention by the scientific community.

A standard has been applied to this field which is unique to modern science and which sets a precedent to hinder acceptance of future novel claims in nuclear physics. This standard involves various outspoken skeptics demanding that nuclear products be identified as the source of the claimed energy before they will accept even that the anomalous energy is real. Moreover, these nuclear products had to fit the behavior found during "hot" fusion. In other words, if the the behavior is not as expected, the entire set of observation is rejected. Application of such a limited and uncompromising approach is not how science has advanced so successfully in the past.

Finally, after 11 years of work, a nuclear source for the anomalous energy has been demonstrated. It is now incumbent upon these same skeptics to acknowledge this work and join in an effort to evaluate the new discoveries in a rational, open-minded, and productive manner. The integrity of these people is now on the line and their reputations as competent scientists may eventually be at risk..