You are on page 1of 8

Cold fusion​ is a hypothesized type of ​nuclear reaction​ that would occur at, or near, ​room

temperature​. This is compared with the ​"hot" fusion​ which takes place naturally within ​stars​,
under immense pressure and at temperatures of millions of degrees, and distinguished from
muon-catalyzed fusion​. There is currently no accepted theoretical model that would allow cold
fusion to occur.
In 1989 ​Martin Fleischmann​ (then one of the world's leading ​electrochemists​) and ​Stanley Pons
reported that their apparatus had produced anomalous heat ("excess heat") of a magnitude they
asserted would defy explanation except in terms of nuclear processes.​[1]​ They further reported
measuring small amounts of nuclear reaction byproducts, including ​neutrons​ and ​tritium​.​ The
small tabletop experiment involved ​electrolysis​ of ​heavy water​ on the surface of a ​palladium​ (Pd)
electrode.​[3]​ The reported results received wide media attention,​[3]​ and raised hopes of a cheap
and abundant source of energy.​[4]

Many scientists tried to replicate the experiment with the few details available. Hopes faded due
to the large number of negative replications, the withdrawal of many reported positive
replications, the discovery of flaws and sources of experimental error in the original experiment,
and finally the discovery that Fleischmann and Pons had not actually detected nuclear reaction
[5]​ [6]​[7]​
byproducts.​ By late 1989, most scientists considered cold fusion claims dead,​ and cold
fusion subsequently gained a reputation as ​pathological science​.​ In 1989 the ​United States
Department of Energy​ (DOE) concluded that the reported results of excess heat did not present
convincing evidence of a useful source of energy and decided against allocating funding
specifically for cold fusion. A second DOE review in 2004, which looked at new research,
reached similar conclusions and did not result in DOE funding of cold fusion.​[10]

Fleischmann–Pons experiment
The most famous cold fusion claims were made by Stanley Pons and Martin Fleischmann in
1989. After a brief period of interest by the wider scientific community, their reports were called
into question by nuclear physicists. Pons and Fleischmann never retracted their claims, but
moved their research program to France after the controversy erupted.
Events preceding announcement
Electrolysis cell schematic

Martin Fleischmann​ of the ​University of Southampton​ and ​Stanley Pons​ of the ​University of Utah
hypothesized that the high compression ratio and mobility of ​deuterium​ that could be achieved
within palladium metal using electrolysis might result in nuclear fusion.​[26]​ To investigate, they
conducted electrolysis experiments using a palladium cathode and heavy water within a
calorimeter, an insulated vessel designed to measure process heat. Current was applied
continuously for many weeks, with the ​heavy water​ being renewed at intervals.​[26]​ Some
deuterium was thought to be accumulating within the cathode, but most was allowed to bubble
out of the cell, joining oxygen produced at the anode.​[27]​ For most of the time, the power input to
the cell was equal to the calculated power leaving the cell within measurement accuracy, and the
cell temperature was stable at around 30 °C. But then, at some point (in some of the
experiments), the temperature rose suddenly to about 50 °C without changes in the input power.
These high temperature phases would last for two days or more and would repeat several times
in any given experiment once they had occurred. The calculated power leaving the cell was
significantly higher than the input power during these high temperature phases. Eventually the
high temperature phases would no longer occur within a particular cell.​

In 1988 Fleischmann and Pons applied to the ​United States Department of Energy​ for funding
towards a larger series of experiments. Up to this point they had been funding their experiments
using a small device built with $100,000 ​out-of-pocket​.​ The grant proposal was turned over for
peer review​, and one of the reviewers was ​Steven Jones​ of ​Brigham Young University​.​ Jones
had worked for some time on ​muon-catalyzed fusion​, a known method of inducing nuclear fusion
without high temperatures, and had written an article on the topic entitled "Cold nuclear fusion"
that had been published in ​Scientific American​ in July 1987. Fleischmann and Pons and
co-workers met with Jones and co-workers on occasion in ​Utah​ to share research and
techniques. During this time, Fleischmann and Pons described their experiments as generating
considerable "excess energy", in the sense that it could not be explained by ​chemical reactions
alone.​[27]​ They felt that such a discovery could bear significant commercial value and would be
entitled to ​patent protection​. Jones, however, was measuring neutron flux, which was not of
[28]​[​clarification needed]​ ​
commercial interest.​ To avoid future problems, the teams appeared to agree
to simultaneously publish their results, though their accounts of their 6 March meeting differ.​[29]

Although the experimental protocol had not been published, physicists in several countries
attempted, and failed, to replicate the excess heat phenomenon. The first paper submitted to
Naturer​ eproducing excess heat, although it passed peer-review, was rejected because most
similar experiments were negative and there were no theories that could explain a positive
result;​[notes 1]​[38]​this paper was later accepted for publication by the journal ​Fusion Technology.​
Nathan Lewis​, professor of chemistry at the ​California Institute of Technology​, led one of the
most ambitious validation efforts, trying many variations on the experiment without success,​
while ​CERN​ physicist Douglas R. O. Morrison said that "essentially all" attempts in Western
Europe had failed.​ Even those reporting success had difficulty reproducing Fleischmann and
Pons' results.​[40]​ On 10 April 1989, a group at ​Texas A&M University​ published results of excess
heat and later that day a group at the ​Georgia Institute of Technology​ announced neutron
production—the strongest replication announced up to that point due to the detection of neutrons
and the reputation of the lab
A cold fusion experiment usually includes:
· ​a metal, such as ​palladium​ or ​nickel​, in bulk, thin films or powder; and
· ​ ​deuterium​, ​hydrogen​, or both, in the form of water, gas or plasma.

Electrolysis cells can be either open cell or closed cell. In open cell systems, the electrolysis
products, which are gaseous, are allowed to leave the cell. In closed cell experiments, the
products are captured, for example by catalytically recombining the products in a separate part of
the experimental system. These experiments generally strive for a steady state condition, with
the electrolyte being replaced periodically. There are also "heat-after-death" experiments, where
the evolution of heat is monitored after the electric current is turned off.
The most basic setup of a cold fusion cell consists of two electrodes submerged in a solution
containing palladium and heavy water. The electrodes are then connected to a power source to
transmit electricity from one electrode to the other through the solution.​[109]​ Even when
anomalous heat is reported, it can take weeks for it to begin to appear—this is known as the
"loading time," the time required to saturate the palladium electrode with hydrogen (see "Loading
ratio" section).
The Fleischmann and Pons early findings regarding helium, neutron radiation and tritium were
never replicated satisfactorily, and its levels were too low for the claimed heat production and
inconsistent with each other.​ Neutron radiation has been reported in cold fusion experiments
at very low levels using different kinds of detectors, but levels were too low, close to background,
and found too infrequently to provide useful information about possible nuclear processes.​

Excess heat and energy production

An excess heat observation is based on an ​energy balance​. Various sources of energy input and
output are continuously measured. Under normal conditions, the energy input can be matched to
the energy output to within experimental error. In experiments such as those run by Fleischmann
and Pons, an electrolysis cell operating steadily at one temperature transitions to operating at a
higher temperature with no increase in applied current.​[27]​ If the higher temperatures were real,
and not an experimental artifact, the energy balance would show an unaccounted term. In the
Fleischmann and Pons experiments, the rate of inferred excess heat generation was in the range
of 10–20% of total input, though this could not be reliably replicated by most
researchers.​[112]​Researcher ​Nathan Lewis​ discovered that the excess heat in Fleischmann and
Pons's original paper was not measured, but estimated from measurements that didn't have any
excess heat.​[113]

Unable to produce excess heat or neutrons, and with positive experiments being plagued by
errors and giving disparate results, most researchers declared that heat production was not a
real effect and ceased working on the experiments.​[114]​ In 1993, after their original report,
Fleischmann reported "heat-after-death" experiments—where excess heat was measured after
the electric current supplied to the electrolytic cell was turned off.​[115]​ This type of report has also
become part of subsequent cold fusion claims.​

Helium, heavy elements, and neutrons

"Triple tracks" in a ​CR-39​ plastic radiation detector claimed as evidence for neutron emission from
palladium deuteride

Known instances of nuclear reactions, aside from producing energy, also produce ​nucleons​ and
particles on readily observable ballistic trajectories. In support of their claim that nuclear
reactions took place in their electrolytic cells, Fleischmann and Pons reported a ​neutron flux​ of
4,000 neutrons per second, as well as detection of tritium. The classical ​branching ratio​ for
previously known fusion reactions that produce tritium would predict, with 1 ​watt​ of power, the
12​ [117]
production of 10​ neutrons per second, levels that would have been fatal to the researchers.​
In 2009, Mosier-Boss et al. reported what they called the first scientific report of highly energetic
neutrons, using ​CR-39​ plastic radiation detectors,​ but the claims cannot be validated without a
quantitative analysis​ of neutrons.​[118]​[119]
Several medium and heavy elements like calcium, titanium, chromium, manganese, iron, cobalt,
copper and zinc have been reported as detected by several researchers, like ​Tadahiko Mizuno​ or
George Miley​. The report presented to the ​United States Department of Energy (DOE)​ in 2004
indicated that deuterium-loaded foils could be used to detect fusion reaction products and,
although the reviewers found the evidence presented to them as inconclusive, they indicated that
those experiments did not use state-of-the-art techniques.​[120]

In response to doubts about the lack of nuclear products, cold fusion researchers have tried to
capture and measure nuclear products correlated with excess heat.​ Considerable attention
4​ [13]​
has been given to measuring ​ He production.​ However, the reported levels are very near to
background, so contamination by trace amounts of helium normally present in the air cannot be
ruled out. In the report presented to the DOE in 2004, the reviewers' opinion was divided on the
evidence for ​ He; with the most negative reviews concluding that although the amounts detected
were above background levels, they were very close to them and therefore could be caused by
contamination from air.​[122]

One of the main criticisms of cold fusion was that deuteron-deuteron fusion into helium was
expected to result in the production of ​gamma rays​—which were not observed and were not
observed in subsequent cold fusion experiments.​[40]​[123]​ Cold fusion researchers have since
claimed to find X-rays, helium, neutrons​[124]​ and ​nuclear transmutations​.[125]​
​ Some researchers
also claim to have found them using only light water and nickel cathodes.​ The 2004 DOE
panel expressed concerns about the poor quality of the theoretical framework cold fusion
proponents presented to account for the lack of gamma rays.​

Proposed mechanisms
Researchers in the field do not agree on a theory for cold fusion.​[126]​ One proposal considers
that hydrogen and its ​isotopes​ can be absorbed in certain solids, including ​palladium hydride​, at
high densities. This creates a high partial pressure, reducing the average separation of hydrogen
isotopes. However, the reduction in separation is not enough by a factor of ten to create the
fusion rates claimed in the original experiment.​[127]​ It was also proposed that a higher density of
hydrogen inside the palladium and a lower potential barrier could raise the possibility of fusion at
lower temperatures than expected from a simple application of ​Coulomb's law​. ​Electron
screening​ of the positive hydrogen nuclei by the negative electrons in the palladium lattice was
suggested to the 2004 DOE commission,​ but the panel found the theoretical explanations not
convincing and inconsistent with current physics theories.​[91]

Criticism of cold fusion claims generally take one of two forms: either pointing out the theoretical
implausibility that fusion reactions have occurred in electrolysis set-ups or criticizing the excess
heat measurements as being spurious, erroneous, or due to poor methodology or controls. There
are a couple of reasons why known fusion reactions are an unlikely explanation for the excess
[text 6]
heat and associated cold fusion claims.​
Repulsion forces
Because nuclei are all positively charged, they strongly repel one another.​[40]​ Normally, in the
absence of a catalyst such as a ​muon​, very high ​kinetic energies​ are required to overcome this
charged repulsion​.[129]​
​ Extrapolating from known fusion rates, the rate for uncatalyzed fusion at
room-temperature energy would be 50 orders of magnitude lower than needed to account for the
reported excess heat.​ In muon-catalyzed fusion there are more fusions because the
presence of the muon causes deuterium nuclei to be 207 times closer than in ordinary deuterium
gas.​ But deuterium nuclei inside a palladium lattice are further apart than in deuterium gas,
and there should be fewer fusion reactions, not more.​[127]

Paneth and Peters in the 1920s already knew that palladium can absorb up to 900 times its own
volume of hydrogen gas, storing it at several thousands of times the ​atmospheric
​ This led them to believe that they could increase the nuclear fusion rate by simply
loading palladium rods with hydrogen gas.​[132]​ Tandberg then tried the same experiment but
used electrolysis to make palladium absorb more deuterium and force the deuterium further
together inside the rods, thus anticipating the main elements of Fleischmann and Pons'
experiment.​[132]​[21]​ They all hoped that pairs of hydrogen nuclei would fuse together to form
helium, which at the time was needed in Germany to fill ​zeppelins​, but no evidence of helium or
of increased fusion rate was ever found.​[132]

This was also the belief of geologist Palmer, who convinced Steven Jones that the helium-3
occurring naturally in Earth perhaps came from fusion involving hydrogen isotopes inside
catalysts like nickel and palladium.​[133]​ This led their team in 1986 to independently make the
same experimental setup as Fleischmann and Pons (a palladium cathode submerged in heavy
water, absorbing deuterium via electrolysis).​[134]​ Fleischmann and Pons had much the same
[135]​ 27​
belief,​ but they calculated the pressure to be of 10​ atmospheres, when cold fusion
experiments only achieve a loading ratio of one to one, which only has between 10,000 and
[text 7]​
20,000 atmospheres.​ ​John R. Huizenga​ says they had misinterpreted the ​Nernst equation​,
leading them to believe that there was enough pressure to bring deuterons so close to each
other that there would be spontaneous fusions.​[136]

Lack of expected reaction products

Conventional deuteron fusion is a two-step process,​[text 6]​ in which an unstable high energy
intermediary is formed:
4​ *​
D​ + D → ​ He​ + 24 ​MeV

Experiments have observed only three decay pathways for this excited-state nucleus, with
the ​branching ratio​ showing the probability that any given intermediate follows a particular
[text 6]​
pathway.​ The products formed via these decay pathways are:
4​ *​ 3​
He​ → ​n​ + ​ He​ + 3.3 MeV (​ratio​=50%)
4​ *​ 3​
He​ → ​p​ + ​ H​ + 4.0 MeV (ratio=50%)
He​*​ → 4​​ He​ + ​γ​ + 24 MeV (ratio=10​−6​)

Only about one in one million of the intermediaries decay along the third pathway, making its
products comparatively rare when compared to the other paths.​[40]​ This result is consistent with
[text 8]​ 18​
the predictions of the ​Bohr model​.​ If one watt (1 W = 1 J/s ; 1 J = 6.242 x 10​ eV = 6.242 x
12​ −19​
10​ MeV since 1 eV = 1.602 x 10​ joule) of nuclear power were produced from ~2.2575 x
10​ deuteron fusion individual reactions each second consistent with known branching ratios, the
3​ [40]​[137]​
resulting neutron and tritium (​ H) production would be easily measured.​ Some
researchers reported detecting ​ He but without the expected neutron or tritium production; such
a result would require branching ratios strongly favouring the third pathway, with the actual rates
of the first two pathways lower by at least five orders of magnitude than observations from other
experiments, directly contradicting both theoretically predicted and observed branching
[text 6]​ 4​
probabilities.​ Those reports of ​ He production did not include detection of ​gamma rays​,
which would require the third pathway to have been changed somehow so that gamma rays are
[text 6]
no longer emitted.​

The known rate of the decay process together with the inter-atomic spacing in a ​metallic crystal
makes heat transfer of the 24 MeV excess energy into the host metal lattice prior to the
intermediary​'s decay inexplicable in terms of conventional understandings of ​momentum​ and
[138]​ [139]​
energy transfer,​ and even then there would be measurable levels of radiation.​ Also,
experiments indicate that the ratios of deuterium fusion remain constant at different energies.​[140]
In general, pressure and chemical environment only cause small changes to fusion ratios.​ An
early explanation invoked the ​Oppenheimer–Phillips process​ at low energies, but its magnitude
was too small to explain the altered ratios.​

Setup of experiments
Cold fusion setups utilize an input power source (to ostensibly provide ​activation energy​), a
platinum group​ ​electrode​, a deuterium or hydrogen source, a ​calorimeter​, and, at times,
detectors to look for byproducts such as helium or neutrons. Critics have variously taken issue
with each of these aspects and have asserted that there has not yet been a consistent
reproduction of claimed cold fusion results in either energy output or byproducts. Some cold
fusion researchers who claim that they can consistently measure an excess heat effect have
argued that the apparent lack of reproducibility might be attributable to a lack of quality control in
the electrode metal or the amount of hydrogen or deuterium loaded in the system. Critics have
further taken issue with what they describe as mistakes or errors of interpretation that cold fusion
researchers have made in calorimetry analyses and energy budgets.

In 1989, after Fleischmann and Pons had made their claims, many research groups tried to
reproduce the Fleischmann-Pons experiment, without success. A few other research groups,
however, reported successful reproductions of cold fusion during this time. In July 1989, an
Indian group from the ​Bhabha Atomic Research Centre​ (​P. K. Iyengar​ and M. Srinivasan) and in
October 1989, ​John Bockris'​ group from ​Texas A&M University​ reported on the creation of
tritium. In December 1990, professor ​Richard Oriani​ of the ​University of Minnesota​ reported
excess heat.​[142]

Groups that did report successes found that some of their cells were producing the effect, while
other cells that were built exactly the same and used the same materials were not producing the
effect.​ Researchers that continued to work on the topic have claimed that over the years
many successful replications have been made, but still have problems getting reliable
replications.​ Reproducibility​ is one of the main principles of the scientific method, and its lack
led most physicists to believe that the few positive reports could be attributed to experimental
[143]​[text 9]​
error.​ The DOE 2004 report said among its conclusions and recommendations:

"Ordinarily, new scientific discoveries are claimed to be consistent and reproducible; as a result,
if the experiments are not complicated, the discovery can usually be confirmed or disproved in a
few months. The claims of cold fusion, however, are unusual in that even the strongest
proponents of cold fusion assert that the experiments, for unknown reasons, are not consistent
and reproducible at the present time. (...) Internal inconsistencies and lack of predictability and
reproducibility remain serious concerns. (...) The Panel recommends that the cold fusion
research efforts in the area of heat production focus primarily on confirming or disproving reports
of excess heat."​[91]

Loading ratio

Michael McKubre working on deuterium gas-based cold fusion cell used by ​SRI

[144]​ [89]​
Cold fusion researchers (​McKubre​ since 1994,​ ​ENEA​ in 2011​ ) have speculated that a cell
that is loaded with a deuterium/palladium ratio lower than 100% (or 1:1) will not produce excess
heat.​ Since most of the negative replications from 1989–1990 did not report their ratios, this
has been proposed as an explanation for failed replications.​ This loading ratio is hard to
obtain, and some batches of palladium never reach it because the pressure causes cracks in the
palladium, allowing the deuterium to escape.​ Fleischmann and Pons never disclosed the
deuterium/palladium ratio achieved in their cells,​ there are no longer any batches of the
palladium used by Fleischmann and Pons (because the supplier uses now a different
manufacturing process),​[144]​ and researchers still have problems finding batches of palladium
that achieve heat production reliably.​

Misinterpretation of data
Some research groups initially reported that they had replicated the Fleischmann and Pons
results but later retracted their reports and offered an alternative explanation for their original
positive results. A group at ​Georgia Tech​ found problems with their neutron detector, and Texas
A&M discovered bad wiring in their thermometers.​[146]​ These retractions, combined with negative
results from some famous laboratories,​ led most scientists to conclude, as early as 1989, that
no positive result should be attributed to cold fusion.​
Calorimetry errors
The calculation of excess heat in electrochemical cells involves certain assumptions.​ Errors
in these assumptions have been offered as non-nuclear explanations for excess heat.
One assumption made by Fleischmann and Pons is that the efficiency of electrolysis is nearly
100%, meaning nearly all the electricity applied to the cell resulted in electrolysis of water, with
negligible ​resistive heating​ and substantially all the electrolysis product leaving the cell
unchanged.​[27]​ This assumption gives the amount of energy expended converting liquid D​2​O into
gaseous D​2​and O​2​.​ The efficiency of electrolysis is less than one if hydrogen and oxygen
recombine to a significant extent within the calorimeter. Several researchers have described
potential mechanisms by which this process could occur and thereby account for excess heat in
electrolysis experiments.​[150]​[151]​[152]

Another assumption is that heat loss from the calorimeter maintains the same relationship with
measured temperature as found when calibrating the calorimeter.​[27]​ This assumption ceases to
be accurate if the temperature distribution within the cell becomes significantly altered from the
condition under which calibration measurements were made.​[153]​ This can happen, for example,
if fluid circulation within the cell becomes significantly altered.​ Recombination of
hydrogen and oxygen within the calorimeter would also alter the heat distribution and invalidate
the calibration.