Imprimir Republish


A test-tube mirage

Failed cold fusion experiment attempted 30 years ago showed how illusory discoveries trigger self-correcting mechanisms in science

Pons (left) and Fleischmann present their experiment in March 1989; On May 8, Time Magazine reported on the doubts of the scientific community

Zuma Press, Inc. / Alamy / Fotoarena

Over the last four years, Google has invested US$10 million into research by physicists, chemists, and engineers to answer a question first asked three decades ago: Is it possible to fuse the nuclei of two atoms at room temperature and generate sustainable energy from the reaction? The mythical idea of “cold fusion,” which promises to free the planet from humankind’s reliance on fossil fuels, has never been achieved, and not for lack of trying. Google’s latest attempt once again failed, according to an article published in the journal Nature in May by a team led by Canadian chemist Curtis Berlinguette of the University of British Columbia in Vancouver, Canada, and engineer Matthew Trevithick of Google Research, USA.

The group attempted to achieve the feat via three different methodologies proposed in experiments carried out in the 1980s and 1990s. None showed any evidence that fusion could occur. But their effort was not considered a complete waste of time. As part of the research, the scientists designed new instruments and studied new materials that could be useful in other experiments in the future. “Evaluating cold fusion led our program to study materials and phenomena that we otherwise might not have considered. We set out looking for cold fusion, and instead benefited contemporary research topics in unexpected ways,” wrote the article authors. Trevithick told Nature that as part of the project, the researchers developed “the best calorimeters in the world,” capable of detecting even the subtlest of heat changes that might suggest nuclear fusion had occurred. Techniques developed to electrically charge palladium to attract heavy hydrogen nuclei—one of the tested methods—could also be used to increase hydrogen storage capacity in fuel cells.

Wikimedia Commons Blondlot (right) announced the discovery of N-rays and Schiaparelli saw canals on Mars: errors and wishful interpretationsWikimedia Commons

By subjecting the experiments to rigorous scrutiny, the Google researchers also brought this field of research back to mainstream science after decades of discredit, with most—but not all—of the scientific community seeing cold fusion as unfeasible. An episode that occurred 30 years ago may explain the scientific purgatory in which cold fusion has found itself trapped. On March 23, 1989, electrochemists Martin Fleischmann of the University of Southampton, England, and Stanley Pons of the University of Utah, USA, announced at a press conference that they had successfully generated thermal energy from the fusion of heavy hydrogen atoms—deuterium—in a simple electrolysis experiment. An electric current produced by two electrodes, one platinum and one palladium, caused deuterium to accumulate on the palladium. Neutrons were detected and the container in which the electrodes were immersed heated up. Fleischmann and Pons believed the only explanation for the reaction was the compression of deuterium atoms on the palladium and their consequent fusion at room temperature.

Fleischmann was a respected researcher and former chair of the International Society of Electrochemistry, as well as being a member of the Royal Society (the British Academy of Sciences), which lent credence to the announcement. But the way the discovery was made public completely ignored good scientific practices. Another group was working on cold fusion at the time, led by Steven Jones of Brigham Young University, also in Utah, and the two teams agreed to publish their findings together by submitting them jointly to Nature. But Pons and Fleischmann later decided to disclose their results early. They withdrew their article from Nature, which had just started its evaluation process, and submitted it an obscure electrochemistry journal called the Journal of Electroanalytical Chemistry, which quickly agreed to publish the results of the experiment. Before it was even printed, they announced in a press conference that they had achieved cold fusion. Years later, Fleischmann claimed that the move was made at the behest of the University of Utah, which wanted to patent and exploit the discovery.

Cato Institute | Susan Ragan/ AP Photo/ Glow images David Rasnick (right) and Peter Duesberg suggested that HIV is not the real cause of AIDS: their alternative hypothesis was discredited by fellow researchersCato Institute | Susan Ragan/ AP Photo/ Glow images

Despite not yet having access to the paper, researchers around the world raced to reproduce the results. It was such a simple experiment that even high-school students tried to repeat it. A few observed some heat production, but most failed—and the pair were accused of fraud and unethical behavior. Two weeks later, when the article was finally published, academics described it as weak and uninformative. The debacle soon turned into a major embarrassment. In May 1989, cold fusion was completely discredited at a meeting of the American Physical Society. To this day, it is unclear how the pair achieved the phenomena they described—and where the heat and neutrons came from. Because nobody was able to consistently reproduce the experiments, the US Department of Energy determined in November 1989 that no more money should be invested in attempts to replicate the results.

But this was not enough to put an end to the field of research. Although nobody has yet been able to demonstrate that cold fusion is possible, nobody has been able to rule it out either. “No one really knows if it might be possible to sustain a fusion reaction at low temperatures, or what those temperature limits might be,” computer scientist Steven Salzberg, from the University of Maryland, wrote in an article in Forbes. Nuclear fusion is a known phenomenon: energy is produced when two atoms combine to form a heavier atom. The process is typically seen in high-energy astrophysical environments and generates less radioactive waste than the fission process conducted at nuclear power plants. Nuclear fusion takes place inside stars: at the Sun’s core, under extreme temperatures and pressures, hydrogen fusion reactions create helium and release enormous amounts of energy.

Oak Ridge National Laboratory Construction of the International Thermonuclear Experimental Reactor in Cadarache, FranceOak Ridge National Laboratory

Science is still trying to harness this process. The most promising approach has nothing to do with Pons and Fleischmann’s room-temperature electrolysis experiments, but with perfecting the use of experimental reactors known as tokamaks, which confine heavy hydrogen isotopes—deuterium and tritium—in powerful magnetic fields and reach extremely high temperatures, causing fusion reactions. The challenge in this line of research is to increase the efficiency of the tokamaks so that the reactions release more energy than they consume, which no machine has yet achieved. Since 2007, a consortium formed by the European Union, China, South Korea, the United States, India, and Japan has been building the International Thermonuclear Experimental Reactor (ITER) in Cadarache, France, at a cost of €20 billion (see Pesquisa FAPESP issue no. 186). The reactor is scheduled to begin operating in 2025, with large-scale energy production expected in the 2040s. To create the conditions needed for nuclear fusion to occur, the reactor will heat the hydrogen to 150,000 degrees Celsius. “The great challenge of nuclear fusion is to achieve a sustainable and continuous reaction,” says physicist Peter Schulz, from the School of Applied Sciences at the University of Campinas (UNICAMP).

Throughout the 1990s, Pons and Fleischmann continued to argue that they had found a way to produce abundant clean energy. They held international conferences on cold fusion, each attended by hundreds of researchers, but the topic was seen as eccentric by the mainstream scientific community. In 1992, the two scientists moved to France to work in a lab that closed in 1998 without producing any concrete results—the funder, automaker Toyota, had invested US$15 million in the venture. Pons later became a French citizen, while Fleischmann retired in 1995 and returned to the UK, where he died in 2012.

EFDA Jet The JET tokamak in Culham, UK: the challenge is to produce continuous and sustainable reactionsEFDA Jet

“Cold fusion has become a fringe science, which may not be a pseudoscience, but certainly flirts with it,” says Schulz, from UNICAMP. “Fringe science has little respect or credibility in the scientific community, and is therefore studied by restricted groups, discussed in closed communities, and published in little-known journals.” Some groups are still engaged in cold fusion research, but they have adopted new terminologies such as “low-energy nuclear reactions” or “chemically assisted nuclear reactions.”

Another example of a fringe science was the theory, suggested at the end of the twentieth century, that the HIV virus does not cause AIDS. Biologist Peter Duesberg of the University of California, Berkeley, and biochemist David Rasnick argued that the disease was caused by other factors, such as malnutrition and recreational drug use. The idea was abandoned when drugs were created that prevented the virus from replicating and eliminated the symptoms of the syndrome. But in the early 2000s, the theory was revived by the South African government, which did not want to fund the purchase of drugs for people with the virus.

Chemist Philip Ball, former editor of the journal Nature, believes cold fusion is an example of a pathological science, a concept formulated by Nobel Prize–winning chemist Irving Langmuir to describe research by scientists who unintentionally deviate from the scientific method and end up producing false results that support their own exaggerated beliefs—known as wishful data interpretation. “This term was coined in the 1950s to describe a striking claim that conflicts with previous experience, that is based on effects that are difficult to detect and that is defended against criticism by ad hoc excuses,” Ball said in an article published in Nature last month.

Toyota invested US$15 million in a cold fusion lab in France, but closed it in 1998 after achieving no conclusive results

Some of the best-known examples of pathological science include the 1903 discovery of so-called N-Rays by French physicist Prosper-René Blondlot, who continued to defend their existence even after being dismissed by other researchers at the time, and the existence of canals on Mars, described by Italian astronomer Giovanni Schiaparelli in 1877 and used as evidence in the late nineteenth century of life on the red planet—they turned out to be nothing but an optical illusion. According to Ball, cold fusion has taught us lessons that remain relevant at a time when psychology and social science studies are in the spotlight because they have not been reproduced, and when discoveries are challenging traditional knowledge based on evidence gathered at the very limit of what our scientific instruments can detect. He says scientists must be given the freedom to look for innovative approaches and to make seemingly controversial claims, just as the scientific community needs to remain united in the spirit of investigation and the ongoing pursuit of self-correction.

“Cold fusion showed us the dangers of polarization, the distorting influence of commercial interests, and the importance of being open about methods, data, and mistakes,” said the chemist. “Would the cold-fusion saga play out differently today, with social media, fake news and an even more urgent need for clean energy? Probably—but not necessarily for the better.”