Since 2015, Google has been funding experiments into the controversial science of cold fusion — the theory that nuclear fusion, the process that powers the Sun, can produce energy in a table-top experiment at room temperature. Two scientists first made sensational claims about achieving the phenomenon — promising endless, cheap energy — 30 years ago, but their results were quickly debunked and the topic is now considered a scientific taboo.
Google’s project — revealed in a peer-reviewed Nature Perspective1 this week — found no evidence that cold fusion is possible, but made some advances in measurement and materials-science techniques that the researchers say could benefit energy research. The team also hopes that its work will inspire others to revisit cold-fusion experiments, even if the phenomenon still fails to materialize.
“This is not just a chase for cold fusion,” says Matthew Trevithick, a research programme manager at Google in Mountain View, California. “If it were, I don’t think we would have maintained an interest of this calibre of team for so long.”
The Google team explored three experimental set-ups that have been proposed to generate cold fusion — two involving palladium and hydrogen, and one involving metallic powders and hydrogen. None found evidence of fusion. The results have been published across 12 papers over the past 2 years: 9 in peer-reviewed journals and 3 on the arXiv preprint server.
Some scientists welcomed the scrutiny brought by the Google project. But Frank Close, a theoretical physicist at the University of Oxford, UK, says that the scientific mainstream has shunned the topic for good reason: no one has managed to independently reproduce the finding and more worthwhile topics have emerged, he says. “There is no theoretical reason to expect cold fusion to be possible, and a vast amount of well-established science that says it should be impossible,” says Close, who was involved in efforts to replicate the original 1989 experiment.
In March 1989, two US-based chemists Stanley Pons and Martin Fleischmann announced that they had seen excess heat and fusion-reaction products — signs of nuclear fusion — when they ran a current across two palladium plates in water laden with deuterium, a heavy isotope of hydrogen. Others quickly pointed out errors in their experimental procedure. Since then, two US Department of Energy reviews have found no evidence of the phenomenon .
But cold fusion — now commonly referred to as low-energy nuclear reactions — has retained a persistent following that continues to claim evidence of success.
Google’s US$10-million project aimed to test the cold-fusion claims rigorously in a field that lacked credible scientific data, says Trevithick. Another goal was also to push methods in challenging experimental conditions. But, he adds: “The fact that the pay-out could be huge is definitely a component of our interest.”
Nuclear fusion is thought to happen only in extreme environments such as the Sun, where high temperatures and pressures can cause hydrogen atoms to overcome their mutual repulsion and fuse into helium, releasing enormous amounts of energy. Some experiments on Earth are trying to replicate the phenomenon, but haven’t yet proved that they can generate enough energy to make up for the vast amounts they need to run.
The probability of atoms fusing at much lower temperatures is thought to be vanishingly small. But, if possible, this phenomenon would bring enormous benefits by doing away with fusion’s vast energy requirements.
Google’s team was made up of 30 researchers who had no strong opinions on cold fusion. All had access to each other’s data and apparatus, and could review each other’s work.
The researchers pursued the three experimental strands that they deemed sufficiently credible. In one, they tried to load palladium with amounts of deuterium hypothesized to be necessary to trigger fusion. But at high concentrations the team was unable to create stable samples.
A second strand followed up on 1990s work by US physicists who claimed to have generated anomalous levels of tritium — another heavy hydrogen isotope, created only through nuclear reactions — by bombarding palladium with pulses of hot deuterium ions. Google’s analysis of nuclear signatures showed no tritium production from this experiment.
A final strand involved heating up metallic powders in a hydrogen-rich environment. Some current proponents of cold fusion claim that the process produces excess and unexplained heat, which they theorize is the result of fusing elements. But across 420 tests, the Google-funded team found no such heat excess.
But the researchers say that both palladium experiments warrant further study. The hypothesized effects in the tritium experiment could be too small to measure with current equipment, they suggest. The team also says that further work could produce stable samples at extremely high deuterium concentrations, where interesting effects might occur.
All the projects pushed the frontier of experimental methods, says Trevithick, including developing “the best calorimeters in the world” to detect even slight excesses of heat under extreme experimental conditions. These could potentially be used to test future claims.
Pushing the envelope
“I think that the authors have done a really good job,” says David Williams, an electrochemist at the University of Auckland in New Zealand — especially in how they have navigated the controversial topic. Pushing the envelope of measurement science is also important, says Williams, whose team did some of the first failed replication studies of the original claim.
The techniques the team developed to load palladium may also help researchers to boost the hydrogen storage capacity of materials being studied for use in batteries and fuel cells, says George Chen, an electrochemist at the University of Nottingham’s China campus in Ningbo.
Trevithick notes that in one case his team has not been able to reach even the hypothesized starting conditions for fusion, so have not fully eliminated the possibility that it occurs.
But Close says that being unable to rule an idea out completely does not mean there is good reason to pursue it. “You cannot prove a negative in science,” he says. If Google wants to invest in cold fusion — that’s up to them, he says. But “if somebody I was investing my money in started doing this, I would withdraw my money”, says Close.
Curtis Berlinguette, a chemist at the University of British Columbia in Vancouver and one of the project’s principal investigators, is sceptical of the “classic” cold-fusion experiments. But he was excited to do the work and thinks that a new generation of creative scientists could develop methods that drive fusion reactions at low temperatures. Some might judge the team harshly but the project simply explored an underexplored space — which was off limits because of prejudice, he says. “This is what we are supposed to do as scientists.”
Nature 569, 611 (2019)