The controversial finding is turning up the heat on an already simmering debate, especially since it is based on re-analysis of old data that has long been used to argue for exactly the opposite: the constancy of the speed of light and other constants.
A varying speed of light contradicts Einstein's theory of relativity, and would undermine much of traditional physics. But some physicists believe it would elegantly explain puzzling cosmological phenomena such as the nearly uniform temperature of the universe. It might also support string theories that predict extra spatial dimensions.
The threat to the idea of an invariable speed of light comes from measurements of another parameter called the fine structure constant, or alpha, which dictates the strength of the electromagnetic force. The speed of light is inversely proportional to alpha, and though alpha also depends on two other constants (see Graphic), many physicists tend to interpret a change in alpha as a change in the speed of light. It is a valid simplification, says Victor Flambaum of the University of New South Wales in Sydney. It was Flambaum, along with John Webb and colleagues, who first seriously challenged alpha's status as a constant in 1998. Then, after exhaustively analysing how the light from distant quasars was absorbed by intervening gas clouds, they claimed in 2001 that alpha had increased by a few parts in 105 in the past 12 billion years.
But then German researchers studying photons emitted by caesium and hydrogen atoms reported last month that they had seen no change in alpha to within a few parts in 1015 over the period from 1999 to 2003 (New Scientist, 26 June, p 15)- though the result does not rule out that alpha was changing billions of years ago.
Throughout the debate, physicists who argued against any change in alpha have had one set of data to fall back on. It comes from the world's only known natural nuclear reactor, found at Oklo in Gabon, West Africa. The Oklo reactor started up nearly two billion years ago when groundwater filtered through crevices in the rocks and mixed with uranium ore to trigger a fission reaction that was sustained for hundreds of thousands of years. Several studies that have analysed the relative concentrations of radioactive isotopes left behind at Oklo have concluded that nuclear reactions then were much the same as they are today, which implies alpha was the same too.
That's because alpha directly influences the ratio of these isotopes. In a nuclear chain reaction like the one that occurred at Oklo, the fission of each uranium-235 nucleus produces neutrons, and nearby nuclei can capture these neutrons.
For example, samarium-149 captures a neutron to become samarium-150, and since the rate of neutron capture depends on the value of alpha, the ratio of the two samarium isotopes in samples collected from Oklo can be used to calculate alpha.
A number of studies done since Oklo was discovered have found no change in alpha over time. "People started quoting the reactor [data] as firm evidence that the constants hadn't changed," says Steve Lamoreaux of Los Alamos National Lab (LANL) in Albuquerque, New Mexico.
Now, Lamoreaux, along with LANL colleague Justin Torgerson, has re-analysed the Oklo data using what he says are more realistic figures for the energy spectrum of the neutrons present in the reactor. The results have surprised him.
Alpha, it seems, has decreased by more than 4.5 parts in 108 since Oklo was live (Physical Review D, vol 69, p121701). That translates into a very small increase in the speed of light (assuming no change in the other constants that alpha depends on), but Lamoreaux's new analysis is so precise that he can rule out the possibility of zero change in the speed of light. "It's pretty exciting," he says.
So far the re-examination of the Oklo data has not drawn any fire. "The analysis is fine," says Thibault Damour of the Institute of Advanced Scientific Studies (IHES) in Bures-sur-Yvette in France, who co-authored a 1996 Oklo study that found no change in alpha. Peter Moller of LANL, who, along with Japanese researchers, published a paper in 2000 about the Oklo reactor that also found no change in alpha, says that Lamoreaux's assumptions are reasonable.
The analysis might be sound, and the assumptions reasonable, but some physicists are reluctant to accept the conclusions. "I can't see a particular mistake," says Flambaum. "However, the claim is so revolutionary there should be many independent confirmations." While Flambaum's own team found that alpha was different 12 billion years ago, the new Oklo result claims that alpha was changing as late as two billion years ago. If other methods confirm the Oklo finding, it will leave physicists scrambling for new theories. "It's like opening a gateway," says Dmitry Budker, a colleague of Lamoreaux's at the University of California at Berkeley. Some physicists would happily accept a variable alpha.
For example, if it had been lower in the past, meaning a higher speed of light, it would solve the "horizon problem". Cosmologists have struggled to explain why far-flung regions of the universe are at roughly the same temperature. It implies that these regions were once close enough to exchange energy and even out the temperature, yet current models of the early universe prevent this from happening, unless they assume an ultra-fast expansion right after the big bang. However, a higher speed of light early in the history of the universe would allow energy to pass between these areas in the form of light.
Variable "constants" would also open the door to theories that used to be off limits, such as those which break the laws of conservation of energy. And it would be a boost to versions of string theory in which extra dimensions change the constants of nature at some places in space-time. But "there is no accepted varying-alpha theory", warns Flambaum.
Instead, there are competing theories, from those that predict a linear rate of change in alpha, to those that predict rapid oscillations. John Barrow, who has pioneered varying-alpha theories at the University of Cambridge, says that the latest Oklo result does not favour any of the current theories. "You would expect alpha to stop [changing] five to six billion years ago," he says. Before Lamoreaux's Oklo study can count in favour of any varying alpha theory, there are some issues to be addressed. For one, the exact conditions at Oklo aren't known.
Nuclear reactions run at different rates depending on the temperature of the reactor, which Lamoreaux assumed was between 227°C and 527°C Damour says the temperature could vary far more than this. "You need to reconstruct the temperature two billion years ago deep down in the ground," he says. Damour also argues that the relative concentrations of samarium isotopes may not be as well determined as Lamoreaux has assumed, which would make it impossible to rule out an unchanging alpha. But Lamoreaux points out that both assumptions about the temperature of the Oklo reactor and the ratio of samarium isotopes were accepted in previous Oklo studies.
Another unknown is whether other physical constants might have varied along with, or instead of, alpha. Samarium-149's ability to capture a neutron also depends on another constant, alpha(s), which governs the strength of the strong nuclear attraction between the nucleus and the neutron. And in March, Flambaum claimed that the ratio of different elements left over from just after the big bang suggests that alpha(s) must have been different then compared with its value today (Physical Review D, vol 69, p 063506).
While Lamoreaux has not addressed any possible change in alpha(s) in his Oklo study, he argues that it is important to focus on possible changes in alpha because the Oklo data has become such a benchmark in the debate over whether alpha can vary. "I've spent my career going back and checking things that are 'known' and it always leads to new ideas," he says.
This article appears in New Scientist issue: 3 JULY 2004.
PLEASE MENTION NEW SCIENTIST AS THE SOURCE OF THIS STORY AND, IF PUBLISHING ONLINE, PLEASE CARRY A HYPERLINK TO: http://www.newscientist.com.
"These articles are posted on this site to give advance access to other authorised media who may wish to quote extracts as part of fair dealing with this copyrighted material. Full attribution is required, and if publishing online a link to http://www.newscientist.com is also required. Advance permission is required before any and every reproduction of each article in full - please contact email@example.com. Please note that all material is copyright of Reed Business Information Limited and we reserve the right to take such action as we consider appropriate to protect such copyright."
Author: Eugenie Samuel Reich
UK CONTACT - Claire Bowles, New Scientist Press Office, London:
Tel: 44-0-20-7331-2751 or email firstname.lastname@example.org
US CONTACT – Toni Marshall, New Scientist Boston office:
Tel: 617-558-4939 or email email@example.com