A dark matter experiment that has tantalized physicists by reporting a tentative positive signal is back online three months after a fire disrupted its data-taking.
The recovery of the CoGeNT experiment in the Soudan mine in Minnesota means there is a very good chance of confirming or rejecting its signal over the next few months, says Juan Collar of the University of Chicago, the experiment’s principal investigator. “As of yesterday [June 7] everything is back to normal,” says Collar, “in a few months we’ll know if this signal is a fluke.”
On June 3, CoGeNT reported a marginal positive signal from 15 months of data showing a seasonal rise and fall in observations of particles hitting nuclei of germanium in its crystalline detector. This is consistent with the hypothesis that the particles are Weakly Interacting Massive Particles (WIMPS), dark matter particles that flow through the Earth as the planet orbits the Sun. Although the result, which has a statistical significance of 2.8 sigma, meaning there is a more than 0.3% probability that it is due to chance, falls well short of that needed to claim a discovery, it is the first indication of independent evidence for a statistically well-supported but so far unreproduced claim by the DAMA experiment at Gran Sasso in Italy, to have observed the same kind of seasonal variation.
Collar says that although CoGeNT’s result was marginal, he chose to release it now because a 17 March fire in the mine caused a power failure, allowing the detector, which needs to be cryogenically cooled, to warm up. For weeks researchers on the experiment were unable to enter the mine, and judged there was only a 50% chance it would be possible to cool the detector down, re-achieve stability, and continue the run. This, combined with the fact that the signal might agree with DAMA, persuaded him to make the data release.
The new result, together with Collar’s fierce attacks on analyses by a competing dark matter collaboration, XENON100, that did not confirm the positive signals from DAMA and CoGeNT, have ruffled feathers in the dark matter community.
Rick Gaitskell, principal investigator of LUX, a dark matter experiment in the Sanford Underground Laboratory at Homestake, South Dakota, says that Collar’s challenges to XENON100 appear to be a rational attempt to reconcile the signal his collaboration is seeing with the other experiment’s null result. “This is science and it’s absolutely fair there should be a debate about strengths and weaknesses” he says. But Gaitskell adds that it really would have been ideal for the CoGeNT collaboration to have had the chance to collect more data before releasing this result. “It would’ve been nicer to have a couple of years of data, and a greater statistical significance in the result. 2.8 sigma results have a habit of evaporating.” he says.