By Ron Cowen of Nature magazine

Physicists last week announced evidence that particles of dark matter--the invisible, hypothetical material believed to make up more than 80 percent of the mass of the Universe--may have a lower mass than suspected. The results, posted on the arXiv preprint server, match findings from some experiments but contradict others.

What are the findings?

The researchers, including Godehard Angholer at the Max Planck Institute of Physics in Munich, Germany, focused on candidate dark-matter particles called WIMPs--weakly interacting massive particles. The team analysed data from the CRESST II experiment housed at the Gran Sasso National Laboratory deep beneath the Italian Apennine Mountains. Of the 67 WIMP-like signals the researchers identified, 27 could not be discounted as background events, they say. And if those signals do turn out to be WIMPs, then the particles would have a mass of roughly 10 billion to 50 billion electronvolts (9-53 times the mass of a proton).

How do these findings compare to other possible dark matter detections?

Leo Stodoksky, who works on CRESST II, calls the findings, which do not meet the criteria for proof in particle physics "intriguing but not definitive" because the scientists can not be sure they have properly identified and accounted for every background event that would mimic a WIMP-like signal. Other experiments using different detectors, including DAMA/LIBRA, also at Gran Sasso, and CoGeNT, located in the Soudan Underground Laboratory in Minnesota, have detected WIMP-like signals in a similar mass range. But all these results are at odds with two other experiments, XENON100 in the Italian laboratory and CDMS II in Soudan, which have found no statistically significant evidence of WIMPs in a similar mass range.

So is there any consensus about the nature of dark matter?

Obviously, says theorist Neal Weiner of New York University in New York, somebody is wrong. But it's unclear, he adds, whether some--or perhaps all--of the experiments are wrong, or if WIMPs are more complex than the relatively simple particles theorists have hypothesized. The dark-matter particles might interact differently with protons than with neutrons, or they may jump from lower internal energy states to higher ones, just as electrons jump from lower to higher energy levels in an atom. Such properties might explain the apparent discrepancies between experiments, but would contradict a popular theory of particle physics called supersymmetry, which posits that every known elementary particle has a heavier counterpart. The simplest version of supersymmetry does not include a WIMP with such properties as two internal energy states. Regardless of the detailed properties of WIMPs, "I'm taking light WIMPs very very seriously at this time," because of the mounting evidence from CRESST and other experiments, says theorist Dan Hooper of the Fermi National Accelerator Laboratory in Batavia, Illinois. "My feet go where the evidence takes me."

Why did physicists think it unlikely that WIMPs had low mass?

Theorists thought that, because low-mass particles are easier to create in particle accelerators, the fingerprints of a low-mass WIMP would have already been seen in machines such as the Large Hadron Collider at CERN, the particle-physics laboratory near Geneva, Switzerland. But, faced with the latest results, scientists are realizing that WIMPs may have a low mass after all, but are difficult to spot. The assumption that WIMPs should have a high mass "was a kind of a fad with no particular justification," says Stodolsky.

After more than two decades of hunting, why is it still so difficult to find dark matter?

In spite of having a mass similar to that of heavier nuclei, such as those in iron atoms, dark-matter particles interact so weakly with ordinary matter that they often travel through Earth, or even the entire Milky Way, without a trace. So detectors need a large area and many kilograms of mass to catch a WIMP. The experiments must be buried underground to shield them from sources, such as cosmic rays, that can produce a cacophony of background signals. The experiments also need additional shielding from radioactive decay from the walls of the underground laboratories and the experimental equipment itself.

As we can't see it and no one has definitively detected it, why do most scientists think dark matter must exist?

The amount of ordinary matter in the Universe does not provide enough gravitational pull to keep rotating galaxies from flying apart. Dark matter is also thought to have been important early in the development of the Universe because it provided the gravitational attraction needed to gather matter into clumps that would eventually become galaxies. In addition, the abundance of WIMPs in the Universe today, as calculated by particle physicists, roughly matches the amount of dark matter required to explain its present structure. This match is known as the "WIMP miracle."

What's next?

Results from XENON100 are expected later this year, and additional data from CRESST II should further reduce errors created by sources that mimic WIMPs. As CRESST II continues, it will be able to look for seasonal variation in the number of WIMPs striking its detectors. Such variation--more WIMPs in summer than in winter--has been evidenced in the DAMA/LIBRA experiments for years, and in CoGeNT earlier this year5. If confirmed, these variations would be consistent with predictions of a cloud of dark-matter particles that remains stationary as the Galaxy rotates. Evidence confirming a cloud of WIMPs in the Milky Way would put the search for dark matter particles on a surer footing.

This article is reproduced with permission from the magazine Nature. The article was first published on September 13, 2011.