
Amid all the null results in the quest to directly detect dark matter, one research group has unwaveringly claimed success. Since 1997, researchers with the DAMA experiment in Italy have contended that their crystals of ultrapure sodium iodide are detecting particles that fit the profile of weakly interacting massive particles (WIMPs), a leading dark-matter candidate. The collaboration reports an unambiguous signal that peaks in the summer and fades in the winter. Though few researchers doubt that DAMA is seeing something, there are reasons to question the collaboration’s interpretation of GeV-mass WIMPs. Experiments using xenon, germanium, and other kinds of detectors—as well as DAMA’s latest results—seem to rule out at least the standard variety of WIMP as the culprit. And until recently, no other research team had been able to acquire NaI crystals with sufficient purity to perform an analogous search (see Physics Today, July 2016, page 28). Now the first of several long-awaited NaI-based dark-matter hunts has released its findings. Though more data are needed to verify the annual modulation of DAMA’s signal, the results strongly disfavor a generic WIMP interpretation.
Working on the same principle as DAMA, the COSINE-100 experiment looks for photon emission from the collisions of WIMPs and atoms in the NaI lattice. The key challenge is weeding out background signals, especially since the 106 kg of its crystals have up to four times the level of impurities as DAMA’s. The team modeled signals from rogue radioactive atoms embedded in the crystals and from cosmic rays and other outside particles. It housed its eight crystals, shown in the photo, at the Yangyang Underground Laboratory in South Korea and surrounded them with a shield of liquid scintillator, copper, and lead. The researchers, who collected data for nearly two months in late 2016, focused on events that traced back to a recoil in any individual crystal. Using the same model of a galactic dark-matter halo that DAMA has used to assess its data, they then simulated the interactions of WIMPs at any of 18 masses between 5 GeV and 104 GeV. Background contributions were sufficient to explain the entire observed signal. With 90% confidence, the researchers ruled out the phase space for the WIMPs claimed by DAMA (red and blue dotted lines in the graph).

The COSINE-100 team will now work on determining whether its detectors observe a seasonal cycle of nonbackground detections. Due to the sensitivity of the crystals and an unexpected shortfall in detector performance, that probably requires about three more years of data collection. But other NaI-based experiments are on the case, and COSINE-100 may share data with one of them, ANAIS in the Spanish Pyrenees, to speed up the process. With further progress on the NaI front, researchers may finally be able to peg DAMA’s signal to instrument error, an overlooked atmospheric or cosmic interloper, or a more exotic variety of dark-matter particle. (COSINE-100 collaboration, Nature 564, 83, 2018.)