About the DAMA-LIBRA result

The new result of the DAMA-LIBRA collaboration, which finds a yearly-modulated signal of interactions in their NaI crystals which is compatible with dark matter in our galactic halo, has caused some ripples in the web. I was surprised to find that a comment I had posted yesterday on the matter was linked by Symmetry magazine as well as by Peter Woit (who was however the true originator of my remark).

I must say I do not particularly enjoy to always sound skeptical. A scientist should keep an open mind, and if the DAMA-LIBRA signal has a cross-section which is apparently already excluded by the CDMS result, as well as orders of magnitude above the estimates for mainstream dark matter candidate models, one should wait before taking a step back, and rather consider questioning the exclusions rather than the signal. Indeed, establishing a signal comes before challenging it with specific models, and the comparison of DAMA’s result with CDMS exclusion contours belongs firmly to the second category. CDMS excludes specific models, while DAMA establishes a yearly frequency in its signal yield which could be due to particles we have still not even conceived.

So, why not focusing on the establishment of the signal ? I tried to build an opinion on the solidity of the DAMA result this morning by talking to a few people who attend the conference, and in particular with a young researcher who now works for CUORE and has been in the DAMA collaboration in the past. She in fact worked at the analysis of data, trying to interpret the modulation in photomultiplier counts with different density models of dark matter in the milky way halo as well as components from our close satellites, the magellanic clouds.

It looks like the phase of the cosine oscillation in signal yield, which is something I had doubts about, is indeed compatible with being zero for a time of the year compatible with June 2nd, which is the date when the Earth travels in the direction of the Sun’s motion in the galaxy (if one forgets about the inclination of the Earth orbit with respect to the Sun’s line of motion). Not perfectly matching June 2nd, but compatible with it.

Another issue is of course the one which was raised at yesterday’s talk: signal efficiency is steeply rising in the region between 2 and 6 keV where the modulated signal is observed. It turns out that the efficiency is not full but it is quite stable – stability is checked weekly – and the reason why it is not 100% is due to specific cuts that are made to exclude a background contamination. This, of course, might make the signal yield dependent on subtleties of the shape of the PMT response (the cuts are made on signal shape form factors), but I see no reason to doubt that the stability is well under control, although extraordinary claims require extraordinary evidence, and the latter implies going after the subtlest of possible non-exotic explanations.

One thing I would have done if I had designed the experiment myself would have been to prepare a mock-up of the active region, wrap it with black paper, and instrument with similar photomultipliers and the accessory set-up, the overpressure, the temperature controls. This would have allowed to have a real-time comparison between PMT counts from the real NaI detector and PMT counts in a fake one. It would dismiss any claim that PMTs are varying their response seasonally without appeal, but unfortunately we do not have such luxury available in the DAMA-LIBRA setup.

So the question remains. If the signal is strong and significant, and if it is not due to instrumental nuisances -I am sure about the former, less so about the latter- what is its source ? Can we get a model of dark matter particles which produce a similar flux ? It should not be too difficult. Maybe the ball is in the theorists’ court in fact. As for me, I keep cool. I still think there is no new particles to be found with these cunning but a bit overoptimistic endeavours.