Monday 30 June 2008

New Old Dama

A week ago Graciela Gelmini gave a seminar about DAMA here at CERN. DAMA is trying to detect dark matter by observing the nuclear recoil when a dark matter particle interacts hits the ordinary matter. Unlike all other similar experiments, DAMA has been claiming that they actually see the signal. Recently, the DAMA-LIBRA collaboration (it was originally called DAMA-VIRGO, but it was renamed) presented new results whose statistical significance is over 8 sigma. DAMA's claim is however largely ignored in the community. The main reason is that, due to the specifics of the experimental set-up, the interpretation of the DAMA results is controversial. The other experiments use at least two independent probes to discriminate the signal from the background, for example, XENON measures the scintillation and ionization signals, while CDMS focuses on ionization and phonons. As DAMA is sensitive only to the ionization signal, they have no handle to reject the hits due to ordinary particles like electrons or neutrons. Instead, DAMA's claim is based on the observation a periodic annual dependence of the number of hits in their detector. The annual modulation can be attributed to dark matter because the flux of dark matter particles pervading our galaxy varies annualy due to the motion of the Earth around the Sun. The problem is that there are more things in heaven and earth that display an annual variation: four seasons, temperature, hormone level, etc., and it is possible that one of those is responsible for the modulation of the background. Although DAMA insists that they carefully checked and excluded all possible sources of the background, dark matter doesn't seem to be the most likely explanation. Moreover, other dark matter experiments have claimed to exclude the region of parameter space favored by DAMA. Last not least, DAMA has gained a bad reputation, because of their agressive presentation strategy.

There seems to be a trend, however, to take DAMA a bit more seriously, and recently I've seen quite a few new preprints on arXiv that had DAMA in the title. The important question is if the DAMA signal can be reconciled with the null results from other experiments. According to Graciela, this is not very difficult. The point is that DAMA has a lower threshold than most other experiments. Therefore a fairly low mass dark matter particle with a fairly large cross section could show up in DAMA but escape detection in other, in principle more sensitive experiments. Graciela had a paper a few years ago pointing out that a 5-9 GeV WIMP with a $10^{-4}$ pb cross-section can explain the DAMA signal. No dirty tricks are needed -- the conventional Maxwellian velocity distribution of dark matter and the spin-independent cross section will do the job. The newest experimental results still leave that window open. The current experimental constraints are more extensively discussed in the new paper by Petriello and Zurek, from which I stole the plot.

Numerology suggests that a 5 GeV WIMP might not be a bad idea after all: the ratio of baryon density to dark matter density is $\Omega_{DM}/\Omega_{B} \sim 5$, which might be the consequence of $m_{DM} \sim 5 m_p \sim 5 $ GeV. Models that predict such a relation do exist, see for example this old idea of David Kaplan. There is another dark matter experiment in Gran Sasso called CRESST that has a lower threshold than DAMA, so that it should be able to close the low-mass window once their statistics is improved. Yet I don't expect a conclusive resolution of the DAMA puzzle soon: DAMA was being excluded many times before, but it always managed to find a new window to slip back in ;-)