At the last Cosmo Coffee, Celine Boehm was discussing the current status of the 511 keV gamma-ray line from the Milky Way center, in light of the recent results from the INTEGRAL satellite. Photons carrying 511 keV energy arise from annihilation of positrons and electrons that are more less at rest. It is not clear which mechanism is responsible for injecting enough positrons into the interstellar medium of the galactic bulge. One hypothesis is that the positrons are scattered remnants of the ILC (Interplanetary Linear Collider) - an unfinished project of a technologically advanced civilization from that region. More contrived explanations involve black holes, radioactive nuclei from supernovae, pulsars and other fluffy toys.
Yet another, quite exciting possibility is that the positrons come from annihilation of dark matter. At first sight this seems quite natural. We are pretty confident that a lot of dark matter is present in the galactic center and its distribution should be roughly spherical, something which agreed quite well with the earlier INTEGRAL observations. Furthermore, it is likely that dark matter can annihilate into ordinary matter so that its present abundance is a thermal relic. But a dark matter particle that could explain the INTEGRAL signal must be quite peculiar. In particular, it cannot be a WIMP: its mass should be in the MeV range. If it were heavier, annihilation would yield too energetic positrons and the 511 keV line would not stand so prominently over the continuum gamma-ray radiation.
On the microscopic level, MeV dark matter can be realized as a scalar particle (so as to avoid the Lee-Weinberg bound) annihilating via an exotic heavy fermion exchange, or via an exotic heavy gauge boson. See the artist's view above. In fact, both diagrams are needed to explain the INTEGRAL signal and, at the same time, derive the correct dark matter abundance from the thermal relic density computation. The first diagram leads to the cross section that goes to a constant at small velocities and this one is relevant for the annihilation of dark matter today. The second leads to the cross section that goes like $\sigma \sim b v^2$, and it's supposed to dominate in the early hot universe.
Recently, the INTEGRAL satellite announced new results which show an asymmetry of emission (by a factor of two) with respect to the central axis of the galaxy. Morevover, the asymmetry seems to be correlated with the distribution of low mass X-ray binaries (LMXB) - systems including a neutron star or a black hole that accretes matter from its companion. LMXBs have long been one of the suspects in the positron case. The general feeling is that the new results make the dark matter explanation unlikely; Julianne on CV readily flushed MeV dark matter down her toilet. Celine, on the other hand, is more reluctant to push the button (she is, in fact, responsible for much of the stuff being thrown into the toilet). She argued that: 1) We still don't know a conventional astrophysical explanation that could account for enough positron emission, 2) While part of the gamma-ray emission maybe due to boring astrophysics, there's still a large, roughly spherical component that could be due to MeV dark matter. In fact, if only a part of the emission is assigned to dark matter, the microscopic models can more readily satisfy constraints from other experiments, for example, from measurements of the electron anomalous magnetic moment.
I guess it's fair to say that, at the moment, a conventional astrophysical explanation seems far more likely. My faith in that signal is further diminished by the fact that astrophysics provides us with too many excesses (EGRET, for example), each one having its dark matter particle that's supposed to explain it. Finally, MeV dark matter is hard to accommodate in our favourite Beyond the Standard Model schemes. But this last argument should always be taken with a grain of salt, since all existing BSM models suck. It's better to keep our eyes open...