Could the cosmological recombination spectrum help us understand annihilating dark matter?

In this paper, we explore the potential effects of dark matter (DM) annihilations on the cosmological recombination spectrum. With this example, we want to demonstrate that the cosmological recombination spectrum in principle is sensitive to details related to possible extra energy release during recombination. We restrict ourselves to DM models which produce a negligible primordial distortion of the cosmic microwave background (CMB) energy spectrum (usually characterized as -and y-type distortions). However, since during the epoch of cosmological recombination (z 1000) a large fraction of the deposited energy can directly go into ionizations and excitations of neutral atoms, both the cosmological recombination spectrum and ionization history can still be affected significantly. We compute the modifications to the cosmological recombination spectrum using our multilevel H I and He I recombination code, showing that additional photons are created due to uncompensated loops of transitions which are induced by DM annihilations. As we illustrate here, the results depend on the detailed branching of the deposited energy into heating, ionizations and excitations. This dependence in principle should allow us to shed light on the nature of the underlying annihilating DM model (or more generally speaking, the mechanism leading to energy injection) when measuring the cosmological recombination spectrum. However, for current upper limits on the potential DM annihilation rate during recombination the cosmological recombination spectrum is only affected at the level of a few per cent. Nevertheless, we argue here that the cosmological recombination spectrum would provide another independent and very direct way of checking for the presence of sources of extra ionizing or exciting photons at high redshifts. This would open a new window to possible (non-standard) processes occurring before, during and between the three epochs of recombination.