The Effects of Dark Matter Decay and Annihilation on the High-Redshift 21 cm Background

TL;DR

This paper investigates how dark matter decay and annihilation during the dark ages could heat and ionize the intergalactic medium (IGM) and thereby modify the high-redshift 21 cm background. It introduces a simple, generic energy-deposition framework that distinguishes optically thick and optically thin regimes and computes the resulting mean IGM histories and 21 cm fluctuations, emphasizing the coupling to Lyα radiation. The main findings show that DM decay with lifetimes around $10^{24}$–$10^{27}$ s can produce order-unity changes in the global 21 cm signal and its power spectrum, with detectable signatures for energy injection rates above about $10^{-24}$ eV cm$^{-3}$ s$^{-1}$, while annihilation effects are generally smaller and occur at higher redshift. The 21 cm probe is therefore complementary to CMB constraints, as it directly probes the thermal and ionization history and can also boost H2 formation, potentially affecting early star formation and reionization.

Abstract

The radiation background produced by the 21 cm spin-flip transition of neutral hydrogen at high redshifts can be a pristine probe of fundamental physics and cosmology. At z~30-300, the intergalactic medium (IGM) is visible in 21 cm absorption against the cosmic microwave background (CMB), with a strength that depends on the thermal (and ionization) history of the IGM. Here we examine the constraints this background can place on dark matter decay and annihilation, which could heat and ionize the IGM through the production of high-energy particles. Using a simple model for dark matter decay, we show that, if the decay energy is immediately injected into the IGM, the 21 cm background can detect energy injection rates >10^{-24} eV cm^{-3} sec^{-1}. If all the dark matter is subject to decay, this allows us to constrain dark matter lifetimes <10^{27} sec. Such energy injection rates are much smaller than those typically probed by the CMB power spectra. The expected brightness temperature fluctuations at z~50 are a fraction of a mK and can vary from the standard calculation by up to an order of magnitude, although the difference can be significantly smaller if some of the decay products free stream to lower redshifts. For self-annihilating dark matter, the fluctuation amplitude can differ by a factor <2 from the standard calculation at z~50. Note also that, in contrast to the CMB, the 21 cm probe is sensitive to both the ionization fraction and the IGM temperature, in principle allowing better constraints on the decay process and heating history. We also show that strong IGM heating and ionization can lead to an enhanced H_2 abundance, which may affect the earliest generations of stars and galaxies.

Figures (8)

• Figure 1: IGM histories for long-lived dark matter. In each panel, the curves take $\xi_{-24}=1,\,0.1,\,10^{-2},\,10^{-3}$, and $0$, from top to bottom. (a): Ionization histories. (b): Thermal histories. Here the thin solid curve shows $T_\gamma$.
• Figure 2: 21 cm signals for long-lived dark matter. Curves take the same parameters as in Fig. \ref{['fig:therm-thick']}. (a): Fluctuation amplitude at $k=0.04 \hbox{Mpc$^{-1}$}$ (note that this scale is arbitrary). (b): Mean (sky-averaged) signal. The thin horizontal dotted line shows $\bar{\delta T}_b=0$.
• Figure 3: As Fig. \ref{['fig:signal-thick']}. The dotted curves take $\xi_{-24}=0$; the rest have $\xi_{-24}=0.1$. The solid curves show the net signal. The short-dashed curves set $\chi_\alpha=0$. The long-dashed curve (shown only in the top panel) assumes $\theta_u=0$.
• Figure 4: As Fig. \ref{['fig:therm-thick']}, except for energy injection in the transparency window. The bottom solid curve in each panel assumes no extra energy injection. The other curves take $\xi_{-24}=1$. The dotted, dot-dashed, short-dashed, and long-dashed curves take $\tau_{100}=1,\,10^{-1},\,10^{-2}$ and $10^{-3}$, respectively.
• Figure 5: 21 cm signals for long-lived dark matter with energy injection in the transparency window. Curves take the same parameters as in Fig. \ref{['fig:therm-thin']}. (a): Fluctuation amplitude at $k=0.04 \hbox{Mpc$^{-1}$}$ (note that this scale is arbitrary). (b): Mean (sky-averaged) signal.