Sub-keV dark matter can strongly ionize molecular clouds

Abstract

We show that the ionization of dense molecular clouds can be used to set strong constraints on dark matter models producing UV/X-ray photons in their annihilation or decay. We place robust and competitive constraints on various dark matter models, such as axion-like particles, scalars and sterile neutrinos, for masses between $\sim30$~eV and $10$~keV, and project forecasts to illustrate the potential of this target. We discuss how these constraints can be significantly improved by considering a more refined sample of molecular clouds near the Galactic Center and above the Galactic plane, a detailed modeling of the cosmic-ray ionization contribution and, potentially, a more refined analysis of the gas density in clouds through dust extinction maps. Thus, ionization of molecular clouds emerges as one of the most powerful tools for probing sub-keV dark matter.

Figures (8)

• Figure 1: Comparison of previous bounds on ALPs in the $g_{a\gamma}-m_{a}$ plane with the bounds derived in this work. The constraints for two well characterized clouds are shown: L1551 (in purple) and the DRAGON cloud (in magenta), where the hatch region denotes its uncertainty. In addition, we show an optimistic forecast based on G1.4-1.8+87 (shown as a brown dashed line to distinguish it from the robust limits). Other constraints come from DESI observations Wang:2023imi (olive), observations of dwarf galaxies by Hubble Space Telescope Todarello:2024qci (HST -- in lime), from reionization Cadamuro:2011fd (orange), CMB spectral distorsions Bolliet:2020ofj (teal), X-ray light Essig:2013goa (blue), heating of gas in the LeoT dwarf at T$_{gas} = 7552$ k Wadekar:2019mpc (dodgerblue) and X-ray observations by XMM-Newton Gewering-Peine:2016yojFoster:2021ngm (yellow).
• Figure 2: Model-independent constraints on the lifetime (left panel) and annihilation rate (right panel) for DM particles producing two photons. We show our constraints from L1551 in purple, from the DRAGON cloud in magenta and the optimistic forecast based on G1.4-1.8+87 as a brown dashed line. We compare with previous results from CMB spectral distortions (teal) Bolliet:2020ofj, LeoT (dodgerblue) and G33.4-8 (green) gas heating at T$_{gas} = 7552$ K Wadekar:2019mpc, and XMM-Newton diffuse Galactic observations (yellow) Essig:2013goaGewering-Peine:2016yojBoyarsky:2018tvuFoster:2021ngm.
• Figure S1: left panel: Product of DM-induced photon flux times ionization cross section, J$_{\chi} \sigma_{H_2}$, for the decay into two photons. This serves as a direct indicator of the ionization rate, for several DM masses, and assumes $\rho_{\chi} =0.4$ GeV/cm$^{3}$, modeling a local MC like L1551. Right panel: Ionization rate as a function of the DM mass, induced by two- and three-body decays (solid lines) with a rate $\Gamma=10^{-25}$ s$^{-1}$ or annihilation (dot-dashed line) for an averaged cross section $\langle\sigma v\rangle=10^{-30}~{\rm cm}^{3}~{\rm s}^{-1}$. These calculations refer to the MC L1551, whose $2\sigma$ upper limit in ionization rate is shown by the purple dashed line.
• Figure S2: Absorption probability of photons emitted by DM as a function of their energy, for each of the clouds studied here.
• Figure S3: Constraints on annihilation rate $\langle \sigma v \rangle$ (left panel) and lifetime $\tau$ (right panel), for the DRAGON cloud (in magenta) and the forecast for G1.4-1.8+87 (in maroon). In every case, the bands indicate the difference between the constraint obtained when assuming a Burkert and a Moore profile, and the solid lines represent the constraint obtained assumed an NFW profile.