Dark Matter Freeze-In and Small-Scale Observables: Novel Mass Bounds and Viable Particle Candidates

TL;DR

The paper investigates how dark matter produced via freeze-in can imprint small-scale structure, and develops a model-independent method to bound freeze-in candidates by mapping their quasi-thermal phase-space distributions to warm dark matter using the second moment $\sigma_q$ and a production scale $T_P$. It derives lower bounds on the dark matter mass for three FI production channels (two-body decays, scatterings, three-body decays) and applies the framework to concrete models, including a Higgs-portal scalar, sterile neutrinos, axion-like particles, and the dark photon portal. The results show that FI scenarios can produce observable suppression of small-scale power consistent with current constraints from MW satellites, JWST lensing, and the Lyman-$\alpha$ forest, often yielding competitive or stronger bounds than standard WDM. The approach is computationally efficient, cross-validates with explicit microscopic realizations, and provides a versatile tool for exploring UV-dominated or multi-component freeze-in scenarios in future work.

Abstract

The suppression of cosmological structure at small scales is a key signature of dark matter (DM) produced via freeze-in in the low-mass regime. We present a comprehensive analysis of its impact, incorporating recent constraints from Milky Way satellite counts, strong gravitational lensing with JWST data, and the Lyman-$α$ forest. We adopt a general strategy to translate existing warm dark matter (WDM) bounds into lower mass limits for a broad class of DM candidates characterized by quasi-thermal phase space distributions. The benefits of this approach include computational efficiency and the ability to explore a wide range of models. We derive model-independent bounds for DM produced via two-body decays, scatterings, and three-body decays, and apply the framework to concrete scenarios such as the Higgs portal, sterile neutrinos, axion-like particles, and the dark photon portal. Results from specific models confirm the validity of the model-independent analysis.

Figures (7)

• Figure 1: The constraints on $m_{\rm WDM}$ at 95% CL, as a function of the number of observable satellite galaxies derived following the framework of Ref. Dekker:2021scf. The dotted lines show the limits adopted in this work.
• Figure 2: WDM mass bounds from small-scale observations: the blue identifies the stringent bound from MW satellites; the orange from strong lensing due to JWST observations; the green from Lyman-$\alpha$ forests; the red the conservative bound from MW satellites; the violet from stellar streams; the brown from UV luminosity. See Sec. \ref{['sec:WDM']} for details.
• Figure 3: Mass bounds on DM with a quasi-thermal PSD in the $(\sigma_q, T_P)$ plane. The dimensionless second moment $\sigma_q$ is defined in Eq. \ref{['eq:sigmaq']}, while the temperature $T_P$ appears in the definition of the comoving momentum, Eq. \ref{['eq:qcom']}. Contours show the minimum allowed DM mass, obtained via the rescaling in Eq. \ref{['eq:warmness_map']} using $m_{\rm WDM}^{\rm min} = 6.8\,{\rm keV}$ as the reference value.
• Figure 4: Lower mass bounds on freeze-in produced DM from: two-body decays (left), binary scatterings (center), and three-body decays (right). The $(\sigma_q, m_1)$ plane is the same as in Fig. \ref{['fig:general_bounds']}, with $T_P$ replaced by the mass of the heaviest particle involved in production, $m_1$. Solid black contours reproduce those of Fig. \ref{['fig:general_bounds']}, corresponding to the reference WDM bound $m^{\rm min}_{\rm WDM} = 6.8$ keV. For each channel, we explore different mass spectra, as indicated in the legend, by varying $m_2/m_1 = \{0, 0.25, 0.5, 0.75, 0.9\}$. Whenever a third bath particle is involved, we always set $m_3 = 0$. Each choice for the mass spectrum identifies a point in the plane; the black contour intersecting the point gives the associated mass bound. Vertical dashed lines show analytically computed values of $\sigma_q$ (see App. \ref{['app:Cs']}), and the thick gray line marks the WDM benchmark $\sigma_q^{\rm WDM} \simeq 3.6$.
• Figure 5: Lower bounds on the DM mass $m_{\chi}^{\rm min}$ for the three production mechanisms considered in this work: two body decays (left), binary scatterings (center), and three body decays (right). The three lines in each panel correspond to the reference WDM mass bounds shown in Fig. \ref{['fig:wdm-consraints']}, using the same color scheme. Matrix elements are taken to be constant and adjusted to reproduce the observed DM relic abundance. The variable $m_1$ denotes the mass of the heaviest particle participating in the process, while ${\cal B}_2$ (and ${\cal B}_3$ when present) are assumed to be massless. The DM mass bound is shown as a function of $m_1$.