Probing displaced (dark)photons from low reheating freeze-in at the LHC

TL;DR

The paper studies a dark-sector extension with a dark photon DM candidate and a long-lived pseudoscalar mediator, focusing on freeze-in production at low reheating temperature $T_{ m RH}$. It derives coupled Boltzmann equations for the yields and shows how Higgs-portal and dark-photon portals shape DM production, with cosmological constraints (BBN and warmness) restricting viable regions. The collider angle leverages displaced-photon searches to constrain the Higgs portal, translating into bounds on $_{HS}$ and ruling out a thermal mediator for the plausible $T_{ m RH}$ window, notably around $T_{ m RH} m\,GeV$. Overall, the work demonstrates a testable, non-thermal DM scenario where LLP-induced displaced photon signatures at the LHC provide meaningful probes of the dark-sector couplings and reheating history.

Abstract

We extend the Standard Model (SM) by introducing a $U(1)'$ gauge boson and a real pseudo-scalar field, both odd under a $\mathbb{Z}_2$ symmetry. The resulting low-energy spectrum consists of a stable vector as the dark matter candidate, and a pseudo-scalar mediator, which interacts with the SM via a Higgs portal coupling and a dimension-five portal connecting it to both the dark and visible photons. We explore the freeze-in of both particles at low reheating temperature, finding a rich yield evolution dynamics in the early Universe. This setup brings a consistent dark matter scenario in which the dark photon relic abundance is generated through freeze-in at low reheating temperatures. In addition to its cosmological viability, the model can be tested at the LHC: Higgs bosons can decay into dark photons and displaced visible photons via the long-lived mediator. These signatures allow us to constrain the Higgs portal coupling using recent searches for non-pointing photons and limits on invisible or undetected Higgs decays. We derive meaningful constraints on the dark matter parameter space, in particular excluding a thermalized mediator in the region compatible with the observed relic abundance.

Figures (9)

• Figure 1: Relevant parameter space to recast displaced photon searches. Cyan, red and orange curves correspond to decay lengths of 0.01, 1, and 100 m, respectively. Solid (dashed) lines show $m_{\gamma'}=0$ ($m_{\gamma'}=m_{\phi}-15$) GeV.
• Figure 2: Leading tree-level processes contributing to the production of the dark states, and the (inverse) decay relevant for freeze-in mechanism at low $T_{\rm RH}$.
• Figure 3: Yield evolution of $\phi$ (gray lines) and $\gamma'$ (black lines) starting from $T_{\rm RH} = 2~\mathrm{GeV}$. In both plots, we fix $m_{\gamma'} = 1~\mathrm{GeV}$ and $m_\phi = 30~\mathrm{GeV}$. The red and orange solid lines correspond to the equilibrium yields, while the red dotted line in the right panel represents the combination $Y_{\gamma'} \, Y_{\phi,\mathrm{eq}} / Y_{\gamma',\mathrm{eq}}$ for the case $\lambda_{HS} = 0.1$. The red dotted line illustrates the quasi-static equilibrium Hall:2009bx, where $\phi$ participates via the process $\phi \leftrightarrow \gamma'\gamma$ before full chemical decoupling.
• Figure 4: Random scan points that yield the correct relic abundance for $m_{\gamma'} = 1$ GeV. Orange and black points indicate the regions where $\phi$ does not thermalize and where it does, respectively. Further details on the scan are provided in the text.
• Figure 5: Portal couplings as a function of the reheating temperature for $m_\phi = 40$ GeV (red) and 50 GeV (black), with $m_{\gamma'} = 1$ GeV. All curves reproduce the correct dark photon relic abundance. (Left) Solid and dashed lines use $\lambda_{ HS} = 10^{-3}$ and $10^{-1}$, respectively. (Right) Solid and dashed lines use $g_D = 10^{-9}$ GeV$^{-1}$ and $10^{-11}$ GeV$^{-1}$, respectively.