The Four Basic Ways of Creating Dark Matter Through a Portal

TL;DR

The paper analyzes DM production from SM particles through two portal mechanisms: kinetic mixing with a hidden U(1)' and the Higgs portal. It identifies four fundamental regimes—freeze-in, reannihilation, hidden-sector freeze-out, and connector freeze-out—and shows that their interplay yields a universal Mesa-shaped relic-density phase diagram, largely determined by the connector strength and the hidden-sector coupling. For kinetic mixing, direct detection can probe freeze-in, while reannihilation and HS freeze-out have distinct phenomenology and cosmological constraints; for the Higgs portal, the same phase structure applies but with mediator-mass effects altering the production channels and testability. The work provides analytic scaling relations and discusses cosmological, astrophysical, and direct-detection constraints, highlighting how a DM hidden sector created from SM particles could be confronted with data across multiple experiments and observations.

Abstract

We consider the possibility that along the thermal history of the Universe, dark matter (DM) would have been created from Standard Model particles, either through a kinetic mixing portal to an extra U(1) gauge field, or through the Higgs portal. Depending solely on the DM particle mass, on the portal and on the DM hidden sector interaction, we show how the observed DM relic density can be obtained. There are four possible freeze-in/reannihilation/freeze-out regimes, which together result in a simple characteristic relic density phase diagram, with the shape of a "Mesa". In the case of the kinetic mixing portal, we show that, unlike other freeze-in scenarios discussed in the literature, the freeze-in regime can be probed by forthcoming DM direct detection experiments. These results are well representative {of} any scenario where a DM hidden sector would be created out of the Standard Model {sector}.

Figures (18)

• Figure 1: Processes that are relevant for the production of DM and thermalization of the hidden sector through kinetic mixing. We work in the basis in which the DM is millicharged so that the hidden photons, $\gamma^\prime$, only couple to DM and not the SM degrees of freedom (see text).
• Figure 2: Evolution of the ratio of the visible ($\rho$) and hidden ($\rho'$) sectors energy densities, for a range of connector parameter, $\kappa=10^{-6,-7,-8,-9,-10,-11}$ (from up to down), and for various DM masses.
• Figure 3: Phase diagrams for the kinetic mixing portal: contours of $Y_{DM}$ as a function of $\kappa$ and $\alpha'$ for $m_{DM}=m_e,\,0.1\,\hbox{GeV},\,10\,\hbox{GeV},\,1\,\hbox{TeV}$. The dashed thick line gives $\Omega_{DM}h^2=0.11$, or in other words $Y_{DM} m_{DM} =4.09\cdot10^{-10}$ GeV. We have drawn the "transition lines" delimiting the 4 phases. Phases I, II, III, IV correspond to the freeze-in, reannihilation, hidden sector freeze-out and connector freeze-out phases respectively. There are transition lines between I and II, II and III, III and IV and I and IV. The solid black line corresponds to $\epsilon={1/\sqrt{4 \pi}}$. Below this line the connector interaction is not expected to be perturbative. Note that, for $m_{DM}=10$GeV, the $Z$ boson leads to a "Mesa" phase diagram that is slightly more complex, but analogous to the one obtained in the case of the Higgs portal (see Section 6 and Appendix D).
• Figure 4: Values of $\kappa$ that give the observed relic density through freeze-in ($\alpha'$ is assumed to be negligible). The continuous line corresponds to the contribution of both the $\gamma$ and $Z$ channels, the dashed line is only for the $\gamma$.
• Figure 5: DM relic abundance $Y_{DM}$ as a function of the connector parameter $\kappa$ for different DM masses $m_{DM}$ and values of the hidden sector interaction, $\log_{10}(\alpha'/\alpha)=1,-1,-3,-5,-7$ (bottom-up).