Ultra-relativistic freeze-out: a bridge from WIMPs to FIMPs (2511.02117v1)
Abstract: We re-examine the case for dark matter (DM) produced by ultra-relativistic freeze-out (UFO). UFO is the mechanism by which Standard Model (SM) neutrinos decouple from the radiation bath in the early universe at a temperature $T_{d} \approx 1$ MeV. This corresponds to chemical freeze-out without Boltzmann suppression, such that the freeze-out (decoupling) temperature $T_{d}$ is much greater than $m_{\nu}$ and the neutrinos are therefore ultra-relativistic at freeze-out. While UFO has historically been rejected as a viable mechanism for DM production due to its association with hot DM and the accompanying incompatibility with $\Lambda$CDM, we show that when the approximation of instantaneous reheating after inflation is lifted, UFO can produce cold DM and account for the entire observed relic density in large regions of parameter space. In fact, DM with masses ranging from sub-eV to PeV scales can undergo UFO and be cold before structure formation, given only a simple perturbative, post-inflationary reheating period prior to radiation domination. For some interactions, such as a contact interaction between the Higgs and DM scalars, there is a seamless transition between the WIMP and FIMP regimes which excludes UFO. However, for many other interactions, such as SM fermions producing fermionic DM via a heavy scalar or vector mediator, the WIMP to FIMP transition occurs \textit{necessarily} via a large intermediate region corresponding to UFO. We characterize the general features of UFO in this paper, while we supply a more detailed analysis in a companion paper. We find that UFO during reheating can produce the correct relic density ($\Omega_{\chi}h2 = 0.12$) for DM masses spanning about 13 orders of magnitude, reheating temperatures spanning 17 orders of magnitude, and beyond the Standard Model (BSM) effective interaction scales spanning 11 orders of magnitude.
Sponsor
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.