Freeze-In of radiative keV-scale neutrino dark matter from a new $\text{U}(1)_\text{B-L}$

TL;DR

This paper proposes a new model where keV-scale neutrino dark matter is produced via freeze-in in a U(1)_B-L gauge symmetry framework, linking neutrino masses and dark matter with specific cosmological and particle physics constraints.

Contribution

It extends the Dirac Scotogenic model with a U(1)_B-L symmetry and introduces a novel dark matter production mechanism compatible with cosmological bounds.

Findings

Dark matter produced via freeze-in with low reheating temperature.

Predicted ΔN_eff below future experimental sensitivity.

Model excludes detectable ΔN_eff signals, constraining future observations.

Abstract

We extend the Dirac Scotogenic model with the aim of realizing neutrino masses together with the mass of a keV-scale dark matter (DM) candidate via the same one-loop topology. Two of the Standard Model (SM) neutrinos become massive Dirac fermions while the third one remains massless. Our particle content is motivated by an anomaly free $\text{U}(1)_\text{B-L}$ gauge symmetry with exotic irrational charges and we need to enforce an additional $\mathcal{Z}_5$ symmetry. The dark matter candidate does not mix with the active neutrinos and does not have any decay modes to SM particles. DM is produced together with dark radiation in the form of right handed neutrinos via out of equilibrium annihilations of the SM fermions mediated by the heavy B-L gauge boson. In order to avoid DM over-production from Higgs decays and to comply with Lyman-$\alpha$ bounds we work in a low temperature reheating scenario with $4\;\text{MeV}\lesssim T_\text{RH}\lesssim 5\;\text{GeV}$. Our setup predicts a contribution to $\Delta N_\text{eff.}$ that decreases for larger DM masses and is below the sensitivity of upcoming precision measurements such as CMB-S4. A future observation of a signal with $\Delta N_\text{eff.}\gtrsim 0.012$ would exclude our scenario. We further sketch how inflation, reheating and Affleck-Dine baryogenesis can also be potentially realized in this unified framework.