From Dark Sectors to the Axion-Neutrino Connection

The Standard Model (SM) of particle physics provides a successful description of fundamental particles and their interactions but fails to explain phenomena such as neutrino oscillations, dark matter (DM), and the baryon asymmetry of the Universe. These clear signs of physics beyond the SM (BSM) motivate extensions introducing new particles and symmetries. Theoretical questions like the flavor puzzle – fermion masses and mixing – and the strong charge-conjugation and parity (CP) problem further guide BSM frameworks.

This thesis explores BSM scenarios based on two guiding principles: constructing unified frameworks that address multiple open problems in (astro)particle physics and cosmology, and emphasizing their experimental testability through detailed phenomenological analyses. This approach uncovers deep connections between seemingly disconnected sectors, offering a more complete view of fundamental physics.

We investigate dark sector models for the origin of neutrino masses, proposing a novel setup: the dark linear seesaw, which radiatively links neutrino mass generation to viable DM candidates and predicts charged lepton flavor violation. In these scenarios, dark sector particles can constitute weakly interacting massive particle (WIMP) DM, testable via direct detection experiments. We study spontaneous CP violation induced by a complex scalar singlet, acting as a common origin of low- and high-energy CP violating effects relevant for leptogenesis.

We also analyze a Nelson-Barr model that solves the strong CP problem, generates a realistic CKM matrix radiatively, and yields scalar WIMP DM. Additionally, we present unified axion frameworks in which a colored sector radiatively generates neutrino masses. These models predict distinctive axion couplings to photons and fermions and accommodate axion DM in both pre- and post-inflationary cosmologies. Finally, we explore minimal flavored Peccei-Quinn symmetries that link the flavor puzzle, neutrino masses and DM within a predictive and testable BSM framework.