Boiling away asymmetries: low-scale phase transitions, gravitational waves and leptogenesis

TL;DR

The study assesses how entropy injection from a late-time, supercooled first-order phase transition can dilute a preexisting baryon asymmetry produced by leptogenesis, focusing on a Type-I seesaw model with 2 right-handed neutrinos (RHN). It introduces two dilution channels, vacuum-decay dilution $d_{vac}$ and reheating dilution $d_{reh}$, with a total dilution factor $d = d_{vac} d_{reh}$, where $d_{reh} = (1+\alpha)^{3/4}$ and $d_{vac}$ is computed from the vacuum entropy change between $T_c$ and $T_{compl}$. The analysis solves quantum kinetic equations for ARS-like leptogenesis using AMIQS, scanning parameters $M \in [0.1,100]$ GeV, $\Delta M/M \in [10^{-15},0.1]$, $y \in [10^{-8},10^{-4}]$, and CP phases, and compares the produced $Y_B$ to the observed value $Y_B^{obs} \approx 8.65 \times 10^{-11}$. It finds that larger dilution shrinks the viable leptogenesis parameter space, with SGWB-best-fit dilutions potentially excluding leptogenesis in some NO/IO scenarios, thereby linking SGWB, FOPT, neutrino masses, and BAU; the framework is generalizable to other dilutions and models. Finally, the work suggests that future measurements of heavy neutral lepton properties could constrain allowable entropy injections, while a detected HNL signal would tightly bound the strength of any late-time FOPT.

Abstract

Leptogenesis is one of the most popular mechanisms to account for the observed baryon asymmetry of the Universe. A generic feature of leptogenesis is a large separation of scales between the epoch of baryon asymmetry production (sphaleron freeze-out at temperature $T \sim 130$ GeV) and the one where it affects the big bang nucleosynthesis processes (BBN at $T \sim 1$ MeV). Any entropy release between these two epochs would lead to a dilution of previously produced relics, such as the baryons. Motivated by the recent evidence of a stochastic gravitational waves background (SGWB) in the nHz frequency range, we consider the case of supercooled first-order phase transition and we study the impact of the induced entropy dilution on the leptogenesis parameter space. We employ the Type-I seesaw with 2 right-handed neutrinos as benchmark scenario, and demonstrate that the viable leptogenesis parameter space is significantly reduced. Interestingly, the values of dilution predicted by the SGWB best fit points in several first order phase transition scenarios would completely exclude the leptogenesis parameter space testable by future experiments, thus establishing a phenomenological interconnection between leptogenesis, SGWB, first order phase transition and neutrino mass generation. Our analysis can be generalised to different leptogenesis models and entropy dilution mechanisms.

Figures (2)

• Figure 1: Leptogenesis parameter space that produces the observed BAU for different values of dilution $d$ in the model in Eq. (\ref{['eq:lagrangian']}), for a normal (NO, left panel) and inverted (IO, right panel) ordering of light neutrino masses. Constraints from laboratory searches in the single flavour dominance limit are reported as blue ($|U_e|^2$), green ($|U_\mu|^2$) and red ($|U_\tau|^2$) dashed regions, and we exclude points that violate at least one of these. Below the See-Saw Limit, presented as solid region, the model cannot reproduce the light neutrino masses. The bound from BBN is given as a transparent grid, but no cut is performed on the points, to underline the fact that it is generally relaxed by the dilution of the HNL population, with the exact revised bounds depending on the specific dilution factor and reheating temperature values (see discussion in the main text). The expected sensitivities of DUNE, SHiP, MATHUSLA, HL-LHC and FCC-ee are shown as coloured dashed lines Izaguirre:2015pgaAntusch:2016ejdAntusch:2017pkqCurtin:2018mvbPascoli:2018hegSHiP:2018xqwBallett:2019bgdDrewes:2019fouGorbunov:2020rjxDrewes:2021nqrAbdullahi:2022jlv.
• Figure 2: Plot of $\Delta S(T,T_c)$ as defined in Eq. \ref{['Srel']} for $T$ in the range between completion ($T_{\text{compl}}$) and critical ($T_{\text{c}}$) temperatures for the BP1 in Costa:2025csj (see also Table \ref{['tab:d_tot']}). For reference we signal also the nucleation ($T_n$) and percolation ($T_p$) temperatures.