Genesis of baryon and dark matter asymmetries through ultraviolet scattering freeze-in

Abstract

We introduce a new mechanism for the simultaneous generation of baryon and dark matter asymmetries through ultraviolet-dominated freeze-in scatterings. The mechanism relies on heavy Majorana neutrinos that connect the visible Standard Model sector to a dark sector through the neutrino portal. Following reheating of the visible sector to a temperature well below the heavy neutrino masses, we show that 2-to-2 scattering processes can populate the dark sector and generate both baryon and dark matter asymmetries. In some parameter regions, the dominant source of baryon asymmetry can be charge transfer from the dark sector, a process we call dark wash-in. We also demonstrate that annihilation of the dark matter to massless states within the dark sector can deplete the symmetric population without destroying the net baryon charge to leave only an asymmetric dark matter abundance today. Depending on the specific model parameters, the observed baryon and dark matter abundances can be attained with heavy neutrino masses $M_N \gtrsim 10^{10}$ GeV, and dark matter masses in the range 0.1 GeV $\lesssim m_χ\lesssim 10^3$ GeV if the dark matter relic abundance is mainly asymmetric and even lower masses if it is symmetric.

Figures (8)

• Figure 1: Schematic of the cosmological history and the evolution of different abundances $Y_Q$ in our setup. Temperature increases from right to left and time runs from left to right. The universe reheats to temperature $T_{\rm RH}$ below the heavy neutrino masses and only populates the SM. The dark sector is immediately populated through UV freeze-in. The asymmetries in the two sectors are dominantly generated at this stage as well. At lower temperatures, the $\phi$ field decays to the DM $\chi$, which subsequently freezes-out into the dark bath $\eta$. If the frozen-out symmetric abundance of $\chi$ (denoted by $Y_{\Sigma \chi}$) is not too far from the asymmetric yield $Y_{\Delta \chi}$, the freeze-out process produces mostly asymmetric dark matter ($Y_{\Delta \chi} \sim Y_{\Sigma \chi}$). The generated SM lepton asymmetry is partially converted into a baryon asymmetry by SM sphalerons throughout this timeline; the conversion stops after the electroweak symmetry breaking.
• Figure 2: Diagrams contributing to asymmetric squared matrix elements at leading non-trivial order. The tree diagrams in the top left (middle and right) apply to reactions that conserve (violate) $L+L_x$ such as $\ell_i H\to \chi \phi$ ($\ell_i H\to \chi^\dagger \phi^\dagger$) and have an even (odd) number of Majorana neutrino mass insertions. The $u$-channel tree-level diagram only arises for $\ell_i H\to \ell_j^\dagger H^\dagger$, $\chi\phi\to\chi^\dagger\phi^\dagger$, and their conjugates. The loop diagrams in the left (right) column apply to processes that conserve (violate) $L+L_x$ and involve two (one) Majorana mass insertion(s).
• Figure 3: Evolution of the dark temperature ratio $\xi = T_x/T$ (left), and charge densities (right) of $|Y_{B-L}|$ (solid) and $|Y_{L_x}|$ (dashed) as functions of the inverse temperature relative to reheating for several values of $|\lambda_1|$. Other parameters are specified in the text, with $M_1=10^{10} \,\text{GeV}$, phase benchmark B1, and $|y_1|$ as large as possible subject to the upper bound on neutrino masses. The dotted black line corresponds to the $|Y_{B-L}|$ value required for the observed baryon asymmetry today. The evolution in the left plot is generic to UV freeze-in models and shows that for the chosen parameters the two sectors do not thermalize ($\xi < 1$). The right plot highlights the possibility of dark wash-in at larger $\lambda_1$ values, whereby some of the asymmetry in the dark sector is transferred to the visible sector via wash-out reactions.
• Figure 4: Evolution of the charge densities $|Y_{B-L}|$ (solid) and $|Y_{L_x}|$ (dashed) as a function of the inverse temperature relative to reheating. In the left panel we show curves for $M_1=10^{8,10,12,14} \,\text{GeV}$ and phases as in benchmark B1, see Eq. \ref{['eq:phasebench']}. In the right panel we show curves for phase benchmarks B1, B2, B3, B4 with $M_1=10^{10} \,\text{GeV}$. For both panels, we also set $x_{\rm RH}=30$, $|\lambda_1|=1$, and $|y_1|$ as large as possible subject to the neutrino mass bound. The horizontal dotted black line corresponds to the $|Y_{B-L}|$ value required for the observed baryon asymmetry today. The left panel shows that increasing $M_1$ reduces wash-out in the dark sector and increases the asymmetries in both sectors. The right panel shows that generic benchmarks B1 and B2, as well as benchmark B3 with zero visible sector source terms, give rise to comparable asymmetries via the dark wash-in mechanism. In benchmark B4 with no dark source term, there is no dark wash-in, but instead a transfer of asymmetry from the visible to the dark sector.
• Figure 5: Late time charge densities $Y_B$ (left) and $Y_{\Delta\chi}$ (right) as functions of $M_1$ and $|\lambda_1|$ obtained in the scenario with maximal lepton coupling $|y_1|$, phase benchmark B1, and $x_{\rm RH}=30$. The shaded regions in the lower left of both plots fail to produce the observed baryon density and are excluded. The dashed white line marks the boundary above which the strong dark wash-out condition of Eq. \ref{['eq:washxx']} is met. The left panel shows that in our setup we can produce the full baryon asymmetry for a wide range of parameters. The right panel demonstrates that both asymmetries tend to be similar in magnitude except in the strong dark wash-out regime where the dark charge can be orders of magnitude lower.