Towards precise baryogenesis in the 2HDM$+a$

TL;DR

The paper investigates baryogenesis in the 2HDM+$a$ framework, leveraging transient CP violation from a nonzero vev of the singlet pseudoscalar to evade EDM constraints. It advances the BAU computation by using the full one-loop finite-temperature effective potential, solving bubble nucleation, and applying transport equations with velocity bounds to connect cosmology to collider and flavour phenomenology. The results show a marked reduction of the predicted BAU relative to prior estimates, requiring larger mixing between the singlet and 2HDM pseudoscalars and favoring collider-accessible regions, while current LHC and flavour constraints further restrict the viable parameter space. Overall, the work sets a tighter, more testable target for EW baryogenesis in 2HDM+$a$ and outlines the experimental prospects for confirming or ruling out this mechanism.

Abstract

We perform a detailed investigation of the viable baryogenesis parameter space of a non-minimal Higgs sector consisting of two Higgs doublets and a singlet pseudoscalar (2HDM$+a$). In such a model, an early Universe period of transient CP violation may occur, driven by a nonvanishing vacuum expectation value of the CP-odd scalar $a$. This naturally avoids the stringent electric dipole moment experimental constraints on beyond-the-Standard-Model sources of CP violation. We provide a state-of-art computation of the baryon asymmetry, providing several important improvements over existing baryogenesis computations for this model. We show that the required thermal history and successful baryogenesis lead to a predictive scenario, testable in the near future by a combination of LHC searches and low-energy flavour measurements. Our improved predictions for the baryon asymmetry find that it is rather suppressed compared to earlier predictions, requiring larger mixing between the singlet and 2HDM pseudoscalars and hence leading to a more easily testable model at colliders.

Figures (12)

• Figure 1: Iso-contours of the effective potential at the critical temperature in the $\rho_1$-$a$ plane for the Hartree (left) and full one-loop effective potential (right), with $\rho_2$ and $\delta$ fixed at their value in the EW minimum. The red curve traces along the valley (the MEP, see text for details) connecting the CPV and EW minima. The figures below the potential landscape show how $\rho_2$, $\varphi$ and $V - V_{\mathrm{min}}$ vary along this valley.
• Figure 2: One-loop bounce profiles at differing values of supercooling $T_n/T_c$ for the same benchmark point in \ref{['fig:crit_temp_figure']}. The solid red curves are the configurations which satisfy the nucleation condition, $S_3/T \approx 140$.
• Figure 3: Sample phase diagrams for the 2HDM$+a$, showing the nucleation temperature, $T_n$ - at fixed mixing angles $s_{\theta}$ - as a function of the singlet mass $m_{a_1}$ for BP1 (left) and BP2 (right). Lines are dashed where $\xi_n < 1$.
• Figure 4: Left panels: The normalised BAU as function of wall velocity for the same benchmark point in \ref{['fig:crit_temp_figure']}. The lower (ballistic) and upper (LTE) bounds for the wall velocity are indicated by dashed blue lines. Right panels: The variation of the chemical potential $\mu$ and plasma velocity $u$ of the relevant species with respect to the distance from the planar bubble wall ($z=0$) at a wall velocity $v_w = 0.4$. In our convention, $z<0$ is the EW phase and $z>0$ is the CPV phase.
• Figure 5: A comparison of the Hartree and one-loop potential with respect to the singlet pseudoscalar mass $m_{a_1}$ and mixing angle $s_{\theta}$ for the first benchmark point in Figure 3 from Huber:2022ndk. The plots on the first row display the phase diagram and whether the transition is one-step (blue) or two-step, and, if the latter is true, whether the second transition is first (green) or second order (red). The yellow region features a potential that is unbounded from below. The second to fourth row demonstrates how key quantities in estimating the BAU vary and the final row shows how the normalised BAU estimate \ref{['bau estimate']} varies, with respect to $m_{a_1}$ and $s_{\theta}$ for the desired FOPTs.