Strong gravitational radiation from a simple dark matter model

TL;DR

This work investigates gravitational waves from a first-order phase transition in a minimal dark sector where dark matter consists of the massive gauge bosons of a dark SU(2)_D that communicates with the Standard Model via a Higgs portal. The relic density fixes the dark gauge coupling $g_D$ (and related scales), linking the GW signal to DM properties, and two production regimes are analyzed: standard freeze-out and radiatively induced (supercool) symmetry breaking with a classically scale-invariant potential. Gravitational-wave spectra are computed from bubble nucleation via $S_3/T$ and templates that incorporate sound waves and turbulence, while astrophysical WD-WD foregrounds are included to assess detectability by LISA, BBO, and the Einstein Telescope. The results show that strong phase transitions in the dark sector can generate detectable GWs, especially in the supercool DM regime, and that future GW observatories can probe significant regions of the model’s parameter space, potentially pointing to TeV-scale new physics if a signal is observed.

Abstract

A rather minimal possibility is that dark matter consists of the gauge bosons of a spontaneously broken symmetry. Here we explore the possibility of detecting the gravitational waves produced by the phase transition associated with such breaking. Concretely, we focus on the scenario based on an $SU(2)_D$ group and argue that it is a case study for the sensitivity of future gravitational wave observatories to phase transitions associated with dark matter. This is because there are few parameters and those fixing the relic density also determine the effective potential establishing the strength of the phase transition. Particularly promising for LISA and even the Einstein Telescope is the super-cool dark matter regime, with DM masses above $\mathcal{O}$(100) TeV, for which we find that the gravitational wave signal is notably strong. In our analysis, we include the effect of astrophysical foregrounds, which are often ignored in the context of phase transitions.

Figures (12)

• Figure 1: The dominant DM annihilation channels for $m_{A} \gg m_{h_D}$ and $\theta \ll 1$.
• Figure 2: Mixing angles excluded by Xenon1T Aprile:2017iyp (shaded area) together with the projected sensitivity from LZ Mount:2017qzi (solid line). Here we assume $m_A\gtrsim \unit[0.1]{TeV}$. Left: assuming the relic density is produced via the freeze-out mechanism. Right: for the classically scale invariant potential, including contours for some choices of the DM mass.
• Figure 3: An example of the gravitational wave spectrum when the symmetry breaking occurs at tree level, together with the white-dwarf white-dwarf binary foreground, and LISA and BBO sensitivity curves. Here we assume $v_{w}=1$.
• Figure 4: The parameter space returning a significant BBO or LISA signal, SNR $>5$, when the symmetry breaking occurs at tree level (standard potential). For LISA we assume $v_{w}=1$ as the BM criterion is fulfilled roughly in this region. For BBO we show contours assuming $v_{w} = 0.1$ and $1$. Only the strongest transitions, close to the point at which no transition occurs at all, can be probed by LISA in this case. In contrast BBO can probe a substantial fraction of the parameter space with a strong first order phase transition. Here we show the SNR with no foreground. If the foreground is included the BBO area remains practically unchanged, while the already small LISA area is approximately halved.
• Figure 5: Similar to Fig. \ref{['fig:largevev']}, but with contours of the peak frequency and the corresponding GW energy density assuming $v_{w}=1$.