Collapsing domain wall networks: impact on pulsar timing arrays and primordial black holes

Unstable domain wall (DW) networks in the early universe are cosmologically viable and can emit a large amount of gravitational waves (GW) before annihilating. As such, they provide an interpretation for the recent signal reported by Pulsar Timing Array (PTA) collaborations. A related important question is whether such a scenario also leads to significant production of Primordial Black Holes (PBH). We investigate both GW and PBH production using 3D numerical simulations in an expanding background, with box sizes up to N = 3240, including the annihilation phase. We find that: i) the network decays exponentially, i.e. the false vacuum volume drops as ∼ exp(-η3), with η the conformal time; ii) the GW spectrum is larger than traditional estimates by more than one order of magnitude, due to a delay between DW annihilation and the sourcing of GWs. We then present a novel semi-analytical method to estimate the PBH abundances: rare false vacuum pockets of super-Hubble size collapse to PBHs if their energy density becomes comparable to the background when they cross the Hubble scale. Smaller (but more abundant) pockets will instead collapse only if they are close to spherical. This introduces very large uncertainties in the final PBH abundance. The first phenomenological implication is that the DW interpretation of the PTA signal is compatible with observational constraints on PBHs, within the uncertainties. Second, in a different parameter region, the dark matter can be entirely in the form of asteroid-mass PBHs from the DW collapse. Remarkably, this would also lead to a GW background in the observable range of LIGO-Virgo-KAGRA and future interferometers, such as LISA and Einstein Telescope.