Towards a Post-Inflationary Composite Axion Model

TL;DR

This work addresses the cosmology of composite axions in a post-inflationary universe by presenting two explicit $N_{ m DW}=1$ constructions that employ center-symmetry identifications (Lazarides–Shafi) to resolve domain-wall issues. It introduces a secondary, short period of inflation to dilute heavy exotic relics (CHAMPs) while preserving axion dark matter production predominantly from the decay of string-wall networks. One model uses a chiral $SU(5)$ moose to realize $N_{ m DW}=1$, and the other uses a gauged $U(1)$ PQ symmetry to achieve the same goal, each with high-dimension PQ-violating operators to address the axion quality problem. The paper maps viable regions in parameter space where relics are diluted and defect-induced axions match the observed dark matter density, while highlighting observational avenues such as CHAMP searches and possible gravitational-wave signatures from confinement transitions, and calling for dedicated simulations to nail down the defect dynamics.

Abstract

Composite axions offer a scenario where the axion emerges as a pion-like state, avoiding fine-tuning of elementary scalars and ameliorating the axion quality problem. Despite these advantages, their post-inflationary cosmology remains largely unexplored, with challenges including the domain wall problem and the presence of exotic relics. We propose two composite axion models with an effective domain wall number $N_\text{DW} = 1$ and study the dilution of relics via a short period of inflation. One model is based on an $SU(5)$ chiral gauge theory, while the other employs a ``gauged'' $U(1)$ Peccei-Quinn symmetry in vector-like $SU(N)$ gauge theories. We identify the viable parameter space in which axion strings re-enter the horizon before or even after the QCD transition and axion dark matter is dominantly produced from the decay of the string-wall network.

Figures (6)

• Figure 1: Moose diagram of the accidental composite axion model. The bifundamental fermions $\chi_{1,1}$ and $\psi_{1,2}$ serve as link fields connecting adjacent gauge nodes. The SM quark fields reside in the $SU(3)$ gauge sector, which is connected to the chiral $SU(N)$ sector through the link fermions $Q$ and $\bar{Q}$.
• Figure 2: Moose diagram of the chiral moose composite model. The fermions $\chi_{1,1}$ and $\psi_{1,2}$ serve as link fields connecting adjacent gauge nodes. The SM quark fields reside in the $SU(3)$ gauge sector, which is connected to the chiral $SU(5)$ sector through the link fermions $Q$ and $\bar{Q}$.
• Figure 3: Physical domain for $N=3$ and $M=4$.
• Figure 4: Viable parameter space in the $T_R$--$F_a$ plane. The color coding is as follows: Gray: CMB constraint on $\Omega_{\rm CHAMP} h^2$, Light Blue: $\Omega_{\rm CHAMP} h^2<10^{-8}$, Blue: $\Omega_{\rm CHAMP} h^2<10^{-10}$, Red: $\Omega_a h^2 > 0.12$, Brown: $T_{\rm max} > 10^{-3} m_B$, Purple: scale ordering different from $F_a > T_i > T_R > T_{\rm ent}$. The lower bound of $F_a\gtrsim 10^8\,{\rm GeV}$ is set by astrophysics Carenza:2019pxuBuschmann:2021juvSpringmann:2024retCaputo:2024oqc.
• Figure 5: Moose diagram for extending the chiral moose model. The link fermion fields are denoted by their $SU(5)$ representations.