Early Structure Formation Constraints on the Ultra-Light Axion in the Post-Inflation Scenario

TL;DR

This paper investigates how late-time symmetry breaking of ultra-light axion-like particles affects small-scale structure formation and cosmological observables, providing new constraints on ALP properties in the post-inflation scenario.

Contribution

It introduces the impact of white-noise density fluctuations from post-inflation symmetry breaking on structure formation and derives observational constraints on ALP parameters.

Findings

Ly$oldsymbol{ m f a}$ forest provides the strongest current constraints.

High-redshift reionization observations can further restrict ALP models.

Future 21cm observations could significantly improve constraints through BAO features.

Abstract

Many works have concentrated on the observable signatures of the dark matter being an ultralight axion-like particle (ALP). We concentrate on a particularly dramatic signature in the late-time cosmological matter power spectrum that occurs if the symmetry breaking that establishes the ALP happens after inflation -- white-noise density fluctuations that dominate at small scales over the adiabatic fluctuations from inflation. These fluctuations alter the early history of nonlinear structure formation. We find that for symmetry breaking scales of $f_A \sim 10^{13}-10^{15}$GeV, which requires a high effective maximum temperature after inflation, ALP dark matter with particle mass of $m_A \sim 10^{-13}-10^{-20}$eV could significantly change the number of high-redshift dwarf galaxies, the reionization history, and the Ly$\alpha$ forest. We consider all three observables. We find that the Ly$\alpha$ forest is the most constraining of current observables, excluding $f_A \gtrsim 10^{15}$GeV ($m_A \lesssim 10^{-17}$eV) in the simplest model for the ALP and considerably lower values in models coupled to a hidden strongly interacting sector ($f_A \gtrsim 10^{13}$GeV and $m_A \lesssim 10^{-13}$eV). Observations that constrain the extremely high-redshift tail of reionization may disfavor similar levels of isocurvature fluctuations as the forest. Future $z\sim 20-30$ 21cm observations have the potential to improve these constraints further using that the supersonic motions of the isocurvature-enhanced abundance of $\sim10^4M_\odot$ halos would shock heat the baryons, sourcing large BAO features.