Testing inflation on all scales: a case study with $\alpha$-attractors

TL;DR

This paper develops a comprehensive methodology to evaluate inflationary models, specifically hybrid α-attractors, across all scales by integrating observational constraints, non-Gaussianity analysis, and primordial black hole and gravitational wave predictions.

Contribution

It introduces a systematic approach to constrain inflationary models on all scales, incorporating reheating uncertainties and non-Gaussianity effects, with specific focus on hybrid α-attractors.

Findings

Large-scale constraints limit parameter space for small-scale phenomenology.

Non-Gaussianity at peak scales is small, justifying linear approximations.

Some models predict detectable primordial black holes and gravitational waves for LISA.

Abstract

A plethora of inflationary models can produce interesting small-scale phenomenology, such as enhanced scalar fluctuations leading to primordial black hole (PBH) production and large scalar-induced GW. Nevertheless, good models must simultaneously explain current observations on all scales. In this work, we showcase our methodology to establish the small-scale phenomenology of inflationary models on firm grounds. We consider the case of hybrid $\alpha$-attractors, and focus on a reduced parameter space featuring the two potential parameters which roughly determine the position of the peak in the scalar power spectrum, $\mathcal{P}_\zeta$, and its amplitude. We first constrain the parameter space by comparing the large-scale predictions for $\mathcal{P}_\zeta$ with current CMB anisotropies measurements and upper limits on $\mu$-distortions. We take into account uncertainties due to the reheating phase, and observe that the parameter-space area compatible with large-scale constraints shrinks for extended reheating stages. We then move to smaller scales, where we find that non-Gaussianity at peak scales is of the local type and has amplitude $f_\text{NL}\sim \mathcal{O}(0.1)$. This ensures that non-linear effects are subdominant, motivating us to employ the tree-level $\mathcal{P}_\zeta$ to compute the abundance of PBHs and the spectrum of induced GWs for models consistent with large-scale tests. The former allows us to further constrain the parameter space, by excluding models which over-produce PBHs. We find that a subset of viable models can lead to significant production of PBHs, and a fraction of these is within reach for LISA, having a signal-to-noise ratio larger than that of astrophysical foregrounds. Our first-of-its-kind study systematically combines tests at different scales, and exploits the synergy between cosmological observations and theoretical consistency requirements.