Isocurvature-Free QCD Axion Dark Matter from Inflaton-Driven Early QCD: the Necessity of Inflationary Plateaus

Katherine Freese, Evangelos I. Sfakianakis, Barmak Shams Es Haghi · 2026

Read Isocurvature-Free QCD Axion Dark Matter from Inflaton-Driven Early QCD: the Necessity of Inflationary Plateaus by Katherine Freese, Evangelos I. Sfakianakis, Barmak Shams Es Haghi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-15192.txt.

Introduction

Katherine Freese, Evangelos I. Sfakianakis, and Barmak Shams Es Haghi examine axion dark matter when the inflaton couples straight to gluons. That coupling can raise the QCD confinement scale during that era, making the axion temporarily heavy and damping isocurvature ripples. As it proceeds that scale relaxes, the axion lightens, and late ripples can supply the seen dark matter. The mechanism selects plateau-like inflation over monomial models, which let that scale grow too fast.

Catalog record

CREATORKatherine Freese, Evangelos I. Sfakianakis, Barmak Shams Es Haghi
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE44595
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.15192

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Isocurvature-Free QCD Axion Dark Matter from Inflaton-Driven Early QCD: the Necessity of Inflationary Plateaus Katherine Freese, Evangelos I. Sfakianakis, Barmak Shams Es Haghi Abstract A direct coupling between the inflaton and Standard Model gluons can dynamically raise the QCD confinement scale during inflation, making the axion temporarily heavy and suppressing axion isocurvature perturbations. As inflation proceeds, the confinement scale relaxes, the axion becomes light, and late-time de Sitter fluctuations can generate the observed dark matter abundance. We analyze this mechanism without specifying an inflationary potential, instead parametrizing the background by ϵ ⁡ ( N ) ∝ 1 / N p \epsilon(N)\propto 1/N^{p} , where N N is the number of e e -folds before the end of inflation. The single parameter p p distinguishes monomial models ( p = 1 p=1 ), standard plateau models ( p = 2 p=2 ), and ultra-flat plateau or hilltop-like models ( p ≥ 3 p\geq 3 ). We analytically show that the mechanism selects plateau-like ( p ≥ 2 p\geq 2 ) inflation: monomial models generically cause the confinement scale to grow too rapidly, while plateau models keep the QCD sector under perturbative control. In the minimal scenario, reheating occurs through the same inflaton-gluon coupling, and viable axion dark matter production is obtained when deconfinement occurs after the CMB window. The early-confinement sector also generically shifts the scalar spectral index to smaller, redder values. Because ultra-flat ( p ≥ 3 p\geq 3 ) models inherently predict overly red spectra, this shift exacerbates their tension with CMB data, leaving p = 2 p=2 plateau models as the phenomenologically viable parameter space (in this parametrization). Introduction. —The Peccei-Quinn (PQ) solution 28 ; 27 to the strong CP problem introduces a global U(1) PQ symmetry spontaneously broken at a high energy scale f_{a} . This yields the QCD axion 34 ; 35 ; 36 ; 8 ; 29 ; 1 ; 9 , an excellent cold dark matter (DM) candidate. The resulting cosmology depends critically on whether the PQ symmetry breaks before or after inflation 26 . In the pre-inflationary scenario, the axion field is homogenized, avoiding topological defects. However, as a light spectator field, it acquires quantum fluctuations of order H/(2\pi) during inflation, where H is the Hubble scale. These super-horizon modes follow a random walk, generating DM isocurvature perturbations. Satisfying the Planck limit ( \beta_{\rm iso}<0.038 ) typically restricts the scale of inflation to H\lesssim 10^{8} GeV, in tension with high-scale inflationary models. A mechanism to escape this bound is to increase the axion mass during inflation, exponentially suppressing its fluctuations. Since the axion mass scales with the QCD confinement scale, m_{a}\propto\Lambda_{\rm QCD}^{2}/f_{a} , an enhanced axion mass can be achieved by making \Lambda_{\rm QCD} dynamical in the early universe 10 ; 21 ; 11 ; 22 ; 18 . The QCD sector is confined during the observable part (in the Cosmic Microwave Background) of inflation with large values of m_{a} and \Lambda_{\rm QCD} ; then deconfined at a later time in inflation and at the high temperatures after inflation; then confined once the temperature of the universe drops to T<\Lambda_{0} , today’s value of the QCD scale. In this Letter, we study the inflaton itself as the driver of this dynamical confinement. We relate the field excursion directly to a generic parameterization of the slow-roll parameter \epsilon(N)\propto 1/N^{p} , where N is the number of e -folds before the end of inflation. The dimensionless parameter p distinguishes monomial models ( p=1 ), standard plateau models ( p=2 ), and ultra-flat plateau or hilltop-like models ( p\geq 3 ). We derive universal conditions required for the mechanism to succeed without spoiling the inflationary background. We demonstrate that this physical mechanism does not work with monomial inflation, whereas it perfectly accommodates standard plateau models. Further, we show that the predicted scalar spectral index ( n_{s} ) of inflaton density perturbations is shifted to lower, redder values. Because ultra-flat plateau models ( p\geq 3 ) already predict overly red spectra, this shift exacerbates their tension with cosmic microwave background (CMB) data. We therefore demonstrate that standard p=2 plateau models are robustly selected as the best candidates for the success of this mechanism. We finally showcase the dependence of the various energy scales on the overall magnitude of the slow-roll parameter \epsilon(N) for plateau and ultra-flat plateau models. An extended analysis will be presented in Ref. 12 . During the completion of the current work (which we began in April 2025), Ref. 11 appeared, using the same interaction term, but within SU(5) grand unified theory. In contrast, we take a phenomenological model-independent approach. We therefore show (a) the necessity of plateau models for the success of this mechanism and (b) the resulting generic negative shift in n_{s} . Inflaton-Driven Confinement. —We introduce a non-renormalizable interaction between the inflaton \phi and the Standard Model (SM) gluon field strength G_{\mu\nu} : {\cal L}_{g}\supset-{1\over 4}\left({1\over g_{s0}^{2}}+{\phi\over M}\right)G_{\mu\nu}G^{\mu\nu}\,, (1) where M is the mass scale of new physics. The above inflaton-gluon coupling can arise for example from integrating out heavy beyond SM fields, which will define the scale M , see e.g. Ref. 11 . During inflation, the classical homogeneous background \bar{\phi}(N) changes the gluon kinetic term. This modifies the running of the strong coupling \alpha_{s} , yielding a confinement scale exponentially sensitive to the field value. We will take a negative field value for the inflaton. Assuming the inflaton rolls toward the origin ( \bar{\phi}<0 ), the elevated confinement scale is 21 \Lambda(\bar{\phi})=\Lambda_{0}\exp\left(\frac{24\pi^{2}}{2n_{f}-33}\frac{\bar{\phi}}{M}\right)\simeq\Lambda_{0}e^{11|\bar{\phi}|/M}\,, (2) where n_{f}=6 is the number of active quarks and \Lambda_{0}\simeq 400 MeV is the standard QCD confinement scale at \bar{\phi}=0 . Because \bar{\phi} modifies the confinement scale, the axion mass m_{a}\sim\Lambda^{2}/f_{a} is correspondingly enhanced during inflation compared to the standard case without gluon coupling. As the inflaton rolls down its potential, one can see that the value of \Lambda decreases following the above equation. In de Sitter space, the causal horizon is R_{H}\sim H^{-1} , corresponding to an ambient Gibbons-Hawking temperature T_{\rm GH}=H/(2\pi) 15 . For QCD to confine, the characteristic size of a hadronic bound state ( \Lambda^{-1} ) must fit within the horizon; equivalently, the background temperature must be colder than the critical scale ( H\lesssim\Lambda ). If \Lambda(\bar{\phi})<H , the spacetime expands too rapidly for quarks to causally communicate, melting the QCD vacuum and exponentially suppressing the instanton effects responsible for the axion mass and rendering the axion effectively massless. Therefore, the axion can only acquire a heavy mass ( m_{a}>3H/2 ) capable of suppressing isocurvature perturbations while \Lambda(\bar{\phi})>H . This is the case at the onset of inflation. We define the deconfinement time, N_{\rm dec} (in e -folds before the end of inflation), as the exact moment when \Lambda(N_{\rm dec})=H . Hence, once the inflaton has rolled down the potential beyond the deconfinement time, i.e. for N<N_{\rm dec} , the axion becomes an effectively massless spectator field and accumulates de Sitter fluctuations. The variance at the end of inflation is dominated by these late-time modes: \sigma_{\theta}^{2}\equiv\langle\theta^{2}\rangle\simeq{H^{2}\over 4\pi^{2}f_{a}^{2}}N_{\rm dec}\,. (3) These fluctuations are innocuous as far as observable isocurvature modes are concerned, because they are on far too small scales to be seen in the CMB. However, they are important as they excite

Read the complete text