Extreme Axions Unveiled: A Novel Fluid Approach for Cosmological Modeling - Abstract and Intro

20 May 2024

This paper is available on arxiv under CC 4.0 license.


(1) HARRISON WINCH, Department of Astronomy & Astrophysics, University of Toronto and Dunlap Institute for Astronomy and Astrophysics, University of Toronto;

(2) RENEE´ HLOZEK, Department of Astronomy & Astrophysics, University of Toronto and Dunlap Institute for Astronomy and Astrophysics, University of Toronto;

(3) DAVID J. E. MARSH, Theoretical Particle Physics and Cosmology, King’s College London;

(4) DANIEL GRIN, Haverford College;

(5) KEIR K. ROGERS, Dunlap Institute for Astronomy and Astrophysics, University of Toronto.



Axion-like particles (ALPs) are a broad class of dark matter (DM) particle candidates that possess both a strong theoretical justification and a variety of potentially observable signatures. While the traditional quantum chromodynamics (QCD) axion is a pseudo-Nambu-Goldstone boson arising from a broken Peccei-Quinn symmetry (Peccei & Quinn 1977), ALPs can arise from broken symmetries more generally, and are produced naturally from a variety of string theories as a result of compactified higher dimensions, making them a well-motivated DM particle candidate (Dine & Fischler 1983; Preskill et al. 1983; Abbott & Sikivie 1983; Svrcek & Witten 2006; Duffy & van Bibber 2009; Arvanitaki et al. 2010; Marsh 2016; Adams et al. 2022). Throughout this work, we will use axion and ALP interchangeably to refer to this broad class of low-mass pseudo-Nambu-Goldstone boson DM candidates.

Some work has been done to model the evolution of the axion field with these extreme starting angles, such as the works of Cedeno et al. ˜ (2017); Zhang & Chiueh (2017b); Leong et al. (2019); Zhang & Chiueh (2017a). However, the rapidly-oscillating nature of these axion fields (both at the background and perturbation level) necessitates extremely high temporal resolution for the computations, requiring long computation times for a brute force solution Zhang & Chiueh (2017b,a). This makes running repeated estimates of the axion evolution, of the sort required for a Markov Chain Monte Carlo (MCMC) or other likelihood sampler method, prohibitively expensive.

In this work, we present a novel method of efficiently and accurately modeling the behaviour of these extreme axions as a cosmological fluid. We follow the structure of the vanilla axion modeling code axionCAMB, explained in more detail in Hlozek et al. ˇ (2015). We implement a number of innovations and improvements to axionCAMB compute predictions for cosmological observables such as the linear MPS. These innovations, described in more detail in the Section 2, include a restructuring of the initial conditions, and a novel effective sound speed of the extreme axion fluid. All of these innovations reduce the runtime to model extreme axions down to ∼ 7 seconds. This opens up new opportunities to put observational constraints on extreme axion models with higher-dimensional MCMC algorithms that require tens of thousands of calls to the axion evolution code.