The Spectrum of Global Axion Strings
Alejandro Vaquero · 2025
Read The Spectrum of Global Axion Strings by Alejandro Vaquero (2025) 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-2502-02398.txt.
Introduction
Mathieu Kaltschmidt and co-authors refine the predicted mass of the axion, a candidate for dark matter, in a paper from 2025. In one leading scenario, the axion's mass is set by how a field broke apart in the early universe. That breaking left string-like defects whose decay shed axions.
The team ran large simulations of the field on a fixed lattice and studied the spectrum of the shed axions. They flag effects past work missed, such as dependence on starting states and wiggles in the spectrum. Correcting for these, the dark matter mass lands between 95 and 450 micro electron volts. That range sits within reach of the next round of lab searches.
Catalog record
| CREATOR | Alejandro Vaquero |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 18 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-02398
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
The Spectrum of Global Axion Strings Alejandro Vaquero Abstract The post-inflationary Peccei-Quinn (PQ) symmetry breaking scenario provides a unique opportunity to pinpoint the QCD axion dark matter mass, which is a crucial input for laboratory experiments that are designed for probing specific mass ranges. Predicting their mass requires a precise knowledge of how axions are produced from the decay of topological defects in the early Universe that are inevitably formed. In this contribution, we present recent results on the analysis of the spectrum of axions radiated from global strings based on large scale numerical simulations of the cosmological evolution of the PQ field on a static lattice [ 1 ] . We highlight several systematic effects that have been overlooked in previous works, such as the dependence on the initial conditions, contaminations due to oscillations in the spectrum, and discretisation effects; some of which could explain the discrepancy in the current literature. Taking these uncertainties into account and performing the extrapolation to cosmologically relevant string tensions, we find that the dark matter mass is predicted to be in the range of 95 μ eV ≲ m a ≲ 450 μ eV 95\,\mu\text{eV}\lesssim m_{a}\lesssim 450\,\mu\text{eV} , which will be probed by some of the next generation direct detection experiments. [a]Mathieu Kaltschmidt 1 Introduction The quantum chromodynamics (QCD) axion 1 1 1 For a thorough discussion of the topic we refer to recent reviews on axion cosmology [ 2 ] , axion models [ 3 ] and experimental searches [ 4 ] . , originally proposed as a solution to the strong CP problem [ 5 , 6 , 7 , 8 ] , is a leading candidate to explain cold dark matter (CDM) in our Universe [ 9 , 10 ] . In the post-inflationary Peccei-Quinn (PQ) symmetry breaking scenario, a network of global cosmic strings inevitably forms in the early Universe when the global continuous PQ symmetry is spontaneously broken, as explained by the Kibble mechanism [ 11 ] . These strings decay, radiating axions that contribute to the dark matter density today. Understanding the axion production from strings offers the unique opportunity to give a precise prediction of the axion dark matter mass, which serves as crucial input for experimental searches, as it determines the frequency ranges for resonant detection methods such as haloscopes. However, accurate predictions are hindered by uncertainties on the axion emission spectrum, particularly its spectral index q , and the evolution of the string network density. Even with extensive numerical simulations, the system can only be studied at unphysical regimes that ultimately require careful extrapolations over many orders of magnitude in the string tension. Previous studies yielded conflicting results, with predictions for the axion mass ranging from a few to \sim 1000\ \mu eV [ 12 , 13 , 14 , 15 , 16 , 17 ] . In this contribution we discuss the relevant parameters that characterize the axion radiation from global strings and several systematic effects that could explain the discrepancy in the existing literature. 2 Simulating Axion Strings To study the dynamics of the network of global axion strings we simulate a complex scalar field \phi with Lagrangian, \mathcal{L}=\frac{1}{2}\left|\partial_{\mu}\phi\right|^{2}-V_{\mathrm{PQ}}(\phi),\quad\text{where}\quad V_{\mathrm{PQ}}(\phi)=\frac{\lambda}{4}\left(|\phi|^{2}-f_{a}^{2}\right)^{2}. (1) Here, f_{a} defines the high-energy scale of the spontaneous symmetry breaking. Expanding the field around the potential minima located at |\phi|=f_{a} yields \phi(x)=\left(f_{a}+r(x)\right)e^{i\theta(x)} . We identify a(x)=\theta(x)f_{a} as the axion , i. e. the massless Goldstone boson, and r(x) as the heavy radial field, dubbed saxion , with mass m_{r}=\sqrt{2\lambda}f_{a} . Strings are located where the modulus |\phi| approaches zero, as the phase \theta(x) winds through a full circle and continuity of the field requires |\phi|=0 at one point within each loop in \theta , where \theta is therefore undefined. These regions have high energy densities as they correspond to maxima of the potential in field space, i. e. they retain residual potential energy from the unbroken phase \langle\phi\rangle=0 . To run simulations of the PQ field during radiation domination in an expanding FRW background, we utilise jaxions 2 2 2 Available on Github: https://github.com/veintemillas/jaxions [ 18 ] , a massively parallel C++ code, originally developed in Ref. [ 19 ] . Performing numerical simulations of global axion strings suffers from an infeasible issue with the dynamical range of the system. This becomes most apparant when looking at the energy per unit length (or tension ), \mu , of the string, which is logarithmically divergent: \mu=\hat{\mu}+\pi f_{a}^{2}\int_{c_{\mathrm{UV}}}^{c_{\mathrm{IR}}}\frac{\mathrm{d}r}{r}\approx\hat{\mu}+\pi f_{a}^{2}\ln\left(\frac{c_{\mathrm{IR}}}{c_{\mathrm{UV}}}\right). (2) In the cosmological scenario of the QCD axion, the UV cutoff c_{\mathrm{UV}} is provided by the width of the string core c_{\mathrm{UV}}\approx m_{r}^{-1} and the IR cutoff scale by the typical inter-string distance c_{\mathrm{IR}}\sim H^{-1} , which is motivated by the fact that the strings are expected to enter a scaling regime with \mathcal{O}(\text{few}) strings per Hubble patch [ 11 , 20 ] . At the latest times f_{a}/H\sim 10^{30} , and therefore \ln\left(\frac{f_{a}}{H}\right)\sim 70 , which is impossible to resolve, even on the largest supercomputers. This poses a major problem as we are forced to simulate at unphysical, low values of the string tension and extrapolate results over several orders of magnitude. From this point we use the short-hand notation \ell\equiv\ln\left(\frac{f_{a}}{H}\right) for the parameter that we use to track the temporal evolution of the network, since \ell\equiv\ell(\tau) grows with conformal time, because H\propto 1/\tau^{2} during radiation domination. 3 String Network Density and Attractor Determining the evolution of the string density parameter \xi=\frac{l_{s}}{\mathcal{V}}t^{2} , with l_{s} the string length and \mathcal{V}=t^{3} the causal volume in conformal units, is very important to minimize the impact of the choice of initial conditions on the axion dark matter mass prediction. Previous studies, cf. Ref. [ 12 ] , observed and introduced the notion of an attractor solution of the string density evolution. We confirmed the existence of such behavior, cf. Fig. 1 , and found that it can also be observed in the energy spectra of both the axion and especially the saxion fields. The attractor evolution, inspired by the study of conformal string networks 3 3 3 Conformal string networks are artificially kept at fixed values of \ell . of Ref. [ 21 ] , can be modeled as, \frac{d\xi}{dt}=\frac{C}{t}\left(\xi_{c}(\ell(t))-\xi(t)\right). (3) Here, \xi_{c}(\ell) , is the equilibrium density of the conformal string network and t/C is the characteristic time scale of the network restoration, approaching the limit \xi_{c}\rightarrow\xi . In modifying the constant C\rightarrow C(x=\xi/\xi_{c}) , two models give satisfactory fits to the conformal data of Ref. [ 21 ] . Applying this to our simulations yields \xi_{c}^{\mathrm{lin}}=-0.19(3)+0.205(7)\ell\quad\text{and}\quad\xi_{c}^{\mathrm{sat}}=\frac{-0.25(15)+0.23(6)\ell}{1+0.02(4)\ell}. (4) Evaluated at \ell=70 , this results in values of \xi^{\operatorname{lin}}(\ell=70)\sim 13.8(5)\quad\text{and}\quad\xi^{\mathrm{sat}}(\ell=70)\sim 7(3), (5) disfavouring an early saturation as proclaimend in Refs. [ 22 , 23 , 15 ] , that find \xi\sim 1.2(2) . A comparison with the results of Refs. [ 13 , 14 ] is displayed in the right panel of Fig. 1 . Figure 1: Left: Evolution of the string density \xi for different initial densities (simulations with N^{3}=2048^{3} grid sites) and for the largest simulations ( N^{3}=11268^{3} ) using initial conditions close to the attractor \xi\approx 0.3 (black dashed line). Right: Comparison of the string density evolution with the results of Refs. [ 13 , 14 ]