DFSZ-Type Axions and Where to Find Them

TL;DR

This paper systematically analyzes non-minimal DFSZ axion models, calculating their axion-photon couplings, classifying models, and exploring their experimental implications, revealing a broad range of couplings and potential for enhanced detectability.

Contribution

It provides a comprehensive calculation and classification of axion-photon couplings in non-minimal DFSZ models, extending the viable parameter space and comparing with KSVZ models.

Findings

Non-minimal DFSZ models can have axion-photon couplings nearly three orders of magnitude larger than minimal models.

Most couplings in non-minimal models are close to minimal model values.

A subset of models avoids the domain wall problem and has enhanced couplings.

Abstract

We systematically calculate the axion-photon coupling for non-minimal DFSZ models. Thereby we can classify every calculated model and study the resulting distributions, relevant for axion experiments like haloscopes, helioscopes or light-shining-through-a-wall experiments. By adding more than one additional Higgs doublet, these non-minimal DFSZ models extend the viable axion parameter space and lead to a large range of axion-photon couplings. We find couplings almost three orders of magnitude larger than the ones of the minimal models. Most of the possible axion-photon couplings, however, lie in the vicinity of the values dictated by the minimal models. We quantify this by introducing a theoretical prior probability distribution for DFSZ-type axions and giving $68\%$ and $95\%$ lower bounds as well as two-sided bands. We compare our results for the DFSZ axion-photon coupling distributions with the KSVZ case, for which a similar analysis has been conducted. Both display similar values as well as a very specific pattern. In order to identify preferred models, we discuss the role of flavour changing neutral currents and the domain wall problem as possible selection criteria. It is possible to construct a large number of non-minimal DFSZ models with a domain wall number of unity and thereby avoiding the domain wall problem. This subset also has a significantly enhanced axion-photon coupling compared to the minimal DFSZ models.