Abstract
We study a class of DFSZ-like models for the QCD axion that can address observed anomalies in stellar cooling. Stringent constraints from SN1987A and neutron stars are avoided by suppressed couplings to nucleons, while axion couplings to electrons and photons are sizable. All axion couplings depend on few parameters that also control the extended Higgs sector, in particular lepton flavor-violating couplings of the Standard Model-like Higgs boson h. This allows us to correlate axion and Higgs phenomenology, and we find that BR(h → τe) can be as large as the current experimental bound of 0.22%, while BR(h → μμ) can be larger than in the Standard Model by up to 70%. Large parts of the parameter space will be tested by the next generation of axion helioscopes such as the IAXO experiment.
Highlights
The axion can account for excessive energy losses observed in several stellar environments [8,9,10,11,12,13], which hints to new cooling channels such as a light axion with large couplings to electrons
In general the axion coupling to electrons depends on free parameters that control lepton flavor-violating (LFV) effects, which are mediated by LFV couplings of both the axion and the physical Higgs scalars
In this article we have explored the correlation of axion and Higgs phenomenology in variant axion models with a light second Higgs doublet
Summary
Field strengths and similar in the gluon sector, E/N is the ratio of EM and color anomaly coefficients and we use the convention 0123 = −1 (for more details see appendix A). The QCD scale the relevant couplings are those to photons, nucleons n, p and electrons, Cγ a fa αem 8π. For the purpose of addressing the stellar cooling anomalies with axions it is helpful to have small couplings to nucleons in order to avoid the stringent constraints from SN1987A and neutron star cooling, cf section 3.3. (2.4) and (2.5) it is clear that axion couplings to nucleons can be suppressed if the UV quark couplings satisfy the approximate relations. As analyzed in detail in ref. [25] (see refs. [35, 36]), one can realize these conditions in the context of “nucleophobic” DFSZ models that we will discuss in more details
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
