Heavy fields and the axion quality problem

The idea that the gravity-induced breaking of global symmetries is encoded in Planck-suppressed operators is not scale-invariant: heavy particles which have nothing to do with the UV completion of gravity can mediate the breaking and produce low-energy operators (partly) suppressed by their own mass scales. Such contributions from heavy fields are typically subdominant with respect to the least Planck-suppressed operators, unless the latter are forbidden, as is usual in solutions to the axion quality problem based on four-dimensional gauge symmetries. Therefore, with a focus on axion physics, I investigate such situations and present toy examples where a non-minimal sector of heavy fields coupling to the axion generates operators whose coefficients are orders of magnitude larger than the naive Planck-suppressed estimates, despite gravity being the only source of breaking. I also stress that the key features of these toy models, namely several families of heavy fields with family-dependent gauge charges, are already present in some non-minimal QCD axion models, using the axiflavon/flaxion and the lighter-than-usual KSVZ axion as examples. This suggests that gauge-symmetry-based solutions to the axion quality problem, or of any quality problem really, need to describe complete UV scenarios or make precise UV assumptions.