Non-perturbative renormalization for lattice massive QED$_2$: the ultraviolet problem

We consider a lattice regularization, preserving Ward Identities (WI) and with a Wilson term, of the Massive QED$_2$, describing a fermion with mass $m$ and charge $\mathsf{e}$ interacting with a vector field with mass $M$, in the regime $m\ll M\ll a^{-1}$ ($a$ being the lattice spacing) which is the suitable one to mimic a realistic 4d massive gauge theory like the Electroweak sector. The presence of the lattice and of the mass $m$ breaks any solvability property. In this paper we prove that the effective action obtained after the integration of the ultraviolet degrees of freedom is expressed by expansions which are convergent for values of the coupling $|\mathsf{e}|\le \mathsf{e}_0$, with $\mathsf{e}_0$ independent on $a$ and $m$, and with cut-off-independent bare parameters. By combining this result with the analysis of the infrared part in previous papers we get a complete construction of the model and a number of properties whose analogous are expected to hold in 4d. The analysis is done by integrating out the bosons and reducing to a fermionic theory; however, with respect to the case with momentum regularizations (which break essential features like the WI), the resulting effective fermionic action has not a simple form and this requires the developments of new methods to get the necessary bounds.

💡 Research Summary

The paper presents a rigorous, non‑perturbative construction of massive quantum electrodynamics in two Euclidean dimensions (massive QED₂) using a lattice regularization that incorporates a Wilson term and preserves Ward Identities (WIs). The authors work in the physically motivated regime where the fermion mass m, the gauge‑boson mass M, and the inverse lattice spacing a⁻¹ satisfy m ≪ M ≪ a⁻¹, a hierarchy that mirrors the situation in the electroweak sector of the Standard Model.

The model is defined on a finite toroidal lattice Λ with spacing a = 2⁻ᴺ and periodic boundary conditions. The bosonic field A_μ is integrated with a Gaussian measure whose propagator contains both transverse and longitudinal components; the longitudinal part is controlled by a gauge‑fixing parameter ξ∈