Finite volume treatment of pi pi scattering and limits to phase shifts extraction from lattice QCD

TL;DR

This paper analyzes finite volume effects on pipi scattering in lattice QCD, comparing different theoretical approaches to quantify errors in phase shift extraction and assessing the validity of standard methods like Luscher's.

Contribution

It compares three scattering approaches to evaluate finite volume effects and quantifies errors in phase shift extraction from lattice QCD data.

Findings

Finite volume effects are negligible above 2.5m_pi^-1 for scalar-isoscalar channel.

Errors in phase shift extraction are around 5% at 1.5-2m_pi^-1 for certain channels.

Standard Luscher method introduces quantifiable approximation errors.

Abstract

We study theoretically the effects of finite volume for pipi scattering in order to extract physical observables for infinite volume from lattice QCD. We compare three different approaches for pipi scattering (lowest order Bethe-Salpeter approach, N/D and inverse amplitude methods) with the aim to study the effects of the finite size of the box in the potential of the different theories, specially the left-hand cut contribution through loops in the crossed t,u-channels. We quantify the error made by neglecting these effects in usual extractions of physical observables from lattice QCD spectra. We conclude that for pipi phase-shifts in the scalar-isoscalar channel up to 800 MeV this effect is negligible for box sizes bigger than 2.5m_pi^-1 and of the order of 5% at around 1.5-2m_pi^-1. For isospin 2 the finite size effects can reach up to 10% for that energy. We also quantify the error made when using the standard Luscher method to extract physical observables from lattice QCD, which is widely used in the literature but is an approximation of the one used in the present work.