Is the nonlinear sigma model the m sigma -->

Abstract

We propose to investigate whether the SO(N) nonlinear sigma model is equivalent to the m/sub sigma/..-->..infinity limit of the linear sigma model by comparing the corresponding one-loop effective-action expansions up to the four-derivative terms and including the symmetry-breaking term. For this purpose we use a new background-field method to calculate the effective-action expansion directly. In the case of the linear sigma model, the renormalization procedure is implemented carefully before the m/sub sigma/..-->..infinity limit is taken. For the nonlinear sigma model we introduce a new and intuitive covariant treatment for the perturbation calculation of the field theory with nonlinear constraint. We do not find any noninvariant terms in either case. We show that the divergent parts of the effective Lagrangians due to m/sub ..pi../..-->..0, m/sub sigma/..-->..infinity, or N..-->..infinity are equivalent in the two models. However, the nonleading finite parts of the effective Lagrangians are different. Therefore, the two operations, taking the m/sub sigma/..-->..infinity limit and calculating the quantum corrections, do not commute. The origin of this difference may be a violation of decoupling.