Finite Size Spectrum of SU(N) Principal Chiral Field from Discrete Hirota Dynamics
Vladimir Kazakov, Sebastien Leurent
TL;DR
The paper develops a finite-volume framework for the SU(N)×SU(N) principal chiral field by translating the Y-system/Hirota dynamics into a finite set of nonlinear integral equations (NLIEs) via Wronskian determinant representations. It provides explicit construction for the U(1) sector, reveals how central node equations and large-volume ABA data feed into the NLIEs, and demonstrates through N=3 numerics that the method captures the full spectrum from infrared to ultraviolet, including Lüscher corrections and conformal limits. The work clarifies subtle bound-state effects for N>2, derives finite-size Bethe equations, and shows how the energy of excited states can be computed with carefully chosen contours; it also discusses the generalization to more complex states and potential applications to AdS/CFT Y-systems. Overall, the approach offers a powerful, broadly applicable route to exact finite-volume spectra in integrable sigma-models and paves the way for extensions to higher N and holographic duals. $
Abstract
Using recently proposed method of discrete Hirota dynamics for integrable (1+1)D quantum field theories on a finite space circle of length L, we derive and test numerically a finite system of nonlinear integral equations for the exact spectrum of energies of SU(N)xSU(N) principal chiral field model as functions of m L, where m is the mass scale. We propose a determinant solution of the underlying Y-system, or Hirota equation, in terms of determinants (Wronskians) of NxN matrices parameterized by N-1 functions of the spectral parameter, with the known analytical properties at finite L. Although the method works in principle for any state, the explicit equations are written for states in the U(1) sector only. For N>2, we encounter and clarify a few subtleties in these equations related to the presence of bound states, absent in the previously considered N=2 case. In particular, we solve these equations numerically for a few low-lying states for N=3 in a wide range of m L.
