Non-equilibrium Dynamics of O(N) Nonlinear Sigma models: a Large-N approach

TL;DR

This paper investigates the non-equilibrium dynamics of the O(N) nonlinear sigma model in 2+1 dimensions using a large-N approach, revealing a dynamical critical point and non-thermalization behavior relevant to holographic duals.

Contribution

It introduces a large-N Schwinger-Keldysh framework to analyze time-dependent mass gaps and identifies a dynamical critical point during coupling evolution.

Findings

Identifies a dynamical critical coupling $g_c^{ m dyn}$ where the gap vanishes.

Shows the system does not thermalize within finite time after a sudden quench.

Derives a criterion for adiabatic breakdown and Kibble-Zurek scaling law.

Abstract

We study the time evolution of the mass gap of the O(N) non-linear sigma model in 2+1 dimensions due to a time-dependent coupling in the large-$N$ limit. Using the Schwinger-Keldysh approach, we derive a set of equations at large $N$ which determine the time dependent gap in terms of the coupling. These equations lead to a criterion for the breakdown of adiabaticity for slow variation of the coupling leading to a Kibble-Zurek scaling law. We describe a self-consistent numerical procedure to solve these large-$N$ equations and provide explicit numerical solutions for a coupling which starts deep in the gapped phase at early times and approaches the zero temperature equilibrium critical point $g_c$ in a linear fashion. We demonstrate that for such a protocol there is a value of the coupling $g= g_c^{\rm dyn}> g_c$ where the gap function vanishes, possibly indicating a dynamical instability. We study the dependence of $g_c^{\rm dyn}$ on both the rate of change of the coupling and the initial temperature. We also verify, by studying the evolution of the mass gap subsequent to a sudden change in $g$, that the model does not display thermalization within a finite time interval $t_0$ and discuss the implications of this observation for its conjectured gravitational dual as a higher spin theory in $AdS_4$.