Generalized isotropic Lipkin-Meshkov-Glick models: ground state entanglement and quantum entropies

TL;DR

This paper introduces a new class of generalized isotropic Lipkin-Meshkov-Glick models with su(m+1) symmetry, analyzing their ground state entanglement properties, entropy scaling, and phase diagram, revealing non-critical behavior and extensive Tsallis entropy under certain conditions.

Contribution

The paper develops a comprehensive analysis of generalized su(m+1) Lipkin-Meshkov-Glick models, deriving closed-form entanglement entropies and characterizing their phase structure, which was not previously known.

Findings

Entanglement entropies scale as a log function with block size L.

Models are non-critical as Rènyi entropy becomes q-independent.

Tsallis entropy can be extensive for specific parameter choices.

Abstract

We introduce a new class of generalized isotropic Lipkin-Meshkov-Glick models with su$(m+1)$ spin and long-range non-constant interactions, whose non-degenerate ground state is a Dicke state of su$(m+1)$ type. We evaluate in closed form the reduced density matrix of a block of $L$ spins when the whole system is in its ground state, and study the corresponding von Neumann and R\'enyi entanglement entropies in the thermodynamic limit. We show that both of these entropies scale as $a\log L$ when $L$ tends to infinity, where the coefficient $a$ is equal to $(m-k)/2$ in the ground state phase with $k$ vanishing su$(m+1)$ magnon densities. In particular, our results show that none of these generalized Lipkin-Meshkov-Glick models are critical, since when $L\to\infty$ their R\'enyi entropy $R_q$ becomes independent of the parameter $q$. We have also computed the Tsallis entanglement entropy of the ground state of these generalized su$(m+1)$ Lipkin-Meshkov-Glick models, finding that it can be made extensive by an appropriate choice of its parameter only when $m-k\ge3$. Finally, in the su$(3)$ case we construct in detail the phase diagram of the ground state in parameter space, showing that it is determined in a simple way by the weights of the fundamental representation of su$(3)$. This is also true in the su$(m+1)$ case; for instance, we prove that the region for which all the magnon densities are non-vanishing is an $(m+1)$-simplex in $\mathbf R^m$ whose vertices are the weights of the fundamental representation of su$(m+1)$.