Fractals and spontaneous symmetry breaking with type-B Goldstone modes: a perspective from entanglement

TL;DR

This paper reveals that the ground state subspace of a ferromagnetic Heisenberg model exhibits an intrinsic fractal structure, which is reflected in the entanglement entropy scaling and linked to the number of type-B Goldstone modes.

Contribution

It demonstrates that the ground state subspace forms a fractal set, with entanglement entropy scaling related to the fractal dimension, providing a new perspective on symmetry breaking and Goldstone modes.

Findings

Entanglement entropy scales logarithmically with block size.

The fractal dimension correlates with the number of type-B Goldstone modes.

Ground state subspace decomposes into a union of fractal regions.

Abstract

The one-dimensional spin-$s$ ${\rm SU}(2)$ ferromagnetic Heisenberg model, as a paradigmatic example for spontaneous symmetry breaking (SSB) with type-B Goldstone modes (GMs), is expected to exhibit an abstract fractal underlying the ground state subspace. This intrinsic abstract fractal is here revealed from a systematic investigation into the entanglement entropy for a linear combination of factorized (unentangled) ground states on a fractal decomposable into a set of the Cantor sets. The entanglement entropy scales logarithmically with the block size, with the prefactor being half the fractal dimension of a fractal, as long as the norm for the linear combination scales as the square root of the number of the self-similar building blocks kept at each step $k$ for a fractal, under an assumption that the maximum absolute value of the coefficients in the linear combination is chosen to be around one, and the coefficients in the linear combination are almost constants within the building blocks. Actually, the set of the fractal dimensions for all the Cantor sets forms a {\it dense} subset in the interval $[0,1]$. As a consequence, the ground state subspace is separated into a disjoint union of countably infinitely many regions, each of which is labeled by a decomposable fractal. Hence, the interpretation of the prefactor as half the fractal dimension is valid for any support beyond a fractal, which in turn leads to the identification of the fractal dimension with the number of type-B GMs for the orthonormal basis states. Our argument may be extended to any quantum many-body systems undergoing SSB with type-B GMs.