Entanglement Entropy and Twist Fields
Michele Caraglio, Ferdinando Gliozzi
TL;DR
This work analyzes entanglement entropy via the replica method in 1+1D CFTs, focusing on the role of branch-point twist fields and the validity of multi-interval formulae. It exposes inconsistencies in the conventional Calabrese–Cardy expressions for disjoint intervals by highlighting missing Schwarzian singularities in conformal mappings, and it introduces a practical numerical approach to compute tr ρ_A^n as a vacuum expectation value in an n-fold replicated system. The authors implement the method in 2D Potts models, confirming predicted twist-field correlator behavior for both unitary and non-unitary cases and demonstrating the approach yields direct access to the central charge through scaling. The results reinforce the reliability of single-interval predictions while outlining the limitations of multi-interval analytic formulas and suggesting broad applicability of the numerical technique to higher dimensions and gauge theories.
Abstract
The entanglement entropy of a subsystem of a quantum system is expressed, in the replica approach, through analytic continuation with respect to n of the trace of the n-th power of the reduced density matrix. This trace can be thought of as the vacuum expectation value of a suitable observable in a system made with n independent copies of the original system. We use this property to numerically evaluate it in some two-dimensional critical systems, where it can be compared with the results of Calabrese and Cardy, who wrote the same quantity in terms of correlation functions of twist fields of a conformal field theory. Although the two calculations match perfectly even in finite systems when the analyzed subsystem consists of a single interval, they disagree whenever the subsystem is composed of more than one connected part. The reasons of this disagreement are explained.
