Notes on non-trivial and logarithmic CFTs with c=0

TL;DR

This paper investigates the structure and properties of two-dimensional logarithmic conformal field theories with zero central charge, proposing new approaches and solutions to longstanding issues like the c=0 catastrophe, with implications for models like percolation.

Contribution

It introduces a novel framework for c=0 logarithmic CFTs using a Jordan cell for the stress energy tensor and explores tensor model solutions, advancing understanding of these theories.

Findings

Derived OPEs and two-point functions for T(z) and t(z)

Proposed a new solution to the c=0 catastrophe via tensor models

Discussed implications for percolation and representation theory

Abstract

We examine the properties of two-dimensional conformal field theories (CFTs) with vanishing central charge based on the extended Kac-table for c_(9,6)=0 using a general ansatz for the stress energy tensor residing in a Jordan cell of rank two. Within this setup we will derive the OPEs and two point functions of the stress energy tensor T(z) and its logarithmic partner field t(z) and illustrate this by a bosonic field realization. We will show why our approach may be more promising than those chosen in the literature so far, including a discussion on properties of the augmented minimal model with vanishing central charge such as full conformal invariance of the vacuum as a state in an irreducible representation, consequences on percolation from null vectors and the structure of representations within the Kac table. Furthermore we will present another solution to the c --> 0 catastrophe based on an logarithmic CFT tensor model. As example, we consider a tensor product of the well-known c=-2 logarithmic CFT with a four-fold Ising model. We give an overview of the possible configurations and various consequences on the two point functions and the OPEs of the stress energy tensor T(z) and its logarithmic partner field t(z). We will motivate that due to the full conformal invariance of the vacuum at c=0, we have to assume a Jordan cell for the identity since t(z) is now a descendant of a new h=0 field.