[Paper Review] Holomorphic Spinor Observables in the Critical Ising Model

Dmitry Chelkak, Konstantin Izyurov|arXiv (Cornell University)|May 28, 2011

Stochastic processes and statistical mechanics17 references4 citations

TL;DR

This paper introduces holomorphic spinor observables on double covers of multiply connected domains to rigorously establish the conformal invariance of spin correlation ratios in the critical 2D Ising model. By defining a discrete fermionic observable with a sign twist based on loop lifting properties, the authors prove its scaling limit converges to a conformally covariant spinor field, yielding exact formulas for spin correlation ratios that match conformal field theory predictions.

ABSTRACT

We introduce a new version of discrete holomorphic observables for the critical planar Ising model. These observables are holomorphic spinors defined on double covers of the original multiply connected domain. We compute their scaling limits, and show their relation to the ratios of spin correlations, thus providing a rigorous proof to a number of formulae for those ratios predicted by CFT arguments.

Motivation & Objective

• To rigorously prove the conformal invariance of spin correlation ratios in the critical 2D Ising model, which had remained out of reach despite progress in related areas.
• To extend Smirnov's discrete holomorphic fermionic observables to multiply connected domains using double covers, enabling the study of spin correlations under different boundary conditions.
• To provide a mathematical foundation for CFT predictions regarding spin correlation ratios by constructing observables whose scaling limits match conformally covariant spinor fields.
• To resolve the long-standing problem of proving conformal covariance of spin correlations by introducing a new class of observables that encode boundary condition dependence through monodromy.

Proposed method

• Define a new discrete holomorphic observable $ F_{ar{\varpi}}(a,z) $ on a double cover $ \widetilde{\Omega}^\delta $ of the original domain, incorporating a sign factor $ (-1)^{l(S) + \mathbf{1}_{\gamma:a\to z}} $ based on loop lifting and path lifting behavior.
• Establish that the observable satisfies discrete holomorphicity and Riemann-type boundary conditions on the double cover, ensuring convergence to a continuous holomorphic spinor in the scaling limit.
• Construct the scaling limit of the observable as a meromorphic spinor on $ \mathbb{C}_+ \setminus \{w_1,\dots,w_m\} $, with prescribed branching, real boundary values, and residue conditions.
• Solve a linear system for parameters $ \lambda_j $ to enforce the correct singular behavior at branch points $ w_j $, ensuring the limit satisfies the required conformal covariance properties.
• Use the solution to derive an explicit formula for the ratio of spin correlations as $ \vartheta(w_1,\dots,w_m) = f(0) $, where $ f $ is the limiting spinor field.
• Fix the branch of the square root via boundary continuity (e.g., from $ \infty $ along $ (-\infty,0) $) to ensure consistency with physical boundary conditions and positive correlation in the limit.

Experimental results

Research questions

• RQ1 Can discrete holomorphic observables be extended to multiply connected domains to study spin correlations under non-trivial boundary conditions?
• RQ2 Do the scaling limits of these observables exhibit conformal covariance, matching predictions from conformal field theory?
• RQ3 What is the precise form of the scaling limit of spin correlation ratios in the critical Ising model on general domains?
• RQ4 How can the monodromy of the double cover be encoded in the observable to capture the correct boundary condition dependence?
• RQ5 Can the ratio of spin correlations be computed exactly in the scaling limit using a holomorphic spinor field with prescribed singularities and boundary values?

Key findings

• The discrete holomorphic spinor observable $ F_{\varpi}(a,z) $ converges in the scaling limit to a continuous holomorphic spinor field $ f $ on $ \mathbb{C}_+ \setminus \{w_1,\dots,w_m\} $, satisfying conformal covariance and boundary conditions.
• The scaling limit satisfies $ f(z) \in \mathbb{R} $ for $ z \in \mathbb{R} $, $ f(z)^2 = O(|z-w_j|^{-1}) $, and $ \mathop{\mathrm{res}}_{z=w_j} f(z)^2 \in i\mathbb{R}_+ $, ensuring correct monodromy and singularity structure.
• For a single point $ w \in \mathbb{C}_+ $, the ratio of spin correlations converges to $ \vartheta(w) = \frac{\mathrm{Re}\,w}{|w|} = \cos[\pi \text{hm}_{\mathbb{C}_+}(w,\mathbb{R}_-)] $, matching CFT predictions.
• In general domains, the ratio $ \frac{\mathbb{E}_{a^\delta b^\delta}[\sigma(w^\delta)]}{\mathbb{E}_{+}[\sigma(w^\delta)]} \to \cos[\pi \text{hm}_\Omega(w, (ab))] $, where $ \text{hm}_\Omega $ is harmonic measure.
• The solution to the linear system for $ \lambda_j $ is unique and non-degenerate, ensuring the existence and uniqueness of the limiting spinor field.
• The explicit formula for the correlation ratio involves a product of Blaschke-type factors and a rational function correction, derived from solving a system of $ m $ equations in $ m $ unknowns.