🔢 mathematics

From the planar Ising model to quasiconformal mappings

This paper establishes that the scaling limits of full-plane Kadanoff-Ceva fermions and energy-energy correlations for generic, non-degenerate $s$ -embeddings of the (near-)critical planar Ising model are described by solutions to conjugate Beltrami equations and Green kernels of uniformly elliptic operators, thereby revealing a richer conformal structure involving quasiconformal mappings and spacelike surfaces in Minkowski space that extends beyond classical Euclidean conformal covariance.

Original authors: Rémy Mahfouf

Published 2026-08-18

📖 7 min read🧠 Deep dive

CC BY 4.0

Original authors: Rémy Mahfouf

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a vast, flat sheet of material, like a piece of fabric or a thin layer of ice, composed of countless tiny atoms. In the world of physics, scientists often study how these atoms interact with one another, particularly when they are arranged in a grid and can exist in one of two states, like a switch that is either on or off. This is known as the Ising model, a mathematical framework used to understand how order emerges from chaos, such as how a magnet suddenly becomes magnetic as it cools down. For decades, researchers have been fascinated by what happens when this system is observed from a great distance, a state called the "scaling limit." In this distant view, the individual atoms blur together, and the system's behavior is usually described by the smooth, predictable laws of classical geometry, much like the way a coastline looks like a simple curve from a plane.

However, this familiar picture assumes the underlying grid of atoms is perfectly regular, like the squares on a chessboard. What happens if the grid is irregular, or if the connections between the atoms vary wildly? For a long time, physicists believed that even in these messy, disordered situations, the large-scale behavior would still settle into the same smooth, predictable patterns found in the regular cases. This paper challenges that assumption by investigating a specific type of irregular grid, known as an s-embedding, where the connections between atoms are not uniform. The researcher set out to see if the rules of geometry change when the grid itself is distorted, and whether the standard laws of physics still apply or if something entirely new takes over.

The study reveals that the old rules do not hold for all irregular grids. When the grid is irregular, the system does not simply smooth out into the familiar Euclidean geometry we learn in school. Instead, the large-scale behavior of the system is governed by a more complex and flexible type of geometry. The researcher found that the patterns formed by the atoms are best described not by simple curves, but by shapes that can be stretched and squeezed in specific ways without tearing. In mathematical terms, these shapes are related to quasiconformal mappings, which are transformations that preserve angles locally but allow for distortion globally. This means that the "fabric" of the system is not rigid; it has an internal flexibility that allows it to adapt to the irregularities of the underlying grid.

To reach this conclusion, the author developed a new way of looking at the connections between the atoms. They focused on specific measurements, called correlations, which tell us how the state of one atom influences another far away. In a regular grid, these correlations follow a simple, well-known formula. But on the irregular grids studied here, the researcher discovered that these formulas become much more complicated. They showed that the correlations are determined by solutions to a specific type of equation that describes how waves or fields behave in a distorted space. This equation is a generalization of the ones used for regular grids, but it accounts for the unevenness of the landscape. The researcher proved that as the grid becomes finer and finer, approaching a continuous surface, these complex correlations converge to a precise, predictable limit. This limit is not the same for every irregular grid; it depends entirely on the specific geometry of the grid itself.

One of the most striking findings is that this new geometric structure is not just a mathematical curiosity; it is a fundamental feature of the system. The researcher demonstrated that the system naturally lives in a richer geometric world than previously thought. They showed that the "distance" between points in this system is not measured by a straight line, but by a path that is influenced by the local shape of the grid. This leads to a new kind of symmetry, where the laws of physics remain consistent even as the grid is stretched and warped, provided the stretching follows certain rules. This discovery confirms a long-standing prediction that the behavior of these systems is far more diverse than the standard models suggested, yet it also reveals a deep universality: for a broad class of critical doubly-periodic graphs, the near-critical scaling exponents and correlation functions match those of the critical square lattice, proving that universality holds even in these periodic settings.

The paper also explores what happens when the system is confined to a specific shape, like a circle or a square, rather than an infinite plane. In these bounded regions, the researcher found that the correlations between atoms follow a twisted version of the usual rules. The way the system responds to the edges of the container depends on the local geometry of the grid, leading to a unique pattern of influence that varies from place to place. This is different from the uniform behavior seen in regular grids, where the influence of the edges is the same everywhere. The researcher provided explicit formulas for these patterns, showing how they can be calculated based on the specific shape of the grid and the nature of the distortions.

Furthermore, the study connects these findings to a broader concept in physics known as conformal field theory, which describes how systems behave at critical points, such as the moment a material changes phase. The researcher showed that the standard version of this theory, which works perfectly for regular grids, is only a special case of a much larger family of theories. The new results reveal that for irregular grids, the theory must be modified to include these quasiconformal mappings. This means that while the "critical" behavior of the system is universal for periodic graphs, it depends on the specific details of the grid's geometry for general irregular embeddings.

The author also looked at the energy of the system, which is a measure of how much work is required to change the state of the atoms. They found that the way energy is distributed across the grid follows a pattern that is directly linked to the geometric distortions. In regions where the grid is highly irregular, the energy distribution changes in a way that reflects the local stretching and squeezing of the space. This provides a direct link between the microscopic arrangement of the atoms and the macroscopic behavior of the system, showing that the geometry of the grid leaves a clear fingerprint on the energy.

In summary, this work fundamentally changes our understanding of how complex systems behave on irregular grids. It proves that the smooth, predictable laws of classical geometry are not the only possibility. Instead, these systems can exhibit a rich variety of behaviors, governed by a flexible geometry that adapts to the underlying structure. The researcher has provided the mathematical tools to describe these behaviors, opening the door to a deeper understanding of phase transitions and critical phenomena in disordered environments. This is not just a refinement of existing theories; it is an expansion of the very framework we use to describe the physical world, showing that nature is capable of far more geometric diversity than we had imagined, while also confirming that deep universal laws persist in specific periodic arrangements.

Technical Summary: From the Planar Ising Model to Quasiconformal Mappings

Problem Statement
The paper addresses the scaling limit of the planar Ising model on generic, non-degenerate $s$ -embeddings. While the critical Ising model on regular lattices (e.g., the square lattice) and isoradial graphs is known to exhibit conformal invariance, and near-critical regimes on specific structures are described by massive Dirac equations, the behavior of the model on general, irregular planar graphs with arbitrary weights remained unresolved. Specifically, the author seeks to identify the continuum limit of Kadanoff–Ceva fermions and energy-energy correlations when the underlying discrete geometry is not uniform or integrable. The central question is whether the scaling limit remains within the realm of standard conformal field theory or if it necessitates a broader geometric framework, such as quasiconformal mappings and Lorentzian geometry.

Methodology
The author employs the framework of $s$ -embeddings, a combinatorial construction introduced by Chelkak that associates a weighted planar graph with a geometric tiling of the plane by tangential quadrilaterals. This embedding encodes both the graph combinatorics and the Ising coupling constants.

Key methodological components include:

Criticality Conditions: The analysis relies on two primary assumptions for the sequence of $s$ $s$ -embeddings $(S_\delta)_{\delta>0}$ $(S_{δ})_{δ > 0}$ :
- Lip $(\kappa, \delta)$ : A condition ensuring the associated "origami map" $Q_\delta$ (a real-valued function defined on the graph vertices) is $\kappa$ -Lipschitz above the scale $\delta$ . This condition characterizes the model as being in a (near-)critical phase, preventing the origami map from becoming space-like in a degenerate manner.
- Exp-Fat $_0(\delta, \rho(\delta))$ : A non-degeneracy condition ensuring that regions with exponentially small geometric features (e.g., tiny incircles in the tangential quadrilaterals) do not percolate to macroscopic scales.
Bosonization and S-graphs: The Ising model is mapped to a dimer model via bosonization. This allows the author to utilize the theory of $t$ -embeddings and $S$ -graphs (a subclass of T-graphs). The Kadanoff–Ceva fermions are related to gradients of harmonic functions on these $S$ -graphs, which support a specific continuous-time random walk.
Green Kernel Identification: A crucial technical innovation is the identification of a primitive of the Kadanoff–Ceva fermion with a full-plane Green kernel for a random walk on the $S$ -graph. Unlike previous approaches on integrable lattices that relied on explicit discrete exponential functions, this method uses the probabilistic properties of the random walk (specifically, its invariant measure and harmonic properties) to determine the correct local scaling factors.
Continuum Limits: The discrete observables are shown to converge to solutions of conjugate Beltrami equations. The limiting origami map $\vartheta$ (the limit of $Q_\delta$ ) defines a space-like surface in Minkowski space $\mathbb{R}^{(2,1)}$ . The scaling limits are described via quasiconformal mappings and, in specific cases, solutions to massive Dirac equations on this surface.

Key Contributions and Results

Full-Plane Scaling Limits: The paper establishes the first rigorous convergence results for Kadanoff–Ceva fermions on generic $s$ -embeddings. The scaling limit is expressed in terms of Green kernels for solutions to conjugate Beltrami equations with prescribed singularities.
- Theorem 2.1: Provides the scaling limit of two-point fermionic correlators $\langle \chi_c \chi_a \rangle$ . The limit is given by a generalized Green kernel involving a function $F_\vartheta$ that satisfies a conjugate Beltrami equation determined by the limiting origami map $\vartheta$ . The scaling factors are local and depend explicitly on the geometry of the embedding (edge lengths and radii of tangential quadrilaterals).
- Theorem 2.2: Derives the scaling limit of energy-energy correlations. The result generalizes the known critical square lattice formula, showing that the covariance is governed by the difference of squared moduli of the Beltrami kernels $F_\vartheta$ and $F^\star_\vartheta$ .
Bounded Domains and Conformal Covariance:
- Theorem 2.3: Proves the conformal covariance of the energy density for critical doubly-periodic graphs. This extends universality results beyond isoradial lattices to general periodic structures, confirming that the energy density scales with the hyperbolic metric element of the domain.
- Theorem 2.4: Establishes a "twisted" conformal covariance rule for energy correlations on maximal surfaces in $\mathbb{R}^{(2,1)}$ . When the limiting origami map $\vartheta$ satisfies the minimal surface equation (maximal surface), the scaling limit involves a metric distortion reflecting the Lorentzian geometry of the lift.
- Theorem 2.5: Extends the results to smooth limiting surfaces where $\vartheta$ is $C^2$ . In this regime, the scaling limit is described by solutions to a massive Dirac equation on the unit disk, where the mass term is geometrically interpreted as the mean curvature of the surface in $\mathbb{R}^{(2,1)}$ .
Universality of Exponents: The paper demonstrates that for critical doubly-periodic graphs, all near-critical scaling exponents match those of the standard square lattice. This is achieved by combining the derived energy covariance results with Edwards-Sokal coupling and known RSW (Russo-Seymour-Welsh) estimates.

Significance and Claims
The paper claims to confirm predictions made in prior works (specifically [19], [26], [25]) that the scaling limits of generic (near-)critical Ising models naturally reside in a richer conformal structure than the classical Euclidean one.

Quasiconformal Nature: The results rigorously realize the picture that Ising fermions and energy observables on generic graphs are described by quasiconformal mappings and solutions to conjugate Beltrami equations, rather than standard holomorphic functions.
Geometric Scaling: A major contribution is the explicit identification of local scaling factors that vary from region to region based on the embedding geometry. This contrasts with previous results on isoradial grids where scaling factors were uniform.
Lorentzian Geometry: The work highlights the relevance of Minkowski geometry $\mathbb{R}^{(2,1)}$ in describing correlations. The limiting origami map $\vartheta$ defines a space-like surface, and the deviation from criticality (or the specific nature of the critical limit) is encoded in the geometry of this surface (e.g., mean curvature for smooth limits, maximality for twisted covariance).
Technical Breakthrough: The author notes that their approach overcomes the obstruction of missing discrete exponential functions on irregular grids. By identifying the fermion primitive with a Green kernel, they provide a robust method for computing scaling limits even in highly degenerate local geometries, a task previously thought to require integrability.

The paper concludes that these findings provide a complete universality statement for critical doubly-periodic graphs and lay the groundwork for understanding the Ising model in random environments and non-integrable settings, though the convergence of spin correlations (as opposed to energy/fermion observables) remains an open problem in this general setting.

More like this