[Paper Review] SLE(6) and the geometry of diffusion fronts

Pierre Nolin|ArXiv.org|Dec 18, 2009

Stochastic processes and statistical mechanics1 references3 citations

TL;DR

This paper rigorously establishes that the geometry of diffusion fronts in a two-dimensional particle diffusion model exhibits fractal interfaces with Hausdorff dimension 7/4, using connections to near-critical percolation and SLE(6). By modeling particle positions via random walks and analyzing the resulting cluster boundaries, the authors prove that the interface fluctuations scale with exponent 5/7 and are described by the Schramm-Loewner Evolution SLE(6), confirming a long-standing conjecture from statistical physics.

ABSTRACT

We study the diffusion front for a natural two-dimensional model where many particles starting at the origin diffuse independently. It turns out that this model can be described using properties of near-critical percolation, and provides a natural example where critical fractal geometries spontaneously arise.

Motivation & Objective

• To mathematically describe the geometry of diffusion fronts arising from many independent random walks starting at the origin.
• To establish that the interface separating occupied and vacant regions in this model is fractal with dimension 7/4.
• To connect the diffusion front model to near-critical percolation and SLE(6) processes, providing a rigorous foundation for earlier numerical and physical conjectures.
• To analyze the scaling behavior of the interface's length and fluctuations in the near-critical regime.

Proposed method

• Model the system as a large number of particles performing independent simple random walks on the triangular lattice.
• Define the occupied set as sites visited by at least one particle at time t, and study the boundary of the macroscopic cluster containing the origin.
• Use the connection between random walk local times and percolation thresholds to map the diffusion process to an inhomogeneous gradient percolation model with site occupation probability p(z) = P(at least one particle visits z).
• Apply results from near-critical percolation theory, including Kesten’s scaling relations and Smirnov’s conformal invariance, to analyze the interface geometry.
• Use SLE(6) theory to characterize the scaling limit of the interface, leveraging known results on the Hausdorff dimension 7/4 for SLE(6) curves.
• Establish asymptotic estimates for the particle occupation probability ρ_t(z) and use integral approximations to derive the radial profile of the front, showing localization near r = r_{μ,t}^*.

Experimental results

Research questions

• RQ1 Does the interface of the diffusion front in the 2D particle diffusion model exhibit fractal geometry with dimension 7/4?
• RQ2 Can the geometry of the diffusion front be rigorously linked to the SLE(6) process and near-critical percolation?
• RQ3 What is the scaling behavior of the interface length and its fluctuations around the macroscopic radius?
• RQ4 How does the interface localize in space, and what determines its radial position in the near-critical regime?

Key findings

• The diffusion front interface has a Hausdorff dimension of exactly 7/4, matching that of SLE(6) curves.
• The interface is localized within an annulus of width t^{2/7+ε} around the circle of radius r_{μ,t}^*, which scales as √t.
• The discrete length L_t of the interface satisfies t^{5/7−ε} ≤ L_t ≤ t^{5/7+ε} for any ε > 0, indicating a power-law scaling with exponent 5/7.
• The radial profile of the occupation probability ρ_t(z) converges to an integral of e^{-u}/u, confirming the localization of the front near the critical percolation threshold.
• The interface fluctuations are larger than t^{2/7−ε} both inward and outward, confirming the roughness of the front.
• The model with a stationary source of particles (e.g., ink stain) yields the same asymptotic interface geometry, with the same scaling exponents and localization behavior.