betaFeedback

WorksMathematics

Mean Curvature Flow with Thermal Fluctuations in Dimensions Two and Higher

Abstract

Surface tension drives interfaces towards smaller area, while thermal motion produces shape fluctuations whose strength is linked to the interfacial mobility through the fluctuation–dissipation relation. These mechanisms lead to mean-curvature flow with thermal noise, whose graph formulation is known as the Kawasaki–Ohta equation. We establish global-in-time existence of energy martingale solutions to the spatially regularized Kawasaki–Ohta equation for periodic graphs in every dimension d≥2d\geq2, with initial data in H1(Td)H^1(\mathbb T^d). No restriction on the noise intensity is needed for d=2d=2, while a smallness condition is required for d≥3d\geq3. The key new challenges arise from the thermal noise, which has a (1+∣∇u∣2)1/4(1+|\nabla u|^2)^{1/4}-type coefficient. This coefficient gives rise to an irregular Stratonovich-to-Itô correction term of the form ∇u⋅D2u∇u/(1+∣∇u∣2)3/2\nabla u\cdot D^2u\nabla u/(1+|\nabla u|^2)^{3/2}. Our main innovation is a compactness argument tailored to the structure of the correction. By combining a factorization of this term with H1(Td)H^1(\mathbb T^d) and area estimates, we establish strong convergence of the normalized gradients and reconstruct the weighted Hessian, which allows us to define the correction at energy regularity and identify its approximation limit.

Paper details

vlnc:W_9wbbhmwt
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
arXiv nonexclusive license
Usage

views · 10

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

ChatGPT 5.6 Pro

View statement
Looking for related papers…