Anisotropic surface tensions for phase transitions in periodic media

DOI10.1007/S00526-022-02216-5MaRDI QIDQ2127346zbMATH OpenOpenAlexWikidataFDO

Authors Rustum Choksi, Jessica Lin, Raghavendra Venkatraman, Irene Fonseca

Publication date 20 April 2022

Published in Calculus of Variations and Partial Differential Equations (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2010.05849

Allen-Cahn energy functional anisotropic surface energy anisotropic surface tension bound homogenization signed distance function

Mathematics Subject Classification ID

Optimality conditions for free problems in two or more independent variables (49K10) Optimality conditions for problems involving partial differential equations (49K20) Variational problems in a geometric measure-theoretic setting (49Q20) Homogenization in equilibrium problems of solid mechanics (74Q05)

Abstract: This paper establishes bounds on the homogenized surface tension for a heterogeneous Allen-Cahn energy functional in a periodic medium. The approach is based on relating the homogenized energy to a purely geometric variational problem involving the large scale behaviour of the signed distance function to a hyperplane in periodic media. Motivated by this, a homogenization result for the signed distance function to a hyperplane in both periodic and almost periodic media is proven.

Review text

The authors consider the Allen-Cahn energy functional \(\mathcal{F} _{\varepsilon }:H^{1}(\Omega )\rightarrow \lbrack 0,\infty ]\) defined as \( \mathcal{F}_{\varepsilon }(u)=\int_{\Omega }[\frac{1}{\varepsilon }a(\frac{x }{\varepsilon })W(u)+\frac{\varepsilon }{2}\left\vert \nabla u\right\vert ^{2}]dx\), where \(\Omega \sqsubseteq \mathbb{R}^{N}\), \(N\geq 2\), is a Lipschitz domain, \(a:\mathbb{R}^{N}\rightarrow \mathbb{R}\) is a continuous, strictly positive and \(\mathbb{T}^{N}\)-periodic function, \(\mathbb{T}^{N}\) being the standard \(N\)-dimensional torus, which satisfies \(0<\theta \leq a(x)\leq \Theta \), for all \(x\in \mathbb{R}^{N}\), and \(W\) is the double-well potential \(W(u)=(1-u^{2})^{2}\). The authors recall some homogenization results concerning this functional. They introduce the anisotropic surface energy \(\sigma :\mathbb{S}^{N-1}\rightarrow \lbrack 0,\infty )\) through the cell formula \(\sigma (\nu )=\lim_{T\rightarrow \infty }\frac{1}{T^{N-1}}\inf \{\int_{TQ_{\nu }}[a(y)W(u)+\frac{1}{2}\left\vert \nabla u\right\vert ^{2}]dy:u\in C(TQ_{\nu })\}\), where \(C(TQ_{\nu })=u\in \{H^{1}(TQ_{\nu }):u=\rho \ast u_{0,\nu }\) on \(\partial (TQ_{\nu })\}\), with \(u_{0,\nu }(y)=-1\) if \(x\cdot \nu \leq 0\), and \(u_{0,\nu }(y)=1\) if \(x\cdot \nu >0\), and \(\rho \in C_{c}^{\infty }(B(0,1))\), with \(0\leq \rho \leq 1\), and \(\int_{ \mathbb{R}^{N}}\rho (x)dx=1\). They also introduce the function \(q:\mathbb{R} \rightarrow \mathbb{R}\) defined by \(q(z)=\tanh (\sqrt{2}z)\), \(z\in \mathbb{R} \), and for \(\nu \in \mathbb{S}^{N-1}\), they define \underline{\(\lambda \)}\( (\nu )=\lim \inf_{T\rightarrow \infty }\frac{1}{T^{N}}\int_{TQ_{\nu }}[a(y)W(q\circ h_{\nu })+\frac{1}{2}2\left\vert \nabla (q\circ h_{\nu })\right\vert ^{2}]dy\) and \(\overline{\lambda }(\nu )=\lim \sup_{T\rightarrow \infty }\frac{1}{T^{N}}\int_{TQ_{\nu }}[a(y)W(q\circ h_{\nu })+\frac{1}{2}2\left\vert \nabla (q\circ h_{\nu }\right\vert ^{2}]dy\) , where \(h_{\nu }(y)=sign(y\cdot \nu )\inf_{z\in \Sigma _{\nu }}d_{\sqrt{a} }(y,z)\), \(\Sigma _{\nu }=\{x\in \mathbb{R}^{N}:x\cdot \nu =0\}\). The first main result proves the existence of a universal constant \(\Lambda _{0}>0\) and of \(\lambda _{0}:\mathbb{S}^{N-1}\rightarrow \lbrack 0,\Lambda _{0}]\) such that \(\overline{\lambda }(\nu )-\lambda _{0}(\nu )\leq \sigma (\nu )\leq \underline{\lambda }(\nu )\). The second main result proves that for each \(\nu \in \mathbb{S}^{N-1}\), there exists a unique \(c(\nu )\in \lbrack \sqrt{\theta },\sqrt{\Theta }]\) such that for every compact \(K\sqsubseteq \mathbb{R}^{N}\), \(\lim_{T\rightarrow \infty }\sup_{x\in K}\left\vert \frac{1}{T}h_{\nu }(Tx)-c(\nu )(x\cdot \nu )\right\vert =0\), and \(c(\nu )=c(-\nu )\). The authors finally also prove a similar result if \(a:\mathbb{R} ^{N}\rightarrow \mathbb{R}\) is a Bohr almost periodic function. For the proof of the first main result, the authors use the standard De Giorgi slicing technique that they recall in an Appendix to prove the upper bound. For the proof of the lower bound, they choose \(\phi =\sqrt{2}\int_{0}^{z} \sqrt{W(s)}ds\) and they prove properties of the function \(h_{\nu }\) in connection with the signed distance \(d_{\sqrt{a}}(y_{1},y_{2})=\inf_{\gamma (0)=y_{1},\gamma (1)=y_{2}}\int_{0}^{1}\sqrt{a(\gamma (t))}\left\vert \overset{.}{\gamma }(t)\right\vert dt\), where \(\gamma \) is a Lipschitz curve \([0,1]\rightarrow \mathbb{R}^{N}\). For the proof of the second and third main results, the authors prove further properties of the function \(h_{\nu }\) and of Bohr almost periodic functions for the third result.

Recommendations

Cites work

Cited in

(12)

This page was built for publication: Anisotropic surface tensions for phase transitions in periodic media

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2127346)