Special Session 9: Stochastics and randomness in physical models

Moments estimates and sharp interface limit for the stochastic Cahn-Hilliard equation
Dimitra Antonopoulou
National and Kapodistrian University of Athens
Greece
Co-Author(s):    
Abstract:
The phase separation process in a binary alloy is parameterized by a small positive parameter $\varepsilon$ representing the order of the width of the evolving layers. Stochastic versions for the equations modelling the evolution of the concentration of the phases arise due to thermal fluctuations, external mass supply or impurities in the alloy. We consider the $\varepsilon$-dependent stochastic Cahn-Hilliard equation with multiplicative and sufficiently regular in space noise with strength of order $O(\varepsilon^\gamma)$, $\gamma > 0$ in dimensions $d = 1,2,3$. We prove $p$-moments estimates in $H^1$, $H^2$ and $L^\infty$ norms. When the initial data are layered by using the energy ($H^1$) estimate we prove that, as $\varepsilon$ tends to $0$, the solution $u\rightarrow \pm 1$ in the $L^2$ norm with probability tending to $1$. This implies the complete separation of the two phases on the sharp interface limit even in the presence of noise.