Large deviations for the macroscopic motion of an interface (Q2628653): Difference between revisions

We have a detailed study about the law that governs the power needed to force a motion of a planar interface between two different phases of a given ferromagnetic sample with a prescribed speed $V$. Usually it is done on a lattice model of Ising-spins with Glauber dynamics by a multi-scale procedure based on the interaction range of the Kac-potential. However, this approach meets some difficulties and the obtained results are biased by several possible errors. In order to avoid this, the authors use a different technique by introducing coarse-grained time-space-magnetization boxes. Fixing a space-time $(\xi,\tau)$ scale and considering the example, where $\xi = 0$ (at $\tau = 0$) changes to a final position $\xi = R$ within a fixed time $T$, the interface moves to a final position. For such a motion with constant velocity $V=R/T$, the needed power is given by $V^2/\mu$, where $\mu$ is a mobility coefficient. This allows to verify the limits of validity of this law in a stochastic model of interacting spins. \par The paper is divided into 7 sections where the model and preliminary results are successively presented: -- Section 2, the main results with Theorem 3.3 -- Section 3. Considerations about extremal cases, i.e., less or more that fixed $N$ spin flips are given in Section 4. Normal cases of time intervals between spin flips are analyzed in Section 5, whereas Section 6 considers again extremal cases. Section 7 gives derivation of the cost functional.