MATLAB Phase-Field Dendritic Growth Simulation

Resource Overview

A MATLAB-based phase-field simulation program for dendritic growth in pure nickel solidification studies, excluding fluid flow coupling. The implementation models 2D grain evolution using finite difference discretization with periodic boundary conditions, featuring tunable thermophysical parameters for adaptation to other pure metals. Key computational parameters include: anisotropy strength p=36, grid spacing dx=l, time step dt=0.1, gradient coefficient k=0.5, latent heat L=1.0, linear coefficient m=1.0, and 300x300 grid nodes.

Detailed Documentation

Application Background This MATLAB-implemented phase-field simulation program models dendritic growth during solidification processes, specifically designed for pure nickel without fluid flow coupling. The code architecture permits adaptation to other pure metals through modification of thermophysical parameters in the material property definitions. Technical Implementation The algorithm simulates 2D grain evolution using finite difference discretization over a square grid domain with periodic boundary conditions. The initial phase-field values are generated as random numbers following normal distribution within (-0.001, 0.001) range, implemented via MATLAB's randn() function with scaling. Secondary phase particles are approximated as circular obstacles through geometric masking functions. Core simulation parameters include: anisotropy coefficient p=36 controlling interfacial energy orientation, spatial discretization dx=l, temporal step dt=0.1 ensuring numerical stability, gradient energy coefficient k=0.5, latent heat parameter L=1.0 governing solidification kinetics, linear coefficient m=1.0, and computational domain resolution of 300x300 grid points. The modular code structure allows for future enhancements such as fluid dynamics coupling through additional Navier-Stokes solvers, enabling more comprehensive solidification behavior analysis. Current implementation employs explicit time integration with spectral methods for efficient phase-field equation solution.