npj Computational Materials (Apr 2017)

A phase field model for snow crystal growth in three dimensions

  • Gilles Demange,
  • Helena Zapolsky,
  • Renaud Patte,
  • Marc Brunel

DOI
https://doi.org/10.1038/s41524-017-0015-1
Journal volume & issue
Vol. 3, no. 1
pp. 1 – 7

Abstract

Read online

Spontaneous patterns: Simulating snowflakes with a softer touch A model that reproduces complex 3D snowflake growth using versatile interface descriptors may benefit other dendritic materials. Snow crystals solidify by expanding outward from an initial seed, capturing water molecules as they travel through the atmosphere. While most simulation methods treat this growing interface as a sharp boundary, Gilles Demange and colleagues from the University of Rouen in France report that a less rigid approach yields highly realistic results. Their technique uses a phase field model to represents the snowflake’s surface as a thin moveable layer where ice and vapour mix, and a new surface tension function to explain the anisotropic crystallisation. Including a special algorithm to simulate 3D crystal faceting enabled the model to duplicate essential snowflake morphologies and potentially predict ice water content in clouds under various weather conditions.