Recent attempts to model the failure of brittle materials via  particle based methods (e.g. molecular dynamics simulations) and  discrete models from statistical physics (random fuse model, random  spring model, etc.) have largely been focused on geometries such as  uniaxial tension (mode I) or anti-plane strain (mode III).  They are  not appropriate for modeling the formation of systems with in-plane  shear (mode II) where one expects qualitatively different emergent  properties.

I will present some preliminary results on a novel technique to allow  for particle-based modeling   of brittle systems in 2D planar shear  geometry.  The model will be shown to exhibit qualitative behavior  which is strikingly similar to behaviors found both in laboratory  experiments on sheared rock and at geologically relevant scales.