MicroFract: An image based code for microstructural crack path prediction

Brittle failure is prevalent in ceramics, fiber composites and metals where the microstructural cracks propagate in the path of least resistance following weak interfaces and grain boundaries. We formulate an image-based methodology for identifying the path of least resistance in the microstructure. In this approach, a user specifies the line direction of the macroscopic crack and a cohesive energy map corresponding to the weak and strong interfaces in the microstructure. The path of least resistance is formulated as a process of minimizing a discretized cost function based on an Ising-type energy model. The main advantage of this algorithm is that it can be exactly solved by the max-flow min-cut theorem. The results are shown via 2D examples from literature and the code and examples are made available to the community. Keywords: Graph cuts, Fracture, Microstructure, Image analysis, Crack path prediction