Crack propagation in disordered materials: how to decipher fracture surfaces

For a half-century, engineers know how to describe and predict the propagation of a crack in a model elastic homogeneous medium. The case of real materials is much more complex. Indeed, we do not know how to relate their lifetime or their resistance to their microstructure. To achieve such a prediction, understanding the role of the microstructural disorder on the behavior of a crack is determinant. Fracture surfaces represent a promising field of investigation to address this question. From the study of various disordered materials, we propose a statistical description of their roughness and determine to which extent their properties are dependent of the material. We show that fracture surfaces display an anisotropic scale invariant geometry characterized by two universal exponents. Glass ceramics is then studied because its microstructure can be tuned in a controlled manner. Their fracture surfaces display the same general anisotropic properties but with surprisingly low exponents independent of the detail of the ceramics microstructure. This suggests the existence of a second universality class in failure problems. Using finally theoretical tools from out-of-equilibrium statistical physics and fracture mechanics, we relate the statistical properties of fracture surfaces with the mechanisms occurring at the microscopic scale during the failure of a material. In particular, we show that the first class of fracture surfaces results from a failure involving damage processes while the second one results from a perfectly brittle failure.