Understanding cracked materials: is Linear Elastic Fracture Mechanics obsolete?

Linear elastic fracture mechanics has enabled the research community to solve a wide variety of problems of practical and scientific interest; however, it has historically suffered from two main shortcomings. Firstly, it predicts physically unrealistic singular stresses and strains at crack tips. Secondly, microstructural effects are lacking, so that a major source of size-dependent behaviour is not captured. Gradient-enriched elasticity overcomes both these shortcomings: singularities are avoided, so that crack-tip stresses can be used to assess integrity, and the inclusion of microstructural terms implies that size effects can be captured. In this investigation, it is shown that gradient-enriched crack tip stresses can directly be used to model the transition from the short to the long crack regime. The accuracy of this approach was validated by a wide range of experimental results taken from the literature and generated under both static and high-cycle fatigue loading. This high level of accuracy was achieved without having to resort to phenomenological model parameters: the extra constitutive coefficient was simply the (average) grain size of the material.