Influence of Microstructural Disorder and Wavefield in Dynamic Fracture

Dynamic fracture and its instabilities have been widely studied but the influence of the finite sample size and subsequent 3D aspects are generally neglected. However, a sample of a few centimeter is a waveguide for the elastodynamic field emitted by the propagating crack front (from 100kHz to a few GHz): It excites the sample's free oscillations (or normal modes), and creates a fluctuating landscape of elastic energy. This may be seen as an effective noise, with an amplitude proportional to the frequency of a given mode, which can reach the same order of magnitude as that of the fracture toughness (In PMMA: 103 J.m-2 for f ∼ MHz). We designed an experiment to evidence this effect in a homogeneous brittle material (PMMA) and subsequently to characterize the possible coupling between the fracture front and its wavefield. Dynamic cracks are driven by means of a wedge splitting geometry which allow us to modulate, over a wide range, the velocity of the crack tip. Spatial geometry and frequency content of the emitted wavefield are modulated by adjusting the geometry of the sample and the loading conditions. Hints of the wavefield are looked in the high-frequency fluctuations of the crack speed, measured on both sides of the specimen via a state-of-the art potential drop method. Fractography and statistical analysis of the post-mortem fracture surfaces are used to characterize the mesoscale/microstructure scale response of the crack front to the wavefield. Experiments performed in PMMA will finally be compared to others performed on heterogeneous materials with controlled defects size (40 - 500µm). This study will permit (i) to shed light on the key role of elastic wavefield in dynamic fracture, and how those are selected by the sample geometry and microstructure and finally and (ii) to give some leads on how to account for these effects by adapting the paradigm of interface growth model to the case of dynamic fracture.