Abstract
Fracture mechanisms in solid materials have been extensively studied. Although cracks are also commonly seen in soft solids, the fracture process is still not very well understood for these materials. In this thesis we choose to study fracture on a particular class of materials: complex fluids. We will focus on one particular family of complex fluids which are self-assembled transient gels. These viscoelastic gels have the property to flow at long timescale while behaving as an elastic solid at short timescales. We have investigated three model systems: a bridged micro emulsion and a entangled solution of wormlike micelle, and a “hybrid” system made of bridged micelles of tunable morphology. These systems are at thermodynamic equilibrium and behave as Maxwell fluids but they differ in microscopic structures. Bridged micro emulsions are made of surfactant-stabilzed oil droplet dispersed in water and bridged by telechelic polymers. Wormlike micelles are long semi flexible aggregates made from the self-assembly of surfactant in a water solution. Lastly, bridged micelles are made of surfactant aggregates of controllable shape (sphere -> cylinder -> worm) in water bridged by telechelic polymers. We choose to study these different systems in a confined geometry: a radial Hele-Shaw cell. The Hele-Shaw cell is made of two glass plates separated by spacers of controllable thickness. A hole is pierced in the center of the cell for injecting the fluids. The experiments consist in the injection at a controlled rate of low viscosity oil inside the highly viscous gel. Because of the high viscosity contrast between the two fluids, the oil/gel interface is unstable. Depending of the injection rate, we observed different instabilities. At lowest rates, an instability of visco-capillary origins appears and the oil/gel interface is deformed leading to a viscous fingering pattern. This instability called Saffman Taylor instability is widely known and has been extensively studied for Newtonian fluids. At highest rates another instability patterns arise of elasto capillary origin where the patterns are vastly different from the previous one and are made of cracks propagating through the gel. We have quantified the difference between the two types of instability. By combining direct visualization using high speed imaging and digital image correlation techniques we have characterized the displacement field of the gel around the crack tip, and in particular how its amplitude decays away from the tip. For bridged microemulsion, we have also evidenced the existence of a velocity discontinuity between the crack velocity and the velocity of the gel near the crack tip whereas no discontinuity occurs in the case of viscous fingering. Using bridged micelles of tunable morphologies we have also shown that the transition between the two instabilities is controlled by the viscoelasticity of the gel. Finally, for gel that can reorient under flow we have measured a birefringence signal associated to these reorganization. By studying this signal at the crack tip we were able to perform a measurement of the size of the “process zone” which could be considered as the first macroscopic quantitative analysis of the ductility of a crack in complex fluids. During complementary experiments which consist of the injection of wormlike micelles in themselves we have reported a new kind of flow instability. This instability is characterized by the transient loss of the radial symmetry during flow and by the apparitions of typical “branches” which propagates at very high speed through the sample and finally distort the air/gel interface.