Mixed-mode fatigue fracture of adhesive joints in harsh environments and nonlinear viscoelastic modeling of the adhesive

A four point bend, mixed-mode, reinforced, cracked lap shear specimen experimentally simulated adhesive joints between load bearing composite parts in automotive components. The experiments accounted for fatigue, solvent and temperature effects on a swirled glass fiber composite adherend/urethane adhesive system. Crack length measurements based on compliance facilitated determination of da/dN curves. A digital image processing technique was also utilized to monitor crack growth from in situ images of the side of the specimen. Linear elastic fracture mechanics and finite elements were used to determine energy release rate and mode-mix as a function of crack length for this specimen. Experiments were conducted in air and in a salt water bath at 10, 26 and 90°C. Joints tested in the solvent were fully saturated. In air, both increasing and decreasing temperature relative to 26°C accelerated crack growth rates. In salt water, crack growth rates increased with increasing temperature. Threshold energy release rate is shown to be the most appropriate design criteria for joints of this system. In addition, path of the crack is discussed and fracture surfaces are examined on three length scales.; Three linear viscoelastic properties were measured for the neat urethane adhesive. Dynamic tensile compliance (D*) was found using a novel extensometer and results were considerably more accurate and precise than standard DMTA testing. Dynamic shear compliance (J*) was determined using an Arcan specimen. Dynamic Poisson's ratio (nu*) was extracted from strain gage data analyzed to include gage reinforcement. Experiments spanned three frequency decades and isothermal data was shifted by time-temperature superposition to create master curves spanning thirty decades. Master curves were fit to time domain Prony series. Shear compliance inferred from D* and nu* compared well with measured J*, forming a basis for finding the complete time dependent material property matrix for this isotropic material.; A constitutive model is introduced which replaces time with internal energy in time-temperature superposition. Internal energy for mechanical loading was calculated from stress history and time domain Prony series representation of compliance. The model also included pressure and volume effects. Ramp loading experiments conducted at strain rates spanning three decades were effectively predicted, but unloading predictions were poor.