Standard methods for testing fracture toughness of homogeneous materials are currently being applied to welds. A weld is, by nature, a heterogeneous structure containing zones with different microstructures and mechanical properties. The validity of using fracture parameters developed for homogeneous materials to characterize fracture of welds has not been established. Tests were conducted to investigate whether the extension of a crack along an interface is governed by the same fracture parameters used to predict fracture in homogeneous materials. The tests involved measuring the displacement fields around a crack tip in the interface between two materials with matched elastic properties and differing yield strengths. The form of the displacement distribution was then compared with the theoretical displacements for linear elasticity and nonlinear elasticity with power-law hardening.

Detailed three-dimensional nonlinear finite element (FE) analyses and experimental moire studies are performed on a plate containing a moderately deep part-through surface crack to establish limits of HRR-dominance. The plate is subjected to predominantly far-field tensile loading. The material under investigation is ASTM A710 steel, which was constitutively modeled by large deformation J 2 flow theory of plasticity. The FE mesh was carefully constructed to resolve both crack front fields (such as J -integral and CTOD) and global fields (such as surface displacements, strains). By comparing the J-integral and CTOD results with an earlier HRR-dominance study using (small strain) deformation theory of plasticity, we found little effect of the different formulations on the crack front fields. The global deformation fields from the numerical simulation are in good agreement with our experimental results. The eventual loss of HRR-dominance is intimately related to the interaction of the global plastic flow fields with those of the crack front.