“…This cohesive surface formulation, Needleman (1987), differs from the classical cohesive zone models of Barenblatt (1962) and Dugdale (1960) in several respects; for example, an initial crack is not required to be present so that issues of nucleation of defects can be addressed and the length of a cohesive zone is not a parameter of the model. This framework has been used to address issues regarding void nucleation (Needleman 1987;Tvergaard 1990;Xu and Needleman 1993), quasi-static crack growth (Needleman 1990;Tvergaard andHutchinson 1992, 1993), stability of the separation process (Suo et al 1992;Levy 1994), reinforcement cracking in metal-matrix composites (Finot et al 1994), dynamic crack growth (Xu and Needleman 1994;Siegmund andNeedleman 1996, 1997), and interface fracture (Tvergaard and Hutchinson 1993;Needleman 1995, 1996).…”