Article Preview
TopIntroduction
Diamond films, using chemical vapor deposition (CVD) technologies, have been exploited to realize to coat commercial tungsten-carbide (WC) cutting tools as an alternative to polycrystalline diamond (PCD) tools because of reduced fabrication costs, though the tool life per cutting tip is still inferior to PCD counterparts. From the literature, experiments testing CVD diamond-coated tool performance have confirmed that coating delamination is the major failure mode of CVD diamond-coated tools (Chou & Liu, 2005).
Interface adhesion is an essential issue in most coating delamination, as has been demonstrated from theoretical and experimental viewpoints. Griffith’s theory of brittle fracture provides the fundamentals of fracture mechanics (Griffith, 1920; Griffith, 1924). His ideas were not limited to brittle fracture, and further, Irwin (1957; 1964) succeeded his work and applied to ductile materials. Fracture toughness is critical to understand energy dissipation mechanisms in the delamination process zone, which characterizes the onset of fracture during the fracture process. Brocks (2005) stated that, generally, two types of mechanisms are developed. One is the emission and motion of dislocations from the crack tip (Rice, 1992; Rice & Thomson, 1974; Rice & Tracey, 1969; Thomason, 1985; Tvergaard, 1982), a mechanism involving the formation, growth, and coalescence of voids. To address the interface fracture, a crack problem, a cohesive zone model is introduced ahead of the crack tip to simulate the material degradation and separation. Barenblatt (1962) and Dugdale (1960) first formulated and applied the concept of cohesive zone models (CZM) with the traction-separation law (Figure 1) to interpret interface decohesion involving the crack initiation, growth, and coalescence. In their model, the interface traction first increases with the separation until it reaches a maximum value, and then it falls due to interface weakening and eventually decreases to zero. Xu and Needleman (1994) presented a cohesive zone model for simulating dynamic crack growths. Their results agreed with a wide range of experiments on fast crack growth in brittle solids. Later, Nakamura and Wang (2001) applied a cohesive zone model to simulate crack propagations in porous materials. The observed numerical errors from the cohesive elements increasing with the model compliance need to be minimized by carefully choosing the parameters for the cohesive model. Gao and Bower (2004) made some improvement in avoiding the convergence problem in finite element simulations of crack nucleation and growth at cohesive interfaces. They solved the convergence problem in quasi-static finite element computations by introducing a small viscosity in the constitutive equations for the cohesive interfaces. Repetto, Radovitzky, and Ortiz (2000) applied a tension-shear cohesive law model to simulate the dynamic fracture and fragmentation of glass rods. They demonstrated that the cohesive law, unlike damage theories, introduces well-defined fracture energy with spurious mesh-dependencies, such that the cohesive models give materials with a characteristic length. Xia et al. (2007) applied a cohesive zone model, including the factor stability found by Gao and Bower (2004), to simulate coating delamination under contact loading. They established delamination mechanism maps for a strong elastic coating on an elastic-plastic substrate subject to contact loading.
Figure 1. Typical traction-separation response (Hu, Chou & Thompson, 2008)