ABSTRACT
Crack growth occurs much more slowly than for the tensile loading case and crack growth eventually arrests. For the imposed loading and with the cohesive surface characterization used in this calculation (the impact velocity amplitude of 10 m/s corresponds to a stress of 25 MPa, which is about 7.5% of the cohesive surface strength), extensive sliding and interpenetration occur somewhat after the stage of deformation shown in Fig. 2. A more accurate cohesive surface constitutive description, i.e. one modeling frictional sliding resistance, accounting for interactions other than through the initial cohesive surface, and a finer mesh would be required for a quantitative prediction of cracking under compressive loading. Nevertheless, the development of brittle fracture in Fig. 2 bears a remarkable resemblance to that seen in the photographs in Horii and NematNasser [11], which pertain to cracking under quasi-static compressive loading. It is worth noting that the fracture mode transition seen in Figs. 1 and 2 is an outcome of the initial-boundary value problem solution and occurs even without details of the crack tip fields being resolved. It should also be noted that the configurations shown in Figs. 1 and 2 and subsequently are obtained from the actual displacement magnitudes calculated from the finite deformation formulation. The remaining figures show results of fine mesh calculations for a center cracked 20 mm×2 mm block for three cases; (i) symmetric tensile loading with an impact velocity amplitude of 10 m/s on the loaded faces, (ii) asymmetric tensile loading with an impact velocity amplitude of 20 m/s on the loaded face and (iii) crack growth constrained to be along the initial crack line. Symmetric tensile loading with an impact velocity amplitude of 10 m/s was also used in the Figure 3: Crack advance, Va, versus time, t, for three calculations for a 20 mm ×2 mm block.
