“…On the other hand, in dynamic fracture simulations large system sizes are necessary because of the boundary effects such as wave reflection and the problem of dislocations near fixed boundaries. Thus, simulations have so far mostly been done by using a simple pair potential in a twodimensional system 6,7 or in a system with otherwise less degrees of freedom. 8 Until recently, three-dimensional ͑3D͒ simulations with a realistic many-body potential for describing the behavior of ductile materials, e.g., copper, could be realized only in small systems.…”
Mechanical properties of copper with various types of defects have been studied with the moleculardynamics method and the effective-medium theory potential both at room temperature and near zero temperature. The loading has been introduced as constant rate straining and the dynamics of the process region of fracture is purely Newtonian. With the model three types of defects were studied: point defects, grain boundary, and an initial void serving as a crack seed. Point defects were seen to decrease the system strength in terms of fracture stress, fracture strain, and elastic modulus. Due to random microstructure, highly disordered systems turned out to be isotropic, which on the other hand seems to increase the elastic modulus. In the case of a grain boundary, the elastic modulus was found to be significantly less than the bulk value of the system. In addition, the critical strain for crack initiation seems to be less at the grain boundary than in the bulk. In the case of an initial void, we studied stress concentration, dislocation propagation, and crack propagation in thin systems. The stress concentration was found to be in surprisingly good agreement with continuum predictions. Dislocation and crack were propagated with a velocity much below the speed of sound and they preferred the ͗110͘ crystal orientation.
“…On the other hand, in dynamic fracture simulations large system sizes are necessary because of the boundary effects such as wave reflection and the problem of dislocations near fixed boundaries. Thus, simulations have so far mostly been done by using a simple pair potential in a twodimensional system 6,7 or in a system with otherwise less degrees of freedom. 8 Until recently, three-dimensional ͑3D͒ simulations with a realistic many-body potential for describing the behavior of ductile materials, e.g., copper, could be realized only in small systems.…”
Mechanical properties of copper with various types of defects have been studied with the moleculardynamics method and the effective-medium theory potential both at room temperature and near zero temperature. The loading has been introduced as constant rate straining and the dynamics of the process region of fracture is purely Newtonian. With the model three types of defects were studied: point defects, grain boundary, and an initial void serving as a crack seed. Point defects were seen to decrease the system strength in terms of fracture stress, fracture strain, and elastic modulus. Due to random microstructure, highly disordered systems turned out to be isotropic, which on the other hand seems to increase the elastic modulus. In the case of a grain boundary, the elastic modulus was found to be significantly less than the bulk value of the system. In addition, the critical strain for crack initiation seems to be less at the grain boundary than in the bulk. In the case of an initial void, we studied stress concentration, dislocation propagation, and crack propagation in thin systems. The stress concentration was found to be in surprisingly good agreement with continuum predictions. Dislocation and crack were propagated with a velocity much below the speed of sound and they preferred the ͗110͘ crystal orientation.
“…Molecular dynamics simulations were also performed to investigate the crack propagation velocity (Sieradzki and Dienes 1988;Abraham and Gao 2000) and crack branching (Zhou et al 1996). According to Abraham and Gao (2000), the propagation velocity is bounded by the Rayleigh wave speed in mode I whereas it could reach the To analyze the sensitivity of the model predictions to this parameter, the closed form solutions of the model are used [Equations (15) to (21)].…”
Abstract:To understand and model damage generated during impact by a penetrator of ultra-high strength concrete targets, edge-on impact tests are performed with the so-called Ductal ® concrete, which is unreinforced or reinforced with short fibers. Two edge-on impact configurations are designed with a dynamic confinement system. The first configuration uses aluminum projectiles and allows us to study the dynamic fragmentation that spreads out within the tile without any confined damage close to the impact point. The fragmentation process is composed of numerous oriented millimetric cracks. In the second configuration, steel projectiles are used with a higher impact velocity. Damaged zones are visualized by using an ultra-high speed camera and a sarcophagus configuration designed to prevent the fragments from moving. The post-mortem studies of impacted tiles enable us to observe an intense fragmentation of the targets and confined damage close to the impact point if steel projectiles are used. Simulations are performed with an anisotropic damage model coupled with a concrete plasticity model. Orientation and crack density are compared with postmortem observations.
“…Marder has pointed out that theory and experiment are at variance with respect to one another and has properly emphasized the fact that cracks have a terminal velocity of about half of theoretical predictions. While detailed comparison between theory and experiment regarding a crack-tip equation of motion has not been possible, we would like to point out that molecular-dynamic (MD) simulations have been used to study these issues in considerable detail [2][3][4]. The results of these simulations suggest that crack dynamics is strongly influenced by the lattice structure of a solid and that the physics which controls the terminal velocity of a crack can only be understood by consideration of the discrete atomic nature of solids.…”
mentioning
confidence: 99%
“…Since the theoretical crack-tip terminal velocity is the long wavelength Rayleigh velocity ( = 0.56i^), this corresponds to about 64% of the value predicted by continuum approaches. Sieradzki et al [3] used MD to study crack propagation in the 2D triangular Johnson solid. The relevant results of that investigation are the following: (1) The measured terminal velocity is independent of sample size and loading condition and is =0.25i^ and (2) cracks rapidly achieve terminal velocity even under loading conditions when the energy available for crack growth barely exceeds the energy required to create the crack faces.…”
mentioning
confidence: 99%
“…Figure 1 shows new MD results which we have recently obtained for the 2D triangular linear elastic and Johnson solids. The sample configurations which we used were identical to those examined earlier [3]. Sample-size effects were only found to affect the details of the partitioning of kinetic and potential energy during crack propagation and not the actual shape of the crack-velocitycrack-length curve.…”
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.