Abstract

Conventional models of plastic deformation are based on linear elastic approximations with respect to the empirical results obtained from pure screw or pure edge dislocations. In real materials, however, mixed dislocations are far more commonly observed and their motion carries the bulk of the plastic deformation. Molecular dynamics (MD) simulations are useful to ‘unlock’ the temperature and structure dependence of the phonon drag effect, and the phonon drag strongly influences the mobility of dislocations in FCC metals. The Burgers vector of a dislocation dictates the necessary orientation of the stress required to control the glide behaviour. Hence, the mobility of mixed dislocations can only be studied once the stress can be carefully controlled for both the τxy and τyz components simultaneously. It is non-trivial to accurately control the global stress components effectively under multi-orientation testing. To this end, the present study evaluates the effectiveness of utilising a well-tuned barostat to show the influence of stress loading orientation on the mobility of a mixed 30° dislocation dipole. Self-consistent results with two different EAM potentials and size-independent steady-state velocities (excluding only the lowest stress and smallest size case), suggest that this approach is valid. The glide stress, τg, is contrasted with the shear component in a perpendicular orientation within the glide plane, referred to as the Escaig stress, τe. When the appropriate stress control regime and cell dimensions were applied, the influence of dipole image forces were mitigated down to stresses as low as 5MPa. The stacking fault width was influenced by τe in an identical manner for the two different EAM potentials, once compensation was made to offset the discrepancy in the stacking fault energy. The stacking fault width was effected in a non-linear relationship with τe, which caused either an increase or decrease depending on the orientation and magnitude of τe. The phonon dissipation and radiative damping exhibited by the mixed dislocations was qualitatively better matched to the characteristics of pure edge dislocations, and the calculated drag coefficient was significantly smaller than has been found for pure screw dislocations in the literature. Results show that when the appropriate methodology is applied, the stress-velocity curve can follow a smoothly linear function of the τg component for mixed stress conditions with any combination of τe.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call