Invited Lecture
Three stages of work hardening in full atomistic details. Seriously.

Over 80 years of dislocation science, many models, theories and even principles have been proposed to describe distinct stages I, II and III observed in straining tests of single crystals. Since early 90’s Dislocation Dynamics (DDD) has been heralded as the method to eventually deliver a much-coveted connection between the macroscopic stress-strain response and the underlying dislocation motion. Yet, owing to still remaining problems pertaining to the method’s fidelity and computability, we find ourselves perpetually “10 years away” from a glorious day when DDD finally delivers a full stress-strain curve with all the hardening stages fully resolved.

Here we present ultra-large scale Molecular Dynamics simulations of aluminum single crystals subjected to uniaxial tension. We show that appearance (or not) of 3-stage hardening depends on the initial crystallographic orientation of the straining axis and results from crystal rotation. In its turn, crystal rotation is a direct consequence of co-axiality forced on the specimen by the testing machine, the view widely accepted in the phenomenological Crystal Plasticity community since the classical studies of Schmid predating dislocations. Thus, because 3-stage hardening is not an intrinsic property of the material but results from a particular geometric constraint imposed in the standard uniaxial straining tests, it makes little sense to look for any specific physical mechanisms to define and control the notorious transitions from stage I to stage II to stage III. Remarkably, stress-strain behaviors, slip system activity and crystal rotations observed in our high-rate MD simulations are in an exact qualitative agreement with quasistatic experiments suggesting that physics of crystal plasticity scales, i.e. dislocation mechanisms defining crystal plasticity response remain the same over 12-1­3 decades of straining rates.