A Unified Model for Average Dislocation Behaviour in BCC Transition Metal Alloys

Description

BCC transition metals (TMs) represent an important class of structure materials. However, they become brittle at low temperatures, posing major risks of catastrophic failure. Brittleness or low ductility is attributed to the behaviour of the screw dislocation, which in turn is dictated by its core structure. The core structure, due to its screw nature, is experimentally featureless and intractable.DFT-based calculations reveal that the screw dislocation adopts a non-degenerate (ND) core structure in all pure BCC TMs, while empirical potentials show both ND-core and degenerate (D)-core, the latter of which is believed to be an artefact. Nevertheless, D-core is seen in DFT calculations of W-Re alloys and experiments show drastically increased ductility with a concomitant {112}-slip associated with the D-core. Despite the importance and controversy of D/ND cores, the physical origin governing their competition is not well-understood.We recently discovered that the ND/D-core structure, Peierls potential and stacking fault energy are related to the energy difference between the FCC and BCC structures. We introduced a new material index χ to capture this critical quantity and demonstrated its predictive capability in BCC W-TM alloys. The project here leverages this latest discovery and aims to develop a unified, quantitative χ-model applicable to all BCC TMs.Our preliminary results show that χ can be effectively changed via solid solution alloying and reduction in χ leads to core transition from ND to D, decrease in Peierls potential and decrease in unstable stacking fault energy, all of which are critical for ductility and/or toughness. To quantify the χ effects, we propose a multiscale approach to (i) determine χ as a function of alloy composition ci and temperature T, i.e., χ = (ci , T) using virtual crystal approximation in DFT; (ii) establish dislocation average properties as a function of χ using atomistic simulations and (iii) examine χ-effects on overall plasticity and crack field evolution in discrete dislocation dynamics simulations.χ is based on crystal geometry and bonding, and is expected to be general and robust. Success of the project will thus lead to a unified χ-model which not only captures alloying effects from quantum mechanics origin, but also predicts average dislocation behaviour and overall plasticity at meso/microscales. The computational model can be used for rapid screening of solute composition favourable for high ductility and toughness.