Atomistic simulations of incident dislocation interactions with nickel grain boundaries

Abstract Grain boundaries strengthen metals and act as hardening agents, impeding plastic flow macroscopically. The interactions between grain boundaries and dislocations are complex and difficult to predict. To understand the connection between resolved shear stresses and transmission events we simulated dislocation-grain boundary interactions in a number of <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mrow> <mml:mrow> <mml:mo>[</mml:mo> <mml:mn>1</mml:mn> <mml:mstyle scriptlevel="0"/> <mml:mover> <mml:mn>1</mml:mn> <mml:mo accent="true">―</mml:mo> </mml:mover> <mml:mstyle scriptlevel="0"/> <mml:mn>2</mml:mn> <mml:mo>]</mml:mo> </mml:mrow> </mml:mrow> </mml:math> asymmetric tilt grain boundaries designed for optimal transmission of dislocations. By shearing the cell containing the grains on either side of the boundary, we drove the dislocation into the grain boundary and observed the interaction. The key findings include: (i) roughly half of the observed dislocation-grain boundary interactions resulted in transmission, which is a lower than expected transmission rate of incident dislocations; (ii) a rejection of a hypothesized monotonic relationship between applied stress and transmission of dislocations based on observations; (iii) significant restructuring of the grain boundaries resulting from the applied stress and incident dislocation interactions; and (iv) a suggestion that transmission events appear to be better described as separate absorption and nucleation events, with each event affecting and affected by the evolving grain boundary structure. Together, these point to continued challenges and opportunities surrounding dislocation-grain boundary interactions. The challenges relate to the difficulty in extracting absorption and nucleation criteria. The opportunities suggest that mesoscale models can treat all these events independently based on relevant criteria if they can be obtained.