Probing the transition from dislocation jamming to pinning by machine learning

Abstract

Collective motion of dislocations is governed by the obstacles they encounter. In pure crystals, dislocations form complex structures as they become jammed by their anisotropic shear stress fields. On the other hand, introducing disorder to the crystal causes dislocations to pin to these impeding elements and, thus, leads to a competition between dislocation-dislocation and dislocation-disorder interactions. Previous studies have shown that, depending on the dominating interaction, the mechanical response and the way the crystal yields change.Here we employ three-dimensional discrete dislocation dynamics simulations with varying density of fully coherent precipitates to study this phase transition − from jamming to pinning − using unsupervised machine learning. By constructing descriptors characterizing the evolving dislocation configurations during constant loading, a confusion algorithm is shown to be able to distinguish the systems into two separate phases. These phases agree well with the observed changes in the relaxation rate during the loading. Our results also give insights on the structure of the dislocation networks in the two phases.

From mechanism-based to data-driven approaches in materials science

“…The training was done independently for each feature. Interestingly the critical precipitate density found for each feature was in reasonable agreement across all three features (Salmenjoki et al 2020).…”

“…The data-driven approaches applied to DDD, apart from a 1D case (Sarvilahti et al 2020) use hand-crafted features in the spirit of traditional phenomenology, although sophisticated inclusion of crystal symmetries or other physical insights have not yet been used (Salmenjoki et al 2018;Salmenjoki et al 2020;Steinberger et al 2019). While working along known phenomenons is beneficial to support and understand existing models, it is unlikely to reveal a lot of new physics.…”

Materials Informatics for Mechanical Deformation: A Review of Applications and Challenges

“…Likewise, initial pre-deformation improves the predictive ability of the model. Similar topological descriptors were employed [ 170 ] to characterize precipitates-mediated jamming-to-pinning transition in terms of the dislocation network topology. Using dislocation structures as input, a confusion algorithm was successfully trained based on the binary classification of states according to the probability of being a member of the pinned or jammed phases.…”

Predicting the failure of two-dimensional silica glasses

“…Modern machine learning methods provide a promising alternative pathway for the systematic development of structure-based predictions, 3 as has been shown in the context of molecular properties, [4][5][6] density functional theory force fields, [7][8][9] governing equations for dynamical systems and flow [10][11][12] and dislocation models for crystal plasticity. [13][14][15] Applications to glasses have been so far restricted to idealized models, so-called Lennard-Jones glasses, that were analyzed with support vector machines (SVM) , [16][17][18] graph neural networks (GNN) 19 and deep learning. 20,21 Predictions based on deep learning methods are becoming increasingly accurate but they are also hard to interpret.…”