Genetic algorithm-guided deep learning of grain boundary diagrams: Addressing the challenge of five degrees of freedom

Abstract

Grain boundaries (GBs) often control the processing and properties of polycrystalline materials.Here, a potentially transformative research is represented by constructing GB property diagrams as functions of temperature and bulk composition, also called "complexion diagrams," as a general materials science tool on par with phase diagrams. However, a GB has five macroscopic (crystallographic) degrees of freedom (DOFs). It is essentially a "mission impossible" to construct property diagrams for GBs as a function of five DOFs by either experiments or modeling. Herein, we combine isobaric semi-grand-canonical ensemble hybrid Monte Carlo and molecular dynamics (hybrid MC/MD) simulations with a genetic algorithm (GA) and deep neural network (DNN) models to tackle this grand challenge. The DNN prediction is ~10 8 faster than atomistic simulations, thereby enabling the construction of the property diagrams for millions of distinctly different GBs of five DOFs. Notably, excellent prediction accuracies have been achieved for not only symmetric-tilt and twist GBs, but also asymmetric-tilt and mixed tilt-twist GBs; the latter are more complex and much less understood, but they are ubiquitous and often limit the performance properties of real polycrystals as the weak links. The data-driven prediction of GB properties as function of temperature, bulk composition, and five crystallographic DOFs (i.e., in a 7D space) opens a new paradigm.

Graph neural networks for an accurate and interpretable prediction of the properties of polycrystalline materials

“…Looking ahead, the present GNN model can be extended to describe the orientations (including three parameters for the crystal misorientation and two parameters for the orientation of the normal axis) and other features of grain boundaries 31 by assigning an additional feature vector e ij to each edge that connects node (grain) i and node (grain) j in the microstructure graph. This is similar to the application of GNN model for predicting the properties of organic molecules and inorganic crystals 23,27 where an edge vector e ij is introduced to describe the features of the chemical bond between atom i and atom j.…”

Data-driven prediction of grain boundary segregation and disordering in high-entropy alloys in a 5D space

“…A recent study suggested that SBO can be used as a descriptor to predict and subsequently tailor GB segregation. 28 For example, if we want to promote segregation of a certain element (e.g., Cr) in HEAs, we can increase the composition of the elements with similar SBO values (e.g., Fe and Co) and/or reduce the composition of those with different SBO values (e.g., Mn and Ni). It is interesting to note that large chemicalaffinity disparity of different elements can foster segregation in HEAs.…”

“…Notably, GB complexion (phase) diagrams, which represent GB thermodynamic states or properties as functions of thermodynamic variables such as the temperature and bulk composition (representing chemical potentials), have been developed as the GB counterparts to the bulk phase diagrams. To date, various GB diagrams have been constructed for binary and ternary systems, 19,20,26,[28][29][30] but they are rarely developed for multicomponent systems, 31 and certainly not for HEAs, owing to the increasing complexity of a large, multi-dimensional compositional space. Furthermore, the more general GBs (asymmetric GBs with mixed twist and tilt features), which are ubiquitous in polycrystalline materials and often the weak links chemically and mechanically, 28,32 are still scarcely studied.…”

“…To date, various GB diagrams have been constructed for binary and ternary systems, 19,20,26,[28][29][30] but they are rarely developed for multicomponent systems, 31 and certainly not for HEAs, owing to the increasing complexity of a large, multi-dimensional compositional space. Furthermore, the more general GBs (asymmetric GBs with mixed twist and tilt features), which are ubiquitous in polycrystalline materials and often the weak links chemically and mechanically, 28,32 are still scarcely studied.…”

Chemist versus Machine: Traditional Knowledge versus Machine Learning Techniques

“…The SOAP descriptor encodes atomic geometries with the help of a local expansion of Gaussian smeared atomic densities. The possibilities offered by these potentials are immense, allowing simulations of phenomena requiring many atoms (e.g., grain boundaries [46,47] or amorphous phases [48,49]) or accelerating the search for new crystal structures [50]. It is expected that they will impact many fields and they have already been used in the investigation of battery and potential thermoelectric materials [51,52].…”