A Machine Learning Framework for Quantifying Chemical Segregation and Microstructural Features in Atom Probe Tomography Data.

Atom probe tomography (APT) is ideally suited to characterize and understand the interplay of segregation and microstructure in modern multi-component materials. Yet, the quantitative analysis typically relies on human expertise to define regions of interest. We introduce a computationally efficient, multi-stage machine learning strategy to identify compositionally distinct domains in a semi-automated way, and subsequently quantify their geometric and compositional characteristics. In our algorithmic pipeline, we first coarse-grain the APT data into voxels, collect the composition statistics, and decompose it via clustering in composition space. The composition classification then enables the real-space segmentation via a density-based clustering algorithm, thus revealing the microstructure at voxel resolution. Our approach is demonstrated for a Sm-(Co,Fe)-Zr-Cu alloy. The alloy exhibits two precipitate phases with a plate-like, but intertwined morphology. The primary segmentation is further refined to disentangle these geometrically complex precipitates into individual plate-like parts by an unsupervised approach based on principle component analysis, or a U-Net-based semantic segmentation trained on the former. Following the composition and geometric analysis, detailed composition distribution and segregation effects relative to the predominant plate-like geometry can be readily mapped from the point cloud, without resorting to the voxel compositions.

Simulation of the θ' Precipitation Process with Interfacial Anisotropy Effects in Al-Cu Alloys.

The effects of anisotropic interfacial properties and heterogeneous elasticity on the growth and ripening of plate-like θ'-phase (Al2Cu) in Al-1.69 at.% Cu alloy are studied. Multi-phase-field simulations are conducted and discussed in comparison with aging experiments. The precipitate/matrix interface is considered to be anisotropic in terms of its energy and mobility. We find that the additional incorporation of an anisotropic interfacial mobility in conjunction with the elastic anisotropy result in substantially larger aspect ratios of the precipitates closer to the experimental observations. The anisotropy of the interfacial energy shows comparably small effect on the precipitate's aspect ratio but changes the interface's shape at the rim. The effect of the chemo-mechanical coupling, i.e., the composition dependence of the elastic constants, is studied as well. We show that the inverse ripening phenomenon, recently evidenced for δ' precipitates in Al-Li alloys (Park et al. Sci. Rep. 2019, 9, 3981), does not establish for the θ' precipitates. This is because of the anisotropic stress fields built around the θ' precipitates, stemming from the precipitate's shape and the interaction among different variants of the θ' precipitate, that disturb the chemo-mechanical effects. These results show that the chemo-mechanical effects on the precipitation ripening strongly depend on the degree of sphericity and elastic isotropy of the precipitate and matrix phases.