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.

A New Class of Cluster-Matrix Nanocomposite Made of Fully Miscible Components.

Nanocomposite materials, consisting of two or more phases, at least one of which has a nanoscale dimension, play a distinctive role in materials science because of the multiple possibilities for tailoring their structural properties and, consequently, their functionalities. In addition to the challenges of controlling the size, size distribution, and volume fraction of nanometer phases, thermodynamic stability conditions limit the choice of constituent materials. This study goes beyond this limitation by showing the possibility of achieving nanocomposites from a bimetallic system, which exhibits complete miscibility under equilibrium conditions. A series of nanocomposite samples with different compositions are synthesized by the co-deposition of 2000-atom Ni-clusters and a flux of Cu-atoms using a novel cluster ion beam deposition system. The retention of the metastable nanostructure is ascertained from atom probe tomography (APT), magnetometry, and magnetotransport studies. APT confirms the presence of nanoscale regions with ≈100 at% Ni. Magnetometry and magnetotransport studies reveal superparamagnetic behavior and magnetoresistance stemming from the single-domain ferromagnetic Ni-clusters embedded in the Cu-matrix. Essentially, the magnetic properties of the nanocomposites can be tailored by the precise control of the Ni concentration. The initial results offer a promising direction for future research on nanocomposites consisting of fully miscible elements.

Dislocation exhaustion and ultra-hardening of nanograined metals by phase transformation at grain boundaries.

The development of high-strength metals has driven the endeavor of pushing the limit of grain size (d) reduction according to the Hall-Petch law. But the continuous grain refinement is particularly challenging, raising also the problem of inverse Hall-Petch effect. Here, we show that the nanograined metals (NMs) with d of tens of nanometers could be strengthened to the level comparable to or even beyond that of the extremely-fine NMs (d ~ 5 nm) attributing to the dislocation exhaustion. We design the Fe-Ni NM with intergranular Ni enrichment. The results show triggering of structural transformation at grain boundaries (GBs) at low temperature, which consumes lattice dislocations significantly. Therefore, the plasticity in the dislocation-exhausted NMs is suggested to be dominated by the activation of GB dislocation sources, leading to the ultra-hardening effect. This approach demonstrates a new pathway to explore NMs with desired properties by tailoring phase transformations via GB physico-chemical engineering.

Nanoglass-Nanocrystal Composite-a Novel Material Class for Enhanced Strength-Plasticity Synergy.

The properties of a material can be engineered by manipulating its atomic and chemical architecture. Nanoglasses which have been recently invented and comprise nanosized glassy particles separated by amorphous interfaces, have shown promising properties. A potential way to exploit the structural benefits of nanoglasses and of nanocrystalline materials is to optimize the composition to obtain crystals forming within the glassy particles. Here, a metastable Fe-10 at% Sc nanoglass is synthesized. A complex hierarchical microstructure is evidenced experimentally at the atomic scale. This bulk material comprises grains of a Fe90 Sc10 amorphous matrix separated by an amorphous interfacial network enriched and likely stabilized by hydrogen, and property-enhancing pure-Fe nanocrystals self-assembled within the matrix. This composite structure leads a yield strength above 2.5 GPa with an exceptional quasi-homogeneous plastic flow of more than 60% in compression. This work opens new pathways to design materials with even superior properties.

Calibration of Atom Probe Tomography Reconstructions Through Correlation with Electron Micrographs.

Although atom probe tomography (APT) reconstructions do not directly influence the local elemental analysis, any structural inferences from APT volumes demand a reliable reconstruction of the point cloud. Accurate estimation of the reconstruction parameters is crucial to obtain reliable spatial scaling. In the current work, a new automated approach of calibrating atom probe reconstructions is developed using only one correlative projection electron microscopy (EM) image. We employed an algorithm that implements a 2D cross-correlation of microstructural features observed in both the APT reconstructions and the corresponding EM image. We apply this protocol to calibrate reconstructions in a Cu(In,Ga)Se2-based semiconductor and in a Co-based superalloy. This work enables us to couple chemical precision to structural information with relative ease.