Search NASA⌕ Search

Engineering topics

Fraser, Hamish L.

Publications and source records attributed to Fraser, Hamish L..

A deep learning approach for semantic segmentation of unbalanced data in electron tomography of catalytic materials

In computed TEM tomography, image segmentation represents one of the most basic tasks with implications not only for 3D volume visualization, but more importantly for quantitative 3D analysis. In case of large and complex 3D data sets, segmentation can be an extremely difficult and laborious task, and thus has been one of the biggest hurdles for comprehensive 3D analysis. Heterogeneous catalysts have complex surface and bulk structures, and often sparse distribution of catalytic particles with relatively poor intrinsic contrast, which possess a unique challenge for image segmentation, including the current state-of-the-art deep learning methods. To tackle this problem, we apply a deep learning-based approach for the multi-class semantic segmentation of a γ-Alumina/Pt catalytic material in a class imbalance situation. Specifically, we used the weighted focal loss as a loss function and attached it to the U-Net’s fully convolutional network architecture. We assessed the accuracy of our results using Dice similarity coefficient (DSC), recall, precision, and Hausdorff distance (HD) metrics on the overlap between the ground-truth and predicted segmentations. Our adopted U-Net model with the weighted focal loss function achieved an average DSC score of 0.96 ± 0.003 in the γ-Alumina support material and 0.84 ± 0.03 in the Pt NPs segmentation tasks. We report an average boundary-overlap error of less than 2 nm at the 90th percentile of HD for γ-Alumina and Pt NPs segmentations. The complex surface morphology of γ-Alumina and its relation to the Pt NPs were visualized in 3D by the deep learning-assisted automatic segmentation of a large data set of high-angle annular dark-field (HAADF) scanning transmission electron microscopy (STEM) tomography reconstructions.

36 MATERIALS SCIENCE↗

Nucleation and growth of α phase in a metastable β-Titanium Ti-5Al-5Mo-5V-3Cr alloy: Influence from the nano-scale, ordered-orthorhombic O" phase and α compositional evolution

We investigated the nucleation and growth of α precipitates in the presence of nano-scale, titanium-rich, ordered-orthorhombic O" precipitates formed during heating at 5 °C/min to 400 °C in a metastable β-Ti alloy, Ti-5Al-5Mo-5V-3Cr. The O" precipitates are found to assist α nucleation by serving as preferential nucleation sites. In the very initial stages of its precipitation, the α phase is found to have formed via a coupled diffusional-displacive mode as titanium-rich plates. At this stage, aluminum partitions equally between the β matrix and the growing α precipitates, but slowly diffuses up-hill into the α phase during isothermal aging at 400 °C for 15 h. The interplay between such pre-formed metastable phases and α can be exploited to tailor the microstructure, by refining the α distribution, and consequently improve the mechanical properties of β-Ti alloys. Finally, our work paves the way for exploiting this cascade of metastable phases for further microstructural engineering of β-Ti alloys.

36 MATERIALS SCIENCE↗

ω-Assisted α nucleation in a metastable β titanium alloy

Here, we explore mechanisms of ω-assisted α nucleation for microstructure refinement in a metastable β titanium alloy, Ti-20V (wt%). In particular, we analyze quantitatively the individual and combined effects of concentration and stress fields associated with a growing isothermal ω precipitate and the potency of the ω/β interface on the activation energy barrier for α nucleation as a function of the ω precipitate size and coherency state. We find that the complex interplay among these three factors determines the most potent ω particle size for inducing α nucleation and the most favorable nucleation site.

36 MATERIALS SCIENCE↗