Developing Data Analytical Tools to Map Crystalline Defects from APT 3D Compositional Point Cloud Maps
EMSL Project ID
60211
Abstract
The atom-by-atom 3D point cloud composition mapping produced by Atom Probe Tomography (APT) provides the potential to directly measure nanocrystalline structure and identify specific lattice defects, such as those produced during radiation damage or even during a catalytic or oxidative surface reaction. However, the lack of computational tools that can be applied to APT data to quantitatively determine crystalline lattice structure down to an atomic- to near-atomic-level hinders the development of dynamic theoretical models aimed at either mitigating or exploiting such defects. Here we propose to develop a Machine Learning (ML) model capable of identifying lattice defects within 3D APT point cloud composition maps. As a general means to develop such ML models, our focus will be centered around analysis of already-existing APT data from neutron and ion beam irradiated materials and provide a means to identify and correlate physical lattice defects with irradiated dose. The outcome of this proposed work will be the development of a general ML framework and computational tools for identifying lattice defects that can be broadly applied broadly to materials relevant to existing and future PNNL programs.
Project Details
Start Date
2021-10-01
End Date
N/A
Status
Active
Released Data Link
Team
Principal Investigator
Team Members