SHARE
Research Highlight

Machine learning-enabled automated experiments in electron microscopy

Topic:

The Science
A team of researchers at Oak Ridge National Laboratory has developed and experimentally implemented a workflow for automated sample exploration in scanning transmission electron microscopy (STEM) experiments. This advance will enable rapid discovery of local structures, symmetry-breaking distortions, and internal electric and magnetic fields in complex materials.

The Impact
The developed approach establishes a paradigm for physics-driven automated STEM experiments that enable probing the physics of strongly correlated systems and quantum materials and devices. It can be purely exploratory or tailored toward the discovery of specific physical phenomena. Further, it is universal and expected to be applicable to probe-based microscopic techniques including other STEM modalities and scanning probe microscopies.
 

Scanning transmission electron microscope at ORNL’s CNMS and schematic depiction of electron beam probing a sample. Schematic of deep kernel learning of correlative relationship between structure and targeted physical property.
Top row: Scanning transmission electron microscope at ORNL’s CNMS and schematic depiction of electron beam probing a sample.
Bottom row: Schematic of deep kernel learning of correlative relationship between structure and targeted physical property.

Funding
This research is sponsored by the INTERSECT Initiative as part of the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory, managed by UT-Battelle, LLC for the US Department of Energy under contract DE-AC05-00OR22725. This effort was based upon work supported by the U.S. Department of Energy (DOE), Office of Science, Basic Energy Sciences (BES), Materials Sciences and Engineering Division. Work was partially supported and conducted using resources supported by Oak Ridge National Laboratory’s Center for Nanophase Materials Sciences (CNMS), which is a U.S. DOE Office of Science User Facility. Electron microscopy was performed using instrumentation within ORNL’s Materials Characterization Core provided by UT-Battelle, LLC, under Contract No. DEAC05- 00OR22725 with the DOE and sponsored by the Laboratory Directed Research and Development Program of Oak Ridge National Laboratory, managed by UT-Battelle, LLC, for the U.S. Department of Energy.

Team Members: Kevin M. Roccapriore (ORNL), Ondrej Dyck (ORNL), Mark P. Oxley (ORNL), Maxim Ziatdinov (ORNL), Sergei V. Kalinin (UTK) 

Publication
K. M. Roccapriore, O. Dyck, M. P. Oxley, M. Ziatdinov, S. V. Kalinin. Automated Experiment in 4D STEM: exploring emergent physics and structural behaviors. ACS Nano, accepted. https://doi.org/10.48550/arXiv.2112.04479