What the study found
The study found that a fully autonomous approach can fabricate atomic-level defects in two-dimensional materials using machine learning and automated electron-beam control. As a proof of concept, the authors achieved controlled fabrication of MoS-nanowire edge structures in a MoS2 monolayer.
Why the authors say this matters
The authors say the approach is material-agnostic, meaning it is not limited to MoS2, and could be extended to other two-dimensional materials. They conclude that it may enable the creation of diverse defect structures and heterostructures beyond MoS2.
What the researchers tested
The researchers combined scanning transmission electron microscopy, or STEM, with high-angle annular dark-field, or HAADF, imaging for feedback-controlled monitoring. They used a machine learning framework with a random forest model and a convolutional neural network to decode HAADF images, identify atomic positions and species, and guide an autonomous decision-making platform and FPGA-controlled beam scanning routine.
What worked and what didn't
The study reports controlled fabrication of MoS-nanowire edge structures by iterative, targeted exposure of a MoS2 monolayer to a focused electron beam. The method selectively ejected sulfur atoms and used automated beam positioning and duration control. The abstract does not describe failures or cases where the method did not work.
What to keep in mind
The abstract presents this as a proof of concept, so the demonstrated result is limited to the reported MoS2 example. The abstract does not provide detailed quantitative performance measures or describe specific limitations beyond the statement that the method is being proposed for broader extension.
Key points
- A fully autonomous method was used to fabricate atomic-level defects in a two-dimensional material.
- The proof-of-concept result was controlled fabrication of MoS-nanowire edge structures in a MoS2 monolayer.
- Machine learning models were used to decode HAADF images and identify atomic positions and species.
- The beam control system selectively ejected sulfur atoms with automated positioning and timing.
- The authors say the method could be extended to other two-dimensional materials.
Disclosure
- Research title:
- Autonomous electron-beam fabrication controlled defect structures in 2D materials
- Authors:
- Zijie Wu, Kevin M. Roccapriore, Ayana Ghosh, Kai Xiao, Raymond R. Unocic, Stephen Jesse, Rama K. Vasudevan, Matthew G. Boebinger
- Institutions:
- Oak Ridge National Laboratory, University of Tennessee at Knoxville, North Carolina State University
- Publication date:
- 2026-04-27
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with peer-reviewed research without reading academic papers — one filtered digest, every Friday.