Kevin Roccapriore,Sergei Kalinin1
The University of Tennessee, Knoxville1
Kevin Roccapriore,Sergei Kalinin1
The University of Tennessee, Knoxville1
Direct fabrication of matter on atomic level is the North Star goal for areas spannign quantum computing, quantum communications, biological sensing, and fundamental physical research. Over the last decade, it has been shown that electron beams in Scanning Transmission Electron Microscopy can be used not only to probe structure and electronic properties of materials on atomic level, but also to modify materials on the atomic level. Harnessing electorn beam changes for direct atomic farication however requires synergy between machine learning methods and microscope control. In this presentation, I will illustarte the progression of automated electron microscopy from real-time data analysis to physics discovery to atomic manipulations. Here, the applications of classical deep learning methods in streaming image analysis are strongly affected by the out of distribution drift effects, and the approaches to minimize though are discussed. The robust approach for real-time analysis of the scanning transmission electron microscopy (STEM) data streams, based on the ensemble learning and iterative training (ELIT) of deep convolutional neural networks, is implemented on an operational microscope, enabling the exploration of the dynamics of specific atomic configurations under electron beam irradiation via an automated experiment in STEM. Combined with beam control, this approach allows studying beam effects on selected atomic groups and chemical bonds in a fully automated mode. We demonstrate atomically precise engineering of single vacancy lines in transition metal dichalcogenides and the creation and identification of topological defects graphene. The ELIT-based approach opens the pathway toward the direct on-the-fly analysis of the STEM data and engendering real-time feedback schemes for probing electron beam chemistry, atomic manipulation, and atom by atom assembly. We further illustarte how deep kernel learning (DKL) methods allow to realize both the exploration of complex systems towards the discovery of structure-property relationship, and enable automated experiment targeting physics (rather than simple spatial feature) discovery. The latter is illustrated via experimental discovery of the edge plasmons in STEM/EELS. Jointly, these developments open the pathway for creation and characterization of designed defect configurations and artificial molecules in 2D materials.