Deep Learning Enabled Strain Mapping of Single-Atom Defects in Two-Dimensional Transition Metal Dichalcogenides with...

Area of Science:

• Materials Science
• Condensed Matter Physics
• Nanotechnology

Background:

• Two-dimensional (2D) materials are excellent for studying atomic defects, but radiation damage hinders electron microscopy.
• Probing atomic-scale strain fields around defects is crucial for understanding material properties.

Purpose of the Study:

• To develop a high-precision method for analyzing atomic-scale strain fields induced by single-atom defects in 2D materials.
• To overcome limitations of electron microscopy caused by radiation damage in beam-sensitive materials.

Main Methods:

• Utilized deep learning algorithms to analyze large datasets of aberration-corrected scanning transmission electron microscopy (STEM) images.
• Developed a technique to generate high signal-to-noise class averages from hundreds of defect images.
• Achieved sub-picometer precision in measuring 2D atomic spacings.

Main Results:

• Successfully probed single-atom defects in monolayer WSe2Te2 with sub-picometer precision.
• Identified complex, oscillating strain fields around selenium (Se) vacancies.
• Observed alternating rings of lattice expansion and contraction induced by Se vacancies.

Conclusions:

• Demonstrated the potential of computer vision and deep learning for high-precision electron microscopy of beam-sensitive 2D materials.
• Revealed intricate strain field patterns around atomic defects, offering new insights into material behavior.
• Paved the way for advanced characterization of defects in novel 2D materials.