Characterization of extended defects in 2D materials using aperture-based dark-field STEM in SEM

This is the raw data for the manuscript:

Characterization of extended defects in 2D materials using aperture-based dark-field STEM in SEM

A readme file containing all descriptions can be found in the main folder. 

Abstract:

Quantitative diffraction contrast analysis with defined diffraction vectors is a wellestablished method in TEM for studying defects in crystalline materials. A comparable transmission techniques is however not available in the more widely used SEM platforms. In this work, we transfer the aperture-based dark-field imaging method from the TEM to the SEM, thus enabling quantitative diffraction contrast studies at lower voltages in SEM. This is achieved in STEM mode by inserting a custom-made aperture between the sample and the STEM detector and centering the hole on a desired reflection. To select individual reflections for dark-field imaging, we use our Low Energy Nanodiffraction (LEND) setup [Schweizer et al., Ultramicroscopy 213, 112956 (2020)], which captures transmission diffraction patterns from a fluorescent screen positioned below the sample. The aperture-based dark-field STEM method is particularly useful for studying extended defects in 2D materials, where (i) stronger diffraction at the lower voltages used in SEM is advantageous, but at the same time (ii) two-beam conditions cannot be established, making quantitative diffraction contrast analysis with standard bright-field and annular dark-field detectors impossible. We demonstrate the method by studying basal plane dislocations in bilayer graphene, which have attracted considerable research interest due to their exceptional structural and electronic properties. Direct comparison of results obtained on identical dislocations by the established TEM method and by the new aperture-based dark-field STEM method in SEM shows that a reliable Burgers vector analysis is possible by  applying the wellknown g·b=0 invisibility criterion. We further use the LEND setup to acquire 4D-STEM data and show that the virtual dark-field images match well with those in aperturebased dark-field STEM images for reliable Burgers vector analysis.