Han, Yimo2024-05-212024-05-212024-052024-04-03May 2024Shi, Chuqiao. Advancing Nanobeam Four-Dimensional Scanning Transmission Electron Microscopy (4D-STEM) for Strain Analysis. (2024). PhD diss., Rice University. https://hdl.handle.net/1911/116133https://hdl.handle.net/1911/116133This dissertation delves into the realm of nanobeam four-dimensional scanning transmission electron microscopy (4D-STEM), and its application for characterizing the intricate microstructures of two-dimensional (2D) materials and nano-catalysts. It leverages recent technological breakthroughs in pixelated, fast direct electron detectors that enable the comprehensive collection of momentum-space data at every scan in STEM. This integration of spatial and momentum dimensions enables the generation of rich 4D datasets. The wealth of information captured by these advanced detectors, though vast, poses interpretative challenges due to its complexity. Consequently, this thesis is dedicated to the development and application of novel analytical methodologies for the extraction of crystallographic information from such voluminous 4D-STEM datasets, addressing key problems in materials science. In Chapter 2, the study applies 4D-STEM to investigate the broad structural characteristics of van der Waals 2D ferroelectric SnSe. The research uncovers significant in-plane lattice distortions and out-of-plane stacking variations. This has led to the discovery of considerable lattice strains and distinctive ferroelectric-to-antiferroelectric domain walls which hold implications for the material's physical properties and potential device applications. Chapter 3 shifts the focus to the surface strain of core-shell nano-catalysts' structure. Through meticulous analysis via 4D-STEM, it is demonstrated that cube-shaped Au@Pd particles with sharp-tipped cores exhibit a coherent, dislocation-free heteroepitaxial interface even when the shell thickness is considerably greater than that of comparable nanocatalysts with rounded cores. This finding suggests a route to enhancing the strain stability of such structures, which is paramount in their application as catalysts. Chapter 4 ventures into the machine leaning methods to process extensive 4D-STEM datasets autonomously. This innovative, data-driven approach effectively discerns various material deformations, including strain, lattice distortions, and bending contours. Such detailed comprehension of lattice alterations is crucial for the advancement of material characterization techniques and the ensuing implications for materials science. Overall, the thesis advances the understanding of complex material systems through innovative 4D-STEM analysis and machine learning, potentially impacting the design and application of nanoscale materials and advancing technological frontiers across various disciplines.application/pdfengCopyright is held by the author, unless otherwise indicated. Permission to reuse, publish, or reproduce the work beyond the bounds of fair use or other exemptions to copyright law must be obtained from the copyright holder.nano material4D-STEMThesis2024-05-21