Events

Yiran Lu presents

 Developing quantitative analysis of 4D-STEM images: a focus on electron ptychography

Tuesday, October 21st 2025 at 14:00

Seminar room – Building A – CNRS

The defence will be in English.

Link visio: https://univ-grenoble-alpes-fr.zoom.us/j/99866450469?pwd=0SbAvAbUozWYv173xBdy2RFISLCpH0.1

Abstract: This thesis explores electron ptychography as one of the data treatment methods for four-dimensional Scanning Transmission Electron Microscopy (4D-STEM), a critical tool in materials science for characterizing properties ranging from crystal structures to electric and magnetic fields.
Electron ptychography is a powerful technique addressing the “phase problem” in electron microscopy— the loss of electron phase information during detection. Unlike conventional STEM techniques, electron ptychography leverages full diffraction patterns to reconstruct both the probe and object functions, making it a promising solution to determine material properties quantitatively.
Despite significant advancements in electron ptychography over the last few decades, which have enabled sub-angstrom resolution, discrepancies persist between experiments and simulations, making it challenging for quantitative analysis of electric properties. This thesis aims to investigate the factors that influence phase quantification. As the first PhD student in the group focusing on this area, I implemented the necessary programs and framework for electron ptychographic reconstructions using existing algorithms. The thesis will review the history and algorithms of electron ptychography, compare different reconstruction algorithms (GEM ePIE, PtyPy’s Difference Map, and ePIE), analyze the impact of probe aberrations on reconstructions, and present a quantitative analysis of the reconstructed phase, comparing it with experimental data to identify the factors limiting accurate quantification.