Sampling metrics for robust reconstructions in multislice ptychography: Theory and experiment
Reviewed by Pithpith:24AK6X7Eopen to challenge →
read the original abstract
While multislice electron ptychography can provide thermal-vibration limited resolution and 3D information, it relies on the proper selection of many intertwined experimental and computational parameters. Here, we outline a theoretical basis for selecting experimental parameters to enable robust ptychographic reconstructions. We develop a series of physically informed metrics to describe the selection of experimental parameters in multislice ptychography. Image simulations are used to comprehensively evaluate the validity of these metrics over a broad range of experimental conditions. We develop two metrics, areal oversampling and Ronchigram magnification, which predict reconstruction success with high accuracy. Lastly, we validate these conclusions with experimental ptychographic data, and demonstrate close agreement between trends in simulated and experimental data. Using these metrics, we achieve experimental multislice reconstructions at a scan step of $1.0 \unit{\AA/px}$, enabling large field-of-view ($>\!18\unit{nm}$), data-efficient reconstructions. These experimental design principles enable the routine and reliable use of multislice ptychography for materials characterization.
This paper has not been read by Pith yet.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.