Projected thickness reconstruction from a single defocused transmission electron microscope image of an amorphous...

Current methods for measuring material thickness in electron microscopy often require multiple images or complex setups. Prior research has shown that phase contrast images can reveal structural details. However, reconstructing thickness from a single image remains a challenge. This gap motivated the development of a new non-iterative algorithm. The algorithm must handle spherical aberration and noise. It was already known that phase contrast is sensitive to thickness variations. No prior work had resolved how to use single defocused images effectively. This paper introduces a technique that simplifies the process while maintaining accuracy.

The study aims to develop a stable, non-iterative algorithm for reconstructing projected thickness from a single defocused image. The goal is to improve the resolution and reliability of thickness mapping in electron microscopy. The technique must account for spherical aberration in the optics. It should work for general single-material objects. The method should detect small thickness variations at the nanometer scale. The researchers propose using this approach for amorphous materials with voids. The algorithm must tolerate input parameter errors. The study also seeks to validate the method using simulations and real data.

The algorithm uses a single defocused transmission electron microscope image as input. It applies a non-iterative reconstruction approach to derive the projected thickness map. The method includes corrections for spherical aberration in the objective lens. Simulations validate the algorithm's accuracy and stability. The researchers test the algorithm on latex sphere specimens. They analyze the tolerance of the method to input parameter errors. The technique measures thickness variations as small as 0.25 nm in diameter. The method also uses autocorrelation to estimate open structure sizes in the material.

The algorithm successfully reconstructs the projected thickness map from a single defocused image. It detects thickness variations with a lateral diameter of ∼0.25 nm. The method works well for single-material objects in the strong phase regime. Simulations confirm the algorithm's quantitative accuracy. The reconstruction is stable even with noise in the input image. The technique identifies voids and free volume in amorphous materials. Autocorrelation analysis reveals open structure sizes from the thickness map. The method shows high tolerance for errors in defocus and specimen thickness.

The study demonstrates a stable, non-iterative algorithm for thickness reconstruction. The method works with a single defocused image and accounts for spherical aberration. It is suitable for single-material objects in the strong phase regime. The algorithm detects thickness variations as small as 0.25 nm in diameter. Simulations and latex sphere data confirm the method's accuracy. The technique is robust to noise and input parameter errors. Autocorrelation analysis provides insights into open structures in the material. The method is ideal for studying voids in amorphous materials at the nanometer scale.