Reverse Monte Carlo reconstruction algorithm for discrete electron tomography based on HAADF‐STEM atom counting

Abstract

In this paper, we propose an algorithm to obtain a three-dimensional reconstruction of a single nanoparticle based on the method of atom counting. The location of atoms in three dimensions has been successfully performed using simulations of high-angle-annular-dark-field images from only three zone-axis projections, [110], [310] and [211], for a face-centred cubic particle. These three orientations are typically accessible by low-tilt holders often used in high-performance scanning transmission electron microscopes.

Three-dimensional atomic models from a single projection using Z-contrast imaging: verification by electron tomography and opportunities

“…The alternative method considered here 11,13 combines atom-counting in ADF STEM [15][16][17][18] with prior knowledge about a material's crystal structure. An energy minimization using ab-initio calculations or a Monte Carlo approach is then performed to relax the nanoparticle's 3D structure 11,13,19 . This method is designed to apply to nanoparticles which are roughly symmetrical along the beam direction in the electron microscope.…”

Hybrid statistics-simulations based method for atom-counting from ADF STEM images

“…By counting the number of atoms in each atomic column from two-dimensional (2D) ADF STEM images recorded under a few different viewing directions, a threedimensional (3D) reconstruction of the structure can be obtained [13,14], allowing for the quantification of the shape and size of the nanoparticle. Furthermore, atomcounting results can be combined with ab-initio calcula-tions or Monte Carlo simulations to study the dynamical behaviour of nanoparticles [15,16,17,18].…”