Using molecular dynamics for multislice TEM simulation of thermal diffuse scattering in AlGaN

Real space method for HAADF image simulation

A new real space HAADF simulation method was described in detail and a Real Space- STEM software was developed based on the new simulation method. The algorithm of the real space method can quickly calculate and simulate the microstructure images of complex crystals. The Real Space-STEM software developed in this paper has the functions of HRTEM and HAADF image simulation based on the real space method. By using this software to simulate high-resolution images of representative crystal materials from each crystal system, the HAADF images are both accurate and efficient. The effect of STEM parameters on HAADF imaging has been discussed using simulation results.

Modeling Temperature-Dependent Electron Thermal Diffuse Scattering via Machine-Learned Interatomic Potentials and Path-Integral Molecular Dynamics

Electron thermal diffuse scattering is shown to be sensitive to subtle changes in atomic vibrations and shows promise in assessing lattice dynamics at nanometer resolution. Here, we demonstrate that machine-learned interatomic potentials (MLIPs) and path-integral molecular dynamics can accurately capture the potential energy landscape and lattice dynamics needed to describe electron thermal diffuse scattering. Using SrTiO_{3} as a test bed at cryogenic and room temperatures, we compare electron thermal diffuse scattering simulations using different approximations to incorporate thermal motion. Only when the simulations are based on quantum mechanically accurate MLIPs in combination with path-integral molecular dynamics that include nuclear quantum effects is there excellent agreement with experiments.

Atom counting based on Voronoi averaged STEM intensities using a crosstalk correction scheme

If quantitative scanning transmission electron microscopy is used for very precise thickness measurements with atomic resolution, it is commonly referred to as »atom counting«. Due to scattering and the finite probe extent the signal recorded in one atomic column is dependent not only on its own height but also on the height of its neighbours. Especially for thicker specimens this crosstalk effect can have significant impact on the measured intensity. If it is not appropriately accounted for in the evaluation, it can result in a deterioration of accuracy that impedes the possibility of actual atom counting. However, as the number of possible neighbour configurations can be excessively large, a comprehensive consideration of all in the evaluation reference is neigh impossible. This work proposes a method that allows for the a-posteriori reduction of crosstalk during the evaluation by algebraic means. Based on a parametric model, which is described in detail in the article, the crosstalk is expressed by an invertible matrix. Applying the inverted matrix to the measurement yields crosstalk corrected intensity values with very little computational effort. These can subsequently be evaluated by direct comparison to simple reference data. The working principle of the method is presented on the example of crystalline gold. The crosstalk parametrisation is found by fitting a model to sets of specifically created multislice simulations. The parameters are given for both aberration corrected and uncorrected STEM. Subsequently the abilities and potential of the technique are assessed in simulative studies on multiple model systems including gold nanoparticles. Overall a significant and robust improvement of the attainable precision can be demonstrated making the proposed method a promising tool for reference-based atom counting.

A comparison of molecular dynamics potentials used to account for thermal diffuse scattering in multislice simulations

Here we investigate electron scattering simulations with thermal displacements incorporated using molecular dynamics potentials. Specifically, we explore the sensitivity of electron scattering to the phonon band structure, or more explicitly interatomic forces. Silicon serves as the model material where we introduce thermal atomic displacements via empirical and machine-learned molecular dynamics interatomic potentials and compare them to finite-temperature density functional theory interatomic forces. We demonstrate that when molecular dynamics potentials do not sufficiently reproduce the correct phonon band structure, significant errors in the simulated diffraction and image intensities can occur. Moreover, for Si, we find that multislice simulations using machine-learned interatomic potentials are more accurate than empirical ones. In addition to the selected atomic potential, we demonstrate that the sensitivity to the phonon band structure also depends on the crystal zone axis, which can be used to enhance sensitivity to thermal displacements. Finally, we provide a sensitivity analysis with angle-resolved scanning transmission electron microscopy (STEM) to enhance image sensitivity to the details of the phonon band structure.

Atomic-scale probing of heterointerface phonon bridges in nitride semiconductor

Interface phonon modes that are generated by several atomic layers at the heterointerface play a major role in the interface thermal conductance for nanoscale high-power devices such as nitride-based high-electron-mobility transistors and light-emitting diodes. Here we measure the local phonon spectra across AlN/Si and AlN/Al interfaces using atomically resolved vibrational electron energy-loss spectroscopy in a scanning transmission electron microscope. At the AlN/Si interface, we observe various interface phonon modes, of which the extended and localized modes act as bridges to connect the bulk AlN modes and bulk Si modes and are expected to boost the phonon transport, thus substantially contributing to interface thermal conductance. In comparison, no such phonon bridge is observed at the AlN/Al interface, for which partially extended modes dominate the interface thermal conductivity. This work provides valuable insights into understanding the interfacial thermal transport in nitride semiconductors and useful guidance for thermal management via interface engineering.

A Fast Frozen Phonon Algorithm Using Mixed Static Potentials

Image simulation in electron microscopy is vital for advanced image analysis but can be prohibitively long. This is especially true when including thermal diffuse scattering through the frozen phonon method, requiring repeat simulations for a number of phonon configurations. Here a method of reducing frozen phonon simulation time is demonstrated by emulating random phonon displacements through randomly mixing a set of precalculated static potentials. This avoids excessive recalculation of atom potentials and can lead to significant time improvement. The validity and limitations of this method are also demonstrated with respect to convergent beam electron diffraction and scanning transmission electron microscopy.

The abTEM code: transmission electron microscopy from first principles

Simulation of transmission electron microscopy (TEM) images or diffraction patterns is often required to interpret experimental data. Since nuclear cores dominate electron scattering, the scattering potential is typically described using the independent atom model, which completely neglects valence bonding and its effect on the transmitting electrons. As instrumentation has advanced, new measurements have revealed subtle details of the scattering potential that were previously not accessible to experiment. We have created an open-source simulation code designed to meet these demands by integrating the ability to calculate the potential via density functional theory (DFT) with a flexible modular software design. abTEM can simulate most standard imaging modes and incorporates the latest algorithmic developments. The development of new techniques requires a program that is accessible to domain experts without extensive programming experience. abTEM is written purely in Python and designed for easy modification and extension. The effective use of modern open-source libraries makes the performance of abTEM highly competitive with existing optimized codes on both CPUs and GPUs and allows us to leverage an extensive ecosystem of libraries, such as the Atomic Simulation Environment and the DFT code GPAW. abTEM is designed to work in an interactive Python notebook, creating a seamless and reproducible workflow from defining an atomic structure, calculating molecular dynamics (MD) and electrostatic potentials, to the analysis of results, all in a single, easy-to-read document.  This article provides ongoing documentation of abTEM development. In this first version, we show use cases for hexagonal boron nitride, where valence bonding can be detected, a 4D-STEM simulation of molybdenum disulfide including ptychographic phase reconstruction, a comparison of MD and frozen phonon modeling for convergent-beam electron diffraction of a 2.6-million-atom silicon system, and a performance comparison of our fast implementation of the PRISM algorithm for a decahedral 20000-atom gold nanoparticle.

The abTEM code: transmission electron microscopy from first principles

Simulation of transmission electron microscopy (TEM) images or diffraction patterns is often required to interpret experimental data. Since nuclear cores dominate electron scattering, the scattering potential is typically described using the independent atom model, which completely neglects valence bonding and its effect on the transmitting electrons. As instrumentation has advanced, new measurements have revealed subtle details of the scattering potential that were previously not accessible to experiment. We have created an open-source simulation code designed to meet these demands by integrating the ability to calculate the potential via density functional theory (DFT) with a flexible modular software design. abTEM can simulate most standard imaging modes and incorporates the latest algorithmic developments. The development of new techniques requires a program that is accessible to domain experts without extensive programming experience. abTEM is written purely in Python and designed for easy modification and extension. The effective use of modern open-source libraries makes the performance of abTEM highly competitive with existing optimized codes on both CPUs and GPUs and allows us to leverage an extensive ecosystem of libraries, such as the Atomic Simulation Environment and the DFT code GPAW. abTEM is designed to work in an interactive Python notebook, creating a seamless and reproducible workflow from defining an atomic structure, calculating molecular dynamics (MD) and electrostatic potentials, to the analysis of results, all in a single, easy-to-read document.  This article provides ongoing documentation of abTEM development. In this first version, we show use cases for hexagonal boron nitride, where valence bonding can be detected, a 4D-STEM simulation of molybdenum disulfide including ptychographic phase reconstruction, a comparison of MD and frozen phonon modeling for convergent-beam electron diffraction of a 2.6-million-atom silicon system, and a performance comparison of our fast implementation of the PRISM algorithm for a decahedral 20000-atom gold nanoparticle.

ab initio description of bonding for transmission electron microscopy

The simulation of transmission electron microscopy (TEM) images or diffraction patterns is often required to interpret their contrast and extract specimen features. This is especially true for high-resolution phase-contrast imaging of materials, but electron scattering simulations based on atomistic models are widely used in materials science and structural biology. Since electron scattering is dominated by the nuclear cores, the scattering potential is typically described by the widely applied independent atom model. This approximation is fast and fairly accurate, especially for scanning TEM (STEM) annular dark-field contrast, but it completely neglects valence bonding and its effect on the transmitting electrons. However, an emerging trend in electron microscopy is to use new instrumentation and methods to extract the maximum amount of information from each electron. This is evident in the increasing popularity of techniques such as 4D-STEM combined with ptychography in materials science, and cryogenic microcrystal electron diffraction in structural biology, where subtle differences in the scattering potential may be both measurable and contain additional insights. Thus, there is increasing interest in electron scattering simulations based on electrostatic potentials obtained from first principles, mainly via density functional theory, which was previously mainly required for holography. In this Review, we discuss the motivation and basis for these developments, survey the pioneering work that has been published thus far, and give our outlook for the future. We argue that a physically better justified ab initio description of the scattering potential is both useful and viable for an increasing number of systems, and we expect such simulations to steadily gain in popularity and importance.

Angle-resolved STEM using an iris aperture: Scattering contributions and sources of error for the quantitative analysis in Si

The angle-resolved electron scattering is investigated in scanning-transmission electron microscopy (STEM) using a motorised iris aperture placed above a conventional annular detector. The electron intensity scattered into various angle ranges is compared with simulations that were carried out in the frozen-lattice approximation. As figure of merit for the agreement of experiment and simulation we evaluate the specimen thickness which is compared with the thickness obtained from position-averaged convergent beam electron diffraction (PACBED). We find deviations whose strengths depend on the angular range of the detected electrons. As possible sources of error we investigate, for example, the influences of amorphous surface layers, inelastic scattering (plasmon excitation), phonon-correlation within the frozen-lattice approach, and distortions in the diffraction plane of the microscope. The evaluation is performed for four experimental thicknesses and two angle-resolved STEM series under different camera lengths. The results clearly show that especially for scattering angles below 50 mrad, it is mandatory that the simulations take scattering effects into account which are usually neglected for simulating high-angle scattering. Most influences predominantly affect the low-angle range, but also high scattering angles can be affected (e.g. by amorphous surface covering).

Influence of plasmon excitations on atomic-resolution quantitative 4D scanning transmission electron microscopy

Scanning transmission electron microscopy (STEM) allows to gain quantitative information on the atomic-scale structure and composition of materials, satisfying one of todays major needs in the development of novel nanoscale devices. The aim of this study is to quantify the impact of inelastic, i.e. plasmon excitations (PE), on the angular dependence of STEM intensities and answer the question whether these excitations are responsible for a drastic mismatch between experiments and contemporary image simulations observed at scattering angles below \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sim $$\end{document}∼ 40 mrad. For the two materials silicon and platinum, the angular dependencies of elastic and inelastic scattering are investigated. We utilize energy filtering in two complementary microscopes, which are representative for the systems used for quantitative STEM, to form position-averaged diffraction patterns as well as atomically resolved 4D STEM data sets for different energy ranges. The resulting five-dimensional data are used to elucidate the distinct features in real and momentum space for different energy losses. We find different angular distributions for the elastic and inelastic scattering, resulting in an increased low-angle intensity (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sim $$\end{document}∼ 10–40 mrad). The ratio of inelastic/elastic scattering increases with rising sample thickness, while the general shape of the angular dependency is maintained. Moreover, the ratio increases with the distance to an atomic column in the low-angle regime. Since PE are usually neglected in image simulations, consequently the experimental intensity is underestimated at these angles, which especially affects bright field or low-angle annular dark field imaging. The high-angle regime, however, is unaffected. In addition, we find negligible impact of inelastic scattering on first-moment imaging in momentum-resolved STEM, which is important for STEM techniques to measure internal electric fields in functional nanostructures. To resolve the discrepancies between experiment and simulation, we present an adopted simulation scheme including PE. This study highlights the necessity to take into account PE to achieve quantitative agreement between simulation and experiment. Besides solving the fundamental question of missing physics in established simulations, this finally allows for the quantitative evaluation of low-angle scattering, which contains valuable information about the material investigated.

Efficient and Versatile Model for Vibrational STEM-EELS

We introduce a novel method for the simulation of the impact scattering in vibrational scanning transmission electron microscopy electron energy loss spectroscopy simulations. The phonon-loss process is modeled by a combination of molecular dynamics and elastic multislice calculations within a modified frozen phonon approximation. The key idea is thereby to use a so-called δ thermostat in the classical molecular dynamics simulation to generate frequency dependent configurations of the vibrating specimen's atomic structure. The method includes correlated motion of atoms and provides vibrational spectrum images at a cost comparable to standard frozen phonon calculations. We demonstrate good agreement of our method with simulations and experiments for a 15 nm flake of hexagonal boron nitride.

Temperature-dependent atomic B factor: an ab initio calculation

The Debye-Waller factor explains the temperature dependence of the intensities of X-ray or neutron diffraction peaks. It is defined in terms of the B matrix whose elements B_αβ are mean-square atomic displacements in different directions. These quantities, introduced in several contexts, account for the effects of temperature and quantum fluctuations on the lattice dynamics. This paper presents an implementation of the B factor (8π²B_αβ) in the thermo_pw software, a driver of Quantum ESPRESSO routines that provides several thermodynamic properties of materials. The B factor can be calculated from the ab initio phonon frequencies and displacements or can be estimated, although less accurately, from the elastic constants, using the Debye model. The B factors are computed for a few elemental crystals: silicon, ruthenium, magnesium and cadmium; the harmonic approximation at fixed geometry is compared with the quasi-harmonic approximation where the B factors are calculated accounting for thermal expansion. The results are compared with the available experimental data.