Transmission Electron Microscopy Image Simulations Research Articles

Trifunctional Graphene-Sandwiched Heterojunction-Embedded Layered Lattice Electrocatalyst for High Performance in Zn-Air Battery-Driven Water Splitting.

Zn-air battery (ZAB)-driven water splitting holds great promise as a next-generation energy conversion technology, but its large overpotential, low activity, and poor stability for oxygen reduction reaction (ORR), oxygen evolution reaction (OER), and hydrogen evolution reaction (HER) remain obstacles. Here, a trifunctional graphene-sandwiched, heterojunction-embedded layered lattice (G-SHELL) electrocatalyst offering a solution to these challenges are reported. Its hollow core-layered shell morphology promotes ion transport to Co3S4 for OER and graphene-sandwiched MoS2 for ORR/HER, while its heterojunction-induced internal electric fields facilitate electron migration. The structural characteristics of G-SHELL are thoroughly investigated using X-ray absorption spectroscopy. Additionally, atomic-resolution transmission electron microscopy (TEM) images align well with the DFT-relaxed structures and simulated TEM images, further confirming its structure. It exhibits an approximately threefold smaller ORR charge transfer resistance than Pt/C, a lower OER overpotential and Tafel slope than RuO₂, and excellent HER overpotential and Tafel slope, while outlasting noble metals in terms of durability. Ex situ X-ray photoelectron spectroscopy analysis under varying potentials by examining the peak shifts and ratios (Co2+/Co3+ and Mo4+/Mo6+) elucidates electrocatalytic reaction mechanisms. Furthermore, the ZAB with G-SHELL outperforms Pt/C+RuO2 in terms of energy density (797Wh kg-1) and peak power density (275.8mW cm-2), realizing the ZAB-driven water splitting.

Quantifying the thickness of WTe2 using atomic-resolution STEM simulations and supervised machine learning.

For two-dimensional (2D) materials, the exact thickness of the material often dictates its physical and chemical properties. The 2D quantum material WTe2 possesses properties that vary significantly from a single layer to multiple layers, yet it has a complicated crystal structure that makes it difficult to differentiate thicknesses in atomic-resolution images. Furthermore, its air sensitivity and susceptibility to electron beam-induced damage heighten the need for direct ways to determine the thickness and atomic structure without acquiring multiple measurements or transferring samples in ambient atmosphere. Here, we demonstrate a new method to identify the thickness up to ten van der Waals layers in Td-WTe2 using atomic-resolution high-angle annular dark-field scanning transmission electron microscopy image simulation. Our approach is based on analyzing the intensity line profiles of overlapping atomic columns and building a standard neural network model from the line profile features. We observe that it is possible to clearly distinguish between even and odd thicknesses (up to seven layers), without using machine learning, by comparing the deconvoluted peak intensity ratios or the area ratios. The standard neural network model trained on the line profile features allows thicknesses to be distinguished up to ten layers and exhibits an accuracy of up to 94% in the presence of Gaussian and Poisson noise. This method efficiently quantifies thicknesses in Td-WTe2, can be extended to related 2D materials, and provides a pathway to characterize precise atomic structures, including local thickness variations and atomic defects, for few-layer 2D materials with overlapping atomic column positions.

TEM Image Analysis and Simulation Physics for Two-Step Recrystallization of Discretely Amorphized C3H5-Molecular-Ion-Implanted Silicon Substrate Surface

In this study, we investigate the initial rapid recrystallization of a discretely amorphized C3H5-molecular-ion-implanted silicon (Si) substrate surface in the subsequent thermal annealing treatment through the analysis of plan-view transmission electron microscopy (TEM) images and technology computer-aided design (TCAD) process simulation. In the approach of the analysis of the plan-view TEM image of the Si substrate surface, we found that initial rapid recrystallization occurs in the intermediate regions between the residual crystalline and discrete amorphous regions formed in the C3H5-molecular-ion-implanted Si substrate surface. In addition, the TCAD process simulation results indicate that the intermediate regions correspond to the amorphous pockets formed around the discrete amorphous regions in the C3H5-molecular-ion-implanted Si substrate surface and are recrystallized preferentially during the short thermal annealing time. These plan-view TEM image analysis and TCAD process simulation results reveal a two-step recrystallization of the discretely amorphized C3H5-molecular-ion-implaned Si substrate surface. After the initial rapid recrystallization of amorphous pockets in the 1st step, the recrystallization of discrete amorphous regions starts in the 2nd step. The incubation period between the 1st and 2nd steps is the time required to recrystallize the amorphous pockets around the discrete amorphous regions completely and redefine the amorphous/crystalline interface.

Asymmetric nanoparticle oxidation observed in-situ by the evolution of diffraction contrast

The use of transmission electron microscopy (TEM) to observe real-time structural and compositional changes has proven to be a valuable tool for understanding the dynamic behavior of nanomaterials. However, identifying the nanoparticles of interest typically require an obvious change in position, size, or structure, as compositional changes may not be noticeable during the experiment. Oxidation or reduction can often result in subtle volume changes only, so elucidating mechanisms in real-time requires atomic-scale resolution or in-situ electron energy loss spectroscopy, which may not be widely accessible. Here, by monitoring the evolution of diffraction contrast, we can observe both structural and compositional changes in iron oxide nanoparticles, specifically the oxidation from a wüstite-magnetite (FeO@Fe3O4) core–shell nanoparticle to single crystalline magnetite, Fe3O4 nanoparticle. The in-situ TEM images reveal a distinctive light and dark contrast known as the ‘Ashby-Brown contrast’, which is a result of coherent strain across the core–shell interface. As the nanoparticles fully oxidize to Fe3O4, the diffraction contrast evolves and then disappears completely, which is then confirmed by modeling and simulation of TEM images. This represents a new, simplified approach to tracking the oxidation or reduction mechanisms of nanoparticles using in-situ TEM experiments.

Defect Detection in Atomic Resolution Transmission Electron Microscopy Images Using Machine Learning

Point defects play a fundamental role in the discovery of new materials due to their strong influence on material properties and behavior. At present, imaging techniques based on transmission electron microscopy (TEM) are widely employed for characterizing point defects in materials. However, current methods for defect detection predominantly involve visual inspection of TEM images, which is laborious and poses difficulties in materials where defect related contrast is weak or ambiguous. Recent efforts to develop machine learning methods for the detection of point defects in TEM images have focused on supervised methods that require labeled training data that is generated via simulation. Motivated by a desire for machine learning methods that can be trained on experimental data, we propose two self-supervised machine learning algorithms that are trained solely on images that are defect-free. Our proposed methods use principal components analysis (PCA) and convolutional neural networks (CNN) to analyze a TEM image and predict the location of a defect. Using simulated TEM images, we show that PCA can be used to accurately locate point defects in the case where there is no imaging noise. In the case where there is imaging noise, we show that incorporating a CNN dramatically improves model performance. Our models rely on a novel approach that uses the residual between a TEM image and its PCA reconstruction.

Comparative image simulations for phase-plate transmission electron microscopy

Numerous physical phase plates (PP) for phase-contrast enhancement in transmission electron microscopy (TEM) have been proposed and studied with the hole-free or Volta PP having a high impact and interest in recent years. This study is concerned with comparative TEM image simulations considering realistic descriptions of various PP approaches and samples from three different fields of application covering a large range of object sizes. The simulated images provide an illustrative characterization of the typical image appearance and common artifacts of the different PPs and the influence of simulation parameters especially important for PP simulations. A quantitative contrast analysis shows the superior phase-shifting properties of the hole-free phase plate for biological applications and the benefits of adjustable phase plates. The application of PPs in high-resolution TEM imaging, especially of weak-phase objects such as (atomically thin) 2D materials, is shown to increase image interpretability. The software with graphical user interface written and used for the presented simulations is available for free usage.

Scanning transmission electron microscopy image simulations of complex dislocation structures generated by discrete dislocation dynamics

Scanning Transmission Electron Microscopy Diffraction Contrast Imaging (STEM-DCI) has been gaining popularity for the identification and analysis of dislocations in crystalline materials due to its ability to supress undesirable image features that are often present in conventional TEM images. However, there does not yet exist a robust body of work demonstrating expected contrast in these imaging conditions. A novel approach for the simulation of STEM-DCI images was developed using a modified form of the scattering matrix formalism. This algorithm was used to simulate a variety of dislocation configurations generated using three-dimensional discrete dislocation dynamics.

Machine Learning to Reveal Nanoparticle Dynamics from Liquid-Phase TEM Videos.

Liquid-phase transmission electron microscopy (TEM) has been recently applied to materials chemistry to gain fundamental understanding of various reaction and phase transition dynamics at nanometer resolution. However, quantitative extraction of physical and chemical parameters from the liquid-phase TEM videos remains bottlenecked by the lack of automated analysis methods compatible with the videos’ high noisiness and spatial heterogeneity. Here, we integrate, for the first time, liquid-phase TEM imaging with our customized analysis framework based on a machine learning model called U-Net neural network. This combination is made possible by our workflow to generate simulated TEM images as the training data with well-defined ground truth. We apply this framework to three typical systems of colloidal nanoparticles, concerning their diffusion and interaction, reaction kinetics, and assembly dynamics, all resolved in real-time and real-space by liquid-phase TEM. A diversity of properties for differently shaped anisotropic nanoparticles are mapped, including the anisotropic interaction landscape of nanoprisms, curvature-dependent and staged etching profiles of nanorods, and an unexpected kinetic law of first-order chaining assembly of concave nanocubes. These systems representing properties at the nanoscale are otherwise experimentally inaccessible. Compared to the prevalent image segmentation methods, U-Net shows a superior capability to predict the position and shape boundary of nanoparticles from highly noisy and fluctuating background—a challenge common and sometimes inevitable in liquid-phase TEM videos. We expect our framework to push the potency of liquid-phase TEM to its full quantitative level and to shed insights, in high-throughput and statistically significant fashion, on the nanoscale dynamics of synthetic and biological nanomaterials.

The formation and accumulation of radiation-induced defects and the role of lamellar interfaces in radiation damage of titanium aluminum alloy irradiated with Kr-ions at room temperature

A transmission electron microscopy (TEM) specimen of a titanium aluminide (TiAl) alloy was irradiated in-situ, at the IVEM-TANDEM facility, with 1 MeV Kr ions to a maximum fluence of 1.25 × 1019 ions m−2 at room temperature. The irradiated microstructure was then investigated ex-situ using advanced analytical TEM in combination with TEM image simulations and molecular dynamics (MD) simulations. The TEM examination showed that dot-like defects first formed in both the α2-Ti3Al and γ-TiAl phases of the irradiated microstructure. With increasing irradiation dose planar defects formed and propagated within the γ phase, and most of the planar defects were accompanied by the dot-like defects. The TEM image simulations and MD irradiation simulations revealed that the origins of the dot-like and planar defects were interstitial clusters and stacking faults, respectively, and large interstitial clusters were surrounded by dislocation loops. The interstitial clusters formed and grew in the irradiated microstructure due to much faster diffusion of interstitials than that of vacancies at room temperature. Local stress concentrations increased near large interstitial clusters and lamellar interfaces, which resulted in the nucleation and propagation of stacking faults. Moreover, the configurations of the lamellar interfaces in the start microstructure were found to play an important role in the formation and accumulation of the radiation-induced interstitial clusters and stacking faults at room temperature.

Multislice image simulations of sheared needle-like precipitates in an Al-Mg-Si alloy.

The image contrast of sheared needle-like precipitates in the Al-Mg-Si alloy system is investigated with respect to shear-plane positions, the number of shear-planes, and the active matrix slip systems through multislice transmission electron microscopy image simulations and the frozen phonon approximation. It is found that annular dark field scanning transmission electron microscopy (ADF STEM) images are mostly affected by shear-planes within a distance ∼6-18 unit cells from the specimen surface, whereas about 5-10 equidistant shear-planes are required to produce clear differences in HRTEM images. The contrast of the images is affected by the Burgers vector of the slip, but not the slip plane. The simulation results are discussed and compared to experimental data. LAY DESCRIPTION: Pure aluminium is too soft to be viable in most structural applications, but this may be remedied by alloying the metal with various elements. Adding small amounts of silicon and magnesium to pure aluminium allows small particles to precipitate during heat treatment. These precipitates resist plastic deformation and can increase the strength of the alloy and make it viable for a range of industrial applications, such as automotive door panels and load-bearing profiles. However, if subjected to large loads, the precipitates are sheared and the strength of the alloy changes dynamically. Designing safe products such as cars or buildings require physically based predictions on this dynamical change. Developing models that can provide such predictions depend in turn on experimental observations of the shearing process. Because the precipitates are nm long, experimental observations must be done by transmission electron microscopy. However, understanding these results sometimes require computer simulations of atomic models. In this work, we have performed image simulations of various models of sheared precipitates and compared the results with earlier experiments. The simulations indicate that certain conditions must be met for the sheared precipitates to appear different from unsheared precipitates. These conditions are most likely to be met if precipitates are sheared several times in a relatively homogeneous manner. This is important for two reasons. First, a localized shearing process would lead to large dynamical changes in precipitate strength during deformation, and in turn drastically reduce the work hardening of the alloy. Secondly, a localized shearing process would have promoted earlier fracture and failure of the alloy during deformation. Finally, our results also show how different slip directions influences the images of precipitates. In the future, these influences can be used to further understand the shearing process of these precipitates. Hence, our results can be used to improve model predictions of strength, work hardening, and fracture. In turn, this may improve alloy design and reduce the use of prototype testing in, e.g. the automotiveindustry.

Detailed investigation of the shearing mechanism of β" precipitates in Al-Mg-Si alloys

The mechanism behind shearing of β11 precipitates in Al-Mg-Si alloys during deformation is investigated by applying advanced transmission electron microscopy (TEM) techniques and frozen phonon multislice TEM image simulations on a selection of shearing configurations. In particular, the results indicate that the needle-like precipitates are sheared several times in single matrix Burgers vector steps. The multislice image simulations suggest that shearing events are most likely achieved in single Burgers vector steps, and there are some experimental evidence that the shearing planes are the matrix glide planes.

Atomistic modelling of diamond-type SixGeyCzSn1−x−y−z crystals for realistic transmission electron microscopy image simulations

The realistic simulation of transmission electron microscopy (TEM) images requires an accurate definition of the positions of all atoms, which are linked to the mechanical properties of the material. This paper proposes an optimized atomistic modeling approach to model the lattice parameters and elastic properties of Si, Ge, diamond, alpha-tin, and related diamond alloys, with an approach compatible with systems bigger than 50000 atoms. In order to compute precisely the elastically strained SixGeyCzSn1−x−y−z diamond crystals, a dedicated parameterization of the Keating force field is provided. An original periodic boundary strategy is provided. Our tool is successfully used to interpret experimental TEM data with a reasonable accuracy and precision in a time scale about 10000 times faster than ab initio methods. The method predicts the correct lattice parameters and elastic constants of elementary compounds and alloys with a deviation inferior to 8.1%. We show that subsequent Monte-Carlo simulations predict original self-ordering effects in C in good agreement with the theory. An original approach is used to quantify the short-range and long-range order in comparison with high-resolution cross-sectional TEM experiments: the projected radial distribution function (p-RDF) appears to be a universal and very sensitive analytical tool to quantify the matching between our atomistic model and the experimental HR(S)TEM results. For our reference Si-Ge multilayer with 20 millions of atoms, a maximum broadening of 100 pm is obtained for the third-nearest neighbor (3nn) simulated peak of the p-RDF compared to the experimental one. The same value is obtained from a template matching analysis of the maximum local displacements between the projected experimental atomic positions and the corresponding simulation.

Fast approximate STEM image simulations from a machine learning model

Accurate quantum mechanical scanning transmission electron microscopy image simulation methods such as the multislice method require computation times that are too large to use in applications in high-resolution materials imaging that require very large numbers of simulated images. However, higher-speed simulation methods based on linear imaging models, such as the convolution method, are often not accurate enough for use in these applications. We present a method that generates an image from the convolution of an object function and the probe intensity, and then uses a multivariate polynomial fit to a dataset of corresponding multislice and convolution images to correct it. We develop and validate this method using simulated images of Pt and Pt–Mo nanoparticles and find that for these systems, once the polynomial is fit, the method runs about six orders of magnitude faster than parallelized CPU implementations of the multislice method while achieving a 1 − R2 error of 0.010–0.015 and root-mean-square error/standard deviation of dataset being predicted of about 0.1 when compared to full multislice simulations.

Model‐based geometry reconstruction of quantum dots from TEM

AbstractThe reconstruction of geometric properties of semiconductor quantum dots (QDs) from imaging of bulk‐like samples (thickness 100‐300 nm) by transmission electron microscopy (TEM) is a difficult problem. A direct reconstruction by solving the tomography problem is not feasible due to the limited image resolution (0.5‐1 nm), the highly nonlinear behavior of the dynamic electron scattering, nonlocal effects due to strain and strong stochastic influences due to uncertainties in the experiment. We present a mathematical model for the numerical simulation of TEM images. This includes elasticity theory to obtain the strain profile coupled with the Darwin‐Howie‐Whelan equations, describing the propagation of the electron wave through the sample. Furthermore, we introduce a novel concept for 3D model‐based geometry reconstruction of QDs from TEM images. The approach includes a geometric model for the QD configuration in real space, a database of simulated TEM images and a statistical procedure for the estimation of the geometric properties and the classification of QD types.

Efficient first principles simulation of electron scattering factors for transmission electron microscopy

Electron microscopy is a powerful tool for studying the properties of materials down to their atomic structure. In many cases, the quantitative interpretation of images requires simulations based on atomistic structure models. These typically use the independent atom approximation that neglects bonding effects, which may, however, be measurable and of physical interest. Since all electrons and the nuclear cores contribute to the scattering potential, simulations that go beyond this approximation have relied on computationally highly demanding all-electron calculations. Here, we describe a new method to generate ab initio electrostatic potentials when describing the core electrons by projector functions. Combined with an interface to quantitative image simulations, this implementation enables an easy and fast means to model electron scattering. We compare simulated transmission electron microscopy images and diffraction patterns to experimental data, showing an accuracy equivalent to earlier all-electron calculations at a much lower computational cost.

Characterization of TEM Moiré Patterns Originating from Two Monolayer Graphenes Grown on the Front and Back Sides of a Copper Substrate by CVD Method

The number of layers affects the electronic properties of graphene owing to its unique band structure, called the Dirac corn. Raman spectroscopy is a key diagnostic tool for identifying the number of graphene layers and for determining their physical properties. Here, we observed moiré structures in transmission electron microscopy (TEM) observations; these are signature patterns in multilayer, although Raman spectra showed the typical intensity of the 2D/G peak in the monolayer. We also performed a multi-slice TEM image simulation to compare the 3D atomic structures of the two graphene membranes with experimental TEM images. We found that the experimental moiré image was constructed with a 9–12 Å interlayer distance between graphene membranes. This structure was constructed by transferring CVD-grown graphene films that formed on both sides of the Cu substrate at once.

Lorentz Transmission Electron Microscopy Image Simulations of Experimental Nano-Chessboard Observations in Co-Pt Alloys.

The magnetization configuration of a novel nano-chessboard structure consisting of L10 and L12 phases in a Co40Pt60 alloy is investigated using Lorentz transmission electron microscopy (LTEM) and micro-magnetic simulations. We show high-resolution LTEM images of nano-size magnetic features acquired through spherical aberration correction in Lorentz Fresnel mode. Phase reconstructions and LTEM image simulations are carried out to fully understand the magnetic microstructure. The experimental Fresnel images of the nano-chessboard structure show zig-zag shaped magnetic domain walls at the inter-phase boundaries between L10 and L12 phases. A circular magnetization distribution with vortex and anti-vortex type arrangement is evident in the phase reconstructed magnetic induction maps as well as simulated maps. The magnetic contrast in experimental LTEM images is interpreted with the help of magnetic induction maps simulated for various relative electron beam-sample orientations inside the TEM.

Identification of Extended Defect Atomic Configurations in Silicon Through Transmission Electron Microscopy Image Simulation

We used atomistic simulation tools to correlate experimental transmission electron microscopy images of extended defects in crystalline silicon with their structures at an atomic level. Reliable atomic configurations of extended defects were generated using classical molecular dynamics simulations. Simulated high-resolution transmission electron microscopy (HRTEM) images of obtained defects were compared to experimental images reported in the literature. We validated the developed procedure with the configurations proposed in the literature for {113} and {111} rod-like defects. We also proposed from our procedure configurations for {111} and {001} dislocation loops with simulated HRTEM images in excellent agreement with experimental images.

Molecular dynamics simulations and morphology analysis of TEM imaged PVDF nanofibers

With the goal of elucidating the structure of polyvinylidene difluoride (PVDF) nanofibers, all-atom molecular dynamics simulations were performed, and the results were compared with structures observed in high resolution transmission electron microscopy (TEM) at the molecular level. Simulation shows that the stability of the β-phase component in a PVDF nanofiber is influenced by the diameter of the nanofiber. In unstretched ultra-thin nanofibers, the β phase is readily transformed into a paraelectric phase. During the TEM imaging process, chain scission due to irradiation can accelerate this process. This transformation illuminates the spindle formation and serpentine motion of molecular segments observed by Zhong et al. (Polymer, 54, 2013, 3745–3756) in irradiated PVDF nanofibers. From a comparison between simulated and experimental TEM images it was possible to identify numerous features that are useful in unveiling inherent structures of PVDF nanofibers. These include the orientations of elongated dots with respect to the chain axis; kinks along the chain profile; varying sizes, shapes, and brightness of dots; and distances between dots. This information allows one to distinguish image of short chain segments having a conformation approximating the α phase from those approximating the β phase.

A fast image simulation algorithm for scanning transmission electron microscopy

Image simulation for scanning transmission electron microscopy at atomic resolution for samples with realistic dimensions can require very large computation times using existing simulation algorithms. We present a new algorithm named PRISM that combines features of the two most commonly used algorithms, namely the Bloch wave and multislice methods. PRISM uses a Fourier interpolation factor f that has typical values of 4–20 for atomic resolution simulations. We show that in many cases PRISM can provide a speedup that scales with f4 compared to multislice simulations, with a negligible loss of accuracy. We demonstrate the usefulness of this method with large-scale scanning transmission electron microscopy image simulations of a crystalline nanoparticle on an amorphous carbon substrate.