We present an unsupervised machine learning approach for segmentation of static and dynamic atomic-resolution microscopy data sets in the form of images and video sequences. In our approach, we first extract local features via symmetry operations. Subsequent dimension reduction and clustering analysis are performed in feature space to assign pattern labels to each pixel. Furthermore, we propose the stride and upsampling scheme as well as separability analysis to speed up the segmentation process of image sequences. We apply our approach to static atomic-resolution scanning transmission electron microscopy images and video sequences. Our code is released as a python module that can be used as a standalone program or as a plugin to other microscopy packages.

The present article reports on a comprehensive atomic-scale investigation on S-series precipitates in 2024 aluminum alloy. Cs-corrected high-angle annular dark-field scanning transmission electron microscopy is applied to reveal the fine atomic-scale structure of precipitates at early ageing state. Geometrics phase analysis is used for elucidating the induced strain field from precipitates. The precipitate sequence of S-series precipitates in 2024 Al alloy is identified as follows: super saturated solid solutions (s.s.s.s.) → clusters (GPB zone) → S′ phase → S phase. The interfaces between precipitates acting as precursor of S′ phase are well characterized. One typical characteristic of S-series phase precipitates is the coexistence of clusters and subsequent metastable phases. Transformation of metastable phases is characterized. Corresponding hardness structure relationship is revealed, and S′ phase is considered as the key strengthening structure in S-series precipitates in 2024 Al alloys.

Aging treatment plays an important role in strengthening of 2198 Al–Li alloy. Through a serious of heat treatment processes, a large amount of precipitates emerge, mainly observed to be θ′(Al2Cu), Al3Zr, and T1(Al2CuLi), among which, T1 turns to be the most important precipitate that contributes to the strengthening of 2198 Al–Li alloy. While the temperature of the aging process is 175 °C, the density and size of T1 phase keep increasing through the process and reach peak in about 18 h. T1 phase tends to have a certain orientation relationship of ${\left( {0001} \right)_{{{\rm{T}}_1}}}//{\left\{ {111} \right\}_{{\rm{Al}}}}$, ${\left\langle {1010} \right\rangle _{{{\rm{T}}_{\rm{1}}}}}//{\left\langle {110} \right\rangle _{{\rm{Al}}}}$ and may have different kinds of multilayered structures. In most of the multilayered structures, the distance between two adjacent copper-rich laths is less than that in classical single-layered phase. Thus, it can be inferred that the microstructure of T1 phase might change in the process of developing from single-layered structure to multilayered structure. In addition, the interactions between different phases become relatively frequent when the density of T1 phase reaches a threshold.

This paper reports on an atomic-scale investigation into the β′ precipitates and the long-period stacking ordered phase (LPSO) in Mg–5Y–2.5Ni–0.5Zr (at.%) alloy, using Cs-corrected high angle annular dark field-scanning transmission electron microscopy (HAADF-STEM). The results displayed that the 18R-type and 14H-type LPSO phases coexisted in the as-cast and the solid solution states, and the 18R-type and 14H-type LPSO structures were thermal stable. After aging treatment, the aging peak hardness reached 138 HV at 225 °C for 48 h. The significant increase in hardness was attributed to the formation of the metastable β′ phase. The lattice parameters of a and b axes for β′ phases are a = 0.65 nm, b = 2.20 nm, and c = 0.52 nm by HAADF-STEM. The interaction between the LPSO phase and the β′ can be found. The atomic-scale interactions between the LPSO and β′ phases are divided into two parts: under-aging and peak-aging conditions between the building blocks.

The distribution of octahedral cations in the two component layers of a 7 Å–14 Å interstratified mineral with a bulk chemical composition (Fe4.122+Mg0.07Mn0.01Al1.69□0.11)(Si2.56Al1.44) O10(OH)8 was investigated using high-angle annular dark field scanning transmission electron microscopy (HAADF-STEM) in combination with the image simulations. In the 14 Å component layers, comparison between the observed and simulated images revealed that the M4 sites of the interlayer sheets were occupied preferentially by Al together with a small amount of Fe; the other M1, M2 and M3 sites were occupied by dominant Fe and residual Al in equal proportions. Two types of octahedral sheets with disordered and ordered cation distributions were recognized in the 7 Å component layers. The two types of sheets were similar to the octahedral sheet of the 2:1 layer and the interlayer sheet in the 14 Å layer above, respectively. Irregular vertical stacking and lateral contact of the different component layers in structure and chemistry characterized the interstratification, which may be caused by rapid precipitation and accretion of the component layers in hydrothermal environments.

The morphology, composition, and structure of precipitates in an Al–Si–Mg–Hf alloy after heat treatment at 560°C for 20 h were studied by means of Cs-corrected high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM), energy dispersive X-ray spectrometry (EDS), high-resolution transmission electron microscopy (HRTEM), and first-principle calculations. Precipitates with three kinds of morphologies were observed. The rectangular and square precipitates were predominantly (Si2−xAlx)Hf phases, while the nanobelt-like precipitate is the Si2Hf phase. First-principle calculations were used to show that the Si6 and Si8 sites were the most favorable sites for Al incorporation in the orthorhombic Si2Hf phase.

We present a simple and robust method to acquire quantitative maps of compositional fluctuations in nanostructures from low magnification high-angle annular dark field (HAADF) micrographs calibrated by energy-dispersive X-ray (EDX) spectroscopy in scanning transmission electron microscopy (STEM) mode. We show that a nonuniform background in HAADF-STEM micrographs can be eliminated, to a first approximation, by use of a suitable analytic function. The uncertainty in probe position when collecting an EDX spectrum renders the calibration of HAADF-STEM micrographs indirect, and a statistical approach has been developed to determine the position with confidence. Our analysis procedure, presented in a flowchart to facilitate the successful implementation of the method by users, was applied to discontinuous InGaN/GaN quantum wells in order to obtain quantitative determinations of compositional fluctuations on the nanoscale.

Properties of gold nanoparticles (AuNPs) are very different from bulk gold, in particular, highly dispersed AuNPs exhibit high catalytic activities on metal oxide supports. Catalytic activities of AuNPs are strongly dependent on: (i) size and morphology; (ii) synthesis methods; (iii) nature of the support; (iv) interaction between AuNPs and the support; and (v) oxidation state of AuNPs in the synthesized catalysts. A goal is to maintain the size and to prohibit aggregation of AuNPs, since aggregations deteriorate catalytic activities. Some strong interactions are therefore required between AuNPs and their supports to prevent the movement of AuNPs. SBA-15 is a promising material for the support of AuNPs since it has ordered two-dimensional hexagonal pore channels, uniform pore size ranging from 5 to 30 nm, narrow pore size distribution, thick amorphous walls ranging from 3 to 6 nm, and high surface area. In this study, SBA-15, TiO2-SBA-15 and TiO2-SBA-15-AuNP nanocomposites were synthesized by the sol-gel method and microstructural characterizations were carried out by both X-ray diffraction analysis and electron microscopy.

Doping of silicon with chalcogens (S, Se, Te) by femtosecond laser irradiation to concentrations well above the solubility limit leads to near-unity optical absorptance in the visible and infrared (IR) range and is a promising route toward silicon-based IR optoelectronics. However, open questions remain about the nature of the IR absorptance and in particular about the impact of the dopant distribution and possible role of dopant diffusion. Here we use electron tomography using a high-angle annular dark-field (HAADF) detector in a scanning transmission electron microscope (STEM) to extract information about the three-dimensional distribution of selenium dopants in silicon and correlate these findings with the optical properties of selenium-doped silicon. We quantify the tomography results to extract information about the size distribution and density of selenium precipitates. Our results show correlation between nanoscale distribution of dopants and the observed sub-band gap optical absorptance and demonstrate the feasibility of HAADF-STEM tomography for the investigation of dopant distribution in highly-doped semiconductors.

We show in this article that it is possible to obtain elemental compositional maps and profiles with atomic-column resolution across an InxGa1−xAs multilayer structure from 5th-order aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) images. The compositional profiles obtained from the analysis of HAADF-STEM images describe accurately the distribution of In in the studied multilayer in good agreement with Muraki's segregation model [Muraki, K., Fukatsu, S., Shiraki, Y. & Ito, R. (1992). Surface segregation of In atoms during molecular beam epitaxy and its influence on the energy levels in InGaAs/GaAs quantums wells. Appl Phys Lett61, 557–559].

Graphene is a single layer of carbon atoms arranged in a hexagonal lattice. The high carrier mobility and mechanical robustness of single layer graphene make it an attractive material for “beyond CMOS” devices. The current work investigates through high-resolution transmission electron microscopy (HRTEM) image simulation the sensitivity of aberration-corrected HRTEM to the different graphene stacking configurations AAA/ABA/ABC as well as bilayers with rotational misorientations between the individual layers. High-angle annular dark field–scanning transmission electron microscopy simulation is also explored. Images calculated using the multislice approximation show discernable differences between the stacking sequences when simulated with realistic operating parameters in the presence of low random noise.

ZrO2:Yb3+ nanocrystalline phosphors with high concentrations of ytterbium ions were prepared using the sol-gel method. X-ray diffraction, high-angle annular-dark-field scanning transmission electron microscopy (HAADF-STEM), energy dispersive X-ray spectroscopy, and high-resolution transmission electron microscopy (HRTEM) were used to characterize the nanocrystalline phosphors annealed at 1000°C. Unit-cell distortion and changes in the crystalline structure of the monoclinic zirconia to tetragonal zirconia, and subsequently cubic zirconia, were observed with increased Yb concentration. Yb ions were randomly distributed into the lattice of the crystalline structure. No segregation of Yb2O3 phase was observed. The substitution of Zr atoms by Yb atoms on different crystalline phases was confirmed by the experimental results and theoretical simulations of HRTEM and HAADF-STEM.