High-angle annular dark-field imaging of self-assembled Ge islands on Si(0 0 1)

Low-resolution high-angle annular dark-field (HAADF) imaging is applied to the study of coherent Ge islands on a Si(0 0 1) substrate. Experimental HAADF images reveal a complicated pattern for a coherent Ge island under (0 0 1) zone axis conditions due to strain-induced interband scattering between different Bloch-wave branches. This complicated pattern varies with objective aperture size and defocus because of the effect from the depth of field. This suggests that the strain field of a coherent Ge island can be mapped out in 3 dimensions using HAADF imaging. When samples are tilted away from dynamical conditions, image contrast agrees with the predictions from atomic number variation (Z contrast). Therefore, quantitatively compositional analysis is feasible under kinematical imaging conditions when strain contrast is suppressed. Simulations using multi-beam Bloch-wave theory agree well with the experimental images on the complicated strain-induced and through-focus images.     

Quantitative atomic resolution mapping using high-angle annular dark field scanning transmission electron microscopy

A model-based method is proposed to relatively quantify the chemical composition of atomic columns using high angle annular dark field (HAADF) scanning transmission electron microscopy (STEM) images. The method is based on a quantification of the total intensity of the scattered electrons for the individual atomic columns using statistical parameter estimation theory. In order to apply this theory, a model is required describing the image contrast of the HAADF STEM images. Therefore, a simple, effective incoherent model has been assumed which takes the probe intensity profile into account. The scattered intensities can then be estimated by fitting this model to an experimental HAADF STEM image. These estimates are used as a performance measure to distinguish between different atomic column types and to identify the nature of unknown columns with good accuracy and precision using statistical hypothesis testing. The reliability of the method is supported by means of simulated HAADF STEM images as well as a combination of experimental images and electron energy-loss spectra. It is experimentally shown that statistically meaningful information on the composition of individual columns can be obtained even if the difference in averaged atomic number Z is only 3. Using this method, quantitative mapping at atomic resolution using HAADF STEM images only has become possible without the need of simultaneously recorded electron energy loss spectra.     

An approach to the systematic distortion correction in aberration-corrected HAADF images

Systematic distortion has been analysed in high‐angle annular dark‐field (HAADF) images which may be caused by electrical interference. Strain mapping techniques have been applied to a strain‐free GaAs substrate in order to provide a broad analysis of the influence of this distortion on the determination of local strain in the heterostructure. We have developed a methodology for estimating the systematic distortion, and we correct the original images by using an algorithm that removes this systematic distortion.     

Quantitative analysis of ultrathin doping layers in semiconductors using high-angle annular dark field images

It is well known that the high-angle annular dark field (HAADF) technique in scanning transmission electron microscopy is an incoherent imaging process in the lateral ( xy ) plane. However, as a consequence of the existence of partial coherence in the z direction, accurate quantitative interpretation of image intensity is difficult. The effects of coherence in the z direction can be reduced by increasing the inner collector angle of the annular detector so that the scattering from atoms in the z direction is essentially incoherent. We thus show that it is feasible to quantify the total As concentration of ultrathin InAs _x P_{1− x} layers in InP in a simple but accurate way using a thickness integrated Bloch wave calculation including phonon scattering with a large inner collector angle of the annular detector of around 150 mrad. We compare the As composition derived from this approach with that from the Fresnel method and high resolution imaging. We also show that the non-linear variation of the HAADF intensity with thickness is consistent with our simpler simulations for such conditions. Therefore, this approach enables us easily and quickly to quantify compositions using HAADF images. The tetragonal distortion due to lattice mismatch is also shown to influence the contrast and has been included in the calculations.     

Deconvolution processing of HAADF STEM images

A deconvolution processing of high-resolution high-angle annular dark field (HAADF) scanning transmission electron microscopy (STEM) images, combined with maximum entropy method, is applied to two experimental [0 11]-Si images; one having unresolved dumbbells and the other having resolved dumbbells and artificial bright spots. The deconvoluted images for these images show bright spots corresponding to the projected atomic columns and no artificial bright spots. Thus, the deconvolution processing provides almost a real projected atomic structure by eliminating effects of the probe function from HAADF STEM images.     

A practical approach for STEM image simulation based on the FFT multislice method

It has been demonstrated that a high-angle annular dark-field (HAADF) STEM technique gives an image resolving atomic columns. Due to the diffusion of this technique and an improvement of its resolution, a practical procedure for image simulation becomes important for a quantitative interpretation of the HAADF image. In this report a new practical scheme for a STEM image simulation is developed based on the FFT multislice algorithm. Here, a HAADF intensity due to thermal diffuse scattering (TDS) is calculated from the absorptive potential corresponding to high-angle TDS and the wave function equivalent to the propagating probe within the sample. Contrary to the commonly used Bloch wave method, a coherent bright-field intensity and a coherent HAADF intensity are also obtained straightforwardly. The HAADF image contrast calculated for GaAs is not simply proportional to Z2 as expected from the Rutherford scattering at high-angle, and the As/Ga contrast ratio depends on the specimen thickness. This suggests that the generation of the HAADF signal is appreciably affected by the coherent dynamical scattering. The developed procedure here will have a definitive advantage over the Bloch wave approach for simulating the HAADF images expected from a defect and interface or amorphous materials, and also the HAADF image obtained by using a Cs-corrected microscope. This is because the former requires a huge super cell, while the latter needs a large objective aperture including a large number of incident beam directions.     

Persistent misconceptions about incoherence in electron microscopy

Incoherence in electron microscopic imaging occurs when during the observation the microscope and the object are subject to fluctuations. In order to speed up the computer simulation of the images, approximations are used that are considered as valid. In this paper we will question the validity of these approximations and show that in specific cases they can lead to erroneous results.It is shown in particular in the case of one single vibrating atom that the thermal diffuse scattering that causes the signal in HAADF STEM is not only dependent on Z but also on the mean square displacement of the atom so that it can even be large for light atoms in soft matter, provided the right HAADF aperture is used.In HREM imaging the diffuse scattering leaks out of the coherent (elastic) wave and is redistributed in the background. This might explain the mismatch in elastic contrast (Stobbs factor) especially for crystals with a thickness beyond the extinction distance, where also the HAADF signal saturates and the elastic (coherent) component vanishes.     

Contributions to the contrast in experimental high-angle annular dark-field images

This paper reports on a study of the contributions to the image contrast of high-angle annular dark field (HAADF) images acquired in scanning transmission electron microscopy. Experimental HAADF images were obtained from a model system consisting of an epitaxial perovskite PbTiO3 film grown on a SrTiO3 single crystal. This sample allowed for the study of the intensities of a wide range of atomic numbers. The main objective of the paper was to quantify the influence of TEM foil thickness on the image contrast, but the effects of the annular detector inner angle and the probe forming lens focus were also studied. Sample thicknesses ranging from approximately 10 nm to more than 400 nm were investigated. The image contrast was relatively insensitive to changes in inner angle. The main impact of sample thickness was a rapid increase in a background intensity that contributed equally to the intensities of the atomic columns and the channels between them. The background intensity and its increase with thickness reflected the average atomic number of the crystal. Subtraction of the background intensity allowed for a quantitative interpretation of image contrast in terms of atomic numbers and comparison with multislice image simulations. The consequences for the analysis of interfaces in terms of atom column occupancies are discussed.     

Effects of specimen tilt in ADF-STEM imaging of a-Si/c-Si interfaces

Annular dark field scanning transmission electron microscopy (ADF-STEM) imaging of a crystal depends strongly on specimen orientation, but for an amorphous sample it is insensitive to orientation changes. To fully investigate the effects of specimen tilt, an interface of amorphous Si (a-Si) and crystalline Si (c-Si) was rotated systematically off a zone axis in a STEM equipped with low-angle ADF (LAADF) and high-angle ADF (HAADF) detectors. The change of relative intensity across the interface shows very different trends in the LAADF and the HAADF images upon tilting. More importantly, it is found that the HAADF signal decreases much more rapidly when tilted off a zone axis than does the LAADF signal. The high-resolution lattice fringes also disappear much faster in the HAADF image than in the LAADF image. These trends reflect the fact that the channeling peaks that are responsible for scattering into the HAADF detector decrease more quickly upon tilting than the lower angle scattering to the LAADF detector does.     

Comparison of intensity distributions in tomograms from BF TEM,ADF STEM,HAADF STEM,and calculated tilt series

The three-dimensional (3D) morphology of a nanometer-sized object can be obtained using electron tomography. Variations in composition or density of the object cause variations in the reconstructed intensity. When imaging homogeneous objects, variations in reconstructed intensity are caused by the imaging technique, imaging conditions, and reconstruction. In this paper, we describe data acquisition, image processing, and 3D reconstruction to obtain and compare tomograms of magnetite crystals from bright field (BF) transmission electron microscopy (TEM), annular dark-field (ADF) scanning transmission electron microscopy (STEM), and high-angle annular dark field (HAADF) STEM tilt series. We use histograms, which plot the number of volume elements (voxels) at a given intensity vs. the intensity, to measure and quantitatively compare intensity distributions among different tomograms. In combination with numerical simulations, we determine the influence of maximum tilt angle, tilt increment, contrast changes with tilt (diffraction contrast), and the signal-to-noise ratio (SNR) as well as the choice of the reconstruction approach (weighted backprojection (WB) and sequential iterative reconstruction technique (SIRT)) on the histogram. We conclude that because ADF and HAADF STEM techniques are less affected by diffraction, and because they have a higher SNR than BF TEM, they are better suited for tomography of nanometer-sized crystals.     

Comparison of intensity distributions in tomograms from BF TEM, ADF STEM, HAADF STEM, and calculated tilt series

The three-dimensional (3D) morphology of a nanometer-sized object can be obtained using electron tomography. Variations in composition or density of the object cause variations in the reconstructed intensity. When imaging homogeneous objects, variations in reconstructed intensity are caused by the imaging technique, imaging conditions, and reconstruction. In this paper, we describe data acquisition, image processing, and 3D reconstruction to obtain and compare tomograms of magnetite crystals from bright field (BF) transmission electron microscopy (TEM), annular dark-field (ADF) scanning transmission electron microscopy (STEM), and high-angle annular dark field (HAADF) STEM tilt series. We use histograms, which plot the number of volume elements (voxels) at a given intensity vs. the intensity, to measure and quantitatively compare intensity distributions among different tomograms. In combination with numerical simulations, we determine the influence of maximum tilt angle, tilt increment, contrast changes with tilt (diffraction contrast), and the signal-to-noise ratio (SNR) as well as the choice of the reconstruction approach (weighted backprojection (WB) and sequential iterative reconstruction technique (SIRT)) on the histogram. We conclude that because ADF and HAADF STEM techniques are less affected by diffraction, and because they have a higher SNR than BF TEM, they are better suited for tomography of nanometer-sized crystals.     

Symmetries in BF and HAADF STEM image calculations

Reductions in bright-field (BF) scanning transmission electron microscopy (STEM) and high-angle annular dark-field (HAADF) STEM image calculations with the aid of Bloch wave symmetry are discussed under assumptions that an absorption potential is written by a local potential and a zero-order Laue zone lies parallel to the crystal surface. Translational symmetry allows us to take only partial incident beams in the first Brillouin zone instead of enormous number of partial incident beams in a large convergent disk. Two dimensional point group confines partial incident beams to an irreducible area in addition to factoring a dispersion matrix into noninteracting submatrices on a high symmetry line using the projection operator. The drastic reductions in computing time and memory enable us to readily calculate various BF STEM and HAADF STEM images. The validity and accuracy are demonstrated in comparisons with high resolution experimental BF STEM and HAADF STEM images.     

Essential experimental parameters for quantitative structure analysis using spherical aberration-corrected HAADF-STEM

The accuracy of quantitative analysis for Z-contrast images with a spherical aberration (C_s) corrected high-angle annular dark-field (HAADF) scanning transmission electron microscope (STEM) using SrTiO₃(0 0 1) was systematically investigated. Atomic column and background intensities were measured accurately from the experimental HAADF-STEM images obtained under exact experimental condition. We examined atomic intensity ratio dependence on experimental conditions such as defocus, convergent semi-angles, specimen thicknesses and digitalized STEM image acquisition system: brightness and contrast. In order to carry out quantitative analysis of C_s-corrected HAADF-STEM, it is essential to determine defocus, to measure specimen thickness and to fix setting of brightness, contrast and probe current. To confirm the validity and accuracy of the experimental results, we compared experimental and HAADF-STEM calculations based on the Bloch wave method.     

Imaging of high-angle annular dark-field scanning transmission electron microscopy and observations of GaN-based violet laser diodes

The first part of this paper is devoted to physics, to explain high‐angle annular dark‐field scanning transmission electron microscopy (HAADF‐STEM) imaging and to interpret why HAADF‐STEM imaging is incoherent, instructing a strict definition of interference and coherence of electron waves. Next, we present our recent investigations of InGaN/GaN multiple quantum wells and AlGaN/GaN strained‐layer superlattice claddings in GaN‐based violet laser diodes, which have been performed by HAADF‐STEM and high‐resolution field‐emission gun scanning electron microscopy.     

Quantitative structural analysis of twin boundary in alpha-Zn7Sb2O12 using HAADF STEM method

The structure and composition of the 1/4{110} twin boundary in alpha-Zn7Sb2O12 have been determined by using quantitative high-angle annular dark field scanning transmission electron microscopy (HAADF STEM) analysis. The noise in the experimental HAADF STEM images is reduced by using the maximum entropy method and average processing, and the parameters used in dynamical simulations are experimentally determined. From the analysis, it has been found that octahedral sites in the twin boundary slightly shift parallel to the [110] direction, and a reduction of the Sb concentration at the octahedral sites on the plane adjacent to the twin boundary was detected. The reduction was measured from three regions in the same twin boundary, and the Sb concentrations were 4 +/- 3, 8 +/- 3 and 19 +/-2 at% from 33 at%.     

Column ratio mapping: a processing technique for atomic resolution high-angle annular dark-field (HAADF) images

An image processing technique is presented for atomic resolution high-angle annular dark-field (HAADF) images that have been acquired using scanning transmission electron microscopy (STEM). This technique is termed column ratio mapping and involves the automated process of measuring atomic column intensity ratios in high-resolution HAADF images. This technique was developed to provide a fuller analysis of HAADF images than the usual method of drawing single intensity line profiles across a few areas of interest. For instance, column ratio mapping reveals the compositional distribution across the whole HAADF image and allows a statistical analysis and an estimation of errors. This has proven to be a very valuable technique as it can provide a more detailed assessment of the sharpness of interfacial structures from HAADF images. The technique of column ratio mapping is described in terms of a [110]-oriented zinc-blende structured AlAs/GaAs superlattice using the 1 angstroms-scale resolution capability of the aberration-corrected SuperSTEM 1 instrument.     

Multislice frozen phonon high angle annular dark-field image simulation study of Mo-V-Nb-Te-O complex oxidation catalyst "M1"

Multislice frozen phonon calculations were performed on a model structure of a complex oxide which has potential use as an ammoxidation catalyst. The structure has 11 cation sites in the framework, several of which exhibit mixed Mo/V substitution. In this paper the sensitivity of high-angle annular dark-field (HAADF) imaging to partial substitution of V for Mo in this structure is reported. While the relationship between the average V content in an atom column and the HAADF image intensity is not independent of thickness, it is a fairly weak function of thickness suggesting that HAADF STEM imaging in certain cases can provide a useful starting point for Rietveld refinements of mixed occupancy in complex materials. The thermal parameters of the various cations and oxygen anions in the model affect the amount of thermal diffuse scattering and therefore the intensity in the HAADF images. For complex materials where the structure has been derived via powder Rietveld refinement, the uncertainty in the thermal parameters may limit the accuracy of HAADF image simulations. With the current interest in quantitative microscopy, simulations need to accurately describe the electron scattering to the very high angles often subtended by a HAADF detector. For this system approximately 15% of the scattering occurs above 200 mrad at 200 kV. To simulate scattering to such high angles, very fine sampling of the projected potential is necessary which increases the computational cost of the simulation.     

Extraction of structural and chemical information from high angle annular dark‐field image by an improved peaks finding method

With the development of spherical aberration (Cs) corrected scanning transmission electron microscopy (STEM), high angle annular dark filed (HAADF) imaging technique has been widely applied in the microstructure characterization of various advanced materials with atomic resolution. However, current qualitative interpretation of the HAADF image is not enough to extract all the useful information. Here a modified peaks finding method was proposed to quantify the HAADF‐STEM image to extract structural and chemical information. Firstly, an automatic segmentation technique including numerical filters and watershed algorithm was used to define the sub‐areas for each atomic column. Then a 2D Gaussian fitting was carried out to determine the atomic column positions precisely, which provides the geometric information at the unit‐cell scale. Furthermore, a self‐adaptive integration based on the column position and the covariance of statistical Gaussian distribution were performed. The integrated intensities show very high sensitivity on the mean atomic number with improved signal‐to‐noise (S/N) ratio. Consequently, the polarization map and strain distributions were rebuilt from a HAADF‐STEM image of the rhombohedral and tetragonal BiFeO₃ interface and a MnO₂ monolayer in LaAlO₃/SrMnO₃/SrTiO₃ heterostructure was discerned from its neighbor TiO₂ layers. Microsc. Res. Tech. 79:820–826, 2016. © 2016 Wiley Periodicals, Inc.     

Nanoscale characterization of gold nanoparticles created by in situ reduction at a polymeric surface

Transmission Electron Microscopy is used as a quantitative method to measure the shapes, sizes and volumes of gold nanoparticles created at a polymeric surface by three different in situ synthesis methods. The atomic number contrast (Z‐contrast) imaging technique reveals nanoparticles which are formed on the surface of the polymer. However, with certain reducing agents, the gold nanoparticles are additionally found up to 20 nm below the polymer surface. In addition, plan‐view high‐angle annular dark‐field scanning transmission electron microscopy images were statistically analyzed on one sample to measure the volume, height and effective diameter of the gold nanoparticles and their size distributions. Depth analysis from high‐angle annular dark‐field scanning transmission electron microscopy micrographs also gives information on the dominant shape of the nanoparticles.     

Supported metal sulphide nanoclusters studied by HAADF-STEM

The morphology of MoS₂ and WS₂ nanoclusters supported on high‐surface area graphitic carbon was investigated using high angular annular dark field scanning transmission electron microscopy (HAADF‐STEM). Most of the MoS₂ (WS₂) nanoclusters contain only a single S‐Mo‐S (S‐W‐S) layer and the most commonly encountered morphology is truncated triangular. This is in contrast to the hexagonal morphology of macroscopic MoS₂ (WS₂) crystals. When in addition to molybdenum (tungsten), nickel is also present, the regular nanoclusters are truncated to a larger extent, indicating that Ni has influenced the morphology by the formation of so‐called Ni‐Mo‐S (Ni‐W‐S) structures. For these structures, the additional truncations are observed to lead to dodecahedral‐like shapes.