Abstract: A new methodology is presented to count the number of atoms in multimetallic nanocrystals by combining energy dispersive X-ray spectroscopy (EDX) and high angle annular dark field scanning transmission electron microscopy (HAADF STEM). For this purpose, the existence of a linear relationship between the incoherent HAADF STEM and EDX images is exploited. Next to the number of atoms for each element in the atomic columns, the method also allows quantification of the error in the obtained number of atoms, which is of importance given the noisy nature of the acquired EDX signals. Using experimental images of an Au@Ag core–shell nanorod, it is demonstrated that 3D structural information can be extracted at the atomic scale. Furthermore, simulated data of an Au@Pt core–shell nanorod show the prospect to characterize heterogeneous nanostructures with adjacent atomic numbers.

Abstract: The charge density is among the most fundamental solid state properties determining bonding, electrical characteristics, and adsorption or catalysis at surfaces. While atomic-scale charge densities have as yet been retrieved by solid state theory, we demonstrate both charge density and electric field mapping across a mono-/bilayer boundary in 2D MoS2 by momentum-resolved scanning transmission electron microscopy. Based on consistency of the four-dimensional experimental data, statistical parameter estimation and dynamical electron scattering simulations using strain-relaxed supercells, we are able to identify an AA-type bilayer stacking and charge depletion at the Mo-terminated layer edge.

Abstract: Electron tomography is essential for investigating the three-dimensional (3D) structure of nanomaterials. However, many of these materials, such as metal-organic frameworks (MOFs), are extremely sensitive to electron radiation, making it difficult to acquire a series of projection images for electron tomography without inducing electron-beam damage. Another significant challenge is the high contrast in high-angle annular dark field scanning transmission electron microscopy that can be expected for nanocomposites composed of a metal nanoparticle and an MOF. This strong contrast leads to so-called metal artifacts in the 3D reconstruction. To overcome these limitations, we here present low-dose electron tomography based on four-dimensional scanning transmission electron microscopy (4D-STEM) data sets, collected using an ultrafast and highly sensitive direct electron detector. As a proof of concept, we demonstrate the applicability of the method for an Au nanostar embedded in a ZIF-8 MOF, which is of great interest for applications in various fields, including drug delivery.

Abstract: We present a computational imaging mode for large scale electron microscopy data, which retrieves a complex wave from noisy/sparse intensity recordings using a deep learning approach and subsequently reconstructs an image of the specimen from the Convolutional Neural Network (CNN) predicted exit waves. We demonstrate that an appropriate forward model in combination with open data frameworks can be used to generate large synthetic datasets for training. In combination with augmenting the data with Poisson noise corresponding to varying dose-values, we effectively eliminate overfitting issues. The U-NET[1] based architecture of the CNN is adapted to the task at hand and performs well while maintaining a relatively small size and fast performance. The validity of the approach is confirmed by comparing the reconstruction to well-established methods using simulated, as well as real electron microscopy data. The proposed method is shown to be effective particularly in the low dose range, evident by strong suppression of noise, good spatial resolution, and sensitivity to different atom types, enabling the simultaneous visualisation of light and heavy elements and making different atomic species distinguishable. Since the method acts on a very local scale and is comparatively fast it bears the potential to be used for near-real-time reconstruction during data acquisition.