A Deep Convolutional Neural Network to Analyze Position Averaged Convergent Beam Electron Diffraction Patterns

We establish a series of deep convolutional neural networks to automatically analyze position averaged convergent beam electron diffraction patterns. The networks first calibrate the zero-order disk size, center position, and rotation without the need for pretreating the data. With the aligned data, additional networks then measure the sample thickness and tilt. The performance of the network is explored as a function of a variety of variables including thickness, tilt, and dose. A methodology to explore the response of the neural network to various pattern features is also presented. Processing patterns at a rate of $sim$0.1 s/pattern, the network is shown to be orders of magnitude faster than a brute force method while maintaining accuracy. The approach is thus suitable for automatically processing big, 4D STEM data. We also discuss the generality of the method to other materials/orientations as well as a hybrid approach that combines the features of the neural network with least squares fitting for even more robust analysis. The source code is available at https://github.com/subangstrom/DeepDiffraction.

In-situ real-space imaging of crystal surface reconstruction dynamics via electron microscopy

Crystal surfaces are sensitive to the surrounding environment, where atoms left with broken bonds reconstruct to minimize surface energy. In many cases, the surface can exhibit chemical properties unique from the bulk. These differences are important as they control reactions and mediate thin film growth. This is particularly true for complex oxides where certain terminating crystal planes are polar and have a net dipole moment. For polar terminations, reconstruction of atoms on the surface is the central mechanism to avoid the so called polar catastrophe. This adds to the complexity of the reconstruction where charge polarization and stoichiometry govern the final surface in addition to standard thermodynamic parameters such as temperature and partial pressure. Here we present direct, in-situ determination of polar SrTiO3 (110) surfaces at temperatures up to 900 C using cross-sectional aberration corrected scanning transmission electron microscopy (STEM). Under these conditions, we observe the coexistence of various surface structures that change as a function of temperature. As the specimen temperature is lowered, the reconstructed surface evolves due to thermal mismatch with the substrate. Periodic defects, similar to dislocations, are found in these surface structures and act to relieve stress due to mismatch. Combining STEM observations and electron spectroscopy with density functional theory, we find a combination of lattice misfit and charge compensation for stabilization. Beyond the characterization of these complex reconstructions, we have developed a general framework that opens a new pathway to simultaneously investigate the surface and near surface regions of single crystals as a function of environment.