Techniques Development
Atomic-Resolution Cryo-STEM Across Continuously Variable Temperature by BH Goodge, E Bianco, N Schnitzer, HW Zandbergen, and LF Kourkoutis, Microsc Microanal | doi:10.1017/S1431927620001427 Atomic-resolution cryogenic STEM provides a path to
probe the microscopic nature of low-temperature phases in a wide variety of materials. Successful high-resolution cryo- experiments have been limited to certain fixed temperatures dictated by the cryogen,
leaving most of a material’s phase
space unexplored. Te novel side-entry continuously variable- temperature (CVT) liquid nitrogen cryo-holder specifically addresses this issue: the combination of liquid nitrogen cooling with local MEMS sample heating allows precise temperature control between ∼100–1000 degrees K (Figure). Additional design considerations including a large-volume cryogen reservoir and active rod temperature compensation further mitigate many challenges of cryo-STEM experiments, dramatically reducing driſt rates and enabling consistent sub-Å imaging resolution. Tese proof-of-concept experiments mark significant progress for the accessibility of variable-temperature cryo-electron microscopy, opening the doors to a new range of high-resolution in situ cryo-experiments including real-time observation of phase transitions, temperature cycling, and access to phases with narrow stable temperature windows.
Single-frame HAADF-STEM images of a standard Au sample using the dual-tilt CVT liquid nitrogen cryo-holder show consistent sub-Å resolution across the ∼100–1000 K temperature range.
Materials Applications
Deep Neural Network Enabled Space Group Identification in EBSD by K Kaufmann, C Zhu, AS Rosengarten, and KS Vecchio, Microsc Microanal | doi:10.1017/S1431927620001506 Electron backscatter diffraction (EBSD) is one of the
primary tools in materials development and analysis. Phase identification and differentiation are necessary components of this technique for analysis of single- and multi-phase samples. Aſter collection, current EBSD pattern indexing techniques can differentiate between a user-selected set of phases only if those contain sufficiently different crystal structures. To address these challenges, we report a machine learning-based technique for space group classification of diffraction patterns without user- selected phases or other inputs. Each diffraction pattern is individually analyzed in real-time by a deep neural network. Real-world performance of the deep neural network is gauged by presenting data from materials it has not encountered before and assessing its predictions. Te ability to classify new data with exceptional accuracy was investigated through heatmaps of localized feature importance (Figure). Tis reveals the importance of features beyond Kikuchi lines and diffraction maxima and enables AI-assisted phase identification, enhanced phase differentiation, and autonomous phase mapping.
66 doi:10.1017/S1551929520001030
A heatmap highlighting the important regions for determining the correct space group (225) for the underlying NbC diffraction pattern.
www.microscopy-today.com • 2020 July
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84
