Center for Functional Nanomaterials Seminar

Despite being almost a century old, transmission electron microscopy (TEM) is arguably developing faster than even, enhancing its capabilities for fundamental and applied research. Scanning TEM (STEM) has overtaken conventional TEM in popularity in the last decades, partly due to the fact that it can locally collect multidimensional data which can nowadays be efficiently processed. Two thriving fields of this development have been momentum-resolved or so called "4D-STEM" and STEM based high-resolution electron energy loss spectroscopy (HR-EELS) for which hardware developments have enabled vibrational spectroscopy. In this talk, I will discuss own contributions to the development of both techniques and applications to materials and life science. 4D-STEM data sets are very versatile and powerful but limited by artifacts of their serial acquisition as well as their bulky sizes. Remedies for these two shortcomings will be presented here. To cope with drift and scan artifacts of 4D-STEM data sets, non-rigid registration schemes known from 2D data, such as HAADF images, are adapted to 4D data sets [1]. The scheme to register series of 4D-STEM data and the resulting high fidelity 4D-STEM maps will be presented. Another issue of 4D-STEM is the cumbersome nature of its huge data sets. It will be shown how live processing can reduce the data size by orders of magnitude before storing it while preserving relevant information [2]. This led to basis-transformations that allow for ptychographic reconstructions at atomic resolution for heavily compressed data. Installing a novel spectrometer prism together with Nion led to an unprecedented combination of spatial and energy resolution, allowing to map phonons at extended defects at atomic resolution. Localized phonon modes at grain boundaries in Si will be evidenced, in excellent quantitative agreement with accompanying first principles calculations, and proposed as phononic waveguides [3]. The experimental reconstruction of phonon dispersion surfaces (two momentum and one energy dimension) will be demonstrated and possible applications highlighted [4]. Cryo vibrational EELS will be established as a tool to study beam damage via bond breaking. Finally, the conjunction of 4D-STEM and EELS in the form of energy-resolved 4D-STEM will be demonstrated and prospects discussed for different energy-loss regimes. [1] C. O'Leary & B. Haas et al., Microsc. Microanal. 28, 1417 (2022). [2] B. Haas et al., Microsc. Microanal. 27 (S1), 994 (2021). [3] B. Haas et al., Nano Lett. 23, 5975 (2023). [4] B. Haas et al., Microsc. Microanal. 29 (S1), 356 (2023). Brief bio: Benedikt Haas started his career in electron microscopy during his master project (physics) at Philipps-Universität in Marburg (Germany) in 2010/11, where he investigated organic semiconductors in the group of Kerstin Volz. He defended his PhD on method development for imaging and diffraction based quantitative STEM at the CEA Grenoble (France) under the supervision of Jean-Luc Rouvière in 2017. Afterwards, he did a postdoc in the group of Christoph Koch at Humboldt-Universität zu Berlin (Germany) until 2020 where he has been a staff scientist in charge of the Nion HERMES microscope since, working on 4D-STEM as well as high-resolution EELS. He is currently PI of a German-French project on resonant phonon pumping in the electron microscope.