Microscopy Conference 2023

Prof. Dr.

Session chairs

Prof.

Modern (direct electron) detectors and advanced data recording schemes have pushed the level to which images and diffraction patterns can be quantified, thus bringing experimental data closer to the realm of simulations and theory. At the same time, simulation methods are capable of handling larger atomic structures and include more experimental parameters; and also data reconstruction algorithms have advanced in speed, accuracy and flexibility, thus bringing theory closer to experimental reality. This session welcomes contributions that present the state-of-the-art in extracting quantitative insight/knowledge from image and diffraction data, but also advances in simulation tools.

Advances in ptychography with a focus on low-dose and 3D applications

Abstract text (incl. figure legends and references) The use of fast pixelated detectors optimized for four-dimensional scanning transmission electron microscopy (4D-STEM) experiments has revolutionized the type and amount of data that can nowadays be acquired at an electron microscope. In combination with the development of new computational hardware and reconstruction algorithms, microscopy techniques, such as ptychography, are about to replace conventional STEM imaging methods due to their superior image resolution of electron radiation-sensitive materials [1,2]. Ptychography in particular has already become a mature electron microscopy technique and has experienced immense advances in the last years [3-5]. The current research to further improve this technique is driven by the desire to investigate thick samples as well as to measure with a low electron dose. Here, we will discuss the advances in ptychography during the last couple of years. We will take a look at how researches have tried to relax the strict scanning constraint, which is required for obtaining information redundancy through consecutively overlapping probes. For example, we will look at one approach that involves a defocused probe rastered across the specimen, with a less dense scan pattern and discuss its drawbacks and suggest alternatives. Further, we will examine the measures that have been taken to generate high resolution 3D images of thicker samples. We will discuss possible improvements of the conditioning of the 3D reconstruction problem in order to extent the depth resolution and we will compare the results to those of other STEM techniques. Finally, we will outline future research directions and unresolved problems for ptychography [6]. References: [1] Jiang, Yi, et al. "Electron ptychography of 2D materials to deep sub-ångström resolution." Nature 559.7714 (2018): 343-349. [2] Chen, Zhen, et al. "Electron ptychography achieves atomic-resolution limits set by lattice vibrations." Science 372.6544 (2021): 826-831. [3] Wakonig, Klaus, et al. "PtychoShelves, a versatile high-level framework for high-performance analysis of ptychographic data." Journal of applied crystallography 53.2 (2020): 574-586. [4] Schloz, Marcel, et al. "Overcoming information reduced data and experimentally uncertain parameters in ptychography with regularized optimization." Optics Express 28.19 (2020): 28306-28323. [5] Pelz, Philipp M., et al. "Solving Complex Nanostructures With Ptychographic Atomic Electron Tomography." arXiv preprint arXiv:2206.08958 (2022). [6] M.S., T.C.P. and C.T.K. acknowledge financial support from the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) - Project-ID 414984028 - SFB 1404. J.M. acknowledges support by the DFG Grant No. KO 2911/12-1, KO 2911/13-1, BR 5095/2-1 and Project-ID 182087777 - SFB951. M.S. and J.M. acknowledge financial support by the Volkswagen Foundation (Initiative: "Experiment!", Project: "Beyond mechanical stiffness").

IM 5: Quantitative image and diffraction data analysis

LS 4: Image analysis of large data sets

Invited talk

Prof. Dr. hab.

Organic crystal structure refinement from 3D electron diffraction data with application of dynamical theory and more accurate scattering factors

Abstract text (incl. figure legends and references) In recent years, one can observe spectacular developments in the 3D electron diffraction (3D ED) methods. Practitioners of this method are learning how to improve data collection, among others how to diminish radiation damage and how to deal with dynamical scattering, data processing, and refinement. Currently, available structures from 3D ED reach R-factors even below 10% and resolution around dmin = 0.5 Å. Refinement with a dynamical approach [1] against such data allows for observation of residual density in bonding paths or lone electron pair regions. Therefore, it is possible and profitable to use more sophisticated aspherical models of atomic electrostatic potential instead of the standard independent atom model (IAM). We have already proposed transferable aspherical atom model (TAAM) refinements against 3D ED data in kinematic approximation [2]. The TAAM was parametrized with the Multipolar Atom Types from Theory and Statistical clustering (MATTS) databank (successor of UBDB2018 [3]). Now TAAM is coupled with dynamical refinement and available in Jana2020 [4]. Here, we present refinements of exemplary organic crystal structures, urea, and α-glycine, with the application of the kinematic approach. Next, we show the refinement of 1-methyluracil crystal structure against dmin = 0.56 Å data with TAAM in the dynamical approach. There is a visible clearing of the residual density maps, also lowering of the maximum and minimum residual values, and a further lowering of R-factors. The combination of various variants of TAAM with the dynamical approach allows us to conclude that quantitative information about electron density distribution due to chemical bonding, the presence of lone electron pairs, partial charges on atoms, and charge density polarization due to intermolecular interactions can be extracted from the 3D ED data (Fig.1). In the future, we will go beyond TAAM and refine the parameters of the multipole model of electron density. Fig. 1. Residual electrostatic potential maps for dynamical refinements of 1-methyluracil crystal structure with application of IAM (left), TAAM (middle), and periodic TAAM (right) scattering models. Contour level 0.13 eÅ-1, yellow: positive, cyan: negative. Thermal ellipsoids with 50% probability. References [1] L. Palatinus, V. Petricek, and C. A. Correa, Acta Crystallogr. A, 71 (2015), 235–244. [2] B. Gruza, M. L. Chodkiewicz, J. Krzeszczakowska, and P. M. Dominiak, Acta Crystallogr. A, 76 (2020), 92–109. [3] P. Kumar, B. Gruza, S. A. Bojarowski, P. M. Dominiak, Acta Crystallogr. A, 75 (2019), 398-408. [4] V. Petrícek, M. Dusek, and L. Palatinus, Zeitschrift für Krist. - Cryst. Mater., 229 (2014), 345–352. Acknowledgments: This research was funded by National Science Centre, Poland 2020/39/I/ST4/02904 and University of Warsaw IDUB grant BOB-IDUB-622-20/2021 (,,Infrastructure for Cryomicroscopy and Electron Diffraction Core Facility").

MS 5: Functional organic materials

DI Dr.

Prof. Dr. rer. nat. habil.

Probing 3D reciprocal space of textured nanocrystalline thin films using electron diffraction tomography

Abstract text (incl. figure legends and references) Texture in crystalline materials dictates their anisotropy properties, e.g. opto-electronic properties in light harvesting polycrystalline semiconductors. Texture is conventionally evaluated by probing the reciprocal space using X-ray and/or electron diffraction (ED) methods. In organic bulk heterojunctions of organic solar cells (OSCs) grazing incidence wide angle X-ray scattering (GIWAXS) is a well-established method. However, real space information is missing to correlate domain structure with crystal texture. TEM can deliver both real and reciprocal space as well as analytical signals. However, ED in OSCs is not well explored largely due to radiation damage challenges. In this work we establish a method to probe the 3D reciprocal space of textured nanocrystalline thin films by means of electron diffraction tomography (EDT), complemented by 4D-STEM analysis. As sample systems a radiation robust AlN thin film (Fig. 1) and a radiation sensitive active layer of OSC blend (Fig. 2) are chosen. A tilt series of ED patterns over ±70° is acquired with a single-axis tilt holder. A 3D reciprocal space representation of the data is reconstructed using Process Electron Tilt Series (PETS) software and the corresponding 2D slices are extracted (cf. Fig. 1b and Fig. 2e) for comparison with GIWAXS. The texture is further studied using 4D-STEM and 4D-SCED [1]. The AlN thin film shows a pronounced fiber texture (columnar grains of ~50 nm diameter and ~500 nm length grown along direction, Fig. 1a) and its beam robustness qualifies it as reference for the described method. Figure 1b shows an overview of the EDT workflow. The 3D reciprocal lattice pattern confirms random in-plane rotation of grains and small out-of-plane tilt. The local grain orientation, mapped with ACOM [2] based on nano-beam ED 4D-STEM data (Fig. 1c), shows that the tilted grains are clustered in groups. For the OSC samples, composed of small molecule donor DRCN5T and acceptor PC71BM, dominating texture of π-stacking face-on and edge-on were identified based on GIWAXS studies (Fig. 2c and ref. [3]). In TEM energy-filtering of SAED (EF-SAED) is necessary to suppress inelastic background scattering. Due to the limited dose budget, each ED pattern in the tilt series was acquired at fresh sample areas. The reconstructed 3D reciprocal lattice is compared with the GIWAXS data showing good agreement. A fully quantitative comparison is underway and will be presented together with pros and cons of the two methods. [1] M. Wu, et. al., Nat. Commun. (2022) 13, 2911 [2] C. Ophus, et. al., Microsc. Microanal. (2022) 28, 390 [3] M. Berlinghof, et. al., Z. Kristallogr. (2020) 235, 15 The authors gratefully acknowledge financial support by the DFG via the Collaborative Research Centre SFB 953 "Synthetic Carbon Allotropes". Figure 1: Characterization of highly textured columnar AlN thin film. (a) BF TEM images. (b) ED patterns from a tilt series, principle of 3D reciprocal space reconstruction and 3D view of the reciprocal space. (c) visualized texture maps based on 4D-STEM data. Figure 2: Characterization of active layer DRCN5T:PC71BM after solvent vapour annealing. (a) molecule structure of donor DRCN5T (sulphur-rich) and acceptor PC71BM (carbon-rich) and scheme of donor crystal structure. (b) EF-SAED, (c) GIWAXS, (d) nano-morphology based on STEM-EELS data. (e) 3D reciprocal space from EF-SAED EDT. (f) face-on (grayscale) and edge-on (coloured) donor domains revealed in 4D-SCED data.

Abstract talk

Parametrized diffraction inversion by gradient-based optimization employing a fully differentiable multislice

Abstract text (incl. figure legends and references) With the availability of momentum-resolved STEM data, a multitude of phase retrieval methods were established. Most can be grouped into two categories; direct inversions employing a single object transmission function (OTF) and iterative techniques capable of reconstructing both single- and multiple slice OTFs as well as the STEM probe. Importantly, these optimization schemes act on the pixel-wise distribution of amplitude and phase of OTF and probe. In fact, the diffraction patterns (DP) correspond to a relatively sparse set [P] of parameters, e.g., aberration coefficients, positions and types of atoms, specimen tilt, coherence parameters and probe positions. The goal of an inversion procedure is the retrieval of [P] iteratively from a starting guess using the gradients with respect to [P]. We introduce a fully differentiable multislice scheme TorchSlice implemented in Pytorch, utilizing GPUs. Gradients are obtained by backpropagation, a method widely used in machine learning. Using backpropagation for inverse multislice was introduced in [1, 2]. Here, we demonstrate the capability of this approach in two steps. First, we employ the availability of gradients to study optimum experimental designs. In particular, a 4D STEM simulation of 2D and bulk MoS2 including vacancies is performed, resembling an experimental data set. A parametrized perfect crystal is then assumed to calculate the gradients with respect to the presence or absence of atoms at their native sites, for different STEM signals, doses, probe foci and aberrations. Fig 1 depicts the success rate for the detection of a vacancy in dependence of dose and focus for the 4D STEM and the STEM bright field signals. It is concluded that 4D STEM is a lot more dose-efficient and robust than BF imaging with respect to vacancy detection. Secondly, we address the retrieval of bonding effects from 4D STEM data of hBN and GaAs. DFT simulations are used to generate DP including electron redistributions due to chemical bonding. Then, a model based on isolated atoms is set up as initialization for a gradient-based optimization. That allows to evaluate a priori what dose levels and experimental setups are most promising to detect bonding effects. The hBN monolayer required a significantly higher dose level, electrons per DP, while the bonding effect of a 12nm GaAs crystal appeared already at electrons per DP. It is conceivable that an elevated thickness helps to detect bonding effects. Fig 2 depicts the successful reconstruction of bonding effects for GaAs. Finally, we address further prospects and show first reconstructions of experimental data of 2D heterostructures and ferroelectrics, where we determined the cation displacements from slightly misoriented PZT crystals. Figure 1: detection probability of a vacancy depending on the dose and the defocus, respectively and from bright field or 4d-stem data, respectively. Probabilities were calculated over 70 runs Figure 2: Phase grating (PG) in the isolated atom approximation for GaAs on the left; Difference between PG with bonding and PG at the end of reconstruction on the right. The PG was recovered at the area covered by the probes [1] W. Van den Broek and C. T. Koch, Physical review letters, 2012, 10.1103/PhysRevLett.109.245502 [2] W. Van den Broek and C. T. Koch, Physical Review B, 2013, 10.1103/PhysRevB.87.184108 [3] Support from Helmholtz under contracts VH-NG 1317, ZT-I-0025 and ZT-I-PF-5-028 is gratefully acknowledged.

IM 6: Phase-related techniques & 4D STEM

Optimized segmentation for quantitative 3D analysis of electron tomography reconstructions

Abstract text (incl. figure legends and references) Nanoparticle assemblies are of great interest due to their tunable structure and properties, and their application in various fields.1,2 For example, the coexistence of Au nanoparticle and quantum dots (QDs) enables dual imaging, fluorescence and light scattering, and/or photothermal treatment of cancer cells.3 By changing the number of particles, and/or the interparticle distances, it becomes possible to reach specifically desired structures and properties for targeted usage.4 Hereby, it is very important to investigate the structure-property relationship of the nanoparticle assemblies. In order to extract quantitative information of the 3D structure, an accurate characterization at the nanometer level is required. Transmission electron microscopy (TEM) is a useful technique to study nanomaterials. However, since conventional TEM is limited to 2D projections, it is important to perform electron tomography experiments to obtain reliable information for 3D structures. An important step to extract quantitative 3D information is the segmentation of the 3D data set. Manual segmentation is hereby the most conventional but also most time consuming technique. To overcome this limitations, advanced reconstruction algorithms, such as the discrete algebraic reconstruction technique (DART),5 and the sparse sphere reconstruction algorithm (SSR) 6 have been developed in which segmentation is part of the 3D reconstruction. However, they are computationally expensive and might not be ideal for assemblies where the nanoparticles are agglomerated or not uniform in size. In this work, we therefore focused on the optimization of the segmentation process and we have compared three different segmentation methods to extract quantitative parameters from the 3D reconstructions. In additional to the manual segmentation (Figure 1), we made use of a watershed function7, which separates the agglomerated particles from their catchment points for each local minima, and a spherical Hough function, which detects each particle separately by defining their radius and center point.8 Results showed that the spherical Hough function, is the most promising technique to extract quantitative information from the 3D reconstructions of the nanoassemblies. In this contribution, I will present the results and comparison of these segmentation methods, including their advantages and limitations. 1. Hanske, C. et al. M. Advanced Materials 30, 1707003 (2018). 2. Colloidal Synthesis of Plasmonic Nanometals. (2020). 3. Jin, Y. &amp; Gao, X. Nature Nanotechnology 4, 571–576 (2009). 4. Hollingsworth, J. A. et al. MRS Bulletin 40, 768–776 (2015). 5. Batenburg, K. J. et al. Ultramicroscopy 109, 730–740 (2009). 6. Zanaga, D. et al. Nanoscale 8, 292–299 (2016). 7. Beucher, S. &amp; Meyer, F. in Mathematical Morphology in Image Processing (CRC Press, 1993). 8. Hough, P. V. C. Method and Means for Recognizing Complex Patterns. (1962). Figure 1: The samples: CdSe/CdS quantum dots encapsulated by polymeric micelles (a,e), Au nanosphere and CDSe/CdS quantum dots encapsulated by polymeric micelles (b,f), Au nanorod and CdSe/CdS quantum dots assembled on silica shell (c,g), and Au nanorod and PbS quantum dots assembled on silica shell (d,h). The first row (a-d); 3D visualizations of the reconstructions of nanoassemblies, and the second row (e-h); manual segmentation of the same structures. (Scale bars represent 30 nm).

IM 5
Quantitative image and diffraction data analysis

Appointment

Session chairs

Description

Programme

IM 5Quantitative image and diffraction data analysis

Appointment

Session chairs

Description

Programme

IM 5
Quantitative image and diffraction data analysis