Microscopy Conference 2023

Prof. Dr.

Session chairs

Understanding the mechanism of molecular precursor condensation by in-situ TEM: towards the rational design of porous carbon materials for CO2 absorption

Abstract text (incl. figure legends and references) Carbon-based materials are sustainable, metal-free functional materials used for various applications; for instance—owning to their porosity—they are often exploited for CO2 capture. Their CO2 uptake is usually enhanced by introducing heteroatoms (e.g., nitrogen) and adjusting the porous structure. While nitrogen doping can be achieved through one-step condensation of cheap nitrogen-containing molecular precursors, the porous structure—if no porogen or template is applied—can be adjusted only by synthesis conditions. The choice of the latter remains rather empirical due to the lack of a fundamental understanding of the condensation mechanism. Our study aims to broaden the fundamental understanding of the condensation process and propose guidelines for designing new porous carbon materials for CO2 absorption. To realize this goal, we performed in-situ heating experiments inside a TEM and followed the condensation ex-situ by analyzing the reaction products formed at different temperatures and pressures and evaluating their adsorption properties. As a model molecular precursor, we use uric acid, which forms carbonaceous materials with high nitrogen content. To distinguish the porosity in bulk and the surface of the studied particles, we combined bright-field STEM (internal structural information) with secondary electron imaging in STEM (morphological information). The in-situ condensation in TEM starts at lower temperatures (250 °C) compared to synthesis in nitrogen atmosphere (425 °C from thermogravimetrical analysis). Figure 1 shows the secondary electron (SE) and bright-field (BF) signals from the same particle during heating. The surface of the particle is smooth at 25 ºC (Figure 1 a, SE) and becomes porous at 250 ºC (Figure 1b, SE, mesopores of ~21 nm), indicating the first temperature-induced changes – the beginning of the condensation process. Upon further isothermal heating at 250 ºC, the subsequent mass loss occurs from within the bulk of the particle through the porous surface resulting in the loss of mass first from the pre-surface area (Figure 1 c, BF) and after that from the bulk of the particle (Figure 1 d, BF). Figure 1. Secondary electron (SE) and bright-field (BF) STEM images of the particle a) at 25 °C before the in-situ heating, b) immediately after reaching 250 °C, c) after 4 minutes at 250 °C, d) after 64 minutes at 250 °C. We found that only samples in vacuum (1.5·10-4 bar) have pores of about 14 nm in diameter at the surface, while in nitrogen (1 bar), the surface of the samples appears smooth. The pressure in the reaction chamber influences the mass transfer from the surface of particles leading to more porous surfaces when synthesizing materials at low pressures, as observed in ex-situ samples in vacuum and during in-situ TEM heating (3·10-10 bar). As a result, vacuum samples demonstrate 2 times higher CO2 uptake at 500 °C than nitrogen samples. Thus, we reported the condensation of a molecular precursor in-situ for the first time. We showed that varying pressures and reaction rates result in particles with different porosity; the porosity of the surface of the particles during the early stages of condensation governs the subsequent release of volatiles and the development of a hierarchical pore structure. We gratefully acknowledge ﬁnancial support by the Max Planck Society and Dr. Janina Kossmann and Dr. Nieves Lopes for the synthesis and CO2 sorption characterization

MS 1: Energy-related materials and catalysts

MS 5: Functional organic materials

Poster

Prof.

Electron pair distribution function (ePDF) analysis of micro- and nanoplastic particles

Abstract text (incl. figure legends and references) Micro- and nanoplastics (MNPs) are considered a possible threat to microorganisms in the aquatic environment. Here, we show that total scattering intensity analysis of electron diffraction (ED) data measured by transmission electron microscopy (TEM), which yields the electron pair distribution function (ePDF),1 is a feasible method for the characterization and identification of MNPs down to 100 nm. Polyethylene (PE), polypropylene (PP), polyethylene terephthalate (PET), and polyamide microparticles, and polystyrene (PS) nanoparticles were examined by TEM and serial ED under low dose, non-cryogenic conditions. Reduced pair density functions (RDF) were calculated from the processed and background subtracted ED data and compared to simulated RDF. Nanobeam diffraction was used for the measurements of the PS and SiO2 nanoparticles. High quality serial ED suitable for PDF analysis were measured on polymer microparticles (Figure 1) and nanoparticles. The ePDF method yields a qualitative agreement of experimentally determined and simulated RDFs (Figure 2). For the nanoparticulate PS, a distinction from SiO2 particles can be made (Figure 3), where the SiO2 particles are representative of the natural inorganic matrix. By analyzing a time series of ED, we are able to show the dose dependent structural changes in the materials in the series of RDFs. Those changes are qualitatively different for the two different dose rates that were applied, which is evident in particular in the peak intensities that can be attributed to C-H-bonds and C-C-bonds. We show that detailed structural information can be gained on highly beam sensitive polymer materials by following a low dose, ePDF approach without applying cryogenic or staining techniques. The systematic appearance of C-H pair distance peaks in our measurements exemplifies an advantage of the ePDF method over X-ray based PDF, which usually cannot gain information about hydrogen. Figure 1 ED (top row) and TEM (bottom row) of cryo ball milled PE, PP, PET, and PA microparticles. Figure 2 Reduced pair density functions derived from serial ED measured on cryo ball milled PE, PP, PET, and PA microplastic particles, respectively. A) Results of measurements taken with a dose rate of ≈ 45 e/(nm2s). B) 3D model structures employed in the RDF simulation. C) Results of measurements taken with a dose rate of a dose rate of ≈ 160 e/(nm2 s). Each graph in A and C shows the experimental time series of RDFs, G(r) (color lines), the calculated RDF of a model structure, G(r) model (grey lines), and the elemental pair distribution function of the model (stacked bar charts). The color scale bar indicates the total exposure time at the end of acquisition of an ED frame. Figure 3 Comparison of ePDF of 100 nm sized PS with 100 nm sized a-SiO2 spheres. A) TEM of PS particles after ED measurements. B) First frame of serial ED measured on a PS sphere, C) Shadow image of the beam (diameter ~75 nm, indicated by the red dashed line) on the PS sphere (indicated by the white dashed line). D) TEM of a-SiO2 particles after ED measurements. E) First frame of serial ED measured on an a-SiO2 sphere F) Shadow image of the beam (diameter ~75 nm) on the a-SiO2 sphere. G) Plots of corresponding RDFs of PS (top) and SiO2 (bottom). H) Structural model of PS I) Structural model of SiO2. (1) Gorelik TE et al. Acta Cryst. 2019;75:532–49.

IM 5: Quantitative image and diffraction data analysis

Liquid phase (S)TEM of DNA origami: gold nanoparticle hybrids

Abstract text (incl. figure legends and references) Introduction DNA origami is a versatile and flexibly adaptable molecular template to control self-assembly of nanoparticles (NPs) and other nanoscopic building blocks [1]. Particle-particle and particle-substrate interactions can be controlled effectively by choosing appropriate functionalization of the NPs. As DNA-based processes require liquid environments, high-resolution microscopy applicable to liquids is needed to characterize, understand, and optimize the underlying physico-chemical mechanisms. Here, we use in-situ (scanning) transmission electron microscopy ((S)TEM) to study Au NPs, which are site-specificly bound to DNA origami structures directly in solution. Objectives The study focuses on imaging individual Au NPs adhered to DNA origami, determining the precision, to which particle sizes and positions can be measured in solution, and the characterization of beam-induced artefacts, which may impede the microscopic view and alter the solution chemistry and assembly. Materials &amp; methods Au NPs (52±1 nm in diameter) were synthesized by seed-mediated growth and covered with DNA oligomers using a freeze-thaw approach [2]. 2 x 40 x 150 nm3 large quasi 2D DNA origami pads were designed and synthesized by scaffolded self-assembly [1,3]: Long ssDNA (M13mp18) is folded by oligonucleotides, some of which were extended by ssDNA, whose sequence is comple&shy;mentary to that on the Au NP surface. Two different sequences were used for positioning of Au NP dimers with defined spacing on the pad. Ex-situ (S)TEM and in-situ liquid phase (S)TEM (LP-(S)TEM) were conducted on a JEOL F-200 micro&shy;scope equipped with a Gatan OneView camera and a HAADF detector at 200 kV. For LP-(S)TEM investigations, a Protochips Poseidon Select holder was used to keep the samples in a liquid environment that was sealed with SiN membranes against the high vacuum of the line tube. Results While ex-situ (S)TEM allows to clearly image Au NPs dried from solution and the individual DNA origami attached to them (despite large differences in contrast, see insert of Fig. 1), resolution and contrast of the LP-STEM images are reduced by beam-induced crystallization (predominantly of salts from solution). Nonetheless, upon optimizing the scanning parameters, a precision measurement of particle positions and distances is also possible in solution and first measurements indicate a preferred inter-particle distance as intended by the specifically &quot;programmed&quot; DNA pads (cf. Fig. 2). Indications are that complementary bright and dark field imaging may allow for a better resolution in STEM than in TEM. Conclusion LPEM bears a great potential for the quantitative characterization of NPs in solution. However, resolution-limiting artefacts and side effects render the continuous conduction of ex-situ control experiments indispensable. Financial support by DFG through RTG 2767 and EU&quot;s Horizon 2020 Programme (grant #964248) is gratefully acknowledged. We thank Gunnar Klös for nanoparticle synthesis and functionalization. [1] J. Zessin et al., Nano Lett. 17 (2017) 5163. [2] G. Klös et al., Zenodo (2022), https://doi.org/10.5281/zenodo.6516945. [3] P.W.K. Rothemund, Nature 440 (2006) 297.

IM 7: In situ/operando electron microscopy

Prof. Dr. hab.

Structures of Vitamin D derivatives by electron diffraction

Abstract text (incl. figure legends and references) Introduction. Vitamins D are a group of fat-soluble secosteroids. Ergocalciferol (vitamin D2) and cholecalciferol (vitamin D3) are major forms of vitamins D which occur in fungi and animal organisms.[1] 1α,25-Dixydroxyvitamin D2 (ercalcitriol, 1,25(OH)2D2) and 1α,25-dihydroxyvitamin D3 (calcitriol, 1,25(OH)2D3) are the most active forms of vitamin D2 and D3, respectively. The biological activity of vitamins D (Fig. 1) is expressed mostly through the interactions with the nuclear vitamin D receptor (VDR). VDR, after binding natural agonists, calcitriol or ercalcitriol, forms the heterodimer with nuclear retinoid X receptor (RXR). This complex binds with DNA and regulate the vitamin D-dependent gene expression. Autoimmune diseases such as type 1 diabetes, multiple sclerosis, rheumatoid arthritis, and others such as influenza or tuberculosis occur more often in countries with limited access to sunlight and thus the lowered vitamin D endogenous production. Objectives and results. We will present details of series of structures of the studied derivatives of the vitamin D resulting from kinematic refinement of the electron diffraction data collected in our Lab., comparison of the ED structures with known X-ray structures of the corresponding vitamin D derivatives, results of HAR and TAAM refinements [2] based on aspherical atomic electrostatic potentials and utilizing results of kinematic refinement, and, hopefully, also results of dynamical refinement [3] for the best ED data sets. Materials. Only a small number of the vitamin D analogues crystalize well enough for structural X-ray studies [4-6]. We have previously solved the X-ray structure of the synthetic precursor of 1,25(OH)2D2, i.e. 1a-hydroxyvitamin D2 (1a-OH-D2) and showed that the molecule adopts exclusively an A-ring chair b-conformation [7]. Most vitamin D compounds form nanocrystals which can be structurally characterized only by more advanced techniques, including electron diffraction (ED). In this communication, we will also present ED structures of 1,25(OH)2D3 analogs, the side-chain, and A-ring modified VDR agonists coded PRI-2204 and PRI-1901, respectively. References 1 A. Norman, Vitamin D; Elsevier, 2012; ISBN 978-0-323-14143-7; S. Kato, J. Biochem. (Tokyo) 127 (2000) 717 2 B. Gruza, M. L. Chodkiewicz, J. Krzeszczakowska, P. M. Dominiak, Acta Cryst.. A76 (2020) 92. 3 L. Palatinus, V. Petricek, C. A. Correa, Acta Cryst., A71 (2015) 235 4 M. Wanat, M. Malinska, A. Kutner, K. Woźniak, Molecules, 27(6) (2022) 1757 5 M. Wanat, M. Malińska, A. Kutner, K. Woźniak, Molecules, 25(8) (2020) 802 6 M. Wanat, M. Malińska, A. Kutner, K. Wozniak, Cryst. Growth Des. 18(6) (2018) 3385 7 WKołodziejski, K. Woźniak, P., J. Herold, M. Dominiak, A. Kutner, J. Mol. Structure,734, (2005) 149 Acknowledgments: This research was funded by the University of Warsaw IDUB grants BOB-IDUB-622-20/2021, IUDUB PSP: 501-D112-20-0006100 and by Polish NCN grant 2020/39/I/ST4/02904. Figure 1. Structures and the numbering system of vitamins D: (a) 1,25(OH)2D2 (ercalcitiol) and (b) 1,25(OH)2D3 (calcitriol) and structures of the vitamin D analogs studied: (c) PRI-2204, (d) PRI-1901 (Solverol), (e) 1a-hydroxyvitamin D2 and (f) PRI-2191 (Tacalcitol). H-atoms omitted for clarity.

LS 1: High-resolution cryo-EM

Polymer to carbon transition: In Situ pyrolysis of 3D printed microstructures

Abstract text (incl. figure legends and references) Carbon has been well known as a compelling material to compete with silicon in the field of microchips. One promising method to fabricate micro- or nanoscale carbon architectures is to pyrolyze a pre-patterned polymer precursor [1]. Pyrolysis is, therefore, a crucial processing parameter. However, a mechanistic understanding of the pyrolysis process for sub-micrometer polymer geometries is still missing. In particular, for the temperature regime, where the highest mass loss and, in turn, the highest shrinkage is observed, a detailed analysis of the pyrolysis-induced shrinkage kinetics, geometry dependence shape deformation, the roles of atmosphere and surface area is still lacking [2]. In this work, we aim to fill this knowledge gap and provide a comprehensive understanding of morphological deformation upon pyrolysis. Here, we report a systematic study of in situ pyrolysis process by varying environment pressure, heating temperature, and sample geometry (surface-to-volume ratio) in an environmental SEM (ESEM). We use direct laser writing of IP-Dip photoresist to print microstructures directly on MEMS heating chips [3]. We focus on the early stage of carbonization from 450 to 550 °C, where polymer precursor experiences the greatest mass loss and structural shrinkage. The structural changes are directly tracked by secondary electron imaging. We reveal the pyrolysis kinetics illustrating the temporal and temperature dependency of the deformation (Fig. 1). When evaluating the data pool by generating model-free master curves, we fully describe the dynamic process and extract the effective activation energy (Ea). After changing the environmental conditions, the shrinkage behavior turns out to be fundamentally different, largely kinetically hindered. A prevalence of the aspect ratio on kinetics and the final size becomes apparent, causing a dramatically lower Ea. To complete the picture of the structural changes during pyrolysis-induced transformation, our ongoing experiments tackle focused ion beam (FIB) micromachined cross sections of the microstruts after isothermal exposure using scanning transmission electron microscopy (STEM). Not only morphological changes can be characterized but also local structural and chemical information by electron energy loss spectroscopy in 4D-spectrum images (EELS-SI), as shown in Fig. 2. In conclusion, our in situ and scale-bridging study paves the way to a thorough understanding of morphological mechanisms upon pyrolysis and correlated property strengthening. The precise tuning of functional metamaterials by pyrolysis will facilitate the development of industrially relevant carbon device fabrication. References: [1] Schueller et al., Chem. Mater. 1997, 9, 6, 1399–1406. [2] Sharma et al., Sci Rep 8, 16282 (2018). [3] Sun et al., Microscopy and Microanalysis 27(S2) (2021) 83-84. [4] The authors acknowledge financial funding by the German Research Foundation (DFG) under Germany"s Excellence Strategy (EXC-2082/1─390761711).

High-contrast electron microscopy imaging of polymer materials for Li-batteries

Abstract text (incl. figure legends and references) Introduction Ion-conducting polymers have attracted significant interest as electrolyte membrane materials for solid-state lithium batteries because of their excellent safety, mechanical stability and flexibility. The most commonly used ion-conducting polymer is polyethylene oxide (PEO) because the excellent solubility of lithium salts leads to high ionic conductivity. Block copolymers containing PEO covalently linked to hydrophobic polymer blocks such as polystyrene, polyisoprene or polybutadiene allow designing ordered nanoscale membranes with high ionic conductivity and mechanical stability via self-assembly. Self-assembly can be controlled via block lengths to tailor nanodomain phase morphology and macroscopic orientation. Electron microscopy allows visualizing the order and orientation of the conductive block copolymer domains to optimize the conductivity of the block copolymer membranes. Objectives To correlate measured ionic conductivities of block copolymer membranes to block copolymer morphologies requires an in-depth structural characterization over length scales from nanometers to hundreds of micrometers, which can be provided by electron microscopy combining SEM and TEM images. Using cryo-ultramicrotomy we aim to show that LiTFSi-filled PEO-block copolymer membranes can be dry sectioned under humidity exclusion. Materials &amp; methods As block copolymers we investigated poly(butadiene-b-ethylene oxide) (PB-PEO) with/without added LiTFSi salt. Samples were solvent-cast into films. These were subsequently dry sectioned by cryo-ultramicrotomy and investigated by a JEOL F200 cryo TEM and Thermo Fisher Apreo Volumescope. Results We obtained thin sections of LiTFSi-loaded PB-PEO using cryo-ultramicrotomy with thickness of about 80 nm. The thin sections can be imaged at high voltages of 200 kV with STEM without staining and at 30kV with STEM in SEM. Thin sections where prepared under cryo conditions and transferred to the TEM using a cryo holder. After investigation of the sections in TEM, they were transferred to the SEM without humidity control. We find that there is no structural change during air contact. Images of loaded and unloaded samples can be compared to identify salt induced changes of the morphology. The block copolymer domains in the ribbons can be imaged by TEM by high- and by SEM by low-voltage STEM without staining. Electron microscopy provides crucial information about the domain morphology, ordering and orientation within the membrane. Conclusion We demonstrate that ultramicrotomed thin sections of block copolymers can be imaged with high contrast by STEM at voltages of 200 kV in TEM and 30 kV in SEM without staining. This allows to visualize the LiTFSi-conduction paths across the block copolymer membranes, which is of high importance to improve the conductivity of solid polymer electrolytes in Li-battery applications. Bright field (figure 1) and dark field (figure 2) image of LiTFSi loaded PB-PEO block-co-polymer

Morphology of polylactide stereocomplexes in blends of poly[(L-lactide)-co-(ε-caprolactone) and -poly(D-lactic acid) with varying molar masses

Abstract text (incl. figure legends and references) Introduction Poly[(L-lactide)-co -(ε-caprolactone)] (PLC) and poly(D-lactic acid) (PDLA) are biodegradable polyesters that can be produced from renewable resources and are now widely used in standard, biomedical and pharmaceutical applications. Improving the mechanical properties and heat resistance of materials based on poly(ε-caprolactone) (PCL) and poly(lactic acid) (PLA)) is a major concern when they are used in biomedical and consumer products [1][2]. One of the most effective methods to improve the mechanical properties and thermal stability of PLA-based materials is the formation of stereocomplexes between poly(L-lactide) (PLLA) and poly(D-lactide). This stereocomplex (SC) formation is defined as stereocomplex crystallization between PLLA and PDLA and contributes significantly to the control of polymer properties [3]. Objectives The aim of this study is to identify and visualize the different morphologies, especially crystallinity and stereocomplexation, in the blends of PLC and PDLA of different compositions prepared by solution casting, in order to understand the SC formation and the associated new properties of the polymer. Materials and methods The obtained polymer blends were morphologically studied by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Supporting analytical methods such as wide angle X-ray scattering (WAXS) and differential scanning calorimetry (DSC), were used to detect crystallization, while polymer characterization by GPC and NMR was employed to relate the morphology to polymer. Tensile tests were performed to determine the mechanical properties. Results Visualization of the morphology allowed identification of the amorphous and (stereo-) crystallizing regions. Together with the additional characterizations, maximal overall crystallization and, specifically, stereocrystallization was quantifiably shown. In polymer blends with too short lactide segment lengths, no SC could be detected. From a number average oligolactide sequence length lLA ≥ 4.2, the formation of SC was observed, which further increased with increasing lactide sequence and lattice content. However, the crystallites show a decrease in SC size, which can be attributed to phase dilution. Mixtures with a high PDLA mol mass yielded smaller SC crystallites than mixtures with lower mol mass. Limited SC formation exists due to steric hindrance of long molecular chains and partial crystallization of the homopolymer. Conclusion The totality of the applied analytical methods indicates that the adjustment of the morphology of the polymer blends at the nanoscale is necessary to achieve the desired macroscopic properties of the blends. References [1] A. T. Neffe et al., Nanomaterials 2021, 11, 1472. [2] V. Izraylit et al., Eur. Polym. J. 2020, 137, 109916 [3] X. Su et al., Polymers 2020, 12, 2515.

MS 4: Functional thin films

Flexible laser-patterned carbon coatings for sensing and energy applications

Abstract text (incl. figure legends and references) Fabricating electronic devices from natural, renewable resources has been a common goal in engineering and materials science for many years. In this regard, carbon-based coatings are of significance due to high availability of the raw materials and their environmental degradability. Carbonized materials and composites thereof have been proven as promising candidates for a wide range of future applications in flexible electronics, optoelectronics, energy storage or catalytic systems. On the industrial scale, however, their application is inhibited by tedious and expensive preparation processes and a lack of control over the processing and material parameters. A promising tool to tackle that challenge is carbon laser-pattering (CLaP) allowing for the defined and site-selective synthesis of functional carbon-based materials for flexible on-chip applications. Versatile inks, based on naturally occurring (molecular) starting materials are used to produce films, which are carbonized with a CO2-laser to obtain functional patterns of conductive porous carbon networks [1]. By chemical and physical fine-tuning of the laser-patterned carbons (LP-C), we developed high-performance flexible resistive chemical and mechanical sensors. Their properties make them applicable in many fields e.g., robotics, bionics and even health monitoring (Fig. 1). In this contribution, we present both, the material&quot;s synthesis and tailored properties as well as in-depth microscopic and spectroscopic cross-sectional analyses of such coatings (Fig. 2), which provide unprecedented insights into the microstructure and local chemistry/bonding of laser-patterned carbon [2,3]. In that regard, we as well discuss the challenges and solutions for cross-sectional preparation of such porous and sensitive structures by advanced ultramicrotomy. The gained insights in structure formation and local functionality pave the way for the utilization of alternative fast, large-scale carbonization methods such as rapid thermal annealing. Part of this work was performed at the Micro-and Nanoanalytics Facility (MNaF) of the University of Siegen.

PS 14
Poster session MS 5: Functional organic materials

Termin

Session chairs

Programm

PS 14Poster session MS 5: Functional organic materials

Termin

Session chairs

Programm

PS 14
Poster session MS 5: Functional organic materials