Microscopy Conference 2023

Prof. Dr.

Probing atomic tilt coupling in oxide superlattices Pr0.7Ca0.3MnO3/SrRuO3 using quantitative electron microscopy

Abstract text (incl. figure legends and references) Perovskite oxides have a wide range of tuneable physical properties and are of interest for electronic and energy-related applications. In superlattices of Pr0.7Ca0.3MnO3/SrRuO3 (PCMO/SRO), the anomalous Hall effect depends sensitively on the crystal structures of the SrRuO3 layers, while the field dependence of the Hall resistivity is related to the PCMO layer thickness. Here, we use integrated differential phase contrast (iDPC) scanning transmission electron microscopy (STEM) to study the influence of layer thickness on the structures of PCMO/SRO superlattices. As high angle annular dark field (HAADF) images have poor sensitivity to light elements such as oxygen, especially in close proximity to heavy elements, we use iDPC imaging for the simultaneous visualisation of oxygen and metal atoms in real space. We determine atomic positions with sub-pixel precision and measure atomic tilt angles quantitatively. We use simulation results to establish an approximately linear relationship between optimal defocus and sample thickness in SRO, which provides a guide for tuning the defocus experimentally during iDPC STEM experiments. Our results show that atomic tilt angles in the SRO layers depend on the thicknesses of the PCMO layers, with thicker PCMO layers resulting in smaller oxygen tilt angles and different magnetic properties.

IMLB: Late breaking abstracts

MSLB: Late breaking abstracts

Poster

Tuning electronic structure and composition of FeNi nanoalloys for enhanced oxygen evolution electrocatalysis via a general synthesis strategy

Abstract text (incl. figure legends and references) Developing low-cost and efficient oxygen evolution electrocatalysts is key to decarbonization. A novel, facile, surfactant-free and gram-level biomass-assisted fast heating and cooling synthesis method is reported for synthesizing a series of carbon-encapsulated dense and uniform FeNi nanoalloys with a single-phase face-centered cubic solid-solution crystalline structure. This method also enables precise control of both size and composition. Atomic resolution scanning transmission electron microscopy (STEM) images show uniform nanoparticles with a narrow size distribution of sub-5 nm. STEM analysis also demonstrates a single-phase face-centered cubic solid-solution crystalline structure, in good agreement with XRD pattern analysis. STEM mapping and line-scan analysis show that FeNi nanoalloy experiences surface restructuring under the OER condition and turns into alloy-core/oxides shell structure. Electrochemical measurements show that among FexNi(1−x) nanoalloys, Fe0.5Ni0.5 has the best performance. DFT calculations support the experimental findings and reveal that the optimally positioned d-band center of O-covered Fe0.5Ni0.5 renders a half-filled anti-bonding state, resulting in moderate binding energies of key reaction intermediates. By increasing the total metal content from 25 to 60 wt%, the 60% Fe0.5Ni0.5/40% C shows an extraordinarily low overpotential of 219 mV at 10 mA cm-2 with a small Tafel slope of 23.2 mV dec-1 for OER, which are much lower than most other FeNi-based electrocatalysts and even the state-of-the-art RuO2. It also shows robust durability in an alkaline environment for at least 50 h. The gram-level fast heating and cooling synthesis method is extendable to a wide range of binary, ternary, quaternary nanoalloys, as well as quinary and denary high-entropy-alloy nanoparticles. [1] [1] Y. Wang, W. Nong, N. Gong, T. Salim, M. Luo, T. L. Tan, K. Hippalgaonkar, Z. Liu, Y. Huang, Small 2022, 18, 2203340.

Simultaneous raman and SEM imaging of mineral sections

Abstract text (incl. figure legends and references) Raman spectroscopy is an excellent tool for analysing the chemical bonding and structure in mineral sections with high specificity. Raman mapping/imaging is typically combined with optical microscopy to provide correlated information on samples. Sometimes, optical imaging provides less contrast for mineralogical samples compared to scanning electron microscopy (SEM) images. However, combined SEM Raman instruments are relatively uncommon and often require the sample to be moved within the SEM chamber to a different analysis position. This movement makes it challenging to ensure the SEM and Raman measurements are accurately co-localised. In this work, we demonstrate a combined SEM Raman system that is capable of simultaneous SEM and Raman imaging from the same sample area. Figure 1 illustrates a SEM image and a Raman image taken from a mineral section. Here, there is clear and accurate overlay between SEM and Raman features. Additionally, Raman imaging detected the presence of biotite which is not visible in the SEM image. Figure 1: a) SEM image of mineral section b) Corresponding Raman image illustrating the distribution of rutile (green), oligoclase (red), biotite (blue) and amphibole (purple)

LS 4: Image analysis of large data sets

Prof.

Characterization of high density (111) textured nanotwinned Ag thin film evaporated on an 8-inch Si wafer

Abstract text (incl. figure legends and references) It is known that the atomic diffusivity of (111)- orientation surface is 3 to 5 order of magnitude higher than (110) and (100) surfaces, leading to the decrease of Cu to Cu bonding temperature to 150℃. In the present study, an 8-inch Si wafer has been evaporated firstly with Ag nanotwinned thin film with a thickness about 4 μm using the assistance of ion beam bombardment. FIB cross-sectional image reveals many columnar grains with an average width about 0.5 μm containing numerous parallelly aligned Ag nanotwins. XRD spectrum showed a strong Ag (111) peak without any other crystal orientations both in the central and edge regions, indicated the (111)- preferred orientation of the Ag nanotwinned films. The results are consistent with the top-view EBSD inverse pole figure analysis showing that a very high density of (111) orientation of 85.2% on the surface of the Si wafer. Furthermore, TEM analyses indicated that the twin characteristics of the grain structure in the Ag thin film can be evidenced from the symmetry relationship between the ( )M/( )M and ( )T/( )T diffraction patterns through the inset selected area electron diffraction (SAED) analysis taken from the [011] zone axis. The twin spacing can also be measured from the HR-TEM micrograph, which are ranged from 2 to 15 nm with an average value about 6 nm. It can also be observed from the HR-TEM analyses that the Ag (111) twin boundaries revealed a very fine atomic spacing about 0.2 nm. Finally, the direct wafer bonding at quite low temperatures ranging from 250 to 100℃ has been performed. Fig.1 XRD and FIB analyses of the Ag nanotwins on Si wafer. Fig.2 EBSD and TEM analyses of the Ag nanotwins on Si wafer and low temperature direct bonding of the wafers.

LSLB: Late breaking abstracts

Atomic study of structural changes of LiNiO2 at high cutoff-voltage in lithium ion batteries

Abstract text (incl. figure legends and references) The demand of high-energy lithium-ion batteries boosts the interest in increasing Ni content in cathodes and/or raising charge cutoff voltage. The Co-free LiNiO2 (LNO) has the advantage of lower production cost, high discharge capacity and high energy density compared to traditional LiCoO2, whilst the underlying mechanisms of their capacity loss and degradation when highly charged are not fully understood. The issues along with high charged states include surface O2 evolution and structural degradation. Finding pathways to unlock extra capacity is contingent upon an exact understanding of their structural behaviours. Several decades of effort worldwide have been devoted to developing characterization techniques across multiscale to study the mechanisms involved LNO charge storage but the knowledge still lacks regarding the nanoscale evolution. Here, we use the atomic-resolution annular dark-field scanning transmission electron microscopy (ADF-STEM) combined with electron energy loss spectroscopy (EELS) to directly examine intrigue structural and chemical change mechanisms during the LNO electrochemical process. We have directly visualised the surface densification, void formation at the bulk, stacking fault evolution of LNO over charging to high cut-off voltage. We also provide detailed EELS spectroscopy analysis of Ni oxidation changes at different charged states, comparing the trends at different distances to the surface. This offers atomic-level insights of structural behaviours and degradation mechanisms of highly charge LNO in lithium-ion batteries.

Jun.-Prof. Dr.

Conductive filler networks in flexible conductive composites

Abstract text (incl. figure legends and references) Flexible, electronic hybrid, and composite materials combine mechanical elasticity and electrical conductivity by forming a conductive particle network. The performance of such "network materials&quot;, i.e., particle-based composites, battery materials, etc., is dominated by the morphology of the conductive filler particle network and the contact resistances between particles. Here, we study the structure of the particle network of conductive silver spheres in a silicone elastomer matrix. A conductive network with macroscopic conductive paths is formed at the concentration required to meet the vol% of the percolation threshold in the electrically insulating. We investigate selected morphological parameters as a function of filler volume fraction and correlate the trends with macroscopic electrical conductivities. Spherical silver particles (average diameter 2.5µm) were dispersed in a polydimethylsiloxane (PDMS) elastomer matrix with different silver volume fractions from 25 - 45 vol%. Macroscopic four-point measurements were performed at different volume fractions to analyze the electrical conductivity of the composites. We use Focused Ion Beam – nanotomography (FIB-nt) of volumes with a statistical number of particles to reveal the 3D particle network structure for each silver volume fraction (Figure 1). Post-processing of the 3D reconstruction provided information such as geometric tortuosity distributions, particle coordination number distributions, percolation fractions, particle-particle contact area distributions, and the number of shortest paths. The mean tortuosity decreased with increased filler volume fraction, indicating shorter conductive paths with fewer interfaces. The mean coordination number increased with the volume fraction, increasing the number of potentially conductive contacts. We observe an increase of the average contact area with volume fraction which indicates an increasingly jammed packing of the particles above the percolation threshold. The overall increase in conductivity could be explained by a more interconnected network with reduced particle contact resistances.

MS 7: Ceramics and composites

Phys

ChemE

Synthesis and morphological and electrochemical characterization of graphene quantum dots

Abstract text (incl. figure legends and references) Graphene quantum dots (GQDots) have outstanding properties such as low toxicity, high biocompatibility, tunable fluorescent emission, high stability, and great versatility due to their abundant edges that allow their chemical functionalization. In this sense, it is necessary to have synthesis methods that allow their production in a controlled manner, being of low cost and low environmental impact. Therefore, in this study GQDots functionalized with amino groups were obtained by means of hydrothermal synthesis. The synthesis starts by obtaining graphene through the electrochemical exfoliation of graphite, its subsequent oxidation by the modified Hummers' method, and finally applying the hydrothermal treatment to this oxide. The materials obtained in this process were characterized by different microscopic and spectroscopic techniques. Atomic force microscopy confirmed that graphene nanosheets of 6.02 ± 2.73 nm wide and 1.04 ± 0.32 μm long were obtained. On the other hand, spectroscopic techniques (UV-Vis and FTIR) allowed for verifying the effectiveness of Hummers' method for the oxidation of graphene nanosheets. After applying the hydrothermal treatment to the graphene oxide, using different reaction times and temperatures, the formation of GQDots was verified by atomic force microscopy. The expected amino functionalization of the GQDots was verified by FTIR spectroscopy. The optical properties of these quantum dots were studied by UV-Vis and fluorescence spectroscopy, showing that absorption mechanisms in UV-Vis spectrum function as secondary mechanisms of fluorescent emission. Also, GQDots with higher reaction times have stronger fluorescent emission. Finally, the GQDots were used to modify the surface of a glassy carbon electrode, studying the change in the electrochemical sensitivity of this electrode using cyclic voltammetry. GQDots under certain synthesis parameters can increase the electrochemical signal detected. However, the phenomenon that shields the interaction of the modified electrode surface is still unknown.

Ti-6Al-4V ELI alloy subjected to mechanical surface modification and gas nitriding

Abstract text (incl. figure legends and references) Titanium alloys, like most widely used Ti-6Al-4V, are well known for their high specific strength, biocompatibility and corrosion resistance. However, relatively low wear resistance hinders its eventual application when a direct contact with other ones is required. Thus, a number of methods have been developed with the aim to increase the wear resistance and extend the life-time of the parts produced of such materials. Nitriding has been confirmed as one of the most promising methods of producing a hard and well adhering TiN coating on the top of Ti-based material. Recent reports indicate that nitriding may be even more effective in temperature-time scale if fast diffusion channels are formed by increasing the density of grain boundaries and/or generating a large amount of dislocations in the substrate subsurface area by its plastic deformation. However, understanding the phase transformations taking place at the surface is possible only though the micro- and nanoscale study carried out with the help of the cutting-edge electron microscopes. Therefore, in this work, the influence of the surface plastic deformation of the Ti-6Al-4V ELI alloy on the formation of TiN coating during gas nitriding was studied. The substrate material was subjected to turning and subsequent low-temperature gas nitriding realized at 540°C for 8, 16 and 24 h. The surface topography was analyzed with Hommel Tester T1000. The microstructure characterization, phase and chemical analysis were carried out with ThermoFisher&quot;s scanning (SEM) and transmission (TEM) electron microscopes i.e. Scios 2 Dual Beam and Themis G2 200 FEG, respectively, both equipped with energy dispersive X-ray spectrometers (EDS). The surface analysis was also done with the scanning Microprobe PHI 5000 Versa Probe II X-ray photoemission spectrometer (XPS). The microhardness of the sample was determined with an FM 7 stand equipped with a Vickers indenter. The SEM, TEM and XPS characterization of the subsurface area allowed to reveal that the whole surface of the as-turned alloy is coated with un-evenly distributed amorphous tribo-layer acting as an effective nitrogen diffusion barrier preventing formation of titanium nitrides. The growth of TiN layer is also hindered by the presence of continuous a-Al2O3 at the tribo-material/substrate. The thickness of TiN layer after 24h of gas nitriding was measured to be ~100 nm. The results of indentation experiments showed an increase in hardness by ~10% measured for as-machined material and after 24 h of gas nitriding treatment. Acknowledgment: This work was financially supported by National Science Centre of Poland, grant number UMO-2020/39/D/ST8/02610

Post-processing study of focused ion beam and conventional TEM specimens

Abstract text (incl. figure legends and references) Focused ion beam sample preparation and conventional sample preparation have an important role in transmission electron microscopy (TEM). New materials (ceramics, thin films, composites, etc.) that are coming have specific properties and have to be precisely treated and usually post-treated. This poses ever-new challenges, constant adjustments, and improvements in advanced focused ion beam sample preparation, conventional sample preparation, and other TEM preparation techniques. In this chapter, we compared two nowadays most frequently used TEM sample preparation techniques; focused ion beam sample preparation and conventional sample preparation. Post-processing treatment was performed using NanoMill® (model 1040, Fischione Instruments, Inc.). The study was created for a discussion about different procedures to achieve the best result for the TEM sample. For the post-processing study of focused ion beam TEM specimens, we study two different approaches. The first TEM sample was pre-thinned using FEI Helios NanoLab NL650 dual-beam Focused Ion Beam (FIB) and cleaned with NanoMill. The second TEM sample was polished using FIB and then thinned and finally cleaned with NanoMill. For the post-processing study of the conventional TEM sample, we prepared the TEM sample in two different ways. The first TEM sample was ion-milled using PIPS (Precision Ion Polishing System, Gatan Inc.) to achieve perforation and finally post-cleaned with NanoMill. The second TEM sample was dimpled down to transparent thickness (Dimple grinder, Gatan Inc.) and then thinned and finally cleaned using NanoMill. We want to expose how important these techniques are and how high the quality of the sample that we prepared with this type of sample preparation is. Structural and analytical analysis of TEM samples were investigated on a JEOL ARM 200CF scanning transmission electron microscope with probe Cs corrector.

Characterization of Pt-rare-earth nanoparticles by STEM tomography and EDX

Abstract text (incl. figure legends and references) The search for sustainable energy sources for transport has intensified recently because of the climate crisis. In addition to batteries for electric motors, fuel cells are gaining increasing interest. Platinum-Rare Earth Metal (Pt-RE) alloy nanoparticles (NP) are promising candidates for use in proton‑exchange membrane fuel cells (PEMFCs), which are a sustainable alternative to combustion engines in automobiles and buses (1). In such PEMFCs, Pt-RE nanoparticles on carbon black can be used as a catalyst for the oxygen reduction reaction and the oxidization of H2 (Fig. 1) (1). They show long-term stability and high catalytic activity, which would be optimized by a strained Pt surface layer over the alloy core, providing optimal reactant binding energies and preventing dealloying (2). Recently, a newly developed chemical synthesis route to prepare Pt-RE nanoalloys with tunable alloy phases and particle sizes has been published (3). However, preliminary results on Pt2Gd have shown that after acid leaching to remove byproducts, nanoporous structures with yet unclear 3D distribution of elements are obtained (2, 3). Here, we present an analysis of the elemental distribution and 3D shape of Pt2Gd nanoalloys. The electron-dispersive X-ray (EDX) map in Figure 2b shows the segregation of Pt and Gd in the particle. The map has been acquired using a Super-X detector at an FEI Osiris S/TEM at 200 kV and a total dose of 3E+6 e-/Å2 from the particle depicted in Figure 2a. Noise in the elemental maps has been reduced by principal component analysis using the hyperspy toolbox (4). A comparison between the HAADF-STEM image and the elemental distribution indicates that darker areas in the STEM image contain more Gd, and appear darker due to the lower atomic number of Gd. HAADF-STEM tilt series for electron tomography of Pt2Gd NPs have been acquired using a probe‑corrected Titan3 80-300 equipped with a Fischione photomultiplier STEM detector and operated at 300kV. Figure 2c shows an orthoslice of the volume reconstructed using a workflow based on the ASTRA toolbox (5), indicating a porous structure with a denser core. A comparison with Figure 2b however suggests that the darker areas observed in Figures 2a and 2c do not correspond to voids but rather locations in the sample where the Gd might have been segregated. The investigation of a larger number of particles, ideally in conjunction with EDX tomography is necessary to further clarify this observation beyond doubt. EDX tomography, however, is challenging due to the beam sensitivity of the material. Fig. 1: Scheme of a proton-exchange membrane fuel cell. Pt-RE nanoparticles on carbon black are used as the yellow-marked anode (left) and cathode (right). Fig. 2: a) HAADF-STEM image b) Overlay elemental map of Pt (red) and Gd (blue) from the same particle as in (a). c) Orthosclice from STEM Tomography from another nanoparticle. References M. Shao, Q. Chang, J.-P. Dodelet, R. Chenitz, Chem. Rev. 116, 3594–3657 (2016). Y. Hu et al., Chem. Mater. 33, 535–546 (2021). Y. Hu, J. O. Jensen, L. N. Cleemann, B. A. Brandes, Q. Li, J. Am. Chem. Soc. 142, 953–961 (2020). F. de la Peña et al., hyperspy/hyperspy: Release v1.7.1 (2022), doi:10.5281/zenodo.6659919. W. van Aarle et al., Ultramicroscopy. 157, 35–47 (2015).

Direct atomic resolution imaging of superstructure in LiCoPO4 cathode material by simultaneous ADF- and iDPC-STEM modes

Abstract text (incl. figure legends and references) LiCoPO4 is a promising cathode material for the high voltage lithium-ion batteries. However, the moderate cycling stability of the material limits its practical applications. One of the major reasons of the capacity fade of LiCoPO4 is related to its structural degradation: Li and Co site exchange leads to the blocking of the Li+ transport channels[1]. The stabilization of the atomic structure of the material directly depends on the capability to gain knowledge on the degradation mechanisms though advanced characterization. Cathode materials are obstinate candidates for TEM investigations due to their poor stability under the electron beam. To deal with radiolysis while imaging LiCoPO4 reduced electron dose should be applied. Here, we exploit simultaneous ADF-STEM and iDPC-STEM[2,3] imaging to investigate the LiCoPO4 cathode material down to atomic scale, as iDPC-STEM allows for extremely low dose imaging of beam sensitive materials[4,5]. By combining the experimental and simulation results a superstructure at the surface of the LiCoPO4 nanoparticle is revealed, where every second Li site is partially occupied by Co atoms (Figure 1). Figure 1. Simultaneously recorded, using HAADF and Panther detectors, high resolution ADF- and iDPC-STEM images (from left to right) of the superstructure at the surface of the LiCoPO4 nanoparticle, viewed along the [001] zone axis, together with the corresponding simulations and proposed surface structure. The probe corrected beam conditions: HT 300kV, Convergence semi-angle 30 mrad, Beam current 15 pA. [1] X. Wu, M. Meledina, J. Barthel, et. al., Energy storage materials 22 (2019) 138 [2] I. Lazić, E.G.T. Bosch and Sorin Lazar, Ultramicroscopy 160 (2016) 265 [3] I. Lazić and E.G.T. Bosch, Advances in Imaging and Electron Physics 199 (2017) 75 [4] B. Shen, H. Wang, H. Xiong, et al., Nature 607 (2022) 703 [5] I. Lazić, M. Wirix, M. L. Leidl et. al., Nature Methods 19, (2022) 1126

Single atom Pt on Pd@Au and Pd shell on AuPt nanowire: application in electrocatalysis

Abstract text (incl. figure legends and references) Off late, research in nanoscience emerged as front-runner due to its tremendous industrial applications including application in renewable energy resources. Due to depleting fossil fuel reserves and ever-growing demand for environment friendly renewable energy resources, researchers are trying to develop efficient catalyst for energy conversion. Catalysts based on nanomaterials is one of the focused areas of research. Specifically, one dimensional nanomaterial is widely explored in catalytic applications due to high surface area. Ultrathin Au nanowires have been explored for several applications, but the fragile nature at high temperature, and in polar medium limits its catalytic applications. Instability of these nanowires under the electron beam possess difficulties for its microstructural study. Alloying of Au nanowire with Pd and Pt increases the stability as well as improves electrocatalytic efficiency with faster kinetics due to bifunctional mechanism. In our current investigation, ultrathin single crystalline AuPdPt trimetallic nanowires with different composition have been synthesized using microwave assisted synthesis at liquid-liquid interface. By changing the order of second and third metal reduction on Au nano wire template, single atom Pt on Pd@Au nanowire and Pd shell on AuPt nanowire have been synthesized. Interestingly, these trimetallic nanowires retains single crystalline nature after successive reduction of second and third metal on Au nanowire. Using z-contrast high angle annular dark field imaging (HAADF-STEM), x-ray photoelectron spectroscopy (XPS) and cyclic voltammetry surface atomic distribution as well as chemical environment have been elucidated. Annealing under control atmosphere, e.g., Ar and CO, was performed to modify the surface of these nanowires. Surface atomic distribution has been further visualized and explained using z-contrast HAADF-STEM, and electrocatalysis. Single atom Pt on Pd@Au nanowire showed highest activity with moderate onset potential compared to bimetallic Pd@Au nanowire. After Ar-annealing, alloying improvement led to vanishing Pt single atom rom surface. Whereas CO-annealing caused PdPt segregation on surface leading to improved potential and activity. Ar-annealed Pd shell on AuPt nanowire showed better kinetics than as-synthesized one resulting lower onset potential. CO-annealed Pd shell on AuPt nanowire showed highest activity with higher potential. These nanowires were explored as catalyst for electrocatalytic methanol oxidation. Further, changes in activity with surface atomic distribution has been visualised and explained by electron microscopy.

Investigation of morphology and cathodoluminescence properties of non-polar aluminum nitride microwires

Abstract text (incl. figure legends and references) Introduction As one of the most successful methods for synthesizing bulk AlN crystal, PVT can also be used to synthesize a high-quality micro- and nanoscale 1D single crystal with unintentionally doped impurity due to its special growth environment. CL spectroscopy is a spectroscopic technique based on electron-excited luminescence. It uses a highly focused electron beam to generate electron-hole pairs in the sample, which are through radiative or non-radiative processes for diffusion and recombination. Micro-area characterizations of micro-scale crystals by CL spectroscopy are an advantageous means to understand the correlation between structural and optical properties of the sample. The study of defect properties is also important for understanding the formation or elimination of various defect centers of bulk aluminum nitride materials. Objective To understand the influence of the growth habit on defect-related luminescence properties in m-oriented AlN Materials and methods The large-scale AlN microwires were synthesized in a homemade RF-heated furnace. A high-purity AlN source material is obtained by sublimation of commercially available AlN powder (TOKUYAMA). The growth was performed at a background nitrogen pressure of 100 mbar and nitrogen gas was used as protective gas at the flow rate of 100 sccm at high temperature. The growth temperature was kept at 2000 °C for about 0.5 h. Results Fig. 1a and b show the SEM morphology of the top surface of the single UW and the corresponding panchromatic CL spectra. The near band edge emission at 200 nm cannot be detected due to the limitation of the detector used at an effective wavelength range from 250 to 600 nm. It is found that there is strong CL emission in the region near both sidewalls of the UW, and this symmetrically distributed CL emission pattern extends into the semipolar plane. Interestingly, the strong CL emission makes the Si substrate outside the UW also show an elevated background emission. As shown in Fig. 1c, the CL emission of the UW is around the wavelengths of 320 nm (VL-1), 345 nm (VL-2), and 450 nm (BL). The typical SEM morphology and panchromatic CL spectra of the N-polar c-plane of UW are shown in Fig. 1d and e. The localized CL luminescence appears as a symmetrical "sandwich structure". Fig. 1f shows the point mode CL spectra for the positions marked in Fig. 1e. Fig. 1. CL characterization of the AlN UW. SEM image of the top surface of the UW (a) and simultaneously recorded panchromatic CL spectra (b). SEM image of the bottom surface of the UW (d) and simultaneously recorded panchromatic CL spectra (e). CL spectra were obtained in point mode on different positions of the UW, as shown in (c) and (f). Conclusion In summary, the high-quality nonpolar AlN microwires with rectangular cross-sections were synthesized by the PVT. The difference in impurity incorporation efficiency on different facets is responsible for the peculiar CL patterns in the AlN UW. In addition, it is confirmed that the radiative recombination mechanism is determined by the O/C content ratio. It may contribute to a more thorough understanding of bulk AlN properties and growth. Reference [1] Chuang Wang, Xiaodong Gao, Jiafan Chen, Luhua Wang, Xionghui Zeng*, et al. Morphology evolution and cathodoluminescence properties of non-polar aluminum nitride microwires, Journal of alloys and compounds, 2023, 937, 168407

IM 7: In situ/operando electron microscopy

Prof

Direct observation of domain-wall-free ferroelectricity in Y-doped HfO2

Abstract text (incl. figure legends and references) Due to the recent discovery of ferroelectricity in doped Hafnium Oxide (HfO2) film, HfO2-based materials have significantly attracted lots of attention in the semiconductor industry because of their strong compatibility with Si-based Complementary Metal-Oxide-Semiconductor (CMOS) and thereby their potential for fully integrated ferroelectric devices. The equilibrium phase diagram of bulk HfO2 consists of non-ferroelectric polymorphs that are the monoclinic (P21/c, O phase), tetragonal (P42/nmc, T phase), and cubic (Fm-3m, C phase) phases. The ferroelectric HfO2 is the metastable, orthorhombic phase (Pca21), where a unit cell is divided into active cell/inactive cell, in which oxygen atoms can/cannot generate electric dipole. Based on the flat phonon band theory, the inactive cell is predicted to be an inherent spacer between neighboring active cells and therefore inactive cells enable electric dipoles to switch without forming a domain wall [1].The domain-wall-free ferroelectric behavior in HfO2 has been directly identified by TEM analyses [2,3], but is still challenging due to the polymorphism in doped HfO2 and pico-metric regime of structural modulation. In this work, we present an atomistic behavior of ferroelectricity in epitaxially grown Yttria-doped HfO2 (YHO). The crystallographic structure of YHO is quantitatively analyzed by scanning transmission electron microscopy combined with computational techniques via deep machine learning. Furthermore, we apply a 4D-STEM ptychography to visualize the electric field distribution in YHO. Combined experimental results show YHO can hold polarization within a width of to a single unit cell. Boundary cells, inactive cells between the up- and down-polarized cells, are investigated by our first-principle calculation, and our atomistic analysis confirmed that boundary cell possesses the same structure with the inactive cell in bulk YHO, indicative of the absence of the domain wall. This result gives the new insight into the realization of unit cell scale control of ferroelectric materials. [1] Lee et al., Science 369 (6509), 1343-1347 (2020) [2] Xu et al., Nature Materials 20, 826–832 (2021) [3] Kang et al., Science 376 (6594), 731-738 (2022)

Multiscale tomography of porous transport electrodes in proton exchange membrane water electrolyzers

Abstract text (incl. figure legends and references) The transition to a carbon-neutral energy sector is linked with the implementation of energy storage solutions. The need for energy storage is due to the fluctuation and intermittency of renewable energy sources. Hydrogen, as an energy carrier produced from electrochemical water splitting, is a promising candidate as a storage solution. Proton exchange membrane water electrolyzers (PEMWEs) are suited to produce hydrogen in a renewable energy-based energy infrastructure due to their flexibility, rapid system response, and scalability. The membrane electrode assembly of PEMWEs consists either of a catalyst-coated membrane (CCM) or a porous transport electrode (PTE) configuration. In the latter, the catalyst ink is directly coated onto the porous substrate, typically a titanium fiber or sintered titanium powder porous transport layer (PTL) on the anode side. To date, the CCM configuration is prevalent, but the PTE configuration is a promising alternative concept for PEMWEs.[1] Its different fabrication approach allows the usage of a wider variety of membranes and can reduce the number of processing steps in manufacturing compared to the fabrication of CCMs. However, the PTE design has higher kinetic overpotentials due to a worse connection between catalyst particles and the membrane.[2] The less efficient catalyst utilization originates from the deep infiltration of the catalyst ink into the PTL. We could prove this catalyst distribution and the associated worse catalyst utilization in our recently submitted paper.[3] Further improvements in PTEs are necessary to make them competitive against CCMs. One essential objective is to understand the whole structure of PTEs to elucidate the transport-activity nexus of these electrodes. Therefore, we used a combination of different tomographic methods to image the structure of the titanium fiber substrate, the IrO2 catalyst layer, and its catalyst distribution. X-ray tomography and focused ion beam scanning electron microscopy (FIB-SEM) were employed to resolve the whole electrode on the µm- and nm-scale for the first time. Based on the reconstructions, we calculated the structural and transport parameters of the PTE. The analysis revealed the distribution and thickness of the catalyst layer and the grain and pore size distributions on the different length scales. Our correlative microscopy approach revealed that future PTE designs should aim to use a PTL with a porous blocking layer or a micro-porous layer. [1] M. Bühler, F. Hegge, P. Holzapfel, M. Bierling, M. Suermann, S. Vierrath, S. Thiele, Journal of Materials Chemistry A 2019, 7, 26984. [2] D. Kulkarni, A. Huynh, P. Satjaritanun, M. O&quot;Brien, S. Shimpalee, D. Parkinson, P. Shevchenko, F. DeCarlo, N. Danilovic, K. E. Ayers, C. Capuano, I. V. Zenyuk, Applied Catalysis B: Environmental 2022, 308, 121213. [3] M. Bierling, D. McLaughlin, B. Mayerhöfer, S. Thiele, Towards understanding catalyst layer deposition processes and distribution in anodic porous transport electrodes in proton exchange membrane water electrolyzers (submitted).

Visualising the spatial distribution of individual molecular dopants of organic semiconductors on the sub-nanometre scale in three dimensions

Abstract text (incl. figure legends and references) Molecular dopants are used to enhance the electronic properties of organic electronics. The three dimensional distribution of the dopants and whether they are present as individual dopants or in clusters determines the electrical properties. In applications, the dopant concentration varies depending on application but it can be up to several molar precents.The dopants are about 1 nm in size and the distance between them is about a few to 10 nm. A detailed information about the spatial distribution is important for the understanding of the conduction mechanisms and how the molecular dopants affect the properties of the surrounding polymer. 1,2 In this study, electron tomography is used to visualize the three-dimensional spatial distribution of individual Mo(tfd-COCF3)3 molecular dopants in the semiconducting polymer p(g42T-T) at the sub-nanometer scale.The thickness of the films is in the range of 13-17 nm and dopant concentrations are 5, 20 and 40 mol%.The acquisition parameters are chosen to obtain sufficient resolution to determine the position of individual dopant molecules in the material. A heterogenous distribution is observed. At dopant concentrations of 5 mol%, the majority of the dopants are positioned individually or in clusters consisting of less than five molecules. At higher concentrations, the dopants form larger sized clusters with increasingly elongated shapes. Despite the larger cluster sizes at higher concentrations, all of the observed dopants remain adjacent to surrounding polymer. This is a crucial aspect for the efficient charge transfer between the dopants and the surrounding polymer. Finally, electrical characterization using a four-point probe setup shows a decrease in thin-film conductivity as larger dopant clusters are formed at high dopant concentrations. 3 1 C.J. Boyle et al., Nat. Commun., 2019, 10, 1-10 2 A. Fediai et al., Phys. Chem. Chem. Phys., 2020, 22, 10256-10264 3 G. Persson et al., Nanoscale, 2022, 14, 15404-15413

MS 5: Functional organic materials

Hierarchical self-assembly involving nonclassical and classical steps in organic crystal growth

Abstract text (incl. figure legends and references) Organic crystal nucleation and growth are complex processes that often do not fit into the scope of the existing crystallization theories. We investigated a crystal growth mechanism of an organic dye, perylene diimide, using high-resolution cryogenic transmission electron microscopy and optical spectroscopy. We were able to demonstrate that the evolution of order in our system had a high level of complexity and involved both classical and nonclassical steps. The crystallization mechanism included a series of supramolecular transformations, where each step defined the next one. The crystal growth started from a molecular self-assembly into π-stacks that eventually underwent intermolecular ordering that optimized interactions within the stacks. The latter transformed at the larger scale, by stack interactions, which formed crystalline domains. Finally, the formed faceted crystals gradually grew through oriented attachment of other crystal, and attachment of residual monomeric molecules from solution (Figure 1). Fig.1 An illusration of order evolution during the crystallization of perylene diimide. Our findings present a detailed insight into organic crystal growth and reveal that both classical and nonclassical mechanisms can operate within a crystallization process. However, classical/nonclassical dichotomy provides only a partial insight into the crystallization mechanism, which may be described as a sequence of supramolecular events, traversing a vast size scale from the optimization of intermolecular interactions to the oriented attachment of crystals. In summary, our work revealed the inherent nature of supramolecular transformations occurring in organic crystallization, thus advancing conceptually our understanding of order evolution in organic matter.1 1. Biran, I.; Rosenne, S.; Weissman, H.; Tsarfati, Y.; Houben, L.; Rybtchinski, B. Cryst. Growth Des. 2022, 22 (11), 6647–6655.

PS 7
Late breaking posters MS

Appointment

Programme

PS 7Late breaking posters MS

Appointment

Programme

PS 7
Late breaking posters MS